Early Human Development 86 (2010) 251–254
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Increased expression of matrix metalloproteinase-9 and hepatocyte growth factor in the cerebrospinal fluid of infants with posthemorrhagic hydrocephalus Toshio Okamoto a,⁎, Satoru Takahashi a, Eiki Nakamura a, Ken Nagaya a, Tokitsugi Hayashi a, Masaru Shirai b, Kenji Fujieda a a b
Department of Pediatrics, Asahikawa Medical College, Hokkaido, Japan Department of Pediatrics, Asahikawa Kosei Hospital, Hokkaido, Japan
a r t i c l e
i n f o
Article history: Received 7 December 2009 Received in revised form 2 March 2010 Accepted 25 March 2010 Keywords: Matrix metalloproteinase-9 Hepatocyte growth factor Intraventricular hemorrhage Posthemorrhagic hydrocephalus Extremely low-birth-weight infants
a b s t r a c t Background: In approximately 60% of infants with posthemorrhagic hydrocephalus (PHH), ventricular dilation resolves by unknown intrinsic mechanisms, without the need for a shunt operation. A pathological hallmark of PHH is extensive deposition of extracellular matrix (ECM) proteins in the subarachnoid space. Our previous study revealed that matrix metalloproteinase (MMP)-9, which degrades ECM proteins, may play an important role in the resolution of ventricular dilation. MMP-9 is known to be induced by hepatocyte growth factor (HGF) in various cell lines. Aims: The aim of this study is to confirm our earlier finding that MMP-9 contributes to the resolution of PHH, and to investigate whether HGF also contributes to this process. Study design: Cerebrospinal fluid (CSF) samples were collected from 13 infants who developed ventricular dilation after intraventricular hemorrhage (IVH). Of these infants, 9 exhibited resolution of ventricular dilation without shunt operation; however, 4 infants had to be treated with shunt operation. The CSF levels of MMP-9 and HGF were measured using an enzyme immunoassay. Results: Significantly higher CSF levels of MMP-9 and HGF were detected in patients in whom the ventricular dilation resolved without shunt operation than in those with progressive ventricular dilation (MMP-9: median, 128 ng/ml; range, 47–900 ng/ml vs median, 50 ng/ml; range, 12–110 ng/ml; p b 0.05; HGF: median, 2.42 ng/ml; range, 0.81–7.04 ng/ml vs median, 1.42 ng/ml; range, 0.67–3.87 ng/ml; p b 0.05). Conclusions: Our results indicate that MMP-9 and HGF may participate in the resolution of ventricular dilation following IVH. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Posthemorrhagic hydrocephalus (PHH) is the most serious complication associated with intraventricular hemorrhage (IVH) in extremely low-birth-weight infants (ELBWIs). A shunt operation is the only definitive therapy for PHH; however, this operation usually cannot be performed at an early stage because of the small size and instability of the patients. Therefore, alternative approaches are required to treat infants with PHH. Approximately 25% infants with a large IVH develop slowly progressive ventricular dilation [1]. Of these, approximately 40% require a shunt operation; however, in the remaining 60%, the ventricular dilation resolves without a shunt operation [1]. However, the mechanisms underlying this phenomenon remain unknown. Thus, elucidating this intrinsic mechanism may contribute to the development of a novel treatment strategy for PHH. ⁎ Corresponding author. Department of Pediatrics, Asahikawa Medical College, 2-11-1 Midorigaoka-Higashi, Asahikawa, Hokkaido 078-8510, Japan. Tel.: + 81 166 68 2481; fax: + 81 166 68 2489. E-mail address:
[email protected] (T. Okamoto). 0378-3782/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2010.03.007
Posthemorrhagic ventricular dilation is thought to develop owing to the formation of multiple small blood clots throughout the central nervous system (CNS). These clots occlude channels that participate in the circulation and absorption of cerebrospinal fluid (CSF) [2]. Furthermore, transforming growth factor (TGF)-β1, a cytokine that is released into the CSF after IVH, mediates the obstruction of CSF circulation [3,4]. This cytokine stimulates the expression of extracellular matrix (ECM) protein components [5,6]. TGF-β1-mediated deposition of the ECM proteins may occur either in the outlet of the fourth ventricle or in the channels that participate in CSF absorption [4]. Thus, the abovementioned factors contribute to the obstruction of CSF circulation, leading to progressive ventricular dilation. Matrix metalloproteinases (MMPs) mediate the degradation of the deposited ECM proteins [7]. MMP-9 degrades fibronectin and collagen—ECM components that are generated through the TGF-β1 signaling pathway [7]. We previously measured MMP-9 activity in the CSF of infants with PHH and found it to be higher in patients who did not require a shunt operation than in those who underwent this operation [8]. Thus, our previous data indicate that MMP-9 might contribute to the resolution of progressive ventricular dilation after
