Neuroprotective intervention by interferon-γ blockade prevents CD8+ T cell–mediated dendrite and synapse loss

Neurons are postmitotic and thus irreplaceable cells of the central nervous system (CNS). Accordingly, CNS inflammation with resulting neuronal damage can have devastating consequences. We investigated molecular mediators and structural consequences of CD8⁺ T lymphocyte (CTL) attack on neurons in vivo. In a viral encephalitis model in mice, disease depended on CTL-derived interferon-γ (IFN-γ) and neuronal IFN-γ signaling. Downstream STAT1 phosphorylation and nuclear translocation in neurons were associated with dendrite and synapse loss (deafferentation). Analogous molecular and structural alterations were also found in human Rasmussen encephalitis, a CTL-mediated human autoimmune disorder of the CNS. Importantly, therapeutic intervention by IFN-γ blocking antibody prevented neuronal deafferentation and clinical disease without reducing CTL responses or CNS infiltration. These findings identify neuronal IFN-γ signaling as a novel target for neuroprotective interventions in CTL-mediated CNS disease.In many inflammatory diseases of the central nervous system (CNS) neuronal damage determines permanent neurological deficits. Novel strategies for neuroprotection are urgently sought and require a detailed understanding of the mechanisms underlying neuronal damage. The key contribution of cytotoxic CD8⁺ T lymphocytes (CTLs) to this process has become increasingly appreciated in recent years (Neumann et al., 2002). CTLs are commonly recruited to the brain in viral infections, paraneoplastic disorders (Albert and Darnell, 2004), and autoimmune diseases such as multiple sclerosis (MS) and Rasmussen’s encephalitis (RE; Hauser et al., 1986; Bien et al., 2005; Friese and Fugger, 2005; Goverman, 2009). Upon engagement of their cognate peptide–MHC class I (MHC-I) complex on target cells, CTLs activate an array of effector functions. Cytotoxicity is typically mediated by perforin-dependent mechanisms and Fas–FasL (CD95/CD95L) interactions (Stinchcombe and Griffiths, 2007), but CTLs also secrete cytokines including IFN-γ and TNF. The relative contribution of each of these pathways to tissue damage varies greatly and depends on the target cell type and tissue (Guidotti et al., 1996; Kägi et al., 1996; Medana et al., 2000). At the same time, CTL cytotoxicity and cytokine secretion can also both contribute to virus control in the CNS (Binder and Griffin, 2001; Shrestha and Diamond, 2007; Pinschewer et al., 2010b).Neurons show limited turnover and regenerative capacity yet serve essential functions. Classical concepts have therefore suggested that neurons are spared from CTL attack. This “immune privilege” has been accredited to limited MHC-I expression (Joly et al., 1991), secretion of immunomodulatory TGF-β (Liu et al., 2006), and expression of Fas-L (Medana et al., 2001a). The same mechanisms are also thought to represent an evolutionary reason why neurons serve as a sanctuary for several RNA and DNA viruses, namely members of the Herpes, Paramyxo-, and Arenavirus families (Brown et al., 1979; Sequiera et al., 1979; ter Meulen et al., 1984; Joly et al., 1991). Recent studies have shown, however, that infected neurons do not escape CTL recognition altogether (McDole et al., 2010). For example, CNS-infiltrating CTLs established stable peptide/MHC-I–specific contacts with Herpes simplex virus– or Borna disease virus–infected neurons, respectively (Khanna et al., 2003; Chevalier et al., 2011). In the Theiler’s murine encephalomyelitis virus model, depletion of CTLs or genetic deficiency in MHC-I or perforin preserved axon integrity and prevented neurological deficits (Murray et al., 1998; Deb et al., 2009, 2010). CTLs can attack neuronal somata (Manning et al., 1987) and axons (Medana et al., 2001b) in primary neuronal cultures and explants, and cultured neurons are sensitive to lysis or silencing by perforin (Rensing-Ehl et al., 1996; Meuth et al., 2009). The Fas/Fas-L pathway can cause cytoskeleton breaks and membrane disruption which eventually cause neuronal death (Medana et al., 2000). However, the morphology, electrical activity, and glial cell environment of neurons differ considerably between in vitro culture conditions and the natural tissue habitat, and all of these factors can affect susceptibility to CTL attack (Neumann et al., 1995). Hence, we still lack a clear understanding of how CTLs damage neurons in vivo and which alterations result from such damage.We have recently established the viral déjà vu model, allowing us to study CTL-mediated neuronal damage and the resulting disease in vivo (Merkler et al., 2006). Neonatal intracranial (i.c.) infection of mice with an attenuated lymphocytic choriomeningitis virus (LCMV) variant (rLCMV/INDG) results in viral persistence selectively in CNS neurons (a status referred to as carrier mice). rLCMV/INDG is not cytolytic and carrier mice are therefore clinically healthy, but they express viral nonself-antigens in neurons. Notably, they are free of CNS-infiltrating T cells, and viral epitope-specific CD8⁺ T cell frequencies in peripheral blood remain below detection limits of peptide/MHC-I tetramer measurements (Merkler et al., 2006). This indicates that neonatal rLCMV/INDG infection fails to trigger a clinically significant CTL response. Upon adult infection with LCMV WT (LCMVwt; referred to as challenge), carrier mice mount vigorous CTL responses against the immunodominant H-2D^b–restricted nucleoprotein-derived epitope NP396 that is shared between rLCMV/INDG and LCMVwt. These CTLs infiltrate the CNS gray matter, attack NP396-expressing rLCMV/INDG-infected neurons, and cause severe disease within 7–10 d after challenge. The topographical distribution and the composition of inflammatory infiltrates in viral déjà vu disease recreate histopathological hallmarks of RE. In this human autoimmune disease, oligoclonal and therefore putatively antigen-specific CD8⁺ T cell populations dominate the histological picture and are typically found in direct contact with neurons (Li et al., 1997; Bien et al., 2005). The clinical presentation of RE is characterized by treatment-refractory epilepsy with consequent intellectual decline and hemiparesis, thus necessitating surgical resection of affected brain regions. A viral contribution to RE pathogenesis has long been suspected but remains to be substantiated (Friedman et al., 1977; Walter and Renella, 1989; Farrell et al., 1991).Here, we show that CTL attack on neurons in vivo induces rapid loss of dendrites and synapses rather than neuronal depletion. This deafferentation and the resulting disease depend on IFN-γ signaling from CTLs to neurons but neither on Fas- nor on perforin-dependent pathways. Accordingly, STAT1 phosphorylation and nuclear translocation together with deafferentation represented hallmarks of both the murine viral déjà vu model and human RE. These findings provide important new insights into the molecular mechanisms and cellular consequences of CTL–neuron interactions in viral and autoimmune CNS disorders. This delineates a promising new strategy for neuroprotective intervention in immune-mediated CNS disease.     

