Neurochem Res (2011) 36:1858–1862 DOI 10.1007/s11064-011-0504-4
ORIGINAL PAPER
Serum Glutamine Synthetase Has No Value as a Diagnostic Biomarker for Alzheimer’s Disease Yannick Vermeiren • Nathalie Le Bastard • Christopher M. Clark • Sebastiaan Engelborghs Peter P. De Deyn
•
Accepted: 10 May 2011 / Published online: 20 May 2011 Ó Springer Science+Business Media, LLC 2011
Abstract In order to test whether serum glutamine synthetase (GS) is of potential diagnostic value for Alzheimer’s disease (AD), we set up a study to compare serum GS concentrations between AD patients and control subjects. The study population (n = 165) consisted of AD patients (n = 94) and age-matched (n = 41) and ageunmatched (n = 30) control subjects. Serum GS analysis was performed by means of ELISA. No significant differences in serum GS levels were found between the AD
group and age-matched controls. Age correlated positively with serum GS concentrations in AD patients and control subjects. This study suggests that serum GS levels have no diagnostic value for AD. Keywords Alzheimer’s disease Glutamine synthetase Biomarkers Serum
Introduction Y. Vermeiren N. Le Bastard S. Engelborghs P. P. De Deyn (&) Laboratory of Neurochemistry and Behavior, Reference Center for Biological Markers of Memory Disorders, Institute Born-Bunge, University of Antwerp, Universiteitsplein 1, 2610 Antwerp, Wilrijk, Belgium e-mail:
[email protected];
[email protected] P. P. De Deyn Biobank, Institute Born-Bunge, University of Antwerp, Universiteitsplein 1, 2610 Antwerp, Belgium S. Engelborghs P. P. De Deyn Department of Neurology and Memory Clinic, Middelheim and Hoge Beuken General Hospitals (ZNA), Lindendreef 1, 2020 Antwerp, Belgium S. Engelborghs P. P. De Deyn Department of Health Care Science, Artesis University College of Antwerp, J. De Boeckstraat 10, 2170 Merksem, Belgium S. Engelborghs Department of Nursing Sciences, Faculty of Medicine, University of Antwerp, Universiteitsplein 1, 2610 Antwerp, Belgium C. M. Clark Department of Neurology and Alzheimer’s Disease Center, University of Pennsylvania, 3615 Chestnut St, Philadelphia, PA 19104, USA
123
Glutamine synthetase (GS) is an important enzyme present in all mammals and responsible for the metabolic conversion of the neurotransmitter glutamate and ammonium to glutamine [1]. In general, GS is located mainly in astrocytes and plays a key role in the detoxification of brain ammonia and in the regulation of neuronal glutamate levels [2]. An age-related decrease of astrocyte-specific GS activity has led to the hypothesis that this reduction might result from oxidative damage [3, 4]. Accordingly, Smith et al. [4] suggested that Alzheimer’s disease (AD) could represent a specific brain vulnerability to age-related oxidation, thereby linking AD pathophysiology directly to the aging process. Furthermore, it has been frequently demonstrated that b-amyloid protein (Ab) accumulation in AD induces major oxidative stress and subsequently diminishes GS activity both in vitro [6–9] and in vivo [4, 5]. Interaction of Ab1–40 with the oxidation-sensitive GS also resulted in an increase of Ab-mediated neurotoxicity in hippocampal cell cultures [7]. In addition, animal studies using senescence accelerated mice (SAM) as a model for AD and Ab accumulation demonstrated decreased brain GS activities in comparison with younger SAM [10, 11]. Moreover, alterations in levels or activity of brain GS have been described in related conditions such as Down syndrome [12].
