Science︱可逆性CD8 T细胞-神经元交互：衰老抑制神经再生的重要机制
Author︱Luming Zhou and Simone Di Giovanni（Department of Brain Sciences, Imperial College London, Division of Neuroscience, London, United Kingdom）
Editor-in-Chief︱Sizhen Wang
Associate Editor︱Yi Lu
Editor︱Jiaxue Zha
Aging has been linked to an increased prevalence of axonal injuries, which are characterized by inadequate regeneration and significant impairment [1-4]. However, the cellular and molecular understanding of the age-dependent regenerative decline is very sparse. Previous studies showed that an age-related impairment in de-differentiation and activation of Schwann cells (SCs) limits axonal regrowth in the injured peripheral nervous system (PNS), impairing sensory and motor recovery [5, 6]. In the central nervous system, following spinal cord injury, deletion of PTEN with an increase in mTOR signaling can only partially limit the aging-dependent axonal regenerative decay of corticospinal tracts [7]. These studies addressed some of the molecular mechanisms underpinning the aging-dependent molecular changes following an injury while aging in itself leads to profound modifications in cell signaling, metabolism, immunity, gene regulation, and protein translation in every tissue affecting homeostasis and predisposing to disease [8-12].
Recently, Zhou et al. from Professor Simone Di Giovanni’s lab in Imperial College London discovered that the inflammatory cytokine lymphotoxin activates NFκB, which induces the neuronal expression of the chemoattractive protein C-X-C motif chemokine ligand 13 (CXCL13) that in turn recruits CXCR5+ CD8+ T cells in proximity of aged DRG neurons expressing MHC-I. CD8 T cells repress axonal regeneration of sensory DRG neurons by inhibiting the regenerative signals pAKT and pS6 via caspase 3 activation. Remarkably, CXCL13 neutralization with monoclonal antibodies prevents CXCR5+ CD8+ T cell recruitment to the DRG and reverses aging-dependent regenerative decline promoting neurological recovery following sciatic nerve injury (SNI). These data propose a novel aging-dependent mechanism restricting the axonal regenerative ability and offer a clinically suitable antibody-based manipulation of neuron-immune cell communication to promote repair. This work was published in Science on 13th May 2022.
In order to identify the aging-dependent gene expression profile, the authors initially performed RNA sequencing from sciatic DRG from 8-10 weeks (young) versus 20-22 months old mice (aged) both preceding (sham) or following asciatic nerve injury (SNI). Gene Ontology (GO), KEGG pathway and STRING analysis using significantly differentially expressed (DE) genes (FDR<0.05) suggested a very significant enrichment in the adaptive immune response including in T and B cell signaling and increased chemotaxis of T and B cells in aged DRG, both preceding and following SNI (Fig. 1A-B). This implies an increased activation of specific subtypes of lymphocytes that might be recruited to aged DRG by the induction of chemo-/cytokines including after nerve injury. This implication further led to identify CXCL13 as by far the most prominently upregulated gene as well as its receptor CXCR5, suggesting the presence of a CXCL13/CXCR5 signaling axis (Fig. 1C).
Fig. 1. Aging induces enrichment in genes associated with chemokine and cytokine expression and adaptive immunity in dorsal root ganglia, including after SNI.
(Zhou et al., Science, 2022)
In addition to observing the neuronal enrichment of CXCL13 and its significant increase in aged DRG before and 3 days after sciatic nerve crush where regeneration was reduced in the aged mice (Fig. 1D-I), the authors identified a robust accumulation of B and T cells including CXCR5+ B and T cells in aged DRG. Interestingly, a considerable number of CD8+ and CXCR5+ CD8+ T cells were CD44+CD69+CD62L–, indicating effector memory T cells (Fig.2D).
Fig. 2. Characterization of B and T cells in DRG.
