Matrix metalloproteinases (MMPs) constitute a family of zinc-binding proteinases that degrade the extracellular matrix proteins.1 MMP-9 is mainly produced by inflammatory cells such as T cells and macrophages, and is capable of degrading collagen IV, a major component of the basement membrane of the cerebral epithelium and responsible for the integrity of the blood–brain barrier.2 The activity of MMPs is further controlled by specific tissue inhibitors of metalloproteinases (TIMPs), and TIMP-1 has a high avidity for MMP-9.3
Influenza-associated encephalopathy is a severe neurologic complication of influenza manifesting seizures and progressive coma with a high-grade fever.4 The pathogenesis of influenza-associated encephalopathy remains unclear. We have reported that several proinflammatory cytokines are elevated in serum and cerebrospinal fluid in patients with influenza-associated encephalopathy and are related to the clinical severity of the disease.5–8 Brain computed tomography abnormalities, such as brain edema and low densities in localized areas, are associated with a poor prognosis in these patients.4 A postmortem examination of 1 fatal case of influenza-associated encephalopathy revealed vasogenic brain edema, suggesting that the vascular endothelial cells were impaired.9 It is possible that the blood–brain barrier is damaged in influenza-associated encephalopathy.
To investigate the role of MMP-9 and TIMP-1 in the pathogenesis of influenza-associated encephalopathy, we determined the relationship between serum concentrations of MMP-9 and TIMP-1 in patients with influenza virus infection with or without neurologic complications.
SUBJECTS AND METHODS
Informed consent was obtained from the parents of the patients and controls enrolled in this study. Serum samples were obtained from influenza-associated encephalopathy patients on admission to Yamaguchi University Hospital and 11 research cooperation hospitals in Japan, from December 1999 to May 2005, and were obtained from the patients with influenza-associated febrile seizures, uncomplicated influenza, Epstein–Barr virus (EBV) infection and respiratory syncytial virus (RSV) infection on admission to Yamaguchi University Hospital from March 2003 to January 2006. The specimens were stored frozen at −80°C until assay.
Influenza-Associated Encephalopathy, Influenza-Associated Febrile Seizures and Uncomplicated Influenza.
The criteria for the diagnosis and the day of onset of influenza-associated encephalopathy, influenza-associated febrile seizures and uncomplicated influenza were described in our previous article.8 Thirty-five patients were enrolled with influenza-associated encephalopathy (22 males and 13 females, aged from 6 months to 19 years: mean, 5.8 years). These patients were divided into 2 groups, ie, those who died (n = 7 of 17) or developed neurologic sequelae (n = 10 of 17), and those who had no sequelae of follow-up (n = 18). Serum samples were obtained from these patients on day 1.0 ± 0.5 (range, 1–3) of the illness.
Thirteen patients were enrolled with influenza-associated febrile seizures (8 males and 5 females, aged from 2 to 10 years: mean, 5.4 years). Serum samples were obtained on day 1.0 ± 0.3 (range, 1–2) of the illness.
Forty-six patients were enrolled with uncomplicated influenza (26 males and 20 females, aged from 11 months to 10 years: mean, 5.2 years), and serum were obtained on days 2.0 ± 1.0 (range, 1–4) of the illness.
Control subjects comprised 8 patients who had EBV infection (3 males and 5 females, aged from 2 to 12 years: mean, 5.8 years) and 37 patients who had RSV infection (21 males and 16 females, aged from 1 to 19 months: mean, 0.7 years). Serum were obtained from the patients with EBV infection on days 7.0 ± 3.5 (range, 2–13) and those with RSV infection on days 4.0 ± 2.5 (range, 2–7) of the illness. Thirty-three healthy children (15 males and 18 females, aged from 2 to 15 years: mean, 6.4 years) also served as controls.
Determination of MMP-9 and TIMP-1 Concentrations.
The serum concentrations of MMP-9 and TIMP-1 were determined with sandwich-type ELISA kits (Amersham, Buckinghamshire, UK). Assays were performed following the manufacturer's instructions. The detection limits were 2.5 ng/mL for MMP-9 and 2.4 ng/mL for TIMP-1.
All values are medians ± SD. The differences in the results between groups were analyzed using the Mann–Whitney U test.
Serum MMP-9 concentrations in influenza-associated encephalopathy with poor prognosis, influenza-associated febrile seizures and uncomplicated influenza were significantly higher than the healthy controls (P < 0.0001, P = 0.0005, P < 0.0001, respectively) (Fig. 1). The serum MMP-9 concentrations in influenza-associated encephalopathy with poor prognosis were significantly higher than those for uncomplicated influenza, and the healthy controls (P = 0.0149, and P < 0.0001, respectively).