Neurochem Res (2011) 36:1858–1862
Taking into account the functional relationship between accumulated Ab and structurally affected GS [13], several authors have suggested that GS consequently might be a potential diagnostic biomarker for AD. Indeed, GS was detected in 38 out of 39 cerebrospinal fluid (CSF) samples of AD patients and in only 1 out of 29 CSF samples of healthy control subjects [14]. Tumani et al. [15] found significantly increased GS concentrations in CSF of AD patients and to a lesser extent in patients with vascular dementia and amyotrophic lateral sclerosis compared to healthy control subjects. Serum GS concentrations were higher in AD patients compared to controls although the level of statistical significance was not achieved [15]. Takahashi et al. [16] measured elevated GS concentrations in serum of AD patients compared to non-AD dementias and Burbaeva et al. [17] found increased levels of GS in the prefrontal cortex of AD patients compared to healthy controls. However, decreased GS levels have also been reported in brains of patients with AD [4, 5, 18], spinocerebellar ataxia type I [19] and hepatic encephalopathy [20], which might indicate a relationship between alterations in GS concentrations, oxidative stress and brain injury [4]. In particular, reduced GS activities were present in AD frontal [4] and temporal [5, 18] brain regions compared to age-matched controls. So even though there exists suggestive evidence for altered GS levels in CSF, serum and brain tissue of AD patients compared to healthy controls, its potential diagnostic value remains to be elucidated. Furthermore, given the lack of a validated peripheral diagnostic biomarker for AD whether in blood [21–23] or in urine [21], it is of high interest to further develop and rigorously validate candidate peripheral biochemical biomarkers for AD [24]. As a consequence, the need for lumbar puncture procedures to obtain CSF for measurements of tau and Ab will be reduced. Thus, considering GS as a potential peripheral biomarker candidate for AD, we compared serum GS concentrations between AD patients and control subjects.
Experimental Procedure Patient Population The AD group comprised 94 patients who were recruited from the neurology department of Middelheim General Hospital and its Memory Clinic, Ziekenhuisnetwerk Antwerp (Antwerp, Belgium). The AD group consisted of patients with definite AD (n = 5), probable AD (n = 87), possible AD (n = 1), and probable AD with cerebrovascular disease (CVD) (n = 1). The diagnosis of probable and possible AD was made according to the NINCDSADRDA criteria [25] though all patients fulfilled the DSMIV criteria [26] as well. AD with CVD was diagnosed when
1859
patients fulfilled the criteria of probable AD according to the NINCDS-ADRDA criteria and in addition displayed CVD on brain CT and/or MRI that however did not meet the criteria of relevant CVD according to the NINCDSAIREN criteria of vascular dementia [27], thus excluding multiple large-vessel infarcts, strategically placed infarcts, multiple basal ganglia and white matter lacunes or extensive white matter lesions. Diagnoses were made following intensive diagnostic work-up that consisted of a general physical and neurological examination, blood screening, structural neuroimaging, functional brain imaging, electroencephalogram and an extensive neuropsychological examination (amongst others the Mini-Mental State Examination (MMSE) [28]). The diagnosis of definite AD was established as described in Engelborghs et al. [29]. The control group consisted of age-matched (n = 41) and age-unmatched (n = 30) controls. Control subjects were recruited among hospitalized patients from the Middelheim General Hospital. The inclusion criteria for the control group were: (1) no organic disease involving the central nervous system; (2) no psychiatric antecedents and (3) no cognitive decline which was ruled out following extensive clinical examination. The local ethics committee approved this study. All patients were of Caucasian origin. All controls, patients and their caregivers gave written informed consent. Serum Sampling and Analysis At inclusion, serum samples were collected according to a standard procedure. Sampling was performed between 8 and 10 a.m. after overnight fasting and after having abstained from smoking for at least 12 h. In total, 15 ml of blood was collected in serum gel tubes (MonovetteÒ, Sarstedt inc., Nu¨mbrecht, Germany). After blood coagulation, the serum gel tubes were immediately centrifuged at 3,000 rpm during 10 min. The supernatant was then distributed in different polypropylene vials (NalgeneÒ, VWR Leuven, Belgium) and immediately frozen in liquid nitrogen. All samples were stored at -75°C until analysis at the facilities of the Biobank of the Institute Born-Bunge. Serum GS analysis was performed by means of an ELISA as described in Takahashi et al. [16]. Statistical Analyses Before statistics were applied, outliers concerning the levels of serum GS were identified by means of box plots and excluded from data analysis (n = 5; resulting in a final population that entered data-analysis: AD: n = 91; agematched controls: n = 40; age-unmatched controls: n = 29). Normality was tested by a Kolmogorov–Smirnov test. Parametric testing was allowed because of exclusion
123
1860
Neurochem Res (2011) 36:1858–1862
of all outliers, the normal distribution of all data and the sufficient number of patients and controls included. A parametric one-way analysis of variance (1wANOVA) with posthoc Bonferroni’s procedure was used for comparison of demographic data and serum GS concentrations between the different diagnostic categories. Chi-square statistics were applied to compare male–female ratios. Comparison of GS concentrations between males and females was performed by an unpaired student t test. Pearson’s product-moment was used to calculate correlations between age, storage time and MMSE scores with serum levels of GS. Probability levels of P \ 0.05 were considered significant. All analyses were performed using SPSSÒ 16.0 for Windows (SPSS Inc., Chicago, USA).
