- Email: [email protected]
CXCL10 Axis in Rasmussen Encephalitis
Pediatric Neurology 49 (2013) 451e457
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Original Article
A Role for the CXCR3/CXCL10 Axis in Rasmussen Encephalitis Isabel Mirones PhD a, Inmaculada de Prada MD, PhD b, Ana M. Gómez PhD c, Alfonso Luque PhD c, Roberto Martín b, M. Ángeles Pérez-Jiménez MD, PhD d, Luis Madero MD, PhD c, Javier García-Castro PhD a, Manuel Ramírez MD, PhD c, * a
Cellular, Biotechnology Unit, Instituto de Salud Carlos III, Majadahonda, Spain Pathology, Hospital Universitario Niño Jesús, Madrid, Spain c Hematology and Oncology, Hospital Universitario Niño Jesús, Madrid, Spain d Epilepsy Unit, Hospital Universitario Niño Jesús, Madrid, Spain b
abstract BACKGROUND: Rasmussen encephalitis is a devastating pediatric syndrome of unknown etiology that is characterized by progressive loss of neurological function and intractable focal epilepsy. Cytotoxic T lymphocytes have an active role in the pathogenic process of Rasmussen encephalitis. We studied the implication of CXCL10-CXCR3, a chemotactic axis involved in the pathogenesis of several cases of immune encephalitis. METHODS: We analyzed surgical specimens of children with Rasmussen encephalitis, and performed functional in vitro assays to test the implications of the pathological findings. RESULTS: We found that cytotoxic T lymphocytes infiltrating the damaged areas of primary biopsies expressed CXCR3, whereas neurons and astrocytes in the same areas expressed CXCL10. The in vitro assays demonstrated we found that astrocytes upregulated the expression of CXCL10 messenger RNA and the release of CXCL10 to the supernatants on stimulation with polyinosinic-polycyticylic acid, a synthetic double-stranded RNA that mimics infections with either RNA or DNA viruses. Activated T lymphocytes responded to the production of CXCL10 by astrocytes by increasing their migration in a transwell assay. Finally, the chemotaxis induced by the stimulated astrocytes was completely abrogated in the presence of a small molecule antagonist of CXCR3. CONCLUSIONS: Our results suggest that the CXCR3-CXCL10 axis has a role in recruiting pathogenic T lymphocytes into the brains of patients with Rasmussen encephalitis. This chemotactic mechanism may be targeted pharmacologically. Keywords: rasmussen encephalitis, CXCL10, T lymphocytes, brain damage, chemokines
Pediatr Neurol 2013; 49: 451-457 Ó 2013 Elsevier Inc. All rights reserved.
Introduction
Rasmussen encephalitis was first reported in 1958.1 It is characterized by progressive loss of neurological capacities in the presence of intractable focal epilepsy. Most patients are children and the disease affects one hemisphere, resulting in hemiplegia and language losses if the dominant hemisphere is affected. The patients with Rasmussen
I.M. and I.d.P. contributed equally to this paper.
Article History: Received 5 March 2013; Accepted in final form 30 July 2013 * Communications should be addressed to: Dr. Ramírez; Oncohematología y Trasplante; Hospital Universitario Niño Jesús; Avenida Menéndez Pelayo, 65; 28009-Madrid, Spain. E-mail address: [email protected] 0887-8994/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pediatrneurol.2013.07.019
encephalitis typically receive antiepileptic and immunosuppressant drugs, intravenous immunoglobulins, and eventually they undergo functional hemispherectomy. The etiology and pathogenic process are unknown, although an immune component is clear based on pathological findings in the biopsies of the patients.2,3 Histologically, the findings of Rasmussen encephalitis resemble those of viral meningoencephalitis. Biopsies show leptomeningeal and parenchymal perivascular inflammation, mononuclear infiltration predominantly of T lymphocytes, diffuse proliferation of microglial cells with focal microglial nodule formation, and neuron destruction. Several stages of cortical damage can be identified,2 reflecting an ongoing and progressive immune-mediated process. T lymphocytes, mainly cytotoxic CD8 T cells, are found in the areas of neuronal destruction, indicating an active role of
452
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
these cells in the pathogenic process of Rasmussen encephalitis.4 Additional work found the same process affecting astrocytes in damaged areas.5 Therefore, a T celledependent immune response contributes to the onset and evolution of the disease. CD8 T lymphocytes recognize and destroy cells expressing antigens in their class-I human leukocyte antigen molecules. Antigen presentation takes place in the lymph nodes, and from there, effector T lymphocytes migrate toward the areas where the antigens localize.6 The presence of CD8 T lymphocytes with the same clonogenic marker has been reported in the brains and peripheral blood of patients with Rasmussen encephalitis,7 confirming the movement of pathogenic T lymphocytes from the periphery. The authors of this work concluded that an antigen-driven human leukocyte antigen class Ierestricted, CD8þ T cellemediated attack against antigens presented on neurons and astrocytes was the dominant pathogenic process in Rasmussen encephalitis. These findings have important implications, one of them being that the effector lymphocytes responsible for the brain destruction must migrate from the periphery to the parenchyma, where they cause the damage. This process may be going on as long as the disease is progressing. There is no information about the molecular basis of the movement of T lymphocytes from the site of antigen presentation to the areas of brain destruction in patients with Rasmussen encephalitis. Lymphocytes ingress into the central nervous system (CNS) through the bloodebrain barrier and infiltrate the parenchyma.8 Several adhesion molecules have been involved in the many steps that the lymphocytes proceed in the process, including chemokines.9 Chemokines are a family of molecules that govern the homing of leukocytes into different organs. A current scenario explains that cells within the organs secrete chemokines that recruit those T lymphocytes that express the specific receptor in their membranes. We have studied CXCL10-CXCR3 in children with Rasmussen encephalitis and found a possible role for this axis in the pathogenesis of the disease.
washing twice with phosphate-buffered saline and the astrocyte layer was covered with fresh Dulbecco’s Modified Eagle Medium (Lonza, Basel, Switzerland) containing 10% fetal calf serum. Cortical astrocyte cultures were plated at 1 106 cells per 10-cm plate and grown to confluence, with medium changes every 3 days.12 Quantitative real-time reverse transcription polymerase chain reaction
Astrocytes plated in T25 tissue culture flasks were incubated in the presence of polyinosinic-polycyticylic acid (poly I:C) (Sigma, Tres Cantos, Madrid, Spain) 100 mg/mL or media alone for 24 hours. Cells were scraped off the flasks, and total RNA was extracted using TRIzol reagent (Invitrogen, Alcobendas, Madrid, Spain) per the manufacturer’s instructions. First-strand complementary DNA was generated from 0.75 mg of RNA using M-MLV Reverse Transcriptase (Invitrogen) and oligo(dT)12-18 primers (Invitrogen). Following complementary DNA synthesis, quantitative polymerase chain reaction was carried out using LightCycler 480 SYBR Green I Master (Roche, Madrid, Spain) in the LightCycler detection system (Roche) for 45 cycles. The primers were synthesized by TIB MOLBIOL (Berlin, Germany) and the sequences are as follows: glyceraldehyde 3-phosphate dehydrogenase: forward primer (50 -CTCTGGAAAGCTGTGGCGTGAT-30 ) and reverse primer (50 CTGGGATGGAATTGTGAGGGGG-30 ); CXCL10: forward primer (50 CCAAGTGCTGCCGTCATTTTC-30 ) and reverse primer (50 -GGCTCGCAGGGATGATTTCAA-30 ). A cycle threshold value in the linear range of amplification was selected for each sample and normalized for level of glyceraldehyde 3-phosphate dehydrogenase expression. The complementary DNA levels in samples were estimated by the relative quantification method: using the formula 2DDCT, where DDCT is the difference between the selected cycle threshold value of a particular poly I:C stimulated sample and the mean of the cycle thresholds of the nonstimulated samples.13 The mean CXCL10 expression level of the nonstimulated samples was assigned an expression value of 1.0 and the fold increase in CXCL10 expression was determined for each control and the stimulated sample. Individual stimulated samples were considered to have significantly different expression of CXCL10 compared with nonstimulated sample when the P value for a t test comparing the two was < 0.05. Experiments were done in triplicate. Analysis of CXCL10 protein expression
Material and Methods
Astrocytes were seeded at 2.5 104 cells in 24-well plates in 1 mL medium containing 20% fetal calf serum. Three days later, cells were treated with poly I:C at 10 mg/mL and 100 mg/mL. Doses selected were determined experimentally to elicit optimal responses. At the indicated times, CXCL10 protein levels in culture medium were measured by enzyme-linked immunosorbent assay (ELISA) using a commercial kit and following the manufacturer’s instructions (Mouse CRG-2/IP-10 Ray Bio ELISA Kit, RayBiotech, Norcross, GA).