252
T. Okamoto et al. / Early Human Development 86 (2010) 251–254
IVH. Interestingly, hepatocyte growth factor (HGF) enhances ECM catabolism by increasing the expression of MMP-9 [9,10]. HGF is expected to be a promising candidate for the treatment of various fibrotic diseases, including chronic renal disease or pulmonary fibrosis [11,12]. Furthermore, a recent study showed that intraventricular administration of human recombinant HGF in mice with TGF-β1induced communicating hydrocephalus reduced the severity of collagen fibrosis in the meninges and normalized the CSF flow [13]. Therefore, HGF may participate in the resolution of PHH by enhancing the production of MMP-9. In the present study, we measured MMP-9 and HGF levels in the CSF of infants with or without PHH and found that both levels increased in the former infants. Most importantly, these levels were higher in patients who exhibited resolution of ventricular dilation than in those who underwent a shunt operation. The CSF levels of MMP-9 correlated well with those of HGF.
was frozen immediately after collection and stored at −80 °C until analysis. In the case of the PHH patients, the CSF samples collected at the first intervention were used for measuring the MMP-9 and HGF levels. This study was performed in compliance with the guidelines for human research in Asahikawa Medical College.
2. Materials and methods
2.2. MMP-9 and HGF assay
2.1. Patients and CSF collection
The CSF levels of MMP-9 were measured using a commercially available sandwich enzyme immunoassay (EIA) (Daiichi Fine Chemical, Toyama, Japan) according to the manufacturer's instructions. The CSF levels of HGF were determined using a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Otsuka, Tokyo, Japan). The sensitivity of the MMP-9 and HGF assays was 3 ng/ml and 0.30 ng/ml, respectively.
The clinical features of infants who participated in this study are summarized in Table 1. The study involved 13 infants who developed IVH and subsequent PHH. Ultrasonographic and clinical examinations confirmed the diagnosis of IVH and ventricular dilation. CSF drainage via lumbar puncture or external ventricular drainage was performed on a median of 17 days after IVH. These interventions were performed when at least one of the following criteria was met: (1) signs of progressive ventricular dilation as detected by ultrasonography and progressive enlargement of the head as detected by daily measurement of the head circumference, (2) bulging fontanel accompanied by widening of the sagittal suture, and (3) clinical signs of increased intracranial pressure [1]. Of the 13 infants with PHH (Table 2), 9 did not require a ventriculoperitoneal (VP) shunt operation because the progressive ventricular dilation was arrested by repeated lumbar punctures or external ventricular drainage. VP shunt operations had to be performed in the remaining 4 infants because the ventricles continued to exhibit progressive dilation even after repeated lumbar punctures. To examine whether the CSF levels of MMP-9 and HGF specifically increased in the infants with PHH, we also analyzed CSF samples obtained from the normal infants and patients with congenital nonhemorrhagic hydrocephalus. During a VP shunt operation, ventricular CSF samples were collected from 10 infants with congenital nonhemorrhagic hydrocephalus. In these patients, hydrocephalus was associated with hydranencephaly (n = 1), X-linked congenital hydrocephalus (n = 1), osteocraniostenosis (n = 1), or myelomeningocele (n = 7). As controls, CSF samples obtained from 10 normal infants who underwent a lumbar puncture because CNS infection was clinically suspected were used. Ultrasonographic and clinical examinations did not reveal any hemorrhage or neurological abnormality in these control infants. All the CSF samples obtained were used for clinical diagnosis or therapeutic indications and routinely analyzed for cell count and glucose and total protein concentrations. In addition, for the determination of MMP-9 and HGF levels, a 0.5-ml aliquot of the CSF
Table 2 Clinical features of the infants with PHH in this study. Characteristic
PHH shunt (−) (n = 9)
PHH shunt (+) (n = 4)
Gestational age (weeks) Birth weight (g) Age at CSF sampling (days after IVH)
26 (25–32) 944 (694–1461) 11 (4–32)
28 (25–39) 1064 (818–3094) 35 (7–69)
NS NS NS
Values are expressed as the median (range). NS, not significant.