Results Demographic Data (Table 1) A significant difference in gender between the AD group and age-matched controls was found (P = 0.018). There was no significant difference in age between AD patients and age-matched controls (P = 0.110), both groups were, however, significantly older than the age-unmatched controls (both P \ 0.0001). In the AD group, storage time of serum samples was significantly shorter compared with agematched and age-unmatched controls (both P \ 0.0001). Samples from age-unmatched controls had a longer storage time than those from age-matched controls (P = 0.029). AD patients had an average MMSE score of 15/30 ± 7 with a range varying from 0/30 to 27/30 (n = 58). Serum GS Concentrations (Table 1) No significant differences in serum GS levels were found between the AD group and age-matched controls
(P = 0.505). AD patients and age-matched controls had significantly higher GS concentrations than age-unmatched controls (P = 0.008 and P = 0.001, respectively). No significant differences in serum GS concentrations between males and females were detected in AD patients (P = 0.275), nor in the age-matched and age-unmatched control groups separately (P = 0.803 and P = 0.843, respectively). Correlations of Age, Storage Time and MMSE Scores with Serum Levels of GS Age correlated positively with serum GS concentrations in the AD group (n = 91; r = 0.240; P = 0.022) and pooled control population (n = 69; r = 0.381; P = 0.001). In the AD group, no significant correlations between storage time of the serum samples and GS concentrations were found (n = 91; r = 0.093; P = 0.383). However, in the agematched and age-unmatched control group storage time of the serum samples correlated significantly with GS serum concentrations (n = 40; r = 0.464; P = 0.003 and n = 29; r = -0.613; P = 0.0004, respectively). No significant correlations between MMSE scores and serum GS levels were found in AD patients (n = 58; r = -0.011; P = 0.933).
Discussion No Diagnostic Value of Serum GS for AD No significant differences in serum GS concentrations between AD patients and age-matched controls were found. Furthermore, no correlation between serum GS levels and severity of AD (i.e. MMSE scores) was found. The differences in serum GS levels between AD patients/ age-matched control subjects and age-unmatched control subjects were attributed to the fact that age-unmatched
Table 1 Demographic data and serum GS concentrations
Male/female
AD (n = 91)
Age-matched controls (n = 40) Age-unmatched controls (n = 29) Statistical analysis
24/67°
19/21°
P = 0.044, df = 2, V2 = 6.268
12/17
?
73.7 ± 7.3*
42.2 ± 10.6 *
P \ 0.0001, df = 2, F = 186.295
(40–93) Storage time (years) 7.3 ± 2.5°,?
(60–87) 9.3 ± 1.9°,*
(24–58) 10.7 ± 0.9?,*
P \ 0.0001, df = 2, F = 30.731
(5.3–11.6)
(8.1–11.4)
Age (years)
77.1 ± 8.4
(3.2–11.7) GS (pg/ml)
?,
53.4 ± 18.0? 57.9 ± 19.4*
42.1 ± 10.2?,*
(19.3–110.7)
(27.2–68.1)
(24.9–122.8)
P = 0.001, df = 2, F = 7.356
Data are given as mean ± SD. Ranges are represented between brackets. For comparison of male–female ratios, Chi-square statistics were used. For all other comparisons, 1w-ANOVA with posthoc Bonferroni’s procedure was applied. Significant differences are indicated with the following symbols: ° AD versus age-matched controls; ? AD versus age-unmatched controls; * age-matched controls versus age-unmatched controls
123
Neurochem Res (2011) 36:1858–1862
controls were significantly younger than AD patients, given the significant and positive correlation between age and serum GS levels. Taken together, our study is the first to report unaltered serum GS concentrations in AD patients compared to age-matched control subjects and therefore suggests that there is no potential diagnostic value of serum GS for AD. This result is in contrast with some previous AD-related GS studies that found alterations of GS concentrations in serum [16], CSF [14, 15] and brain tissue [4, 5, 17]. However, these studies did not share the same outcome as in serum and in CSF an increase of GS levels was found ([16] and [15], respectively) whereas in brain tissue diverse outcomes were observed, with increases [17] but also with decreases [4, 5] of reported GS levels. Also, Tumani et al. [15] did not include age-matched controls but a control group with a wide variety of ages. Moreover, Burbaeva et al. [17], Le Prince et al. [5], Smith et al. [4], Takahashi et al. [16] and Tumani et al. [15] included only 11, 8, 16, 24 and 8 AD patients, respectively. Most studies also included less control subjects, with exception of the study of Takahashi et al. [16]. So, the large group of clinically wellcharacterized AD patients and controls renders this study unique and valuable. Besides serum GS, our Reference Center for Biological Markers of Memory Disorders (Antwerp, Belgium) has also examined the diagnostic performance of full-length and N-truncated plasma Ab forms in AD patients and nonAD disease dementia as compared to healthy control subjects [23] in order to further elucidate the concept of peripheral markers of neurodegenerative disorders. However, no significant differences in plasma Ab isoforms were detected between AD patients and non-AD dementia patients or between AD patients and controls. Also, Le Bastard et al. [22] could not find any correlation between plasma and CSF Ab isoforms of patients with AD, nonAD, mild cognitive impairment (MCI) and healthy control subjects. This hampers the diagnostic utility of plasma Ab levels as peripheral biomarkers for dementia whereas CSF Ab levels have proven to be validated biomarkers for MCI and AD [30]. In contrast, Fei et al. [31] recently found a relationship between lowered plasma Ab1–42 levels and the plasma Ab1–42/Ab1–40 ratio and the progression from MCI to incipient AD, although no changes in plasma Ab1–40 levels were observed.