Patients and biopsies
Lymphocyte activation
Brain specimens from six patients (aged 2-14) who underwent hemispherectomy for treatment of Rasmussen encephalitis were examined. The tissues were fixed in 10% neutral formalin. Paraffin-embedded section were cut at 3 mm and stained for hematoxylin-eosin. We used the disease stage established and adapted for Rasmussen encephalitis to define pathologic features of the disease.2,10 The following polyclonal and monoclonal antibodies were used: Glial fibrillary acidic protein (rabbit polyclonal; prediluted, Dako), NeuN1 (A60 mouse monoclonal; 1:100, Chemicon, Billerica, MA), CD8 (C8/144 B mouse monoclonal; prediluted, Dako), CXCR3 (mouse polyclonal; 1:50, BD Pharmingen, San Agustin de Guadalix, Madrid, Spain), and CXCL10 (goat polyclonal; RD Systems, Minneapolis, MN). Lymphoid tissue from tonsil served as positive control for CXCR3 and CD8; papillary carcinoma of thyroid for CXCL10 and normal brain tissue for GFAP and NeuN1.
T lymphocytes were enriched from murine splenocytes by using the Pan-T-cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) and were cultured in triplicate in flat-bottom 24-well plates at 1 106 cells/mL, in RPMI medium (Lonza) containing 10% fetal calf serum. They were stimulated with immobilized anti-CD3 (7 mg/mL) and soluble antiCD28 (1 mg/mL) antibodies (BD Pharmingen). Transwell cultures
Resting or activated astrocytes were cultured at the lower wells of 24-well Transwell plates with 3 mm pores (Costar, Cambridge, MA). Activated T lymphocytes were added to the upper chambers, with or without the CXCR3 antagonist TAK-799 (obtained from the National Institutes of Health AIDS Research and Reference Reagent Program Division of AIDS, NIAID, NIH: TAK-779, 500 nM). The numbers of T CD8 lymphocytes in the lower wells were counted using a flow cytometer (FACS Canto II, BD). All conditions were performed in triplicate.
Astrocyte isolation
Statistics
Neonatal C57BL/6 mice aged 1-2 days were used for the isolation of astrocytes. Primary astrocytes cultures were obtained from brain cortices by a modification of a published method.11 Oligodendrocytes and microglia were removed by shaking cultures at 200 rpm overnight and
Wilcoxon test (Mann-Whitney statistic) was applied to compare the different groups in experiments. Differences were considered statistically significant when P < 0.05. All graphics present the mean standard error.
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
Results Biopsies
We retrospectively studied brain biopsies from children with a diagnosis of Rasmussen encephalitis who underwent surgery. The Table shows the main characteristics of these patients. The expression of several molecules (see Materials and Methods) was assessed by immunohistochemistry. We found the characteristic features of lesions previously described in these patients: moderate chronic inflammation with pan laminar distribution (Supplemental Fig S1A), microglial nodules and perivascular cuffing by round cell (Supplemental Fig S1B) spreading to the white matter, accumulation of lymphocytes (Supplemental Fig S1C) with neuronophagia (Supplemental Fig S1D), decrease of neuronal population, activated microglia, reactive
453
astrocytosis, and cortical vacuolation. In addition, we found that the main lymphocyte population corresponded to CD8 T lymphocytes (Fig 1A), with few CD4 T lymphocytes (Supplemental Fig S2). Many of these T lymphocytes expressed the chemokine receptor CXCR3 (Fig 1A). We looked for the expression of CXCL10, one of the ligands for CXCR3. We found CXCL10 expressed in the same areas of lymphocyte accumulation (Fig 1B). In correlative sections of the brains, we found that mostly astrocytes, and in a lesser amount neurons, were the main cell populations expressing CXCL10 in the brains of children with Rasmussen encephalitis (Fig 1C). These findings suggest that the axis CXCR3CXCL10 may be involved in the recruitment of activated T lymphocytes into the areas of brain damage in these children. We next set up in vitro experiments to rule out this possibility.
TABLE. Patient’s characteristics
Patient Gender Clinical Events Before Epilepsy Onset
Age at Hemisphere Affected Epilepsy Epilepsy Course Onset
Neurological/Cognitive Deficit Before Surgery
Immunologic Treatmenty
Left hemiparesis Psychomotor arrest
Prednisone 3 yr, Gammaglobulin class I Moderate cognitive delay
Mild left hemiparesis Mild cognitive deterioration
4 yr, Prednisone Gammaglobulin class III, Exitus 3 mo after surgery (epileptic status) Prednisone 6 yr, Gammaglobulin Significant cognitive recovery. Language transference to the right hemisphere
1
Girl
Isolated “febrile seizure” at 16 mo of age
2 yr
2
Girl
No clinical events
3 yr
3
Boy
Pharyngoamygdalitis 5 yr 3 weeks before onset
4
Boy
Uveitis
8 yr
5
Girl
Mild cranial trauma 3 mo before onset
10 yr
6
Girl
No clinical events*
6 yr
Left hemisphere Rapid progression into severe epilepsy with polymorphic seizures, status epilepticus, Epilepsia Partialis Continua (EPC) Right hemisphere Rapid progression into severe epilepsy with polymorphic seizures, status epilepticus, EPC Left hemisphere Rapid progression into severe epilepsy with polymorphic seizures, status epilepticus, EPC Right hemisphere Focal motor seizures: > delayed progression into severe epilepsy with polymorphic seizures, EPC Right hemisphere Focal motor seizures: > delayed progression into severe epilepsy with polymorphic seizures, status epilepticus, EPC Right hemisphere Partial complex seizures: > delayed progression into severe epilepsy with polymorphic seizures, status epilepticus
Mild right hemiparesis, ataxia, Mild cognitive deterioration of left hemispheric cognitive functions Mild left hemiparesis, mild cognitive deterioration of right hemisphere cognitive functions
Hemispherectomy Age/ Outcome (Engel’s Class)/ Neuropsychological Status
Prednisone 13 yr, Gammaglobulin class I Tacrolimus Significant cognitive recovery
Mild left hemiparesis, Mild cognitive deterioration of right hemisphere cognitive functions
Prednisone 13 yr, Gammaglobulin class II Moderate cognitive recovery
Mild left body functional deficit, mild deterioration of right hemisphere cognitive functions
Prednisone 8 yr, Gammaglobulin class II Moderate cognitive recovery
Mother suffers from severe psoriasis. All patients were treated according to Bien CG, et al. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis: a European consensus statement. Brain. 2005;128:454-471. * y
454
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
FIGURE 1. T lymphocytes in primary lesions of children with Rasmussen encephalitis are mostly CD8 and express CXCR3, whereas neuron and astrocytes express CXCL10. (A) Immunohistochemistry with anti-CD8 (left) and anti-CXCR3 (right) monoclonal antibodies in two consecutive sections of the same brain area. The blood vessel serves as a reference. (B) Immunohistochemistry with anti-CXCL10 monoclonal antibody. (C) Immunohistochemistry with anti-NeuN1 (left), anti-CXCL10 (middle), and anti-glial fibrillary acidic protein (right) monoclonal antibodies in consecutive slices of the same brain area. The blood vessel serves as a reference. Most CXCL10 positive cells corresponded to astrocytes (glial fibrillary acidic protein positive), although some neurons (NeuN1 positive) also expressed CXCL10.