2.3. Statistical analysis The Mann–Whitney U test was used to compare 2 groups. Spearman's correlation coefficient by rank test was performed to investigate the correlation between the MMP-9 and HGF expression levels in the CSF samples obtained from patients with PHH. Statistical values of p b 0.05 were considered to be significant. 3. Results The CSF levels of MMP-9 were detectable in all the patients with PHH (median value, 102 ng/ml; range 12–900 ng/ml) but in only 2 patients with congenital nonhemorrhagic hydrocephalus (MMP-9 levels of 3 and 6 ng/ml). The levels of MMP-9 were significantly higher in the patients with PHH than in those with congenital nonhemorrhagic hydrocephalus (p b 0.01). None of the CSF samples obtained from the control infants had detectable levels of MMP-9. Similarly, the CSF levels of HGF were detectable in all the patients with PHH (median value, 2.08 ng/ml; range, 0.67–7.04 ng/ml) but in only 5 patients with congenital nonhemorrhagic hydrocephalus (median, 1.08 ng/ml; range, 1.02–4.01 ng/ml). Among these 5 patients, 1 had Xlinked congenital hydrocephalus and 4 had myelomeningocele. None of the CSF samples obtained from the control infants had detectable levels of HGF. The levels of HGF were significantly higher in patients with PHH than in those with congenital nonhemorrhagic hydrocephalus (p b 0.05). It was noteworthy that the CSF levels of MMP-9 were significantly higher in PHH patients who exhibited resolution of ventricular dilation without a shunt operation than in those with progressive
Table 1 Clinical features of the infants in this study. Characteristic
Control (n = 10)
Nonhemorrhagic hydrocephalus (n = 10)
Posthemorrhagic hydrocephalus (n = 13)
Gestational age (weeks) Birth weight (g) Age at CSF sampling (days after birth)
39 (38–41) 3100 (2365–3450) 1.5 (1–5)
37 (27–38)⁎ 2726 (827–3122) 14.5 (13–136)⁎
26.5 (25–39)⁎⁎ 951 (694–3094)⁎⁎ 19 (6–72)⁎⁎
Values are expressed as the median (range). ⁎ p b 0.05 (control vs nonhemorrhagic hydrocephalus). ⁎⁎ p b 0.01 (control vs posthemorrhagic hydrocephalus).
T. Okamoto et al. / Early Human Development 86 (2010) 251–254
ventricular dilation (p b 0.05) (Fig. 1). Similarly, the CSF levels of HGF were significantly higher in infants who exhibited resolution of the ventricular dilation without a shunt operation than in those who showed progressive ventricular dilation (p b 0.05) (Fig. 2). The CSF samples from the PHH patients who required a shunt operation were obtained at a median of 35 days after IVH whereas the patients who did not need a shunt operation were sampled at median of 11 days after IVH (Table 2). Although this difference in timing of CSF sampling day was not statistically significant, the CSF protein concentrations were significantly higher in the non-shunt infants than in the shunt infants (median, 235 mg/dl; range, 168–678 mg/dl vs median, 140 mg/dl; range, 85–180 mg/dl; p b 0.05). However, both the CSF levels of MMP-9 and HGF were not correlated with the CSF protein levels in the PHH patients. This suggested that the higher levels of MMP-9 and HGF in the non-shunt infants might not attribute to the earlier sampling of the CSF. HGF has been shown to increase the expression of MMP-9 in various human cell lines [9,10]. Therefore, we hypothesized that the MMP-9 expression noted in the CSF of the PHH patients was induced by HGF. To verify this hypothesis, Spearman's correlation coefficient by rank test was performed. In the infants with PHH, the CSF level of MMP-9 was significantly correlated with that of HGF (r = 0.733, p b 0.001) (Fig. 3). 4. Discussion The present study revealed that the CSF levels of MMP-9 were higher in patients with PHH than in those with congenital nonhemorrhagic hydrocephalus as well as the normal infants. Among the PHH patients, significantly higher CSF levels of MMP-9 were detected in those who exhibited resolution of the ventricular dilation without a shunt operation than in those with progressive ventricular dilation. These data, which were obtained using the EIA, were consistent with those of our previous study in which MMP-9 activity in the CSF was measured by gelatin zymography [8]. Ventricular dilation after IVH has been shown to be caused by extensive deposition of ECM proteins in the subarachnoid space [4]; these proteins can be degraded by MMP-9 [7]. Therefore, MMP-9 possibly plays an important role in the resolution of progressive ventricular dilation following IVH. In addition, the present study revealed that the CSF levels of HGF increased in patients with PHH, especially those in whom the ventricular dilation resolved without a shunt operation. The CSF
Fig. 1. Cerebrospinal fluid (CSF) concentrations of matrix metalloproteinase (MMP)-9 in patients with posthemorrhagic hydrocephalus (PHH). Nine infants exhibited resolution of the ventricular dilation without a shunt operation; however, 4 infants required a ventriculoperitoneal (VP) shunt operation because of progressive ventricular dilation. Note that CSF samples obtained from the PHH patients in whom the ventricular dilation resolved without a shunt operation had significantly higher levels of MMP-9. The asterisk indicates p b 0.05 (Mann–Whitney's U test).
