Matrix Metalloproteinases and Their Tissue Inhibitors in Cerebrospinal Fluid and Serum of Patients with Japanese Encephalitis

Objective: Japanese Encephalitis (JE) is one kind of central nervoussystem’s acute infectious disease, which is caused by Japanese encephalitisvirus (JEV) infection of parenchymal. JE is one of the most importantendemics in the world, especially in East and Southeast Asia. Clinically, thepatients have symptoms like high fever, headache, seizures and consciousdisturbance. Their are estimated50,000patients, with15,000deaths annually.Despite its high incidence and the present situation that there is no specifictreatment, little is known about the specifically pathogenesis of humaninfection with JEV. Studies have shown that JEV infection may damagecentral nervous system astrocytes, neurons and the protective role of the bloodbrain barrier (BBB). JEV induces microglial activation in the brain thatincreases the expression of many immunologically relevant proteins such ascomplement factors, cytokines (IL-1,IL-6, TNF-α, IL-18), vascularendothelial growth factor, chemokines (MCP-1, MIP-1α, MIP-1β, RANTES),lymphotoxin, and matrix metalloproteinases (MMPs), all of which lead to thedamage of integrity of the BBB. MMPs involve in remodeling theextracellular matrix, the imbalance between MMPs and their tissue inhibitions(TIMPs) connects with many diseases. In this research, we measured thelevels of MMPs and TIMPs in serum and cerebrospinal of patients with JE tounderstand the value in pathogenesis, illness monitoring and prognosticevaluation, provide new targets for the treatment of JE.Methods: We collected24patients who were admitted at the NeurologyDepartment of the Second Affiliated Hospital of Hebei Medical Universitybetween August and November2013. The serum of the patients were shippedto the HeBei Provincial Center for Disease Control and Prevention, the JE specific IgM antibodies were all positive.10non-central system infectioncases were selected, the CSF of them were collected as CSF disease control(DC). Since doing a lumbar puncture in a normal person was consideredinappropriate, we did not set up a CSF healthy control (HC) group underethical considerations. Another10patients whose JE IgM antibodies werenegative were collected, their serum samples were the serum DC and theserum samples of10healthy people were the serum HC. Patients were dividedinto two groups of ordinary type and heavy type according to clinicalmanifestation. All CSF samples were tested CSF routine and biochemistry,and detected the levels of MMP-9and TIMP-1in the CSF and serum byELISA.Results:1Clinical dataJE patients present acute onset of fever,headache, seizures, consciousdisturbance, dystonia and meningeal irritation. All of the JE patients had feverand disturbance of conscious (100%),12patients presented headache (50%),seizures was manifested in8patients (33.33%), dystonia was observed in13patients (54.17%) and meningeal irritation was occured in21patients (87.5%).10non-central system infection cases were selected, the CSF of them werecollected as CSF DC. Another10patients whose JE IgM antibodies werenegative were collected, their serum samples were the serum DC and theserum samples of10healthy people were the serum HC.2CSF routine, biochemistry resultsCompared with CSF DC, JE patients had higher CSF pressure, whiteblood cells and protein content, the differences were statistically significant(P<0.01), there were no significant differences in the levels of glucose andchloride. JE recovery phase compared with acute phase, white blood cells andprotein content were lower， the content of glucose was increased (P<0.05),but no significant differences were observed in CSF pressure and the contentof chloride between them (P>0.05).3The levels of MMP-9in CSF and serum The content determination of MMP-9in CSF: the level range of MMP-9in JE cerebrospinal fluid was9.3～72.9ng/ml, average content was37.06±3.83ng/ml, the level range of MMP-9in CSF disease control was5.3～34.3ng/ml, average content was17.73±3.67ng/ml, the differences weresignificance in statistics between the two groups (P<0.01). The CSFconcentration of MMP-9in convalescence was4.7～92.2ng/ml, averagecontent was46.81±10.09ng/ml, no significant difference was observed in thelevels of MMP-9between acute phase and recovery phase. Patients weredivided into two groups of ordinary type and heavy type according to