(Zhou et al., Science, 2022)
Subsequently, to address whether in vivo neuronal expression of CXCL13 would attract CXCR5+ B and T cells, CXCL13 was overexpressed in DRG neurons by infecting sciatic nerves of young mice with AAV-GFP or AAV-CXCL13-GFP. Strickingly, both CXCR5+ B and CD8+ T cells were significantly increased in CXCL13 overexpressing DRG compared to GFP control (Fig. 3A-D). Although the number of migratory CXCR5+ CD4+ T cells in DRG was significantly increased, the percentage with respect to total T cells did not change (Fig. 3E-F). Importantly, sciatic nerve regeneration after AAV-CXCL13 overexpression to find a significant reduction in axonal regeneration with respect to AAV-GFP injected mice (Fig. 3G-I). These data indicate that DRG neuronal overexpression of CXCL13 recruits CXCR5+ B and T cells and it inhibits axonal regeneration partially phenocopying the aging phenotype.
Fig. 3. Overexpression of CXCL13 in DRG neurons in young mice facilitates the recruitment of CXCR5+ B and T cells to the DRG and reduces sciatic nerve regeneration.
(Zhou et al., Science, 2022)
To explore CXCL13 expression signaling, the transcription factors (TFs) putatively involved in driving aging-dependent CXCL13 expression were analyzed in the RNAseq dataset and only Nfkb2 was significantly upregulated in aged DRG becoming a natural candidate for further investigation (Fig. 4A). Several cytokines stimulating NFκB signaling including TNFα, Il1β and lymphotoxin beta subunit (LTα1β2) [13] were screened in primary DRG cell culture and only LTα1β2 significantly enhanced the expression of both Nfkb2 and Cxcl13, which could be attenuated by selective inhibition of the NFκB kinase IKK by using the IKK inhibitor PS1145 (Fig. 4B-C). Both expression and NFκB2 phosphorylation were also induced in aged DRG in vivo. When PS1145 inhibited the phosphorylation of NFkB2, the increase of CXCL13 expression was blocked. These data suggest that aging-related CXCL13 expression is driven by lymphotoxin beta-triggered NFkB2 signaling(Fig. 4D-E).
Fig. 4. CXCL13 expression is modulated by lymphotoxin-beta induced NFkB2 signaling in DRG.
(Zhou et al., Science, 2022)
Are CD8, CD4 T and/or B cells required to impair axonal regeneration of aged sciatic DRG neurons? To address this question, specifically antibody-mediated cell depletion against CD8, CD4 T or B cells respectively was performed in aged mice followed by a sciatic nerve crush for 3 days. Interestingly, CD8 monoclonal antibody promoted nerve regeneration while depletion of CD4+ T cells or B cells did not alter the aging-dependent regenerative decline following sciatic nerve injury (Fig. 5A-C).
Fig. 5. CD8+ T cells and neuronal MHC I are required for age-dependent regenerative decline after SNI through pAKT and pS6.
(Zhou et al., Science, 2022)
MHC-I is expressed by antigen-presenting cells (APCs) to present antigenic peptides on the cell membrane to activate CD8+ T cells after engaging with T cell receptors (TCRs) [14]. Given the increased expression and membrane localization of MHC-I in aged DRG neurons (Fig. 5D-F), the authors hypothesized the requirement of MHC-I expression for the activation of CD8+ T cells to limit nerve regeneration after injury. Thus, an AAV-based approach was applied to express the virally encoded peptide sequence GAr, which inhibits MHC-I antigen presentation evading CD8+ T cell immune responses [15], in sciatic DRG of aged mice. GAr was linked to a reverse tetracycline transactivator (rtTA) responsive to tetracycline/doxycycline inducible expression with a luciferase reporter [16-18]. Following viral infection in sciatic DRG and a sciatic nerve injury for 3 days in aged mice, MHC-I and cleaved caspase 3 expression were significantly reduced and sciatic axonal regeneration was enhanced upon AAV-GAr-rtTA infection (Fig. 5G-J). Together, these data suggest that CD8+ T cells and neuronal MHC-I are required for aging-dependent regenerative decline following SNI.
Based upon the evidence of enhanced sciatic nerve regeneration when caspase 3 was inactivated in GAr-expressed DRG neurons where MHC-I was disrupted, the authors further discovered a significant induction of cleaved caspase 3 in aged DRG neurons after nerve injury (Fig. 5K-L). Strikingly, the phosphorylation of two outstanding regeneration-associated regulators, AKT and S6, was decreased in aged DRG which was significantly reversed by depletion of CD8+ T cells (Fig. 5M-N). Together these data indicate that CD8+ T cell-dependent activation of caspase 3 is required to inhibit pAKT and pS6 in aged DRG neurons after SNI and that inhibition of caspase 3 activation promotes axonal regeneration.