Serum TIMP-1 concentrations in influenza-associated encephalopathy with poor prognosis were significantly higher than those in influenza-associated encephalopathy without sequelae, influenza-associated febrile seizures and the healthy controls (P = 0.0087, 0.018 and 0.0062, respectively). The serum TIMP-1 concentrations in influenza-associated encephalopathy without sequelae and influenza-associated febrile seizures were significantly lower than those in uncomplicated influenza (P < 0.0001 and <0.0001, respectively). The MMP-9/TIMP-1 ratios in influenza-associated encephalopathy with poor prognosis, influenza-associated encephalopathy without sequelae and influenza-associated febrile seizures were significantly higher than those for uncomplicated influenza (P = 0.0099, 0.0079 and 0.0053, respectively) and the healthy controls (P = 0.0134, 0.012 and 0.0093, respectively).
The serum MMP-9 concentrations and MMP-9/TITMP-1 ratios for uncomplicated influenza were significantly higher than those in EBV infection (P = 0.0006 and P < 0.0001, respectively) and RSV infection (P = 0.0002 and P < 0.0001, respectively), and serum TIMP-1 concentrations were significantly lower than those in EBV infection (P = 0.0001) and RSV infection (P = 0.0017).
Results from our present study suggest that MMP-9 and TIMP-1 are related to the pathogenesis of influenza-associated encephalopathy on the basis of the elevated serum MMP-9 and MMP-9/TIMP-1 ratios, and associated with the clinical severity of influenza-associated encephalopathy. The serum MMP-9 levels and MMP-9/TIMP-1 ratios for uncomplicated influenza were significantly higher than those in EBV and RSV infections, and the serum TIMP-1 concentrations for uncomplicated influenza were significantly lower. These findings suggest that TIMP-1 was not sufficiently produced in influenza virus infection compared with RSV and EBV infections, though MMP-9 was abundantly produced in influenza virus infection. However, EBV and RSV infections may be inappropriate disease controls. EBV infection affects lymphocytes, but not airway epithelial cells, and RSV infects young infants. Another hypothesis is that TIMP-1 is overly consumed in influenza virus infection and, as a result, the serum TIMP-1 values are not elevated. In vitro, influenza virus produced MMP-9, but had little effect on TIMP-1, in cultured cells.10 Moreover, the serum TIMP-1 levels in influenza-associated encephalopathy without sequelae and influenza-associated febrile seizures were significantly lower than those for uncomplicated influenza. One hypothesis is that patients with neurologic complications of influenza could not produce a sufficient amount of TIMP-1. The serum TIMP-1 concentrations in influenza-associated encephalopathy with poor prognosis were not lower than those for the Flu. Interleukin-10 (IL-10) is markedly elevated in the serum of influenza-associated encephalopathy with poor prognosis.7,8 IL-10 stimulates TIMP-1 production.3 We hypothesize that the influenza-associated encephalopathy patients with poor prognoses could not produce a sufficient amount of TIMP-1, but a large amount of IL-10 strongly stimulated TIMP-1 production.
In summary, an imbalance between MMP-9 and TIMP-1 occurred in influenza virus infection, especially with neurologic complications. The imbalance may damage the blood–brain barrier and promote febrile seizure or encephalopathy.
1. Chandler S, Miller KM, Clements JM, et al. Matrix metalloproteinases, tumor necrosis factor and multiple sclerosis: an overview. J Neuroimmunol
2. Lukes A, Mun-Bryce S, Lukes M, Rosenberg GA. Extracellular matrix degradation by metalloproteinases and central nervous system diseases. Mol Neurobiol
3. Lacraz S, Nicod LP, Chicheportiche R, Welgus HG, Dayer JM. IL-10 inhibits metalloproteinase and stimulates TIMP-1
production in human mononuclear phagocytes. J Clin Invest
4. Morishima T, Togashi T, Yokota S, et al. Encephalitis and encephalopathy associated with an influenza epidemic in Japan. Clin Infect Dis
5. Ichiyama T, Nishikawa M, Yoshitomi T, Hayashi T, Furukawa S. Tumor necrosis factor-α, interleukin-1β and interleukin-6 in cerebrospinal fluid from children with prolonged febrile seizures. Comparison with acute encephalitis/encephalopathy. Neurology
6. Ichiyama T, Isumi H, Ozawa H, Matsubara T, Morishima T, Furukawa S. Cerebrospinal fluid and serum levels of cytokines and soluble tumor necrosis factor receptor in influenza virus-associated encephalopathy. Scand J Infect Dis
7. Ichiyama T, Endo S, Kaneko M, Isumi H, Matsubara T, Furukawa S. Serum cytokine concentrations of influenza-associated acute necrotizing encephalopathy. Pediatr Int
8. Ichiyama T, Morishima T, Isumi H, Matsufuji H, Matsubara T, Furukawa S. Analysis of cytokine levels and NF-κB activation in peripheral blood mononuclear cells in influenza virus-associated encephalopathy. Cytokine
9. Togashi T, Matsuzono Y, Narita M, Morishima T. Influenza-associated acute encephalopathy in Japanese children in 1994–2002. Virus Res
10. Yeo SJ, Kim SJ, Kim JH, Lee HJ, Kook YH. Influenza A virus infection modulates the expression of type IV collagenase in epithelial cells. Arch Virol