Abstract：
BACKGROUND: The presence of Blood-Brain Barrier (BBB) dysfunction is defined by albumin quotient (QALB) and characterize a group of Multiple Sclerosis (MS) patients at clinical onset. We evaluated the concentration in cerebrospinal fluid (CSF) of 87 cytokines, to better characterize the CSF inflammatory pattern in presence of BBB damage.
METHODS: In an exploratory cohort, CSF cytokines were evaluated by means of Multiplex technology (Bio-Plex Pro-Human Cytokine, GF and Diabetes 27-Plex Panel, Bio-Plex Pro-Human Chemokines 40-Plex Panel, Bio-Plex Pro-Human Inflammation Assays 37-Plex Panel) in a cohort of Other Not Inflammatory Neurological Disorders (ONIND) and in cohort of patients with MS, stratified according to BBB damage into QALB+ and QALB- MS patients. In the validation cohort, we evaluated the relevant molecules in a cohort of MS patients, stratified again into QALB+ and QALB-, including also Neurofilament Light (NfL) and Chitinase 3-like 1 (CHI3L1) CSF concentration.
RESULTS: While MIP-1α, CXCL-13, and CCL-22 CSF concentrations were higher in both MS groups compared to ONIND, in QALB+ MS CSF concentrations of CXCL-9 (17.85 ± 4.69 pg/mL), CXCL-10 (476.5 ± 324.3 pg/mL), and IL-16 (96.08 ± 86.17 pg/mL) were higher than in QALB- MS (8.98 ± 5368 pg/mL, p < 0.005, 281.0 ± 180.9 pg/mL, p < 0.05, and 47.35 ± 36.87 pg/mL, p < 0.005, respectively) and ONIND (8.98 ± 5368 pg/mlL, p < 0.005, 281.0 ± 180.9 pg/mL, p < 0.005, and 47.35 ± 36.87 pg/mL, p < 0.001, respectively). A strong correlation was observed between CXCL-9 and CXCL-10 in all MS groups (all r>0.75, all p < 0.001). In the validation cohort again CXCL-10 CSF concentration were higher in QALB+ MS than in QALB- MS (94.25 ± 64.75 vs 153.8 ± 99.52, p < 0.05), while no difference was observed in serum. CSF NfL (1642 ± 1963 vs 3231 ± 3492 pg/mL, p < 0.05) and CHI3L1 (183.9 ± 86.62 vs 262 ± 137.5 ng/mL, p < 0.05) were increased in QALB+ MS.
CONCLUSIONS: BBB damage in MS is linked to a specific CSF cytokines pattern (CXCL-9, CXCL-10, IL-16), that are also involved in astrocyte-microglia interaction. To what extent their continuous production in the CNS may mark a more severe disease course merits to be investigated.
METHODS: In an exploratory cohort, CSF cytokines were evaluated by means of Multiplex technology (Bio-Plex Pro-Human Cytokine, GF and Diabetes 27-Plex Panel, Bio-Plex Pro-Human Chemokines 40-Plex Panel, Bio-Plex Pro-Human Inflammation Assays 37-Plex Panel) in a cohort of Other Not Inflammatory Neurological Disorders (ONIND) and in cohort of patients with MS, stratified according to BBB damage into QALB+ and QALB- MS patients. In the validation cohort, we evaluated the relevant molecules in a cohort of MS patients, stratified again into QALB+ and QALB-, including also Neurofilament Light (NfL) and Chitinase 3-like 1 (CHI3L1) CSF concentration.
RESULTS: While MIP-1α, CXCL-13, and CCL-22 CSF concentrations were higher in both MS groups compared to ONIND, in QALB+ MS CSF concentrations of CXCL-9 (17.85 ± 4.69 pg/mL), CXCL-10 (476.5 ± 324.3 pg/mL), and IL-16 (96.08 ± 86.17 pg/mL) were higher than in QALB- MS (8.98 ± 5368 pg/mL, p < 0.005, 281.0 ± 180.9 pg/mL, p < 0.05, and 47.35 ± 36.87 pg/mL, p < 0.005, respectively) and ONIND (8.98 ± 5368 pg/mlL, p < 0.005, 281.0 ± 180.9 pg/mL, p < 0.005, and 47.35 ± 36.87 pg/mL, p < 0.001, respectively). A strong correlation was observed between CXCL-9 and CXCL-10 in all MS groups (all r>0.75, all p < 0.001). In the validation cohort again CXCL-10 CSF concentration were higher in QALB+ MS than in QALB- MS (94.25 ± 64.75 vs 153.8 ± 99.52, p < 0.05), while no difference was observed in serum. CSF NfL (1642 ± 1963 vs 3231 ± 3492 pg/mL, p < 0.05) and CHI3L1 (183.9 ± 86.62 vs 262 ± 137.5 ng/mL, p < 0.05) were increased in QALB+ MS.
