Functionality of major histocompatibility complex class II molecules in mice doubly deficient for invariant chain and H-2M complexes

By combining two previously generated null mutations, Ii degrees and M degrees , we produced mice lacking the invariant chain and H-2M complexes, both required for normal cell-surface expression of major histocompatibility complex class II molecules loaded with the usual diverse array of peptides. As expected, the maturation and transport of class II molecules, their expression at the cell surface, and their capacity to present antigens were quite similar for cells from Ii degrees M degrees double-mutant mice and from animals carrying just the Ii degrees mutation. More surprising were certain features of the CD4(+) T cell repertoire selected in Ii degrees M degrees mice: many fewer cells were selected than in Ii+M degrees animals, and these had been purged of self-reactive specificities, unlike their counterparts in Ii+M degrees animals. These findings suggest (i) that the peptides carried by class II molecules on stromal cells lacking H-2M complexes may almost all derive from invariant chain and (ii) that H-2M complexes edit the peptide array displayed on thymic stromal cells in the absence of invariant chain, showing that it can edit, in vivo, peptides other than CLIP.

Selection of a broad repertoire of CD4+ T cells in H-2Ma0/0 mice

According to past reports, H-2Ma0/0 mice express a single major histocompatiblity complex class II molecule, A(b), heavily loaded with a single peptide derived from the invariant chain, CLIP. Despite the highly restricted diversity of the class II:peptide complexes expressed on thymic stromal cells in the mutant animals, a large and diverse population of CD4+ T cells is positively selected. However, two important issues remained unresolved and are addressed here: Just how preponderant is CLIP occupancy of the class II molecules from H-2M0/0 mice? How extensive and functionally competent is the CD4+ population selected in the mutant animals? Our results argue that a single class II:peptide complex can select a very broad, though not complete, repertoire of CD4+ T cells.

Targeted complementation of MHC class II deficiency by intrathymic delivery of recombinant adenoviruses

De novo differentiation of CD4+ T cells was provoked in mice lacking major histocompatibility complex (MHC) class II molecules by intrathymic injection of adenovirus vectors carrying class II genes. This permits a new approach to questions concerning the dynamics of CD4+ T cell compartments in the thymus and peripheral lymphoid organs. Here two issues are explored. First, we show that mature CD4+ CD8- cells reside in the thymus for a protracted period before emigrating to the periphery, highlighting the potential importance of, and our ignorance of, the postselection maturation period. Second, we demonstrate that the survival of CD4+ cells in peripheral lymphoid organs is markedly curtailed when class II molecules are absent and is not further reduced in the absence of both class II and class I molecules, raising the possibility that MHC-mediated selection may continue in the periphery.

Functional in vivo MHC class II loading by endogenously synthesized glycoprotein during viral infection

MHC class II presentation of antigenic peptides derived from soluble proteins is usually preceded by antigenic uptake via (nonreceptor-mediated) endocytosis by professional APCs, followed by processing in endosomal compartments. Although in vitro alternative pathways for MHC class II loading have been described for certain intracellularly synthesized proteins, the importance of these pathways has not been assessed in vivo. We have shown previously that endogenously produced membrane-associated glycoprotein (GP) of lymphocytic choriomeningitis virus (LCMV), a noncytopathic virus, can be presented in vitro on MHC class II molecules in the absence of the invariant chain (Ii), whereas the cytosolic LCMV nucleoprotein (LCMV-NP) failed to be presented under the same conditions. Taking advantage of this system, we analyzed presentation of LCMV-GP and LCMV-NP in vivo in Ii-deficient mice and followed the induced Th cell and B cell responses. At early time points after LCMV infection of li-deficient mice, we found in vivo MHC class II loading exclusively by the endogenously synthesized LCMV-GP, whereas no MHC class II loading by LCMV-NP could be detected. As a direct consequence, LCMV-specific Th cells exhibited initially only LCMV-GP specificity. In contrast, both LCMV-GP- and LCMV-NP-derived epitopes were presented in comparable amounts on APCs upon LCMV infection of normal mice, and LCMV-GP- as well as LCMV-NP-specific Th cells were comparably induced in vivo. Thus, cell internal MHC class II loading pathways are functional in vivo and may become dominant if the usual Ag presentation pathways are hampered.