Stimulation of astrocytes
The etiology of Rasmussen encephalitis is unknown, but a viral cause has been previously suggested.3 Viruses may induce the production of CXCL10 by neurons in the West Nile Virus encephalitis model.14 We focused on astrocytes, to see whether this cell type could also produced CXCL10 after “viral-like” stimuli. Poly (I:C) is a synthetic dsRNA which mimics infections with either RNA or DNA viruses15 and activates TLR3, a receptor expressed by astrocytes. Astrocytes upregulated the expression of CXCL10 mRNA in a poly (I:C) dose-dependent fashion (Fig 2A). The release of CXCL10 into the supernatants was quantitated by ELISA and showed the production of the protein in parallel to the
upregulation of the messenger RNA (Fig 2B). These experiments showed that TLR3-stimulated astrocytes produced CXCL10, suggesting that a viral infection may induce the production of this chemokine by astrocytes, as it has been reported with neurons. Cocultures
We tested whether the production of CXCL10 by astrocytes had functional implications for T lymphocytes. Purified T lymphocytes were stimulated with immobilized anti-CD3 and soluble anti-CD28 antibodies for 96 hours. The maximal upregulation of CXCR3 was detected 48 hours after stimulation (not shown); therefore, the coculture
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
455
FIGURE 3. TLR3-activated astrocytes induced chemotaxis in activated T lymphocytes. Anti-CD3/anti-CD28-activated T lymphocytes with/without TAK779, a CXCR3 antagonist, were placed in the upper chamber of transwell cultures, with/without TLR3-stimulated astrocytes at the bottom wells. Chemotaxis was quantitated by flow cytometry (AU ¼ Arbitrary units), and values were normalized (AU ¼ 1) to chemotaxis induced by nonstimulated astrocytes. n ¼ 3.
FIGURE 2. Astrocytes produce CXCL10 upon TLR3 stimulation. (A) Quantitative realtime polymerase chain reaction results for expression of CXCL10 in stimulated astrocytes (black column) and control cells (white column). Astrocyte cultures were treated with medium alone and polycyticylic acid (poly [I:C]) 100 mM for 24 hours. Data are plotted as the relative expression of CXCL10 of a particular sample relative to the mean CXCL10 expression level of the nonstimulated astrocytes. Data shown are the average standard error of the mean (SEM) of three separate experiments. Levels are significantly major in poly (I:C)etreated astrocytes (asterisks indicate P < 0.01). (B) CXCL10 chemokine expression by poly (I:C)-treated astrocytes. Cell cultures were treated with medium alone or with 10 mM and 100 mM poly (I:C). Culture supernatants were collected at 24, 48, and 72 hours posttreatment and analyzed for CXCL10 expression by enzyme-linked immunosorbent assay. Data shown are the average SEM of three separate experiments. Levels are significantly major in poly (I:C)etreated astrocytes (asterisks P value <.01).
experiments were done after a 48-hour stimulation of T lymphocytes. Next we stimulated astrocytes with poly (I:C) on the lower wells of transwell culture plates, and seeded T lymphocytes (previously activated with anti-CD3 and antiCD28 antibodies) at the upper chambers. Chemotaxis of T CD8 lymphocytes was quantitated by flow cytometry. The results showed that activated T lymphocytes responded to the production of CXCL10 by poly (I:C)-stimulated astrocytes by increasing their migration in the transwell assay (Fig 3). These results showed that poly (I:C)stimulated astrocytes induced chemotaxis on activated T lymphocytes. Inhibition of chemotaxis by a CXCR3-antagonist small molecule
We set up transwell cultures as explained previously, with or without a CXCR3 antagonist molecule. The
chemotaxis of activated T lymphocytes induced by the TLR3-stimulated astrocytes was completely abrogated in the presence of the antagonist (Fig 3). These results suggest that the recruitment of activated T lymphocytes by stimulated astrocytes may be specifically inhibited by small drugs targeting CXCR3. Discussion
This study shows that the axis CXCR3-CXCL10 may have an active role in recruiting T lymphocytes into the brain parenchyma of children with Rasmussen encephalitis. Biopsies obtained from the damaged brains of children had clearly shown the destruction of neurons3,4 and astrocytes5 by cytotoxic T lymphocytes.4 We found similar CD8 T lymphocyte infiltration in the damaged areas of the brains we analyzed. Our patients represent the prototype of Rasmussen encephalitis, both clinically and pathologically. The finding of CD8 T cells in the brain biopsy specimens of individuals with Rasmussen encephalitis suggests, but does not prove, that these cells are responsible for the primary pathology of the disease. It has been shown that the lymphocytes found in the brains of patients with Rasmussen encephalitis represented the clonal expansion of cytotoxic T cells.7 Interestingly, the T-cell clones isolated from the brains had the same CDR3 genomic sequence of T-cell clones in peripheral blood. These T cells were detected in the patients along the course of the disease. All these findings have important implications, one of them being that the effector lymphocytes responsible for the brain destruction must migrate from the periphery to the brain parenchyma where they cause the damage. We found that CD8 lymphocytes at the sites of tissue destruction expressed CXCR3 on its membranes, and that the expression of CXCL10 in the brain corresponded to the same areas to those of lymphocyte infiltration. Both neurons and
456
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
astrocytes, mostly the latter, produced CXCL10 in the damaged areas. All these findings support the hypothesis that T lymphocytes were recruited through the CXCR3CXCL10 axis in these patients. CXCR3 and CXCL10 have been previously involved in the trafficking of T lymphocytes during encephalomyelitis,16 but not in cases of Rasmussen encephalitis. T lymphocytes upregulate the expression of CXCR3 upon activation,17,18 and CXCR3þ T lymphocytes are needed for an effective immune response at the CNS.19 Interestingly, CXCR3þ T lymphocytes have been involved in the regional antiviral response within the CNS,20,21 underscoring the role of this receptor in the chemotaxis of T lymphocytes to discrete areas of the CNS. CXCL10 is one of the known ligands for CXCR3; the sources of the chemokine at the brain are neurons and astrocytes. Neurons infected during viral encephalitis produce CXCL10.14 Although the etiology of Rasmussen encephalitis is unknown, a viral infection has been proposed. The histological similarities between Rasmussen encephalitis and viral encephalitis have been invoked, linking the disease to a virally triggered process. Several hypotheses have been proposed to explain autoimmune T cellemediated diseases induced by a viral infection.22 Target cells may share self structures with pathogen antigens (antigen mimicry hypothesis); cell destruction in the context of a viral infection may result in an epitope spreading to tissue-specific epitopes (epitope spreading hypothesis); or nonspecific inflammatory stimuli may cause the activation of bystander auto reactive T cells (bystander damage and fertile field hypothesis). An additional hypothesis, known as viral déjà vu,22 proposed that the autoimmunity phenomenon happened after successive antigenically related viral infections. Interestingly, the murine model described in the viral déjà vu theory of viral infection in the brain, is the only animal model related to Rasmussen encephalitis.22 Based on this hypothesis, we tested if astrocytes could produce CXCL10 after a stimulus resembling a viral infection. Astrocytes express TLR323 and produced CXCL10 after stimulating them with poly (I:C).13 In the absence of a known virus for Rasmussen encephalitis, we found that astrocytes upregulated the production of CXCL10 upon TLR3 activation. These results are in line with the work on Japanese encephalitis,24 a model of murine viral encephalitis in which astrocytes produced CXCL10 after infection. In our hands, the production of CXCL10 by astrocytes resulted in the recruitment of stimulated T lymphocytes, as shown in transwell assays. Moreover, we could inhibit the chemotaxis induced by TLR3-activated astrocytes with the use of small molecules that antagonize CXCR3 signaling.25,26 The potential of these molecules as therapeutic anti-CXCR3 drugs have been shown in vitro and in vivo in models of immune and neoplastic diseases, and have been already used in humans.27 In summary, we propose a scenario for Rasmussen encephalitis in which astrocytes and neurons recruit pathogenic T lymphocytes into areas of the brain through the CXCL10-CXCR3 axis. We show that this may be pharmacologically targeted with antagonists of CXCR3. Our findings may help in designing new therapeutic strategies based on immune modulation, specifically directed toward molecules involved in the pathogenesis of the disease.
