Original Paper Received: January 3, 2014 Accepted: April 4, 2014 Published online: September 20, 2014
Eur Neurol 2014;72:234–240 DOI: 10.1159/000362876
Circulating Matrix Metalloproteinase-9 Level Is Associated with Cerebral White Matter Hyperintensities in Non-Stroke Individuals Yoon Kim a Yu-Kyung Kim b Nam Keun Kim b Sang-Heum Kim c Ok-Joon Kim a Seung-Hun Oh a
Department of Neurology, b Institute for Clinical Research, and c Department of Radiology, CHA Bundang Medical Center, CHA University, Seongnam, Republic of Korea
Key Words Cerebral white matter hyperintensities · Matrix metalloproteinase-9 · Blood-brain barrier · Small vessel disease
Abstract Backgrounds: The pathogenesis of cerebral white matter hyperintensities (WMH) has been poorly understood. Our aim was to investigate the association of circulating proteins, the biomarkers of inflammation, blood-brain barrier (BBB) dysfunction, and thrombosis with WMH in non-stroke individuals. Methods: Demographic, laboratory, and brain magnetic resonance imaging parameters were prospectively analyzed in 137 subjects. The relationship between plasma interleukin-6, tumor necrosis factor-α, matrx-metalloproteinase-9 (MMP-9), plasminogen activator inhibitor-1 and overt WMH (Fazekas grading score ≥2) was analyzed. Results: In univariate analysis, old age, high blood pressure, history of hypertension, and elevated plasma MMP-9 level were associated with overt WMH. In multivariate analysis, plasma MMP-9 still maintained a significant association with WMH. Plasma MMP-9 level was weakly but significantly associated with WMH volume (r = 0.232, p = 0.006). All the other circulating proteins examined failed to demonstrate a significant relationship with WMH. Conclusions: Plasma MMP-9 is associated with pathophysiology of WMH development.
Introduction
Cerebral white matter hyperintensities (WMH), previously regarded as a benign finding associated with the natural aging process, have turned out to reflect the underlying pathologic process of small vessel disease (SVD) at work [1]. With the progression of WMH, the risk of dementia and stroke increases [2]. In addition, WMH severity turned out to be the one of the strongest predictors of physical disability including falls [3] and urinary incontinence [4], independent of other confounders and vascular lesions of the brain. Clinicopathological studies have revealed WMH, the diffuse cerebral ischemic lesions to arise from multiple occlusions of smaller (40∼200 μm of diameter) perforating cerebral arteries [5–7]. Attempts have been made to explain their pathomechanism in terms of microatheroma, lipohyalinosis of perforating artery, and/or emboli originating from major cerebral arteries [8]. The affected arteriole is thickened and its lumen distorted, rendering it prone to thrombosis, occlusion, and infarction [9]. Multiple mechanisms such as endothelial failure [10, 11], blood-brain barrier (BBB) dysfunction [12, 13], and inflammation [14, 15] have been proposed to explain the pathogenesis of WMH. Endothelial failure, recently proposed as the key precipitant of SVD, exposes the vessel
© 2014 S. Karger AG, Basel © 2014 S. Karger AG, Basel 0014–3022/14/0724–0234$39.50/0 E-Mail [email protected] www.karger.com/ene
Seung-Hun Oh, MD Department of Neurology, CHA University CHA Bundang Medical Center, 351, Yatap-dong, Bundang-gu Seongnam-si, Gyunggi-do, 463-712 (Republic of Korea) E-Mail ohsh72 @ chamc.co.kr
Downloaded by: Thammasat University 203.131.222.1 - 10/8/2014 4:50:39 AM
a
Materials and Methods
study. The Institutional Ethics Committee of the CHA Bundang Medical Center approved this study (IRB approval no. 2010–083). All subjects underwent a thorough interview about their past history of medico-surgical illness. Hypertension was defined as a high baseline blood pressure (BP) (systolic ≥140 mm Hg or diastolic ≥90 mm Hg) or a history of antihypertensive medication. Diabetes mellitus (DM) was defined as a fasting plasma glucose of ≥126 mg/dl or a history of insulin or oral hypoglycemic therapy. Smoking was defined as the smoking status at the time of examination. Hyperlipidemia was defined as fasting serum total cholesterol of ≥220 mg/dl or a history of treatment with a statin. The category of ischemic heart disease (IHD) included a history of myocardial infarction, unstable angina, coronary angioplasty, or coronary bypass graft surgery. Measurement of Cerebral White Matter Hyperintensity Severity and Volume Brain MR images of all study subjects were examined for WMH lesions. The severity and volume of WMH were analyzed from fluid attenuated inversion recovery (FLAIR) images. Sixteen axial images were obtained using a slice thickness of 5 mm with a 2 mm inter-slice gap. The FLAIR imaging protocol consisted of TR/TE = 9,000/105 ms and inversion time = 2,500 ms. The presence of and severity of WMH had to be agreed upon by one neurologist and one radiologist. The WMH