Functional in vivo MHC class II loading by endogenously synthesized glycoprotein during viral infection

MHC class II presentation of antigenic peptides derived from soluble proteins is usually preceded by antigenic uptake via (nonreceptor-mediated) endocytosis by professional APCs, followed by processing in endosomal compartments. Although in vitro alternative pathways for MHC class II loading have been described for certain intracellularly synthesized proteins, the importance of these pathways has not been assessed in vivo. We have shown previously that endogenously produced membrane-associated glycoprotein (GP) of lymphocytic choriomeningitis virus (LCMV), a noncytopathic virus, can be presented in vitro on MHC class II molecules in the absence of the invariant chain (Ii), whereas the cytosolic LCMV nucleoprotein (LCMV-NP) failed to be presented under the same conditions. Taking advantage of this system, we analyzed presentation of LCMV-GP and LCMV-NP in vivo in Ii-deficient mice and followed the induced Th cell and B cell responses. At early time points after LCMV infection of li-deficient mice, we found in vivo MHC class II loading exclusively by the endogenously synthesized LCMV-GP, whereas no MHC class II loading by LCMV-NP could be detected. As a direct consequence, LCMV-specific Th cells exhibited initially only LCMV-GP specificity. In contrast, both LCMV-GP- and LCMV-NP-derived epitopes were presented in comparable amounts on APCs upon LCMV infection of normal mice, and LCMV-GP- as well as LCMV-NP-specific Th cells were comparably induced in vivo. Thus, cell internal MHC class II loading pathways are functional in vivo and may become dominant if the usual Ag presentation pathways are hampered.

Positive selection of T cells: fastidious or promiscuous?

Several studies reported during the past year, most of which exploited novel in vivo positive selection systems, have addressed the basis of peptide involvement in positive selection of T cells. The very flexible, yet specific, requirements the studies demonstrate differ somewhat from the very specific requirements reported in earlier experiments relying on in vitro selection systems.

Positive selection of T cells induced by viral delivery of neopeptides to the thymus

The relation between an antigenic peptide that can stimulate a mature T cell and the natural peptide that promoted selection of this cell in the thymus is still unknown. An experimental system was devised to address this issue in vivo-mice expressing neopeptides in thymic stromal cells after adenovirus-mediated delivery of invariant chain-peptide fusion proteins. In this system, selection of T cells capable of responding to a given antigenic peptide could be promoted by the peptide itself, by closely related analogs lacking agonist and antagonist activity, or by ostensibly unrelated peptides. However, the precise repertoire of T cells selected was dictated by the particular neopeptide expressed.

Organ-specific disease provoked by systemic autoimmunity

Rheumatoid arthritis (RA) is a chronic joint disease characterized by leukocyte invasion and synoviocyte activation followed by cartilage and bone destruction. Its etiology and pathogenesis are poorly understood. We describe a spontaneous mouse model of this syndrome, generated fortuitously by crossing a T cell receptor (TCR) transgenic line with the NOD strain. All offspring develop a joint disease highly reminiscent of RA in man. The trigger for the murine disorder is chance recognition of a NOD-derived major histocompatibility complex (MHC) class II molecule by the transgenic TCR; progression to arthritis involves CD4+ T, B, and probably myeloid cells. Thus, a joint-specific disease need not arise from response to a joint-specific antigen but can be precipitated by a breakdown in general mechanisms of self-tolerance resulting in systemic self-reactivity. We suggest that human RA develops by an analogous mechanism.

A powerful nonviral vector for in vivo gene transfer into the adult mammalian brain: polyethylenimine

Nonviral gene transfer into the central nervous system (CNS) offers the prospect of providing safe therapies for neurological disorders and manipulating gene expression for studying neuronal function. However, results reported so far have been disappointing. We show that the cationic polymer polyethylenimine (PEI) provides unprecedentedly high levels of transgene expression in the mature mouse brain. Three different preparations of PEI (25-, 50-, and 800-kD) were compared for their transfection efficiencies in the brains of adult mice. The highest levels of transfection were obtained with the 25-kD polymer. With this preparation, DNA/PEI complexes bearing mean ionic charge ratios closest to neutrality gave the best results. Under such conditions, and using a cytomegalovirus (CMV)-luciferase construction, we obtained up to 0.4 10(6) RLU/microgram DNA (equivalent to 0.4 ng of luciferase), which is close to the values obtained using PEI to transfect neuronal cultures and the more easily transfected newborn mouse brain (10(6) RLU/microgram DNA). Widespread expression (over 6 mm3) of marker (luciferase) or functional genes (bcl2) was obtained in neurons and glia after injection into the cerebral cortex, hippocampus, and hypothalamus. Transgene expression was found more than 3 months post-injection in cortical neurons. No morbidity was observed with any of the preparations used. Thus, PEI, a low-toxicity vector, appears to have potential for fundamental research and genetic therapy of the brain.