We thank Dr. Fernando Neria for astrocyte isolation assistance. We gratefully acknowledge the helpful suggestions and expert advice in molecular biology techniques of the article by Dr. Miguel Ángel Rodríguez-Milla.
Supplementary data
Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.pediatrneurol.2013.07.019. References 1. Rasmussen T, Olszewski J, Lloyd-Smith DL. Focal seizures due to chronic localized encephalitis. Neurology. 1958;8:435-455. 2. Pardo CA, Vining EP, Guo L, Skolasky RL, Carson BS, Freeman JM. The pathology of Rasmussen syndrome: stages of cortical involvement and neuropathological studies in 45 hemispherectomies. Epilepsia. 2004;45:516-526. 3. Prayson RA, Frater JL. Rasmussen encephalitis: a clinicopathologic and immunohistochemical study of seven patients. Am J Clin Pathol. 2002;117:776-782. 4. Bien CG, Bauer J, Deckwerth TL, et al. Destruction of neurons by cytotoxic T cells: a new pathogenic mechanism in Rasmussen’s encephalitis. Ann Neurol. 2002;51:311-318. 5. Bauer J, Elger CE, Hans VH, et al. Astrocytes are a specific immunological target in Rasmussen’s encephalitis. Ann Neurol. 2007;62: 67-80. 6. Goverman J. Autoimmune T cell responses in the central nervous system. Nat Rev Immunol. 2009;9:393-407. 7. Schwab N, Bien CG, Waschbisch A, et al. CD8þ T-cell clones dominate brain infiltrates in Rasmussen encephalitis and persist in the periphery. Brain. 2009;132:1236-1246. 8. Wilson EH, Weninger W, Hunter CA. Trafficking of immune cells in the central nervous system. J Clin Invest. 2010;120:1368-1379. 9. Ransohoff RM, Kivisäkk P, Kidd G. Three or more routes for leukocyte migration into the central nervous system. Nat Rev Immunol. 2003;3:569-581. 10. Robitaille Y, Rasmussen T, Dubeau F, Tampieri D, Kemball K. Histopathology of nonneoplastic lesions in frontal lobe epilepsy. Review of 180 cases with recent MRI and PET correlations. Adv Neurol. 1992;57:499-513. 11. McCarthy KD, de Vellis J. Alpha-adrenergic receptor modulation of beta-adrenergic, adenosine and prostaglandin E1 increased adenosine 3’:5’-cyclic monophosphate levels in primary cultures of glia. J Cyclic Nucleotide Res. 1978;1:15-26. 12. Canellada A, Ramirez BG, Minami T, Redondo JM, Cano E. Calcium/ calcineurin signaling in primary cortical astrocyte cultures: Rcan1-4 and cyclooxygenase-2 as NFAT target genes. Glia. 2008;56:709-722. 13. Dussault AA, Pouliot M. Rapid and simple comparison of messenger RNA levels using real-time PCR. Biol Proced Online. 2006;8:1-10. 14. Klein RS, Lin E, Zhang B, et al. Neuronal CXCL10 directs CD8þ T-cell recruitment and control of West Nile virus encephalitis. J Virol. 2005;79:11457-11466. 15. Jacobs BL, Langland JO. When two strands are better than one: the mediators and modulators of the cellular responses to doublestranded RNA. Virology. 1996;219:339-349. 16. Müller M, Carter S, Hofer MJ, Campbell IL. Review: the chemokine receptor CXCR3 and its ligands CXCL9, CXCL10 and CXCL11 in neuroimmunityea tale of conflict and conundrum. Neuropathol Appl Neurobiol. 2010;36:368-387. 17. Bonecchi R, Bianchi G, Bordignon PP, et al. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J Exp Med. 1998;187:129-134. 18. Qin S, Rottman JB, Myers P, et al. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory reactions. J Clin Invest. 1998;101:746-754. 19. Christensen JE, Nansen A, Moos T, et al. Efficient T-cell surveillance of the CNS requires expression of the CXC chemokine receptor 3. J Neurosci. 2004;24:4849-4858. 20. Müller M, Carter SL, Hofer MJ, et al. CXCR3 signaling reduces the severity of experimental autoimmune encephalomyelitis by controlling the parenchymal distribution of effector and regulatory T cells in the central nervous system. J Immunol. 2007;179:2774-2786.
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457 21. Zhang B, Chan YK, Lu B, Diamond MS, Klein RS. CXCR3 mediates regionspecific antiviral T cell trafficking within the central nervous system during West Nile virus encephalitis. J Immunol. 2008;180:2641-2649. 22. Merkler D, Horvath E, Bruck W, Zinkernagel RM, Del la Torre JC, Pinschewer DD. “Viral déjà vu” elicits organ-specific immune disease independent of reactivity to self. J Clin Invest. 2006;116:1254-1263. 23. Farina C, Krumbholz M, Giese T, Hartmann G, Aloisi F, Meinl E. Preferential expression and function of Toll-like receptor 3 in human astrocytes. J Neuroimmunol. 2005;159:12-19. 24. Bhowmick S, Duseja R, Das S, Appaiahgiri MB, Vrati S, Basu A. Induction of IP-10 (CXCL10) in astrocytes following Japanese encephalitis. Neurosci Lett. 2007;414:45-50.
457
25. Johnson M, Li AR, Liu J, et al. Discovery and optimization of a series of quinazolinone-derived antagonists of CXCR3. Bioorg Med Chem Lett. 2007;17:3339-3343. 26. Baba M, Nishimura O, Kanzaki N, et al. A small-molecule, nonpeptide CCR5 antagonist with highly potent and selective anti-HIV1 activity. Proc Natl Acad Sci U S A. 1999;96:5698-5703. 27. Tonn GR, Wong SG, Wong SC, et al. An inhibitory metabolite leads to dose- and time-dependent pharmacokinetics of (R)-N-{1[3-(4-ethoxy-phenyl)-4-oxo-3,4-dihydro-pyrido[2,3-d]pyrimidin-2-yl] -ethyl}-N-pyridin-3-yl-methyl-2-(4-trifluoromethoxy-phenyl)acetamide (AMG 487) in human subjects after multiple dosing. Drug Metab Dispos. 2009;37:502-513.
457.e1
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
SUPPLEMENTAL FIGURE S1. Histological features of the biopsies of patients with Rasmussen encephalitis. The figure shows several characteristic findings in the hematoxylin-eosin examination of the brain biopsies. (A) Low magnification showing the areas of moderate chronic inflammation with pan laminar distribution and microglial nodules. (B) Perivascular accumulation of mononuclear cells. (C) Detailed view of an area with lymphocyte infiltration in close apposition to neurons. (D) Accumulation of lymphocytes with neuronophagia.
SUPPLEMENTAL FIGURE S2. CD4 T lymphocytes in brain biopsies of patients with Rasmussen encephalitis. Low magnification picture of a brain biopsy showing accumulation of CD4 cells in the parenchyma.