severity was assessed using the scoring system of Fazekas et al. [23] and divided into 4 grades: grade 0 = absent; 1 = punctuate; 2 = early confluent; and 3 = confluent lesions in the bilateral periventricular and subcortical white matters. Among 137 subjects, nine subjects had no WMH lesion (Fazekas’ score = 0) and 73 subjects had minimal WMH lesion (Fazekas’ score = 1). The remaining 55 subjects showed clear presence of WMH lesion (Fazekas’ score = 2 and 3). Minimal presence of WMH (Fazekas’ score = 1) was regarded as normal aging, and overt WMH was defined as Fazekas’ score of 2 or higher. The WMH size at each slice was initially measured using a semi-computerized, intensity-threshold technique (Image J software, NIH, Bethesda, MA). The WMH volume was calculated as the WMH size multiplied by the inter-slice thickness (5 mm), and the total WMH volume was determined by summating the WMH area at each slice.
Study Population and Risk Factors Assessment The study was designed as a prospective analysis of non-stroke individuals who visited the outpatient neurology clinic at CHA Bundang Medical Center between March 2008 and December 2010. All of the subjects showed normal cognitive function and normal activities of daily living. The subjects sought medical attention for minor complaints such as headache and dizziness. They had either underlying cardiovascular risk factors or a family history of stroke but had no history of neurological disease. We included a total of 137 subjects who agreed to undergo blood sampling for blood marker measurements and brain MRI for the present study. A neurologist performed neurological examination at the time of the outpatient visit. The exclusion criteria were as follows: (1) age under 50 years old; (2) previous history of neurological disease including cerebrovascular accident; (3) presence of focal neurological deficits at the time of neurological examination; (4) presence of systemic infection; and (5) refusal to be included in the
Measurements of Circulating Proteins Whole blood (10cc) was drawn on the time of outpatient visit using a tube containing ethylenediaminetetraacetic acid (EDTA). The samples were immediately delivered to the laboratory. Plasma was quickly prepared from the whole blood by centrifugation at 3,000 g for 15 min at room temperature. The plasma samples were carefully transferred into appropriately labeled micro-centrifuge tubes by using a sterile transfer pipette tip. The samples were apportioned into 1 ml aliquots from 5 to 7 vials and stored at –80 ° C for later analysis. The method of measuring each marker is described further in the online supplement 1. Briefly, plasma concentrations of IL-6 (IL-6 Quantikine, R&D Systems, MN, USA), total MMP-9 (92 kDa- proMMP-9 and 82 kDa-active MMP-9, e-bioscience, Abingdon, UK), TNF-α (TNF-α Quantikine, R&D Systems, Minn., USA) were measured using a quantitative sandwich ELISA kit. The plasma concentration of PAI-1 was measured using a commercially available active PAI-1 functional assay kit (Innovative Research, Novi, Mich., USA). Using the ELISA reader (Emax®, Molecular
Plasma MMP-9 and Cerebral White Matter Hyperintensities
Eur Neurol 2014;72:234–240 DOI: 10.1159/000362876
235
Downloaded by: Thammasat University 203.131.222.1 - 10/8/2014 4:50:39 AM
wall of arterioles, making it susceptible to thrombosis. Altered BBB permeability induces leakage of plasma components and inflammatory cells into the vessel wall and the surrounding tissue accelerating the process of inflammation and thrombosis. Even though one mechanism may not necessarily contradict another but rather complement it, determining the main mechanism of action will prove invaluable in understanding SVD. Several studies have evaluated the circulating markers that are possibly related to the pathophysiology of WMH [16–19]. A longitudinal study found higher C-reactive protein (CRP) level at baseline was associated with the presence and progression of WMH [16]. A cross-sectional study disclosed that circulating CRP and interleukin-6 (IL-6), both markers of inflammation, were found increased in those with small silent brain infarction, another manifestation of SVD [17]. Higher level of homocysteine, a marker of endothelial dysfunction, was found significantly associated with greater WMH volume [18]. In addition, circulating matrx-metalloproteinase-9 (MMP-9), a marker of ECM remodeling, showed a significant correlation with the higher prevalence of WMH [19]. In the present study, we simultaneously measured several circulating proteins possibly related to the pathomechanism of cerebral vascular diseases: (1) inflammatory cytokines: IL-6 [17] and tumor necrosis factor-α (TNF-α) [20]; (2) BBB dysfunction marker: MMP-9 [21]; and (3) thrombotic/haemostatic marker: plasminogen activator inhibitor-1 (PAI-1) [22] to determine the candidate blood markers for WMH and its severity in non-stroke individuals.