Fig. 6. CXCR5+CD8+ T cells cause aging-dependent regenerative decline after SNI.
(Zhou et al., Science, 2022)
To determine whether CXCR5+ CD8+ T cells are directly responsible for axonal regenerative decline after sciatic nerve injury, WT or CXCR5-/- CD8+ T cells were isolated and adoptively transferred to the young and aged OT-I Rag-/- mice, which are B/T cell deficient, and mice were sacrificed 3 days after nerve injury. The results showed a robust infiltration of adoptively transferred CXCR5+ CD8+ T cells in aged DRG, however these cells were barely detectable in the young recipients (Fig. 6A-B). On the contrary, transferred CXCR5+ CD8+ T cells accumulated more abundantly in the young spleen versus the old (Fig. 5A-B). These migratory patterns of CXCR5+ CD8+ T cells were in line with the differential CXCL13 levels in the young versus aged DRG and spleen tissues (Fig. 6C). Additionally, adoptive transfer of CXCR5+ CD8+ T cells led to a significantly reduced sciatic nerve regeneration after injury only in aged mice where significantly increased perforin and decreased pS6 were observed in DRG neurons (Fig. 6D-G). It suggests that CXCR5+ CD8+ T cells drive aging-dependent axonal regenerative decline after sciatic nerve injury.
Fig. 7. CXCL13 neutralization enhances axonal regeneration, epidermal reinnervation, and sensory functional recovery after SNI in aged mice.
(Zhou et al., Science, 2022)
Lastly, the authors investigated whether blocking CXCL13-dependent pathway could offer a therapeutic opportunity to rejuvenate the sciatic nerve regenerative capacity. Compared to control IgG, they observed that injection of monoclonal antibody against CXCL13 prevented the recruitment of CXCR5+ CD8+ T cells and significantly promoted sciatic nerve regeneration at acute phase after injury and chronically sensory recovery only in aged mice (Fig. 7).
Fig. 8. Aging-dependent regenerative failure in injured sensory neurons.
(Zhou et al., Science, 2022)
More interestingly, an inverse relationship in the expression levels of CXCL13 between spleen and DRG tissues in young and aged mice with corresponding opposite tissue accumulation of CXCR5+ CD8+ T cells. This might not only underpin the mechanism by which these cells accumulate in peripheral tissues, but it might also contribute to a CXCL13-dependent deterioration of immune defense in secondary lymphoid organs that is concomitant with an increase in immune-mediated neuronal damage. Whether this underlines a general phenomenon that goes beyond the peripheral nervous system remains to be determined.
CXCL13 neutralization effectively promoted axonal regeneration and neurological recovery in aged mice while systemic delivery of anti-CXCL13 antibody might affect CXCR5+ T an B cell homing in the secondary lymph tissues, with a risk of increased susceptibility to viral infection and neoplasia. The best timing and route of antibody delivery will be considered for future targeted interventions.
The first author-Dr. Luming Zhou（left）; The co-first author-Dr. Guiping Kong （middle）; The corresponding author-Prof. Simone Di Giovanni（right）.
(Photo credit: Di Giovanni’s lab)
Job opportunity: Exceptionally motivated individual at the PhD or post-doctoral level are invited to apply to Di Giovanni’s lab, email: s.digiovanni@imperial.ac.uk, to study the fundamental mechanisms and translational potential of cell-cell communication for repair in the nervous system across the lifespan.