Analysis of RhCE variants among 806 individuals in France: considerations for transfusion safety, with emphasis on patients with sickle cell disease

BACKGROUND: DNA testing has enabled the documenting of numerous variants of RHCE alleles, especially in individuals of African origin. The risk for production of clinically significant alloantibodies to Rh antigens of patients carrying variant RHCE alleles has led us to analyze the different RhCE variants investigated by molecular biology. Alloimmunization was analyzed regarding the RHCE genetic profile. STUDY DESIGN AND METHODS: Samples from 806 individuals with altered expression of RhCE antigens and/or producing anti‐RhCE in the presence of the corresponding antigen were analyzed. RESULTS: A total of 572 individuals were shown to express RhCE variants. Variant RHCE*ce alleles and RH haplotypes were identified in 83% of cases, the most frequent ones being the R^N haplotype, the ceMO allele, the (C)ce^s haplotype/ce^s1006 allele, and the ceAR allele identified in 36, 23, 20, and 17% of the tested samples, respectively. The absence of a high‐prevalence Rh antigen was documented in 93 individuals. Partial C and partial e were expressed by 53% of individuals with RhCE variants. Rh antibodies were identified in 127 (20%) of 623 patients. They were found to be alloantibodies in 48 (38%) of these 127 patients. Alloimmunization against a high‐prevalence Rh antigen was detected in 25% of cases. CONCLUSION: The challenge in clinical red blood cell (RBC) transfusion of patients with sickle cell disease, notably, would be to provide not only phenotypically matched, but also genetically matched, RBC units regarding RhCE variants.     

Early plasmapheresis and rituximab for acute humoral rejection after ABO-compatible liver transplantation

Acute humoral rejection （AHR） is uncommon after ABO- compatible liver transplantation. Herein, we report two cases of AHR treated with plasmapheresis and rituximab in two ABO-compatible liver-transplant patients with preformed anti-human leukocyte antigen donor-specific antibodies. Patient 1 experienced a biopsy-proven AHR at day 10 post-transplant. She was treated by steroid pulses, and OKT3. Because of persisting signs of biopsy-proven AHR at day 26, she was treated by plasmapheresis and rituximab. Uver enzyme levels did not improve, and she died on day 41. Patient 2 experienced a biopsy-proven AHR on day 10 post-transplant. She was treated by steroid pulses, plasmapheresis, and rituximab. Liver enzymes returned to within normal range 18 d after diagnosis. Uver biopsies, at 3 and 9 mo post-transplant, showed complete resolution of AHR. We conclude that plasmapheresis should be started as soon as AHR is diagnosed, and be associated with a B-cell depleting agent. Rituximab may be considered as a first-line therapy.     

An online educational and supportive care application for rectal cancer survivors with low anterior resection syndrome: A mixed methods pilot study

Aim

Restorative proctectomy is commonly associated with significant bowel dysfunction, known as low anterior resection syndrome (LARS), which has a negative impact on patients' quality of life. We developed an online patient-centred application on LARS (eLARS) for rectal cancer survivors. The primary objective of this study was to assess the feasibility of eLARS for rectal cancer survivors with LARS following restorative proctectomy. The secondary objective was to explore participants' experiences with LARS and the eLARS application.

Methods

This was a mixed methods study, which included a feasibility and qualitative analysis. Participants were rectal cancer survivors who underwent restorative proctectomy for rectal cancer within 3 years, completed all adjuvant treatment, and suffered from bowel dysfunction postoperatively. Participants were given access to the application over a 2-month study period. Feasibility was defined as 75% of study participants using the application ≥4 times per month. Semi-structured interviews were conducted with participants after the study period and were analysed using thematic analysis.

Results

Our sample included eight rectal cancer survivors, five women and three men. The median age was 58.5 years (56.5–64.5). Most participants (75%) were >1-year post-restorative proctectomy. 75% of study participants used the application ≥4 times per month for 2 months. Our thematic analysis revealed that participants felt that they lacked access to credible information and emotional support around the time of ileostomy closure, and found that eLARS addressed these challenges.

Conclusion

eLARS is a feasible educational and supportive care intervention for patients with LARS and has the potential to improve patients' quality of life.