Table 1. Baseline characteristics of the study subjects
Study subjects, n Age, years Gender (female %) BMI, kg/m2 Hypertension (%) Diabetes mellitus (%) Hyperlipidemia (%) IHD (%) Smoking (%) Systolic BP, mm Hg Diastolic BP, mm Hg Creatinine, mg/dl Cholesterol, mg/dl Glucose, mg/dl
Total subjects
No WMH (Fazekas score ≤1)
Overt WMH (Fazekas score ≥2)
p value
137 64.9±8.1 79 (57.6) 25.0±5.7 85 (62.0) 32 (23.3) 35 (25.5) 5 (3.6) 38 (27.3) 129.1±17.3 78.7±9.0 0.98±0.32 194.1±44.5 131±50.8
82 62.9±6.6 51 (62.2) 24.4±3.9 43 (52.4) 19 (23.2) 22 (26.8) 1 (1.2) 22 (26.8) 125.9±14.7 77.4±9.1 0.94±0.25 198.6±45.8 131±50.5
55 68.0±9.2 28 (50.1) 25.8±7.7 42 (76.3) 13 (23.6) 13 (23.6) 4 (7.2) 16 (29.1) 133±19.9 80.8±8.6 1.06±0.39 187.9±42.4 130±51.7
–
Eur Neurol 2014;72:234–240 DOI: 10.1159/000362876
Circulating Matrix Metalloproteinase-9 Level Is Associated with Cerebral White Matter Hyperintensities in Non-Stroke Individuals Yoon Kim a Yu-Kyung Kim b Nam Keun Kim b Sang-Heum Kim c Ok-Joon Kim a Seung-Hun Oh a
Department of Neurology, b Institute for Clinical Research, and c Department of Radiology, CHA Bundang Medical Center, CHA University, Seongnam, Republic of Korea
Key Words Cerebral white matter hyperintensities · Matrix metalloproteinase-9 · Blood-brain barrier · Small vessel disease
Abstract Backgrounds: The pathogenesis of cerebral white matter hyperintensities (WMH) has been poorly understood. Our aim was to investigate the association of circulating proteins, the biomarkers of inflammation, blood-brain barrier (BBB) dysfunction, and thrombosis with WMH in non-stroke individuals. Methods: Demographic, laboratory, and brain magnetic resonance imaging parameters were prospectively analyzed in 137 subjects. The relationship between plasma interleukin-6, tumor necrosis factor-α, matrx-metalloproteinase-9 (MMP-9), plasminogen activator inhibitor-1 and overt WMH (Fazekas grading score ≥2) was analyzed. Results: In univariate analysis, old age, high blood pressure, history of hypertension, and elevated plasma MMP-9 level were associated with overt WMH. In multivariate analysis, plasma MMP-9 still maintained a significant association with WMH. Plasma MMP-9 level was weakly but significantly associated with WMH volume (r = 0.232, p = 0.006). All the other circulating proteins examined failed to demonstrate a significant relationship with WMH. Conclusions: Plasma MMP-9 is associated with pathophysiology of WMH development.