1861
compared to three month old rats, finding an age-related increase of GS concentrations. According to Finch and Cohen [3] and Smith et al. [4], this specifically age-related alteration of GS levels could result from gradually increasing oxidative damage. Nonetheless, as opposed to our results and those of Danh et al. [32], Smith et al. [4] described the exact opposite, i.e. reduced GS activities in frontal cerebral cortex of aged persons compared to their younger counterparts. Finally, two previous studies in man did not find an association between age and GS levels in serum [16] and CSF [15]. On the whole, conflicting results concerning GS levels in relation to aging can be observed, but caution is advised when interpreting results depending on analyses of different biological samples such as brain tissue [4, 32], CSF [15] or serum [16]. Strengths and Weaknesses of the Study AD patients were clinically well-characterized, amongst others by inclusion of the available clinical and neuropsychological follow-up data that contributed to diagnostic certainty and by post-mortem neuropathological confirmation of the clinical diagnosis in 5 AD patients. There was also a larger number of AD patients included as compared to other similar GS studies. Moreover, the standardized serum sampling procedure was a strength of this study, meanwhile ruling out possible circadian rhythm effects on GS levels. Nonetheless, this study had several limitations: (1) control subjects were not matched with regard to gender although gender did not influence serum GS levels; (2) follow-up information was missing for 23 age-matched (and 15 age-unmatched) control subjects and therefore undiagnosed preclinical AD in some of these age-matched controls might have been overlooked as age is one of the main risk factors for developing dementia [33]; (3) serum sample storage time correlated significantly with serum GS concentrations in control subjects; (4) simultaneous analyses of CSF biomarkers such as CSF levels of Ab that could strengthen characterization of AD patients and secondly could allow to potentially associate levels of CSF Ab with serum GS according to their functional relationship described above, were lacking. Conclusions
Association Between Serum GS Levels and Age In the AD and pooled control groups, serum GS concentrations correlated positively with age. These findings are in accordance with findings in rodents reported by Danh et al. [32] who measured GS activity in whole brain and different brain areas of 23- to 26-month-old rats as
We investigated the possible diagnostic value of serum GS levels for AD as compared to control subjects, including a large population of clinically well-characterized patients as well as age-matched and age-unmatched controls. As serum GS levels were comparable between AD patients and agematched controls, serum GS has no diagnostic value for AD.
123
1862
This finding awaits confirmation in an independent study that preferably comprises GS analyses in serum, CSF and brain tissue from one and the same study population, given the evidence for a discrepancy in diagnostic value of GS in serum, CSF and brain tissue. Additionally, simultaneous analyses of CSF Ab levels are recommended as they are likely to influence GS concentrations.
Neurochem Res (2011) 36:1858–1862
14.
15. 16.
Acknowledgments This research was supported by the Special Research Fund of the University of Antwerp; Stichting Alzheimer Onderzoek; the Thomas Riellaerts Research Fund; the Institute BornBunge; the agreement between the Institute Born-Bunge and the University of Antwerp; the central Biobank facility of the Institute Born-Bunge/University of Antwerp; Neurosearch Antwerp; the Fund for Scientific Research - Flanders (FWO-F); the Interuniversity Attraction Poles (IAP) program P6/43 of the Belgian Federal Science Policy Office; the Methusalem excellence grant of the Flemish Government, Belgium. NLB is a PhD fellow of the Research Foundation - Flanders (FWO-F). The authors acknowledge the technical assistance of E. De Leenheir, G. Van de Vijver, F. Franck and J. Luyckx, the administrative assistance of S. Hicketick, W. Wittebolle, and A. Eykens, and the clinical staff involved.