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Original Article
A Role for the CXCR3/CXCL10 Axis in Rasmussen Encephalitis Isabel Mirones PhD a, Inmaculada de Prada MD, PhD b, Ana M. Gómez PhD c, Alfonso Luque PhD c, Roberto Martín b, M. Ángeles Pérez-Jiménez MD, PhD d, Luis Madero MD, PhD c, Javier García-Castro PhD a, Manuel Ramírez MD, PhD c, * a
Cellular, Biotechnology Unit, Instituto de Salud Carlos III, Majadahonda, Spain Pathology, Hospital Universitario Niño Jesús, Madrid, Spain c Hematology and Oncology, Hospital Universitario Niño Jesús, Madrid, Spain d Epilepsy Unit, Hospital Universitario Niño Jesús, Madrid, Spain b
abstract BACKGROUND: Rasmussen encephalitis is a devastating pediatric syndrome of unknown etiology that is characterized by progressive loss of neurological function and intractable focal epilepsy. Cytotoxic T lymphocytes have an active role in the pathogenic process of Rasmussen encephalitis. We studied the implication of CXCL10-CXCR3, a chemotactic axis involved in the pathogenesis of several cases of immune encephalitis. METHODS: We analyzed surgical specimens of children with Rasmussen encephalitis, and performed functional in vitro assays to test the implications of the pathological findings. RESULTS: We found that cytotoxic T lymphocytes infiltrating the damaged areas of primary biopsies expressed CXCR3, whereas neurons and astrocytes in the same areas expressed CXCL10. The in vitro assays demonstrated we found that astrocytes upregulated the expression of CXCL10 messenger RNA and the release of CXCL10 to the supernatants on stimulation with polyinosinic-polycyticylic acid, a synthetic double-stranded RNA that mimics infections with either RNA or DNA viruses. Activated T lymphocytes responded to the production of CXCL10 by astrocytes by increasing their migration in a transwell assay. Finally, the chemotaxis induced by the stimulated astrocytes was completely abrogated in the presence of a small molecule antagonist of CXCR3. CONCLUSIONS: Our results suggest that the CXCR3-CXCL10 axis has a role in recruiting pathogenic T lymphocytes into the brains of patients with Rasmussen encephalitis. This chemotactic mechanism may be targeted pharmacologically. Keywords: rasmussen encephalitis, CXCL10, T lymphocytes, brain damage, chemokines
Pediatr Neurol 2013; 49: 451-457 Ó 2013 Elsevier Inc. All rights reserved.
Introduction
Rasmussen encephalitis was first reported in 1958.1 It is characterized by progressive loss of neurological capacities in the presence of intractable focal epilepsy. Most patients are children and the disease affects one hemisphere, resulting in hemiplegia and language losses if the dominant hemisphere is affected. The patients with Rasmussen
I.M. and I.d.P. contributed equally to this paper.
Article History: Received 5 March 2013; Accepted in final form 30 July 2013 * Communications should be addressed to: Dr. Ramírez; Oncohematología y Trasplante; Hospital Universitario Niño Jesús; Avenida Menéndez Pelayo, 65; 28009-Madrid, Spain. E-mail address: [email protected] 0887-8994/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pediatrneurol.2013.07.019
encephalitis typically receive antiepileptic and immunosuppressant drugs, intravenous immunoglobulins, and eventually they undergo functional hemispherectomy. The etiology and pathogenic process are unknown, although an immune component is clear based on pathological findings in the biopsies of the patients.2,3 Histologically, the findings of Rasmussen encephalitis resemble those of viral meningoencephalitis. Biopsies show leptomeningeal and parenchymal perivascular inflammation, mononuclear infiltration predominantly of T lymphocytes, diffuse proliferation of microglial cells with focal microglial nodule formation, and neuron destruction. Several stages of cortical damage can be identified,2 reflecting an ongoing and progressive immune-mediated process. T lymphocytes, mainly cytotoxic CD8 T cells, are found in the areas of neuronal destruction, indicating an active role of
452
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
these cells in the pathogenic process of Rasmussen encephalitis.4 Additional work found the same process affecting astrocytes in damaged areas.5 Therefore, a T celledependent immune response contributes to the onset and evolution of the disease. CD8 T lymphocytes recognize and destroy cells expressing antigens in their class-I human leukocyte antigen molecules. Antigen presentation takes place in the lymph nodes, and from there, effector T lymphocytes migrate toward the areas where the antigens localize.6 The presence of CD8 T lymphocytes with the same clonogenic marker has been reported in the brains and peripheral blood of patients with Rasmussen encephalitis,7 confirming the movement of pathogenic T lymphocytes from the periphery. The authors of this work concluded that an antigen-driven human leukocyte antigen class Ierestricted, CD8þ T cellemediated attack against antigens presented on neurons and astrocytes was the dominant pathogenic process in Rasmussen encephalitis. These findings have important implications, one of them being that the effector lymphocytes responsible for the brain destruction must migrate from the periphery to the parenchyma, where they cause the damage. This process may be going on as long as the disease is progressing. There is no information about the molecular basis of the movement of T lymphocytes from the site of antigen presentation to the areas of brain destruction in patients with Rasmussen encephalitis. Lymphocytes ingress into the central nervous system (CNS) through the bloodebrain barrier and infiltrate the parenchyma.8 Several adhesion molecules have been involved in the many steps that the lymphocytes proceed in the process, including chemokines.9 Chemokines are a family of molecules that govern the homing of leukocytes into different organs. A current scenario explains that cells within the organs secrete chemokines that recruit those T lymphocytes that express the specific receptor in their membranes. We have studied CXCL10-CXCR3 in children with Rasmussen encephalitis and found a possible role for this axis in the pathogenesis of the disease.
washing twice with phosphate-buffered saline and the astrocyte layer was covered with fresh Dulbecco’s Modified Eagle Medium (Lonza, Basel, Switzerland) containing 10% fetal calf serum. Cortical astrocyte cultures were plated at 1 106 cells per 10-cm plate and grown to confluence, with medium changes every 3 days.12 Quantitative real-time reverse transcription polymerase chain reaction
Astrocytes plated in T25 tissue culture flasks were incubated in the presence of polyinosinic-polycyticylic acid (poly I:C) (Sigma, Tres Cantos, Madrid, Spain) 100 mg/mL or media alone for 24 hours. Cells were scraped off the flasks, and total RNA was extracted using TRIzol reagent (Invitrogen, Alcobendas, Madrid, Spain) per the manufacturer’s instructions. First-strand complementary DNA was generated from 0.75 mg of RNA using M-MLV Reverse Transcriptase (Invitrogen) and oligo(dT)12-18 primers (Invitrogen). Following complementary DNA synthesis, quantitative polymerase chain reaction was carried out using LightCycler 480 SYBR Green I Master (Roche, Madrid, Spain) in the LightCycler detection system (Roche) for 45 cycles. The primers were synthesized by TIB MOLBIOL (Berlin, Germany) and the sequences are as follows: glyceraldehyde 3-phosphate dehydrogenase: forward primer (50 -CTCTGGAAAGCTGTGGCGTGAT-30 ) and reverse primer (50 CTGGGATGGAATTGTGAGGGGG-30 ); CXCL10: forward primer (50 CCAAGTGCTGCCGTCATTTTC-30 ) and reverse primer (50 -GGCTCGCAGGGATGATTTCAA-30 ). A cycle threshold value in the linear range of amplification was selected for each sample and normalized for level of glyceraldehyde 3-phosphate dehydrogenase expression. The complementary DNA levels in samples were estimated by the relative quantification method: using the formula 2DDCT, where DDCT is the difference between the selected cycle threshold value of a particular poly I:C stimulated sample and the mean of the cycle thresholds of the nonstimulated samples.13 The mean CXCL10 expression level of the nonstimulated samples was assigned an expression value of 1.0 and the fold increase in CXCL10 expression was determined for each control and the stimulated sample. Individual stimulated samples were considered to have significantly different expression of CXCL10 compared with nonstimulated sample when the P value for a t test comparing the two was < 0.05. Experiments were done in triplicate. Analysis of CXCL10 protein expression
Material and Methods
Astrocytes were seeded at 2.5 104 cells in 24-well plates in 1 mL medium containing 20% fetal calf serum. Three days later, cells were treated with poly I:C at 10 mg/mL and 100 mg/mL. Doses selected were determined experimentally to elicit optimal responses. At the indicated times, CXCL10 protein levels in culture medium were measured by enzyme-linked immunosorbent assay (ELISA) using a commercial kit and following the manufacturer’s instructions (Mouse CRG-2/IP-10 Ray Bio ELISA Kit, RayBiotech, Norcross, GA).