Introduction
Cerebral white matter hyperintensities (WMH), previously regarded as a benign finding associated with the natural aging process, have turned out to reflect the underlying pathologic process of small vessel disease (SVD) at work [1]. With the progression of WMH, the risk of dementia and stroke increases [2]. In addition, WMH severity turned out to be the one of the strongest predictors of physical disability including falls [3] and urinary incontinence [4], independent of other confounders and vascular lesions of the brain. Clinicopathological studies have revealed WMH, the diffuse cerebral ischemic lesions to arise from multiple occlusions of smaller (40∼200 μm of diameter) perforating cerebral arteries [5–7]. Attempts have been made to explain their pathomechanism in terms of microatheroma, lipohyalinosis of perforating artery, and/or emboli originating from major cerebral arteries [8]. The affected arteriole is thickened and its lumen distorted, rendering it prone to thrombosis, occlusion, and infarction [9]. Multiple mechanisms such as endothelial failure [10, 11], blood-brain barrier (BBB) dysfunction [12, 13], and inflammation [14, 15] have been proposed to explain the pathogenesis of WMH. Endothelial failure, recently proposed as the key precipitant of SVD, exposes the vessel
© 2014 S. Karger AG, Basel © 2014 S. Karger AG, Basel 0014–3022/14/0724–0234$39.50/0 E-Mail [email protected] www.karger.com/ene
Seung-Hun Oh, MD Department of Neurology, CHA University CHA Bundang Medical Center, 351, Yatap-dong, Bundang-gu Seongnam-si, Gyunggi-do, 463-712 (Republic of Korea) E-Mail ohsh72 @ chamc.co.kr
Downloaded by: Thammasat University 203.131.222.1 - 10/8/2014 4:50:39 AM
a
Materials and Methods
study. The Institutional Ethics Committee of the CHA Bundang Medical Center approved this study (IRB approval no. 2010–083). All subjects underwent a thorough interview about their past history of medico-surgical illness. Hypertension was defined as a high baseline blood pressure (BP) (systolic ≥140 mm Hg or diastolic ≥90 mm Hg) or a history of antihypertensive medication. Diabetes mellitus (DM) was defined as a fasting plasma glucose of ≥126 mg/dl or a history of insulin or oral hypoglycemic therapy. Smoking was defined as the smoking status at the time of examination. Hyperlipidemia was defined as fasting serum total cholesterol of ≥220 mg/dl or a history of treatment with a statin. The category of ischemic heart disease (IHD) included a history of myocardial infarction, unstable angina, coronary angioplasty, or coronary bypass graft surgery. Measurement of Cerebral White Matter Hyperintensity Severity and Volume Brain MR images of all study subjects were examined for WMH lesions. The severity and volume of WMH were analyzed from fluid attenuated inversion recovery (FLAIR) images. Sixteen axial images were obtained using a slice thickness of 5 mm with a 2 mm inter-slice gap. The FLAIR imaging protocol consisted of TR/TE = 9,000/105 ms and inversion time = 2,500 ms. The presence of and severity of WMH had to be agreed upon by one neurologist and one radiologist. The WMH severity was assessed using the scoring system of Fazekas et al. [23] and divided into 4 grades: grade 0 = absent; 1 = punctuate; 2 = early confluent; and 3 = confluent lesions in the bilateral periventricular and subcortical white matters. Among 137 subjects, nine subjects had no WMH lesion (Fazekas’ score = 0) and 73 subjects had minimal WMH lesion (Fazekas’ score = 1). The remaining 55 subjects showed clear presence of WMH lesion (Fazekas’ score = 2 and 3). Minimal presence of WMH (Fazekas’ score = 1) was regarded as normal aging, and overt WMH was defined as Fazekas’ score of 2 or higher. The WMH size at each slice was initially measured using a semi-computerized, intensity-threshold technique (Image J software, NIH, Bethesda, MA). The WMH volume was calculated as the WMH size multiplied by the inter-slice thickness (5 mm), and the total WMH volume was determined by summating the WMH area at each slice.
