Blake Middleton

Islet transplantation offers a potential therapy to restore glucose homeostasis in type 1 diabetes patients. A method to image transplanted islets noninvasively and repeatedly would greatly assist studies of islet transplantation. Using recombinant adenovirus, we show that isolated rodent and human islets can be genetically engineered to express luciferase and then imaged after implantation into NOD-scid mice using a cooled charge-coupled device. The magnitude of the signal was dependent on the islet dose. Adenovirus-directed luciferase expression, however, rapidly attenuated. We next tested lentivirus vectors that should direct the long-term expression of reporter genes in transduced islets. Transplanted lentivirus-transduced islets restored euglycemia long term in streptozotocin-treated NOD-scid mice. The signal from implanted lentivirus-transduced islets was related directly to the implanted islet mass, and the signal did not attenuate over the observation period. Viral transduction, luciferase expression, and repeated imaging had no apparent long-term deleterious effects on islet function after implantation. These data demonstrate that the introduction of reporter genes into an isolated tissue allows the long-term monitoring of its survival following implantation. Such imaging technologies may allow earlier detection of graft rejection and the adjustment of therapies to prolong graft survival posttransplantation.

There is a growing interest in using vaccination with CNS antigens to induce autoreactive T cell responses that home to damaged areas in the CNS and ameliorate neurodegenerative disease. Neuroprotective vaccine studies have focused on administering oligodendrocyte antigens or Copaxone® in complete Freund's adjuvant (CFA). Theoretical considerations, however, suggest that vaccination with a neuronal antigen may induce more robust neuroprotective immune responses. We assessed the neuroprotective potential of vaccines containing tyrosine hydroxylase (a neuronal protein involved in dopamine synthesis) or Copaxone® in CFA in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mouse model of Parkinson's disease. Surprisingly, we observed that the main beneficial factor in these vaccines was the CFA. Since the major immunogenic component in CFA is Mycobacterium tuberculosis, which closely related to the bacille Calmette-Guérin (BCG) that is used in human vaccines, we tested BCG vaccination in the MPTP mouse model. We observed that BCG vaccination partially preserved markers of striatal dopamine system integrity and prevented an increase in activated microglia in the substantia nigra of MPTP-treated mice. These results support a new neuroprotective vaccine paradigm in which general (nonself-reactive) immune stimulation in the periphery can limit potentially deleterious microglial responses to a neuronal insult and exert a neurorestorative effect in the CNS. Accordingly, BCG vaccination may provide a new strategy to augment current treatments for a wide range of neuropathological conditions.

We previously showed that, in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mouse model of Parkinson's disease (PD), vaccination with bacillus Calmette-Guerin (BCG) prior to MPTP exposure limited the loss of striatal dopamine (DA) and dopamine transporter (DAT) and prevented the activation of nigral microglia. Here, we conducted BCG dose studies and investigated the mechanisms underlying BCG vaccination's neuroprotective effects in this model. We found that a dose of 1 × 10(6) cfu BCG led to higher levels of striatal DA and DAT ligand binding (28% and 42%, respectively) in BCG-vaccinated vs. unvaccinated MPTP-treated mice, but without a significant increase in substantia nigra tyrosine hydroxylase-staining neurons. Previous studies showed that BCG can induce regulatory T cells (Tregs) and that Tregs are neuroprotective in models of neurodegenerative diseases. However, MPTP is lymphotoxic, so it was unclear whether Tregs were maintained after MPTP treatment and whether a relationship existed between Tregs and the preservation of striatal DA system integrity. We found that, 21 days post-MPTP treatment, Treg levels in mice that had received BCG prior to MPTP were threefold greater than those in MPTP-only-treated mice and elevated above those in saline-only-treated mice, suggesting that the persistent BCG infection continually promoted Treg responses. Notably, the magnitude of the Treg response correlated positively with both striatal DA levels and DAT ligand binding. Therefore, BCG vaccine-mediated neuroprotection is associated with Treg levels in this mouse model. Our results suggest that BCG-induced Tregs could provide a new adjunctive therapeutic approach to ameliorating pathology associated with PD and other neurodegenerative diseases.

In the non-obese diabetic (NOD) mouse, a Th1-biased autoimmune response arises spontaneously against glutamic acid decarboxylase, concurrent with the onset of insulitis. Subsequently, Th1-type autoreactivity spreads intra- and intermolecularly to other beta-cell autoantigens (beta CAAs), suggesting that a spontaneous Th1 cascade underlies disease progression. Induction of Th2 immunity to a single beta CAA results in the spreading of Th2-type T-cell and humoral responses to other beta CAAs in an infectious manner. Thus, both Th1 and Th2 autoimmunity can evolve in amplificatory cascades defined by site-specific, but not antigen-specific, positive feedback circuits. Despite the continued presence of Th1 autoimmunity, the induction of Th2 spreading is associated with active tolerance to beta CAAs and reduced disease incidence. With disease progression there is an attenuation of beta CAA-inducible Th2 spreading, presumably because of a reduced availability of uncommitted beta CAA-reactive precursor T cells. We discuss the implications of these findings for the rational design of antigen-based immunotherapeutics.