The role of CD8+ T cells in the initiation of insulin-dependent diabetes mellitus

While it is generally accepted that T cells are critical for the development of diabetes in the non-obese diabetic (NOD) mouse, the precise functions of the CD4+ and CD8+ subsets remain ill-defined. Transfer experiments have provided evidence that CD4+ cells are the disease initiators, provoking massive mononuclear leukocyte infiltration into the pancreatic islets, while CD8+ cells play an effector role, responsible for the final destruction of islet beta cells. It was surprising, then, to find that NOD mice carrying a null mutation at the beta 2-microglobulin (beta 2-mu) locus, and thereby lacking major histocompatibility complex class I molecules and CD8+ T cells, developed neither insulitis nor diabetes. Here, we argue that the absence of insulitis in these animals results from their lack of CD8+ cells because islet infiltration is also absent when NOD mice are treated with an anti-CD8 monoclonal antibody (mAb) at a young age. Interestingly, the anti-CD8 effect is only observed when the mAb is injected during a discrete age window--2 to 5 weeks after birth. Transfer experiments indicate that the lack of CD8+ cells during this period somehow alters the phenotype of CD4+ cells, preventing them from expressing their insulitis potential. This is not because they are generally immuno-incompetent nor because they are generally more prone to differentiating into cells with Th2 characteristics. Given that neither the beta 2-mu mutation nor anti-CD8 treatment affect insulitis in a T cell receptor transgenic (tg) mouse strain with a CD4+ T cell repertoire highly skewed for an anti-islet cell reactivity, the most straight-forward interpretation of these observations is that CD8+ cells are required for effective priming and expansion of autoreactive CD4+ cells.

The immunosuppressant 15-deoxyspergualin [correction of 1,5-deoxyspergualin] reveals commonality between preT and preB cell differentiation

15 [correction of 1,5] deoxyspergualin (DSG) is a potent immunosuppressant whose mechanism of action is still somewhat of a mystery. We have studied the generation of lymphocytes in mice treated with this drug. The differentiation of T cells in the thymus was blocked at an important early control point: the CD4-8- --> CD4+8+ transition, known to depend on the expression of a preTCR complex that includes the variable TCR-beta, but not TCR-alpha, chain. In clear contrast, a later control point, the CD4+8+ --> CD4+8- or CD4-8+ transition, dependent on the display of a conventional alpha:beta TCR complex, appeared unaffected, as did activation of mature T cells both in vitro and in vivo. Interestingly, preB cell differentiation in the bone marrow was blocked at a precisely equivalent point: the A-C --> C' transition, controlled by expression of a pre-receptor complex containing the Ig heavy, but not light, chain. Mature B cells seemed unperturbed. These findings have theoretical implications, suggesting common signaling pathways in early lymphocytes that are distinct from those employed by more mature cells, and are also of practical interest, to be considered in the design of DSG treatment protocols.

Gene transfer by naked DNA into adult mouse brain

Nonviral gene transfer into the central nervous system could provide a basis for therapeutic uses and fundamental research. We show that naked DNA injected intracerebrally into the mouse brain can provide expression of a reporter protein. Expression is dose dependent, being maximal for 150 mu g DNA injected. We observed less than 5 days expression of the luciferase transgene, which is not improved with plasmid preparations virtually free of lipopolysaccharide. Thus, the adult brain behaves as striated muscle for naked DNA uptake and transcription, albeit at a much lower efficiency. In neither adult brain nor muscle did complexation of DNA with cationic lipid improve transgene expression. Double immunolabeling using cell-specific markers shows that both neurons and glia are transfected by naked DNA gene transfer methodology.

LAG-3 is not responsible for selecting T helper cells in CD4-deficient mice

The product of the LAG-3 gene is a cell surface protein with significant homology to CD4. It has been suggested that it can serve as a functional equivalent of CD4 and account for the MHC class II-restricted responses which persist in CD4-deficient mice. To test this hypothesis, we have created CD4/LAG-3 double-deficient mice by successive homologous recombinations in embryonic stem cells. These animals turn out to be indistinguishable from CD4 single-deficient mice in their lymphocyte populations and responses that are controlled by MHC class II molecules. LAG-3 thus does not explain class II-restricted lymphocyte selection and function in the absence of CD4, strengthening the idea that these phenomena can occur independently of co-receptor signalling.

Independent modes of natural killing distinguished in mice lacking Lag3

The LAG3 protein has several features in common with CD4, suggesting that it may be important in controlling T cell reactivity. However, mice with a Lag3 null mutation have now been shown to exhibit a defect in the natural killer cell, rather than the T cell, compartment. Killing of certain tumor targets by natural killer cells from these mice was inhibited or even abolished, whereas lysis of cells displaying major histocompatibility complex class I disparities remained intact. It appears that LAG3 is a receptor or coreceptor that defines different modes of natural killing.