253
Fig. 2. CSF concentrations of hepatocyte growth factor (HGF) in the patients with PHH. Nine infants exhibited resolution of the ventricular dilation without a shunt operation; however, 4 infants required a VP shunt operation. Note that significantly high levels of HGF were detected in the CSF of PHH patients in whom the ventricular dilation resolved without a shunt operation. The asterisk indicates p b 0.05 (Mann–Whitney's U test).
levels of HGF positively correlated with those of MMP-9 in the PHH patients. We have previously shown that TGF-β1, a central molecule in the pathogenesis of PHH [4], induces MMP-9 expression in cultured human meningeal cells [14]. Further, our preliminary data indicated that the HGF receptor, c-Met, is expressed by human meningeal cells (Okamoto and Takahashi, unpublished). Taken together with the previous studies showing that HGF induces MMP-9 expression in various human cell lines [9,10], our findings support the hypothesis that the MMP-9 expression noted in the CSF of PHH patients might be induced by HGF. However, further studies are required to confirm the relationship between MMP-9 and HGF. In this study, MMP-9 and HGF were not detected in the CSF samples obtained from normal infants, consistent with previous reports [15–17]. Furthermore, the CSF of patients with congenital nonhemorrhagic hydrocephalus barely exhibited detectable levels of MMP-9. Whereas, HGF was detected in the CSF of 50% of the patients with congenital nonhemorrhagic hydrocephalus but at lower levels than in the PHH patients. The CSF levels of HGF have been reported to increase in various CNS diseases, including Alzheimer's disease [17], bacterial meningitis [18], and ischemic brain injury [19]. In mice brain with ischemic injury, HGF has been shown to exert a neuroprotective effect [19,20]. Therefore, HGF production is considered to reflect damage as well as repair in the brain [17,19,20]. The CSF levels of HGF in the patients with nonhemorrhagic hydrocephalus, including those with X-linked congenital hydrocephalus (n = 1) and myelomeningocele (n = 4), probably reflected the degree of damage in the brain or spinal cord.
Fig. 3. Correlation between MMP-9 and HGF concentrations in the CSF of the patients with PHH. Spearman's rank correlation coefficient test revealed that the CSF levels of MMP-9 positively correlated with those of HGF.
254
T. Okamoto et al. / Early Human Development 86 (2010) 251–254
In conclusion, our data indicated that the CSF levels of MMP-9 and HGF increased in patients with PHH. The elevated CSF levels of MMP-9 were specifically observed in the patients with PHH; MMP-9 was barely detected in the CSF of the patients with congenital nonhemorrhagic hydrocephalus. The CSF levels of MMP-9 and HGF especially increased in the patients who exhibited resolution of ventricular dilation without a shunt operation; this finding suggests that both MMP-9 and HGF in the CSF may serve as prognostic factors for progressive ventricular dilation after IVH. Our present data may provide an impetus for the development of therapies targeting MMP9 or HGF, which will eliminate the need for a shunt operation in PHH patients.
