Peripheral Proinsulin Expression Controls Low-Avidity Proinsulin-Reactive CD8 T Cells in Type 1 Diabetes

Low-avidity autoreactive CD8 T cells (CTLs) escape from thymic negative selection, and peripheral tolerance mechanisms are essential for their regulation. We report the role of proinsulin (PI) expression on the development and activation of insulin-specific CTLs in the NOD mouse model of type 1 diabetes. We studied insulin B-chain-specific CTL from different T-cell receptor transgenic mice (G9Cα^-/-) expressing normal PI1 and PI2 or altered PI expression levels. In the absence of PI2 (Ins2^-/-), CTL in pancreatic lymph nodes (PLNs) were more activated, and male G9Cα^-/- mice developed T1D. Furthermore, when the insulin-specific CTLs developed in transgenic mice lacking their specific PI epitope, the CTLs demonstrated increased cytotoxicity and proliferation in vitro and in vivo in the PLNs after adoptive transfer into NOD recipients. Dendritic cell-stimulated proliferation of insulin-specific T cells was reduced in the presence of lymph node stromal cells (LNSCs) from NOD mice but not from mice lacking the PI epitope. Our study shows that LNSCs regulate CTL activation and suggests that exposure to PI in the periphery is very important in maintenance of tolerance of autoreactive T cells. This is relevant for human type 1 diabetes and has implications for the use of antigen-specific therapy in tolerance induction.

A reproducible method for the expansion of mouse CD8+ T lymphocytes

Murine adoptive CD8 + T-cell immunotherapy studies require the generation of large numbers of high viability CD8 + cells. Here we report a tissue culture protocol for the reliable expansion of CD8 + T-cells derived from murine spleen to give a 20-fold expansion after 4 days in culture. The cells were transfected with an mRNA GFP construct and transferred into NOD mice. GFP positive cells could be detected 7 days after transfer thus confirming that the cells survive and are functional for up to 1 week.

•

A technique for reliable expansion of murine CD8 + T cells

•

Excellent cell viability

•

Allows for electroporation of cells with mRNA constructs and detection of expression up to 7 days in vivo

Activation of insulin-reactive CD8 T-cells for development of autoimmune diabetes

OBJECTIVE

We have previously reported a highly diabetogenic CD8 T-cell clone, G9C8, in the nonobese diabetic (NOD) mouse, specific to low-avidity insulin peptide B15-23, and cells responsive to this antigen are among the earliest islet infiltrates. We aimed to study the selection, activation, and development of the diabetogenic capacity of these insulin-reactive T-cells.

RESEARCH DESIGN AND METHODS

We generated a T-cell receptor (TCR) transgenic mouse expressing the cloned TCR Valpha18/Vbeta6 receptor of the G9C8 insulin-reactive CD8 T-cell clone. The mice were crossed to TCRCalpha-/- mice so that the majority of the T-cells expressed the clonotypic TCR, and the phenotype and function of the cells was investigated.

RESULTS

There was good selection of CD8 T-cells with a predominance of CD8 single-positive thymocytes, in spite of thymic insulin expression. Peripheral lymph node T-cells had a naïve phenotype (CD44lo, CD62Lhi) and proliferated to insulin B15-23 peptide and to insulin. These cells produced interferon-gamma and tumor necrosis factor-alpha in response to insulin peptide and were cytotoxic to insulin peptide-coated targets. In vivo, the TCR transgenic mice developed insulitis but not spontaneous diabetes. However, the mice developed diabetes on immunization, and the activated transgenic T-cells were able to transfer diabetes to immunodeficient NOD.scid mice.

CONCLUSIONS

Autoimmune CD8 T-cells responding to a low-affinity insulin B-chain peptide escape from thymic negative selection and require activation in vivo to cause diabetes.

Activated Insulin-Reactive CD8 T cells in NOD mice Cause Diabetes. (129.46)

We have previously generated highly diabetogenic insulin-reactive CD8 T cell clones designated G9C8 from the NOD mouse. The aim of our study was to investigate T cell selection and activation of CD8 T cells from a TCR transgenic mouse expressing the cloned TCRαβ receptor of the G9C8 clone. The mice were crossed to TCR.Cα−/− mice so that all the T cells express the clonotypic TCR. There was a predominance of CD8+ single positive thymocytes and peripheral CD8 TCR transgenic cells, compared with CD4 T cells. Peripheral lymph node T cells express a naïve phenotype (CD44lo, CD62Lhi). They proliferate to the insulin B15-23 peptide, produce interferon-γ and are cytotoxic to targets coated with the insulin peptide. In vivo, the TCR transgenic mice developed insulitis but surprisingly spontaneous diabetes did not occur. However, the TCR transgenic mice can be induced to develop diabetes on immunization. They will also cause diabetes on adoptive transfer after activation. In summary, these mice provide a good model for studying selection and activation of insulin-reactive CD8 T cells in the NOD mouse.