Study Population and Risk Factors Assessment The study was designed as a prospective analysis of non-stroke individuals who visited the outpatient neurology clinic at CHA Bundang Medical Center between March 2008 and December 2010. All of the subjects showed normal cognitive function and normal activities of daily living. The subjects sought medical attention for minor complaints such as headache and dizziness. They had either underlying cardiovascular risk factors or a family history of stroke but had no history of neurological disease. We included a total of 137 subjects who agreed to undergo blood sampling for blood marker measurements and brain MRI for the present study. A neurologist performed neurological examination at the time of the outpatient visit. The exclusion criteria were as follows: (1) age under 50 years old; (2) previous history of neurological disease including cerebrovascular accident; (3) presence of focal neurological deficits at the time of neurological examination; (4) presence of systemic infection; and (5) refusal to be included in the
Measurements of Circulating Proteins Whole blood (10cc) was drawn on the time of outpatient visit using a tube containing ethylenediaminetetraacetic acid (EDTA). The samples were immediately delivered to the laboratory. Plasma was quickly prepared from the whole blood by centrifugation at 3,000 g for 15 min at room temperature. The plasma samples were carefully transferred into appropriately labeled micro-centrifuge tubes by using a sterile transfer pipette tip. The samples were apportioned into 1 ml aliquots from 5 to 7 vials and stored at –80 ° C for later analysis. The method of measuring each marker is described further in the online supplement 1. Briefly, plasma concentrations of IL-6 (IL-6 Quantikine, R&D Systems, MN, USA), total MMP-9 (92 kDa- proMMP-9 and 82 kDa-active MMP-9, e-bioscience, Abingdon, UK), TNF-α (TNF-α Quantikine, R&D Systems, Minn., USA) were measured using a quantitative sandwich ELISA kit. The plasma concentration of PAI-1 was measured using a commercially available active PAI-1 functional assay kit (Innovative Research, Novi, Mich., USA). Using the ELISA reader (Emax®, Molecular
Plasma MMP-9 and Cerebral White Matter Hyperintensities
Eur Neurol 2014;72:234–240 DOI: 10.1159/000362876
235
Downloaded by: Thammasat University 203.131.222.1 - 10/8/2014 4:50:39 AM
wall of arterioles, making it susceptible to thrombosis. Altered BBB permeability induces leakage of plasma components and inflammatory cells into the vessel wall and the surrounding tissue accelerating the process of inflammation and thrombosis. Even though one mechanism may not necessarily contradict another but rather complement it, determining the main mechanism of action will prove invaluable in understanding SVD. Several studies have evaluated the circulating markers that are possibly related to the pathophysiology of WMH [16–19]. A longitudinal study found higher C-reactive protein (CRP) level at baseline was associated with the presence and progression of WMH [16]. A cross-sectional study disclosed that circulating CRP and interleukin-6 (IL-6), both markers of inflammation, were found increased in those with small silent brain infarction, another manifestation of SVD [17]. Higher level of homocysteine, a marker of endothelial dysfunction, was found significantly associated with greater WMH volume [18]. In addition, circulating matrx-metalloproteinase-9 (MMP-9), a marker of ECM remodeling, showed a significant correlation with the higher prevalence of WMH [19]. In the present study, we simultaneously measured several circulating proteins possibly related to the pathomechanism of cerebral vascular diseases: (1) inflammatory cytokines: IL-6 [17] and tumor necrosis factor-α (TNF-α) [20]; (2) BBB dysfunction marker: MMP-9 [21]; and (3) thrombotic/haemostatic marker: plasminogen activator inhibitor-1 (PAI-1) [22] to determine the candidate blood markers for WMH and its severity in non-stroke individuals.
Table 1. Baseline characteristics of the study subjects
Study subjects, n Age, years Gender (female %) BMI, kg/m2 Hypertension (%) Diabetes mellitus (%) Hyperlipidemia (%) IHD (%) Smoking (%) Systolic BP, mm Hg Diastolic BP, mm Hg Creatinine, mg/dl Cholesterol, mg/dl Glucose, mg/dl
Total subjects
No WMH (Fazekas score ≤1)
Overt WMH (Fazekas score ≥2)
p value
137 64.9±8.1 79 (57.6) 25.0±5.7 85 (62.0) 32 (23.3) 35 (25.5) 5 (3.6) 38 (27.3) 129.1±17.3 78.7±9.0 0.98±0.32 194.1±44.5 131±50.8
82 62.9±6.6 51 (62.2) 24.4±3.9 43 (52.4) 19 (23.2) 22 (26.8) 1 (1.2) 22 (26.8) 125.9±14.7 77.4±9.1 0.94±0.25 198.6±45.8 131±50.5
55 68.0±9.2 28 (50.1) 25.8±7.7 42 (76.3) 13 (23.6) 13 (23.6) 4 (7.2) 16 (29.1) 133±19.9 80.8±8.6 1.06±0.39 187.9±42.4 130±51.7
–