[7] [8]
[9]
[10]
[11] [12]
Conflict of interest statement [13]
Authors have no conflict of interest in connection with this manuscript. Acknowledgements We sincerely thank Dr. Shinya Tanaka, Dr. Akihiko Kawase, and Dr. Yuko Tanaka for their support in the collection of CSF samples. References [1] Volpe JJ. Intracranial hemorrhage: germinal matrix-intraventricular hemorrhage of the premature infants. In: Volpe JJ, editor. Neurology of the newborn. Philadelphia: Saunders; 2008. p. 517–88. [2] Hill A, Shackelford GD, Volpe JJ. A potential mechanism of pathogenesis for early posthemorrhagic hydrocephalus in the premature newborn. Pediatrics 1985;73: 19–21. [3] Whitelaw A, Cherian S, Pople I. Transforming growth factor-β1: a possible signal molecule for posthemorrhagic hydrocephalus? Pediatr Res 1999;46:576–80. [4] Whitelaw A, Cherian S, Thoresen M, Pople I. Posthaemorrhagic ventricular dilation: new mechanisms and new treatment. Acta Paediatr Suppl 2004;444: 11–4. [5] Ignotz RA, Massague J. Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J Biol Chem 1986;261:4337–45. [6] Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakefield LM, et al. Transforming growth factor type β rapid induction of fibrosis and angiogenesis in
[14]
[15]
[16]
[17] [18]
[19]
[20]
vivo and stimulation of collagen formation in vitro. Proc Natl Acad Sci U S A 1986;83:4167–71. Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res 2003;92:827–39. Okamoto T, Takahashi S, Nakamura E, Nagaya K, Hayashi T, Shirai M, et al. Matrix metalloproteases in infants with posthemorrhagic hydrocephalus. Early Hum Dev 2008;84:137–9. Gong R, Rifai A, Tolbert EM, Centracchio JN, Dworkin LD. Hepatocyte growth factor modulates matrix metalloproteinases and plasminogen activator/plasmin proteolytic pathway in progressive renal interstitial fibrosis. J Am Soc Nephrol 2003;14:3047–60. Liu Y, Rajur K, Tolbert E, Dworkin LD. Endogenous hepatocyte growth factor ameliorates chronic renal injury by activating matrix degradation pathways. Kidney Int 2000;58:2028–43. Matsumoto K, Nakamura T. Hepatocyte growth factor: renotropic role and potential therapeutics for renal diseases. Kidney Int 2001;59:2023–38. Hattori N, Mizuno S, Yoshida Y, Chin K, Mishima M, Sisson TH, et al. The plasminogen activation system reduces fibrosis in the lung by a hepatocyte growth factor-dependent mechanism. Am J Pathol 2004;164:1091–8. Tada T, Zhan H, Tanaka Y, Hongo K, Matsumoto K, Nakamura T. Intraventricular administration of hepatocyte growth factor treats mouse communicating hydrocephalus induced by transforming growth factor beta 1. Neurobiol Dis 2006;21:576–86. Okamoto T, Takahashi S, Nakamura E, Nagaya K, Hayashi T, Fujieda K. Transforming growth factor-β1 induces matrix metalloproteinase-9 expression in human meningeal cells via ERK and Smad pathways. Biochem Biophys Res Commun 2009;383:475–9. Yushchenko M, Weber F, Mader M, Scholl U, Maliszewska M, Tumani H, et al. Matrix metalloproteinase-9 (MMP-9) in human cerebrospinal fluid (CSF): elevated levels are primary related to CSF cell count. J Neuroimmunol 2000;110:244–51. Sulik A, Wojtkowska M, Oldak E. Elevated levels of MMP-9 and TIMP-1 in the cerebrospinal fluid of children with echovirus type 30 and mumps meningitis. Scand J Immunol 2008;68:323–7. Tsuboi Y, Kakimoto K, Akatsu H, Daikuhara Y, Yamada T. Hepatocyte growth factor in cerebrospinal fluid in neurologic disease. Acta Neurol Scand 2002;106:99–103. Nayeri F, Nilsson I, Hagberg L, Brudin L, Roberg M, Soderstrom C, et al. Hepatocyte growth factor levels in cerebrospinal fluid: a comparison between acute bacterial and nonbacterial meningitis. J Infect Dis 2000;181:2092–4. Miyazawa T, Matsumoto K, Ohmichi H, Katoh H, Yamashita T, Nakamura T. Protection of hippocampal neurons from ischemia-induced delayed neuronal death by hepatocyte growth factor: a novel neurotrophic factor. J Cereb Blood Flow Metab 1998;18:345–8. Honda S, Kagoshima M, Wanaka A, Tohyama M, Matsumoto K, Nakamura T. Localization and functional coupling of HGF and c-Met/HGF receptor in rat brain: implication as neurotrophic factor. Brain Res Mol Brain Res 1995;32:197–210.