CONCLUSIONS: BBB damage in MS is linked to a specific CSF cytokines pattern (CXCL-9, CXCL-10, IL-16), that are also involved in astrocyte-microglia interaction. To what extent their continuous production in the CNS may mark a more severe disease course merits to be investigated.
摘要：
背景：血脑屏障（BBB）功能障碍的存在由白蛋白商（QALB）定义，并表征一组临床发作的多发性硬化（MS）患者。我们评估了87种细胞因子在脑脊液（CSF）中的浓度，以更好地表征存在BBB损伤的CSF炎症模式。
方法：在探索性队列中，CSF细胞因子通过多重技术（Bio-PlexPro-Human细胞因子，GF和糖尿病27-Plex小组，Bio-PlexPro-Human趋化因子40-Plex面板，Bio-PlexPro-人类炎症分析37-Plex小组）在其他非炎症性神经疾病（ONIND）和MS患者队列中，根据BBB损害分层为QALB+和QALB-MS患者。在验证队列中，我们评估了一组MS患者的相关分子，再次分层为QALB+和QALB-，还包括神经丝光(NfL)和几丁质酶3样1(CHI3L1)CSF浓度。
结果：而MIP-1α，与ONIND相比，两个MS组的CXCL-13和CCL-22CSF浓度均较高，在QALB+MSCSF中CXCL-9的浓度（17.85±4.69pg/mL），CXCL-10（476.5±324.3pg/mL），和IL-16（96.08±86.17pg/mL）高于QALB-MS（8.98±5368pg/mL，p<0.005,281.0±180.9pg/mL,p<0.05，和47.35±36.87pg/mL，分别为p<0.005)和ONIND(8.98±5368pg/mlL，p<0.005,281.0±180.9pg/mL,p<0.005，和47.35±36.87pg/mL，p分别<0.001)。在所有MS组中观察到CXCL-9和CXCL-10之间的强相关性(所有r>0.75,所有p<0.001)。在验证队列中，QALB+MS的CXCL-10CSF浓度高于QALB-MS（94.25±64.75vs153.8±99.52，p<0.05），而血清没有观察到差异。CSFNfL（1642±1963vs3231±3492pg/mL，p<0.05）和CHI3L1（183.9±86.62vs262±137.5ng/mL，p<0.05)在QALB+MS中增加。
结论：MS中的BBB损伤与特定的CSF细胞因子模式（CXCL-9，CXCL-10，IL-16）有关，也参与星形胶质细胞-小胶质细胞的相互作用。在多大程度上，它们在CNS中的连续产生可能标志着更严重的疾病进程值得研究。
方法：在探索性队列中，CSF细胞因子通过多重技术（Bio-PlexPro-Human细胞因子，GF和糖尿病27-Plex小组，Bio-PlexPro-Human趋化因子40-Plex面板，Bio-PlexPro-人类炎症分析37-Plex小组）在其他非炎症性神经疾病（ONIND）和MS患者队列中，根据BBB损害分层为QALB+和QALB-MS患者。在验证队列中，我们评估了一组MS患者的相关分子，再次分层为QALB+和QALB-，还包括神经丝光(NfL)和几丁质酶3样1(CHI3L1)CSF浓度。
结果：而MIP-1α，与ONIND相比，两个MS组的CXCL-13和CCL-22CSF浓度均较高，在QALB+MSCSF中CXCL-9的浓度（17.85±4.69pg/mL），CXCL-10（476.5±324.3pg/mL），和IL-16（96.08±86.17pg/mL）高于QALB-MS（8.98±5368pg/mL，p<0.005,281.0±180.9pg/mL,p<0.05，和47.35±36.87pg/mL，分别为p<0.005)和ONIND(8.98±5368pg/mlL，p<0.005,281.0±180.9pg/mL,p<0.005，和47.35±36.87pg/mL，p分别<0.001)。在所有MS组中观察到CXCL-9和CXCL-10之间的强相关性(所有r>0.75,所有p<0.001)。在验证队列中，QALB+MS的CXCL-10CSF浓度高于QALB-MS（94.25±64.75vs153.8±99.52，p<0.05），而血清没有观察到差异。CSFNfL（1642±1963vs3231±3492pg/mL，p<0.05）和CHI3L1（183.9±86.62vs262±137.5ng/mL，p<0.05)在QALB+MS中增加。
结论：MS中的BBB损伤与特定的CSF细胞因子模式（CXCL-9，CXCL-10，IL-16）有关，也参与星形胶质细胞-小胶质细胞的相互作用。在多大程度上，它们在CNS中的连续产生可能标志着更严重的疾病进程值得研究。