Patients and biopsies
Lymphocyte activation
Brain specimens from six patients (aged 2-14) who underwent hemispherectomy for treatment of Rasmussen encephalitis were examined. The tissues were fixed in 10% neutral formalin. Paraffin-embedded section were cut at 3 mm and stained for hematoxylin-eosin. We used the disease stage established and adapted for Rasmussen encephalitis to define pathologic features of the disease.2,10 The following polyclonal and monoclonal antibodies were used: Glial fibrillary acidic protein (rabbit polyclonal; prediluted, Dako), NeuN1 (A60 mouse monoclonal; 1:100, Chemicon, Billerica, MA), CD8 (C8/144 B mouse monoclonal; prediluted, Dako), CXCR3 (mouse polyclonal; 1:50, BD Pharmingen, San Agustin de Guadalix, Madrid, Spain), and CXCL10 (goat polyclonal; RD Systems, Minneapolis, MN). Lymphoid tissue from tonsil served as positive control for CXCR3 and CD8; papillary carcinoma of thyroid for CXCL10 and normal brain tissue for GFAP and NeuN1.
T lymphocytes were enriched from murine splenocytes by using the Pan-T-cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) and were cultured in triplicate in flat-bottom 24-well plates at 1 106 cells/mL, in RPMI medium (Lonza) containing 10% fetal calf serum. They were stimulated with immobilized anti-CD3 (7 mg/mL) and soluble antiCD28 (1 mg/mL) antibodies (BD Pharmingen). Transwell cultures
Resting or activated astrocytes were cultured at the lower wells of 24-well Transwell plates with 3 mm pores (Costar, Cambridge, MA). Activated T lymphocytes were added to the upper chambers, with or without the CXCR3 antagonist TAK-799 (obtained from the National Institutes of Health AIDS Research and Reference Reagent Program Division of AIDS, NIAID, NIH: TAK-779, 500 nM). The numbers of T CD8 lymphocytes in the lower wells were counted using a flow cytometer (FACS Canto II, BD). All conditions were performed in triplicate.
Astrocyte isolation
Statistics
Neonatal C57BL/6 mice aged 1-2 days were used for the isolation of astrocytes. Primary astrocytes cultures were obtained from brain cortices by a modification of a published method.11 Oligodendrocytes and microglia were removed by shaking cultures at 200 rpm overnight and
Wilcoxon test (Mann-Whitney statistic) was applied to compare the different groups in experiments. Differences were considered statistically significant when P < 0.05. All graphics present the mean standard error.
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
Results Biopsies
We retrospectively studied brain biopsies from children with a diagnosis of Rasmussen encephalitis who underwent surgery. The Table shows the main characteristics of these patients. The expression of several molecules (see Materials and Methods) was assessed by immunohistochemistry. We found the characteristic features of lesions previously described in these patients: moderate chronic inflammation with pan laminar distribution (Supplemental Fig S1A), microglial nodules and perivascular cuffing by round cell (Supplemental Fig S1B) spreading to the white matter, accumulation of lymphocytes (Supplemental Fig S1C) with neuronophagia (Supplemental Fig S1D), decrease of neuronal population, activated microglia, reactive
453
astrocytosis, and cortical vacuolation. In addition, we found that the main lymphocyte population corresponded to CD8 T lymphocytes (Fig 1A), with few CD4 T lymphocytes (Supplemental Fig S2). Many of these T lymphocytes expressed the chemokine receptor CXCR3 (Fig 1A). We looked for the expression of CXCL10, one of the ligands for CXCR3. We found CXCL10 expressed in the same areas of lymphocyte accumulation (Fig 1B). In correlative sections of the brains, we found that mostly astrocytes, and in a lesser amount neurons, were the main cell populations expressing CXCL10 in the brains of children with Rasmussen encephalitis (Fig 1C). These findings suggest that the axis CXCR3CXCL10 may be involved in the recruitment of activated T lymphocytes into the areas of brain damage in these children. We next set up in vitro experiments to rule out this possibility.
TABLE. Patient’s characteristics
Patient Gender Clinical Events Before Epilepsy Onset
Age at Hemisphere Affected Epilepsy Epilepsy Course Onset
Neurological/Cognitive Deficit Before Surgery
Immunologic Treatmenty
Left hemiparesis Psychomotor arrest
Prednisone 3 yr, Gammaglobulin class I Moderate cognitive delay
Mild left hemiparesis Mild cognitive deterioration
4 yr, Prednisone Gammaglobulin class III, Exitus 3 mo after surgery (epileptic status) Prednisone 6 yr, Gammaglobulin Significant cognitive recovery. Language transference to the right hemisphere
1
Girl
Isolated “febrile seizure” at 16 mo of age
2 yr
2
Girl
No clinical events
3 yr
3
Boy
Pharyngoamygdalitis 5 yr 3 weeks before onset
4
Boy
Uveitis
8 yr
5
Girl
Mild cranial trauma 3 mo before onset
10 yr
6
Girl
No clinical events*
6 yr
Left hemisphere Rapid progression into severe epilepsy with polymorphic seizures, status epilepticus, Epilepsia Partialis Continua (EPC) Right hemisphere Rapid progression into severe epilepsy with polymorphic seizures, status epilepticus, EPC Left hemisphere Rapid progression into severe epilepsy with polymorphic seizures, status epilepticus, EPC Right hemisphere Focal motor seizures: > delayed progression into severe epilepsy with polymorphic seizures, EPC Right hemisphere Focal motor seizures: > delayed progression into severe epilepsy with polymorphic seizures, status epilepticus, EPC Right hemisphere Partial complex seizures: > delayed progression into severe epilepsy with polymorphic seizures, status epilepticus
Mild right hemiparesis, ataxia, Mild cognitive deterioration of left hemispheric cognitive functions Mild left hemiparesis, mild cognitive deterioration of right hemisphere cognitive functions
Hemispherectomy Age/ Outcome (Engel’s Class)/ Neuropsychological Status
Prednisone 13 yr, Gammaglobulin class I Tacrolimus Significant cognitive recovery
Mild left hemiparesis, Mild cognitive deterioration of right hemisphere cognitive functions
Prednisone 13 yr, Gammaglobulin class II Moderate cognitive recovery
Mild left body functional deficit, mild deterioration of right hemisphere cognitive functions
Prednisone 8 yr, Gammaglobulin class II Moderate cognitive recovery
Mother suffers from severe psoriasis. All patients were treated according to Bien CG, et al. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis: a European consensus statement. Brain. 2005;128:454-471. * y
454
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
FIGURE 1. T lymphocytes in primary lesions of children with Rasmussen encephalitis are mostly CD8 and express CXCR3, whereas neuron and astrocytes express CXCL10. (A) Immunohistochemistry with anti-CD8 (left) and anti-CXCR3 (right) monoclonal antibodies in two consecutive sections of the same brain area. The blood vessel serves as a reference. (B) Immunohistochemistry with anti-CXCL10 monoclonal antibody. (C) Immunohistochemistry with anti-NeuN1 (left), anti-CXCL10 (middle), and anti-glial fibrillary acidic protein (right) monoclonal antibodies in consecutive slices of the same brain area. The blood vessel serves as a reference. Most CXCL10 positive cells corresponded to astrocytes (glial fibrillary acidic protein positive), although some neurons (NeuN1 positive) also expressed CXCL10.