【1】“ 逻辑神经科学 ”诚聘副主编/编辑/运营岗位（在线办公）
【2】Nat Neurosci | 脱髓鞘后新生少突细胞具有更强的髓鞘再生能力【3】Biol Psychiatry︱高天明团队揭示星形胶质细胞通过乙酰胆碱受体M1对成年海马神经发生及记忆的调控作用【4】Curr Biol︱陈楠楠等揭示突触区翻译引起CaMKII蛋白不对称分布，及其在记忆形成中的作用【5】Front Aging Neurosci 综述︱樊东升团队聚焦肌萎缩侧索硬化中外周与中枢免疫系统的交互作用【6】NPP︱陆伟团队揭示GABAA受体辅助亚基磷酸化修饰调控神经行为的新机制【7】Cereb Cortex︱模式僵化，腹内侧前额叶皮层的时间动态性对反刍与抑郁的作用【8】Front Aging Neurosci︱PTAFR作为一种新型生物标志物助力阿尔兹海默症诊断与治疗【9】PNAS︱冯国平实验室揭示前端丘脑环路在工作记忆中的重要作用【10】Mol Psychiatry︱叶克强课题组揭示炎症激活的C/EBPβ/AEP信号通路介导高脂饮食诱发的糖尿病及阿尔兹海默症
参考文献（上下滑动阅读）
1.Verdú, E., et al., Influence of aging on peripheral nerve function and regeneration. Journal of the Peripheral Nervous System, 2000. 5(4): p. 191-208.
2.Vaughan, D.W., Effects of advancing age on peripheral nerve regeneration. Journal of Comparative Neurology, 1992. 323(2): p. 219-237.
3.Pestronk, A., D.B. Drachman, and J.W. Griffin, Effects of aging on nerve sprouting and regeneration. Experimental neurology, 1980. 70(1): p. 65-82.
4.DeVivo, M.J. and Y. Chen, Trends in new injuries, prevalent cases, and aging with spinal cord injury. Archives of physical medicine and rehabilitation, 2011. 92(3): p. 332-338.
5.Painter, M.W., et al., Diminished Schwann cell repair responses underlie age-associated impaired axonal regeneration. Neuron, 2014. 83(2): p. 331-343.
6.Kang, H. and J.W. Lichtman, Motor axon regeneration and muscle reinnervation in young adult and aged animals. Journal of Neuroscience, 2013. 33(50): p. 19480-19491.
7.Geoffroy, C.G., et al., Evidence for an age-dependent decline in axon regeneration in the adult mammalian central nervous system. Cell reports, 2016. 15(2): p. 238-246.
8.Pomatto, L.C. and K.J. Davies, The role of declining adaptive homeostasis in ageing. The Journal of physiology, 2017. 595(24): p. 7275-7309.
9.Barzilai, N., et al., The critical role of metabolic pathways in aging. Diabetes, 2012. 61(6): p. 1315-1322.
10.Weiskopf, D., B. Weinberger, and B. Grubeck‐Loebenstein, The aging of the immune system. Transplant international, 2009. 22(11): p. 1041-1050.
11.Lardenoije, R., et al., The epigenetics of aging and neurodegeneration. Progress in neurobiology, 2015. 131: p. 21-64.
12.Taylor, R.C. and A. Dillin, Aging as an event of proteostasis collapse. Cold Spring Harbor perspectives in biology, 2011. 3(5): p. a004440.
13.Xiao, X., et al., Lymphotoxin β receptor-mediated NFκB signaling promotes glial lineage differentiation and inhibits neuronal lineage differentiation in mouse brain neural stem/progenitor cells. Journal of neuroinflammation, 2018. 15(1): p. 49.
14.Zinkernagel, R.M., On cross‐priming of MHC class I‐specific CTL: rule or exception? European journal of immunology, 2002. 32(9): p. 2385-2392.
15.Zaldumbide, A. and R. Hoeben, How not to be seen: immune-evasion strategies in gene therapy. Gene therapy, 2008. 15(4): p. 239-246.
16.Zaldumbide, A., et al., A potentially immunologically inert derivative of the reverse tetracycline-controlled transactivator. Biotechnology letters, 2010. 32(6): p. 749-754.
17.Hoyng, S.A., et al., Developing a potentially immunologically inert tetracycline-regulatable viral vector for gene therapy in the peripheral nerve. Gene therapy, 2014. 21(6): p. 549-557.
18.Burnside, E.R., et al., Immune-evasive gene switch enables regulated delivery of chondroitinase after spinal cord injury. Brain, 2018. 141(8): p. 2362-2381.
本文完