clinicalmanifestation. One result in ordinary type was undetected, the level range ofMMP-9in other8patients was9.3～58.1ng/ml, average content was31.56±6.83ng/ml, the MMP-9were measured as39.99±4.60ng/ml in heavygroups, the levels of each group showed no significant difference（P>0.05).The content determination of MMP-9in serum: the serum MMP-9levelsof JE, serum DC and serum HC were measured as591.77±41.91ng/ml,424.61±32.33ng/ml，331.64±41.92ng/ml respectively, there were significantdifference of each group (P<0.01). The serum MMP-9concentrations ofordinary type and heavy group were detected as678.97±57.10ng/ml and539.45±54.57ng/ml, there was no statistically significant difference betweenthe groups （P>0.05).4The levels of TIMP-1in CSF and serumThe content determination of TIMP-1in CSF: the results of2CSFsamples were undetected, the cerebrospinal fluid TIMP-1levels of other22JEpatients were71.00±4.58ng/ml, which was significantly higher than the levelsof the DC (15.22±2.89ng/ml, P<0.01). The cerebrospinal fluid TIMP-1levelsin ordinary type and heavy group were measured as60.03±3.19ng/ml and78.60±6.76ng/ml, there were significant differences between the groups（P<0.05). In the recovery phase, the CSF TIMP-1concentration range was26.87～103.74ng/ml, average value was48.44±7.58ng/ml，which was lowerthan the concentration of acute phase (P<0.05).The content determination of TIMP-1in serum: the serum TIMP-1levels of JE, serum DC and serum HC were detected as55.41±1.79ng/ml,56.19±2.76ng/ml and54.73±6.82ng/ml, the difference was insignificantamong groups. The serum TIMP-1levels of ordinary type and heavy groupwere54.58±2.11ng/ml and55.90±2.62ng/ml, the levels of each group showedno significant difference（P>0.05).5Correlations between levels of MMP-9and TIMP-1in CSF and serumNo significant correlation was found between CSF levels of MMP-9andTIMP-1in JE patients. Similarly, no significant correlation was found betweenserum levels of MMP-9and TIMP-1. There were no significant correlationbetween the CSF levels of MMP-9and TIMP-1and the serum levels of thesebiomarkers. No significant correlations were observed between serum IgMantibody titer and serum concentrations of MMP-9or TIMP-1in JE.6Correlations between levels of MMP-9and TIMP-1in CSF and thedamage of blood-brain barrier (BBB).We calculated the albumin ratio [CSF albumin (mg/l)/serum albumin(g/l)]and used it as an indicator of BBB damage. No significant correlationwas found between CSF levels of MMP-9and TIMP-1and the disruption ofBBB.Conclusion:1The cerebrospinal MMP-9levels of JE were significantly higher thanDC. The serum MMP-9levels of JE was higher than serum DC and HC,however, there was no significant difference in the levels of cerebrospinalfluid between DC and HC.No significant difference was observed in the levelsof MMP-9between acute phase and recovery phase. Patients were dividedinto two groups of ordinary type and heavy type according to clinicalmanifestation, no significant differences were observed in the CSF and serumMMP-9levels between the two groups.2The cerebrospinal fluid TIMP-1levels of JE was significantly higherthan the levels of the DC, the difference was statistically significant. The CSFTIMP levels of heavy group were higher than that of ordinary group, whichshowed that detecting the levels of CSF TIMP-1has important significance for judging the severity. However, no significant differences were found in theserum TIMP-1concentrations between the two groups. The CSF concentrationof TIMP-1in convalescence was lower than that in acute phase, there was astatistically significant difference, this showed that dynamic viewing the levelsof the TIMP-1in CSF was important for the illness monitoring and prognosisevaluation of JE patients.There was no statistically significant difference in theserum TIMP concentration among the JE, DC and HC3No significant correlation was found between CSF levels of MMP-9and TIMP-1in JE patients.Similarly, no significant correlation was foundbetween serum levels of MMP-9and TIMP-1. There were no significantcorrelation between the CSF levels of MMP-9and TIMP-1and the serumlevels of these biomarkers. No significant correlations were observed betweenserum IgM antibody titer and serum concentrations of MMP-9and TIMP-1inJE.4No significant correlation was found between CSF levels of MMP-9and TIMP-1and the disruption of BBB. More cases are needed to furtherprospectively investigated.