Stimulation of astrocytes
The etiology of Rasmussen encephalitis is unknown, but a viral cause has been previously suggested.3 Viruses may induce the production of CXCL10 by neurons in the West Nile Virus encephalitis model.14 We focused on astrocytes, to see whether this cell type could also produced CXCL10 after “viral-like” stimuli. Poly (I:C) is a synthetic dsRNA which mimics infections with either RNA or DNA viruses15 and activates TLR3, a receptor expressed by astrocytes. Astrocytes upregulated the expression of CXCL10 mRNA in a poly (I:C) dose-dependent fashion (Fig 2A). The release of CXCL10 into the supernatants was quantitated by ELISA and showed the production of the protein in parallel to the
upregulation of the messenger RNA (Fig 2B). These experiments showed that TLR3-stimulated astrocytes produced CXCL10, suggesting that a viral infection may induce the production of this chemokine by astrocytes, as it has been reported with neurons. Cocultures
We tested whether the production of CXCL10 by astrocytes had functional implications for T lymphocytes. Purified T lymphocytes were stimulated with immobilized anti-CD3 and soluble anti-CD28 antibodies for 96 hours. The maximal upregulation of CXCR3 was detected 48 hours after stimulation (not shown); therefore, the coculture
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
455
FIGURE 3. TLR3-activated astrocytes induced chemotaxis in activated T lymphocytes. Anti-CD3/anti-CD28-activated T lymphocytes with/without TAK779, a CXCR3 antagonist, were placed in the upper chamber of transwell cultures, with/without TLR3-stimulated astrocytes at the bottom wells. Chemotaxis was quantitated by flow cytometry (AU ¼ Arbitrary units), and values were normalized (AU ¼ 1) to chemotaxis induced by nonstimulated astrocytes. n ¼ 3.
FIGURE 2. Astrocytes produce CXCL10 upon TLR3 stimulation. (A) Quantitative realtime polymerase chain reaction results for expression of CXCL10 in stimulated astrocytes (black column) and control cells (white column). Astrocyte cultures were treated with medium alone and polycyticylic acid (poly [I:C]) 100 mM for 24 hours. Data are plotted as the relative expression of CXCL10 of a particular sample relative to the mean CXCL10 expression level of the nonstimulated astrocytes. Data shown are the average standard error of the mean (SEM) of three separate experiments. Levels are significantly major in poly (I:C)etreated astrocytes (asterisks indicate P < 0.01). (B) CXCL10 chemokine expression by poly (I:C)-treated astrocytes. Cell cultures were treated with medium alone or with 10 mM and 100 mM poly (I:C). Culture supernatants were collected at 24, 48, and 72 hours posttreatment and analyzed for CXCL10 expression by enzyme-linked immunosorbent assay. Data shown are the average SEM of three separate experiments. Levels are significantly major in poly (I:C)etreated astrocytes (asterisks P value <.01).
experiments were done after a 48-hour stimulation of T lymphocytes. Next we stimulated astrocytes with poly (I:C) on the lower wells of transwell culture plates, and seeded T lymphocytes (previously activated with anti-CD3 and antiCD28 antibodies) at the upper chambers. Chemotaxis of T CD8 lymphocytes was quantitated by flow cytometry. The results showed that activated T lymphocytes responded to the production of CXCL10 by poly (I:C)-stimulated astrocytes by increasing their migration in the transwell assay (Fig 3). These results showed that poly (I:C)stimulated astrocytes induced chemotaxis on activated T lymphocytes. Inhibition of chemotaxis by a CXCR3-antagonist small molecule
We set up transwell cultures as explained previously, with or without a CXCR3 antagonist molecule. The
chemotaxis of activated T lymphocytes induced by the TLR3-stimulated astrocytes was completely abrogated in the presence of the antagonist (Fig 3). These results suggest that the recruitment of activated T lymphocytes by stimulated astrocytes may be specifically inhibited by small drugs targeting CXCR3. Discussion
This study shows that the axis CXCR3-CXCL10 may have an active role in recruiting T lymphocytes into the brain parenchyma of children with Rasmussen encephalitis. Biopsies obtained from the damaged brains of children had clearly shown the destruction of neurons3,4 and astrocytes5 by cytotoxic T lymphocytes.4 We found similar CD8 T lymphocyte infiltration in the damaged areas of the brains we analyzed. Our patients represent the prototype of Rasmussen encephalitis, both clinically and pathologically. The finding of CD8 T cells in the brain biopsy specimens of individuals with Rasmussen encephalitis suggests, but does not prove, that these cells are responsible for the primary pathology of the disease. It has been shown that the lymphocytes found in the brains of patients with Rasmussen encephalitis represented the clonal expansion of cytotoxic T cells.7 Interestingly, the T-cell clones isolated from the brains had the same CDR3 genomic sequence of T-cell clones in peripheral blood. These T cells were detected in the patients along the course of the disease. All these findings have important implications, one of them being that the effector lymphocytes responsible for the brain destruction must migrate from the periphery to the brain parenchyma where they cause the damage. We found that CD8 lymphocytes at the sites of tissue destruction expressed CXCR3 on its membranes, and that the expression of CXCL10 in the brain corresponded to the same areas to those of lymphocyte infiltration. Both neurons and
456
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
astrocytes, mostly the latter, produced CXCL10 in the damaged areas. All these findings support the hypothesis that T lymphocytes were recruited through the CXCR3CXCL10 axis in these patients. CXCR3 and CXCL10 have been previously involved in the trafficking of T lymphocytes during encephalomyelitis,16 but not in cases of Rasmussen encephalitis. T lymphocytes upregulate the expression of CXCR3 upon activation,17,18 and CXCR3þ T lymphocytes are needed for an effective immune response at the CNS.19 Interestingly, CXCR3þ T lymphocytes have been involved in the regional antiviral response within the CNS,20,21 underscoring the role of this receptor in the chemotaxis of T lymphocytes to discrete areas of the CNS. CXCL10 is one of the known ligands for CXCR3; the sources of the chemokine at the brain are neurons and astrocytes. Neurons infected during viral encephalitis produce CXCL10.14 Although the etiology of Rasmussen encephalitis is unknown, a viral infection has been proposed. The histological similarities between Rasmussen encephalitis and viral encephalitis have been invoked, linking the disease to a virally triggered process. Several hypotheses have been proposed to explain autoimmune T cellemediated diseases induced by a viral infection.22 Target cells may share self structures with pathogen antigens (antigen mimicry hypothesis); cell destruction in the context of a viral infection may result in an epitope spreading to tissue-specific epitopes (epitope spreading hypothesis); or nonspecific inflammatory stimuli may cause the activation of bystander auto reactive T cells (bystander damage and fertile field hypothesis). An additional hypothesis, known as viral déjà vu,22 proposed that the autoimmunity phenomenon happened after successive antigenically related viral infections. Interestingly, the murine model described in the viral déjà vu theory of viral infection in the brain, is the only animal model related to Rasmussen encephalitis.22 Based on this hypothesis, we tested if astrocytes could produce CXCL10 after a stimulus resembling a viral infection. Astrocytes express TLR323 and produced CXCL10 after stimulating them with poly (I:C).13 In the absence of a known virus for Rasmussen encephalitis, we found that astrocytes upregulated the production of CXCL10 upon TLR3 activation. These results are in line with the work on Japanese encephalitis,24 a model of murine viral encephalitis in which astrocytes produced CXCL10 after infection. In our hands, the production of CXCL10 by astrocytes resulted in the recruitment of stimulated T lymphocytes, as shown in transwell assays. Moreover, we could inhibit the chemotaxis induced by TLR3-activated astrocytes with the use of small molecules that antagonize CXCR3 signaling.25,26 The potential of these molecules as therapeutic anti-CXCR3 drugs have been shown in vitro and in vivo in models of immune and neoplastic diseases, and have been already used in humans.27 In summary, we propose a scenario for Rasmussen encephalitis in which astrocytes and neurons recruit pathogenic T lymphocytes into areas of the brain through the CXCL10-CXCR3 axis. We show that this may be pharmacologically targeted with antagonists of CXCR3. Our findings may help in designing new therapeutic strategies based on immune modulation, specifically directed toward molecules involved in the pathogenesis of the disease.