Measurement of pre- and post-synaptic proteins in cerebral cortex: effects of post-mortem delay

Assessments of synaptic density in human brain are often based on measurements of synaptic proteins. Little information is available on their post-mortem stability. We have investigated this by ELISAs of the pre-synaptic proteins syntaxin and synaptophysin, and the post-synaptic protein PSD-95, in rat and human cortex. The rat brains were cooled in situ from 37 to 20 or 4 degrees C over 3 h, and then kept at 20 or 4 degrees C for a further 24-72 h, to simulate post-mortem storage at room temperature or in a mortuary refrigerator. Synaptophysin and PSD-95 levels in rat cerebral cortex were not significantly decreased after 72 h of incubation at 20 degrees C. Syntaxin was stable for 24 h but decreased by 39-44% at 48-72 h. Storage at 4 degrees C resulted in a similar reduction of syntaxin levels over 72 h. In human brain tissue from 160 people aged 24-102 years, post-mortem delay had little effect on synaptic protein levels in superior temporal cortex, but was associated with a decline in PSD-95 and syntaxin in mid-frontal cortex after 24 h. The more robust stability of synaptophysin may be related to its multi-transmembrane structure.

An immunodominant epitope from mycobacterial 65-kDa heat shock protein protects against pristane-induced arthritis

Previous studies showed that mice with pristane-induced arthritis (PIA) and those protected from the disease by preimmunization with mycobacterial 65-kDa heat shock protein (hsp65) possess raised immune responses to hsp65. Additionally, T cells from hsp65-protected mice, but not from pristane-injected or normal mice, produced the Th2-associated cytokines IL-4, IL-5, and IL-10 in response to stimulation with hsp65. Here we demonstrate that the specificity of the immune response to hsp65 and related heat shock protein (hsps) differs between protected and PIA mice. T cells from hsp65-protected mice respond to the bacterial hsps tested but not to the mammalian homologue, hsp58. Similarly, they exhibit high serum titers of anti-hsp65 Abs, yet they have virtually undetectable levels of anti-hsp58 IgG. By contrast, both cellular and humoral immune responses are detectable to bacterial and mammalian hsps in mice with PIA. An immunodominant T cell epitope has been identified in hsp65-protected mice corresponding to amino acids 261-271 from hsp65. Immunization of mice, either before or after the induction of arthritis, with this bacterial peptide, but not its mammalian homologue, protects mice from the development of PIA, and protection is associated with the production of Th2-type cytokines. Other experiments revealed that T cells primed with bacterial 261-271 or the mammalian homologue do not cross-react at the proliferative or cytokine level. These results demonstrate that an hsp65 peptide-specific Th2 response confers protection from PIA but do not support the idea that protection is mediated by a cross-reaction with self hsp58 in the joints.

Effect of T-helper cytokine environment on specificity of T-cell responses to mycobacterial 65,000 MW heat-shock protein

The purpose of this work was to determine if the fine specificity of T cells differed between mice immunized with an antigen in a T helper 1 (Th1) cytokine-dominated environment as compared with a T helper 2 (Th2) cytokine-dominated environment. It was found that splenic T cells from mice immunized with mycobacterial heat-shock protein (hsp 65) and interleukin-12 (IL-12) produced less interleukin-4 (IL-4) and more interferon-gamma (IFN-gamma) in response to stimulation with hsp 65 in vitro than did T cells from mice immunized with hsp 65 alone. The T-cell proliferative response to hsp 65 did not differ between the two groups of mice, although the responses were higher than those of T cells from non-immunized mice. Strikingly, T cells from mice given hsp 65 and IL-12 gave significantly higher responses to six peptides (corresponding to the sequence of hsp 65) to which T cells from mice immunized with hsp 65 alone did not respond. It is considered that different epitopes are presented to T cells (possibly owing to changes in antigen processing) if the environment is shifted, by IL-12, from Th2 towards Th1 cytokines.