Checkpoints in the progression of autoimmune disease: lessons from diabetes models

In the last few years, data from experiments employing transgenic models of autoimmune disease have strengthened a particular concept of autoimmunity: disease results not so much from cracks in tolerance induction systems, leading to the generation of anti-self repertoire, as from the breakdown of secondary systems that keep these cells in check. T cells with anti-self specificities are readily found in disease-free individuals but ignore target tissues. This is also the case in some transgenic models, in spite of overwhelming numbers of autoreactive cells. In other instances, local infiltration and inflammation result, but they are well tolerated for long periods of time and do not terminally destroy target tissue. We review the possible molecular and cellular mechanisms that underlie these situations, with a particular emphasis on the destruction of pancreatic beta cells in transgenic models of insulin-dependent disease.

Mice lacking H2-M complexes, enigmatic elements of the MHC class II peptide-loading pathway

We have generated mice lacking H2-M complexes, critical facilitators of peptide loading onto major histo-compatibility complex class II molecules. Ab molecules in these mice matured into stable complexes and were efficiently expressed at the cell surface. Most carried a single peptide derived from the class II-associated invariant chain; the diverse array of peptides normally displayed by class II molecules was absent. Cells from mutant mice presented both whole proteins and short peptides very poorly. Surprisingly, positive selection of CD4+ T cells was quite efficient, yielding a large and broad repertoire. Peripheral T cells reacted strongly to splenocytes from syngeneic wild-type mice, no doubt reflecting the unique peptide complement carried by class II molecules in mutant animals.

Biosynthesis of major histocompatibility complex molecules and generation of T cells in Ii TAP1 double-mutant mice

Major histocompatibility complex (MHC) class I and II molecules are loaded with peptides in distinct subcellular compartments. The transporter associated with antigen processing (TAP) is responsible for delivering peptides derived from cytosolic proteins to the endoplasmic reticulum, where they bind to class I molecules, while the invariant chain (Ii) directs class II molecules to endosomal compartments, where they bind peptides originating mostly from exogenous sources. Mice carrying null mutations of the TAP1 or Ii genes (TAP10) or Ii0, respectively) have been useful tools for elucidating the two MHC/peptide loading pathways. To evaluate to what extent these pathways functionally intersect, we have studied the biosynthesis of MHC molecules and the generation of T cells in Ii0TAP10 double-mutant mice. We find that the assembly and expression of class II molecules in Ii0 and Ii0TAP10 animals are indistinguishable and that formation and display of class I molecules is the same in TAP10 and Ii0TAP10 animals. Thymic selection in the double mutants is as expected, with reduced numbers of both CD4+ CD8- and CD4- CD8+ thymocyte compartments. Surprisingly, lymph node T-cell populations look almost normal; we propose that population expansion of peripheral T cells normalizes the numbers of CD4+ and CD8+ cells in Ii0TAP10 mice.

Antiviral immune responses of mice lacking MHC class II or its associated invariant chain

Induction of T-helper cells and T-B cell interaction have been considered to critically depend upon recognition of major histocompatibility complex (MHC) class II molecules by the T cell receptor. Mice lacking either MHC class II molecules (class II(0/0) mice) or its associated invariant chain (Ii0/0 mice) provide new opportunities to test this premise. Immune responses to some protein antigens have been studied in these mice; little is known about their ability to withstand viral infections. We therefore tested CD8+ effector T cells and CD4+ T-cell-dependent B cell function during different viral infections. The vesicular stomatitis virus (VSV)-specific primary cytotoxic T cell response which is largely T-helper-dependent was diminished in Ii(0/0) and absent in class II(0/0) mice. The usually less T-helper-dependent cytotoxic vaccinia or lymphocytic choriomeningitis virus (LCMV)-specific CD8+ T cell responses were reduced up to ninefold in class II(0/0) and up to threefold in Ii(0/0) mice. In class II(0/0) mice, the T-helper-independent neutralizing IgM response against the glycoprotein of VSV was within normal ranges but, in contrast to previous results on CD4(0/0) mice, the T-helper-dependent IgG response was absent. Ii(0/0) mice exhibited a normal neutralizing IgM response; in contrast to class II(0/0) mice, they mounted a significant, though reduced specific IgG response. Similar results were obtained for antibody responses against the nucleoprotein of VSV. Although the T-helper-cell response upon infection with VSV seemed diminished only a little in Ii(0/0) mice, presentation of VSV-G to a class II-restricted specific hybridoma was greater than 300-fold reduced in the absence of Ii. This suggests that local protein concentrations reached during viral infection in the host are high enough to override the Ii deficiency of antigen-presenting cells in vivo.