We thank Dr. Fernando Neria for astrocyte isolation assistance. We gratefully acknowledge the helpful suggestions and expert advice in molecular biology techniques of the article by Dr. Miguel Ángel Rodríguez-Milla.
Supplementary data
Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.pediatrneurol.2013.07.019. References 1. Rasmussen T, Olszewski J, Lloyd-Smith DL. Focal seizures due to chronic localized encephalitis. Neurology. 1958;8:435-455. 2. Pardo CA, Vining EP, Guo L, Skolasky RL, Carson BS, Freeman JM. The pathology of Rasmussen syndrome: stages of cortical involvement and neuropathological studies in 45 hemispherectomies. Epilepsia. 2004;45:516-526. 3. Prayson RA, Frater JL. Rasmussen encephalitis: a clinicopathologic and immunohistochemical study of seven patients. Am J Clin Pathol. 2002;117:776-782. 4. Bien CG, Bauer J, Deckwerth TL, et al. Destruction of neurons by cytotoxic T cells: a new pathogenic mechanism in Rasmussen’s encephalitis. Ann Neurol. 2002;51:311-318. 5. Bauer J, Elger CE, Hans VH, et al. Astrocytes are a specific immunological target in Rasmussen’s encephalitis. Ann Neurol. 2007;62: 67-80. 6. Goverman J. Autoimmune T cell responses in the central nervous system. Nat Rev Immunol. 2009;9:393-407. 7. Schwab N, Bien CG, Waschbisch A, et al. CD8þ T-cell clones dominate brain infiltrates in Rasmussen encephalitis and persist in the periphery. Brain. 2009;132:1236-1246. 8. Wilson EH, Weninger W, Hunter CA. Trafficking of immune cells in the central nervous system. J Clin Invest. 2010;120:1368-1379. 9. Ransohoff RM, Kivisäkk P, Kidd G. Three or more routes for leukocyte migration into the central nervous system. Nat Rev Immunol. 2003;3:569-581. 10. Robitaille Y, Rasmussen T, Dubeau F, Tampieri D, Kemball K. Histopathology of nonneoplastic lesions in frontal lobe epilepsy. Review of 180 cases with recent MRI and PET correlations. Adv Neurol. 1992;57:499-513. 11. McCarthy KD, de Vellis J. Alpha-adrenergic receptor modulation of beta-adrenergic, adenosine and prostaglandin E1 increased adenosine 3’:5’-cyclic monophosphate levels in primary cultures of glia. J Cyclic Nucleotide Res. 1978;1:15-26. 12. Canellada A, Ramirez BG, Minami T, Redondo JM, Cano E. Calcium/ calcineurin signaling in primary cortical astrocyte cultures: Rcan1-4 and cyclooxygenase-2 as NFAT target genes. Glia. 2008;56:709-722. 13. Dussault AA, Pouliot M. Rapid and simple comparison of messenger RNA levels using real-time PCR. Biol Proced Online. 2006;8:1-10. 14. Klein RS, Lin E, Zhang B, et al. Neuronal CXCL10 directs CD8þ T-cell recruitment and control of West Nile virus encephalitis. J Virol. 2005;79:11457-11466. 15. Jacobs BL, Langland JO. When two strands are better than one: the mediators and modulators of the cellular responses to doublestranded RNA. Virology. 1996;219:339-349. 16. Müller M, Carter S, Hofer MJ, Campbell IL. Review: the chemokine receptor CXCR3 and its ligands CXCL9, CXCL10 and CXCL11 in neuroimmunityea tale of conflict and conundrum. Neuropathol Appl Neurobiol. 2010;36:368-387. 17. Bonecchi R, Bianchi G, Bordignon PP, et al. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J Exp Med. 1998;187:129-134. 18. Qin S, Rottman JB, Myers P, et al. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory reactions. J Clin Invest. 1998;101:746-754. 19. Christensen JE, Nansen A, Moos T, et al. Efficient T-cell surveillance of the CNS requires expression of the CXC chemokine receptor 3. J Neurosci. 2004;24:4849-4858. 20. Müller M, Carter SL, Hofer MJ, et al. CXCR3 signaling reduces the severity of experimental autoimmune encephalomyelitis by controlling the parenchymal distribution of effector and regulatory T cells in the central nervous system. J Immunol. 2007;179:2774-2786.
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457 21. Zhang B, Chan YK, Lu B, Diamond MS, Klein RS. CXCR3 mediates regionspecific antiviral T cell trafficking within the central nervous system during West Nile virus encephalitis. J Immunol. 2008;180:2641-2649. 22. Merkler D, Horvath E, Bruck W, Zinkernagel RM, Del la Torre JC, Pinschewer DD. “Viral déjà vu” elicits organ-specific immune disease independent of reactivity to self. J Clin Invest. 2006;116:1254-1263. 23. Farina C, Krumbholz M, Giese T, Hartmann G, Aloisi F, Meinl E. Preferential expression and function of Toll-like receptor 3 in human astrocytes. J Neuroimmunol. 2005;159:12-19. 24. Bhowmick S, Duseja R, Das S, Appaiahgiri MB, Vrati S, Basu A. Induction of IP-10 (CXCL10) in astrocytes following Japanese encephalitis. Neurosci Lett. 2007;414:45-50.
457
25. Johnson M, Li AR, Liu J, et al. Discovery and optimization of a series of quinazolinone-derived antagonists of CXCR3. Bioorg Med Chem Lett. 2007;17:3339-3343. 26. Baba M, Nishimura O, Kanzaki N, et al. A small-molecule, nonpeptide CCR5 antagonist with highly potent and selective anti-HIV1 activity. Proc Natl Acad Sci U S A. 1999;96:5698-5703. 27. Tonn GR, Wong SG, Wong SC, et al. An inhibitory metabolite leads to dose- and time-dependent pharmacokinetics of (R)-N-{1[3-(4-ethoxy-phenyl)-4-oxo-3,4-dihydro-pyrido[2,3-d]pyrimidin-2-yl] -ethyl}-N-pyridin-3-yl-methyl-2-(4-trifluoromethoxy-phenyl)acetamide (AMG 487) in human subjects after multiple dosing. Drug Metab Dispos. 2009;37:502-513.
457.e1
I. Mirones et al. / Pediatric Neurology 49 (2013) 451e457
SUPPLEMENTAL FIGURE S1. Histological features of the biopsies of patients with Rasmussen encephalitis. The figure shows several characteristic findings in the hematoxylin-eosin examination of the brain biopsies. (A) Low magnification showing the areas of moderate chronic inflammation with pan laminar distribution and microglial nodules. (B) Perivascular accumulation of mononuclear cells. (C) Detailed view of an area with lymphocyte infiltration in close apposition to neurons. (D) Accumulation of lymphocytes with neuronophagia.
SUPPLEMENTAL FIGURE S2. CD4 T lymphocytes in brain biopsies of patients with Rasmussen encephalitis. Low magnification picture of a brain biopsy showing accumulation of CD4 cells in the parenchyma.