CD4+ Th2 cells specific for mycobacterial 65-kilodalton heat shock protein protect against pristane-induced arthritis

Previous studies showed that mice with pristane-induced arthritis (PIA) and those protected from the disease by preimmunization with mycobacterial 65-kDa heat shock protein (hsp65), possess raised immune responses to hsp65. Thus, a paradox exists whereby T cells from both arthritic and hsp65-protected animals proliferate vigorously in response to the same Ag. Here we demonstrate that T cells from mice with PIA and hsp65-protected mice produce different cytokines in vitro in response to hsp65. The use of a sensitive CelELISA to measure Ag-driven lymphokine production revealed that spleen cells from hsp65-protected mice, but not those from pristane-injected or normal mice, produced the Th2-associated cytokines IL-4, IL-5, and IL-10 in response to stimulation with hsp65. By contrast, the Th1-associated cytokines IL-2 and IFN-gamma were produced by spleen cells from mice of all groups in response to hsp65. Furthermore, there was a dramatic increase in the IgG1 to IgG2a ratio of anti-hsp65 Abs from arthritic to protected mice. Thus, it appears that a Th2 response is protective against PIA. To examine this theory, a regimen of IL-12 administration which polarizes the hsp65-specific (Th2) immune response toward Th1 was identified. This regime abolished hsp65-mediated protection against PIA. Other experiments revealed that the specificity of the response to hsp65 was important, as other bacterial proteins known not to protect against PIA induced similar Th2-associated cytokines in vitro. It is considered that the protection afforded by hsp65 preimmunization is mediated by Th2-associated cytokines produced by hsp65-specific CD4+ T cells.

CD4+ Th2 cells specific for mycobacterial 65-kilodalton heat shock protein protect against pristane-induced arthritis

Previous studies showed that mice with pristane-induced arthritis (PIA) and those protected from the disease by preimmunization with mycobacterial 65-kDa heat shock protein (hsp65), possess raised immune responses to hsp65. Thus, a paradox exists whereby T cells from both arthritic and hsp65-protected animals proliferate vigorously in response to the same Ag. Here we demonstrate that T cells from mice with PIA and hsp65-protected mice produce different cytokines in vitro in response to hsp65. The use of a sensitive CelELISA to measure Ag-driven lymphokine production revealed that spleen cells from hsp65-protected mice, but not those from pristane-injected or normal mice, produced the Th2-associated cytokines IL-4, IL-5, and IL-10 in response to stimulation with hsp65. By contrast, the Th1-associated cytokines IL-2 and IFN-gamma were produced by spleen cells from mice of all groups in response to hsp65. Furthermore, there was a dramatic increase in the IgG1 to IgG2a ratio of anti-hsp65 Abs from arthritic to protected mice. Thus, it appears that a Th2 response is protective against PIA. To examine this theory, a regimen of IL-12 administration which polarizes the hsp65-specific (Th2) immune response toward Th1 was identified. This regime abolished hsp65-mediated protection against PIA. Other experiments revealed that the specificity of the response to hsp65 was important, as other bacterial proteins known not to protect against PIA induced similar Th2-associated cytokines in vitro. It is considered that the protection afforded by hsp65 preimmunization is mediated by Th2-associated cytokines produced by hsp65-specific CD4+ T cells.

Major histocompatibility complex regulation of the class of the immune response: the H-2d haplotype determines poor interferon-gamma response to several antigens

The lymph node cells of CBA (H-2k), but not BALB/c (H-2d) mice, release interferon (IFN)-gamma into the supernatant when immunized with picryl chloride epicutaneously and then exposed to antigen (haptenized cells) in vitro 4 days later. The failure in IFN-gamma production maps to the major histocompatibility complex (MHC; H-2d) in the congenic BALB/c, BALB/k and BALB/b mice. The evidence that this is an MHC regulation of the class of response to a range of antigens and not a classical Ir gene effect is (a) the difference is seen with several antigens including picryl chloride, "oxazolone" and purified protein derivative of tuberculin and (b) BALB/c mice, which fail to produce IFN-gamma, show excellent contact sensitivity to picryl chloride. It was also found that the crosses between responder and nonresponder strains (CBA x BALB/c)F1 respond to antigen on responder cell but not on nonresponder cells. This influence of MHC on the class of the immune response is a possible basis for some of the associations of MHC with disease.