17.
References
22.
1. Meister A (1974) Glutamine synthetase of mammals. In: Boyer PD (ed) The enzymes, vol 10. Academic Press, New York, pp 699–754 2. Sua´rez I, Bodega G, Ferna´ndez B (2002) Glutamine synthetase in brain: effect of ammonia. Neurochem Int 41:123–142 3. Finch CE, Cohen DM (1997) Aging, metabolism, and Alzheimer disease: review and hypotheses. Exp Neurol 143:82–102 4. Smith CD, Carney JM, Starke-Reed PE et al (1991) Excess brain protein oxidation and enzyme dysfunction in normal aging and Alzheimer disease. Proc Natl Acad Sci USA 88:10540–10543 5. Le Prince G, Delaere P, Fages C et al (1995) Glutamine synthetase (GS) expression is reduced in senile dementia of the Alzheimer type. Neurochem Res 20:859–862 6. Aksenov MY, Aksenova MV, Harris ME et al (1995) Enhancement of beta-amyloid A beta(1–40)-mediated neurotoxicity by glutamine synthetase. J Neurochem 65:1899–1902 7. Aksenov MY, Aksenova MV, Butterfield DA et al (1996) Glutamine synthetase-induced enhancement of beta-amyloid peptide A beta(1–40) neurotoxicity accompanied by abrogation of fibril formation and A beta fragmentation. J Neurochem 66:2050–2056 8. Aksenov MY, Aksenova MV, Carney JM et al (1997) Oxidative modification of glutamine synthetase by amyloid beta peptide. Free Radic Res 27:267–281 9. Varadarajan S, Yatin S, Aksenova M et al (2000) Review: Alzheimer’s amyloid b-peptide-associated free radical oxidative stress and neurotoxicity. J Struct Biol 130:184–208 10. Poon HF, Joshi G, Sultana R et al (2004) Antisense directed at the Abeta region of APP decreases brain oxidative markers in aged senescence accelerated mice. Brain Res 1018:86–96 11. Sato E, Oda N, Ozaki N et al (1996) Early and transient increase in oxidative stress in the cerebral cortex of senescence-accelerated mouse. Mech Ageing Dev 86:105–114 12. Contestabile A, Fila T, Bartesaghi R et al (2006) Choline acetyltransferase activity at different ages in brain of Ts65Dn mice, an animal model for Down’s syndrome and related neurodegenerative diseases. J Neurochem 97:515–526 13. Butterfield DA, Hensley K, Cole P et al (1997) Oxidatively induced structural alteration of glutamine synthetase assessed by
123
18. 19.
20.
21.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
analysis of spin label incorporation kinetics: relevance to Alzheimer’s disease. J Neurochem 68:2451–2457 Gunnersen D, Haley B (1992) Detection of glutamine synthetase in the cerebrospinal fluid of Alzheimer diseased patients: a potential diagnostic biochemical marker. Biochemistry 89:11949–11953 Tumani H, Shen G, Peter JB et al (1999) Glutamine synthetase in cerebrospinal fluid, serum, and brain. Arch Neurol 56:1241–1246 Takahashi M, Stanton E, Moreno JI et al (2002) Immunoassay for serum glutamine synthetase in serum: development, reference values, and preliminary study in dementias. Clin Chem 48:375–378 Burbaeva GS, Boksha IS, Tereshkina EB et al (2005) Glutamate metabolizing enzymes in prefrontal cortex of Alzheimer’s disease patients. Neurochem Res 30:1443–1451 Robinson SR (2001) Changes in the cellular distribution of glutamine synthetase in Alzheimer’s disease. J Neurosci Res 66:972–980 Kish S, Chang L, Dixon L et al (1994) Cerebellar glutamate metabolizing enzymes in spinocerebellar ataxia type I. Metab Brain Dis 9:97–103 Lavoie J, Giguere J, Layrargues G et al (1987) Activities of neuronal and astrocytic marker enzymes in autopsied brain tissue from patients with hepatic encephalopathy. Metab Brain Dis 2:283–290 Aluise CD, Sowell RA, Butterfield DA (2008) Peptides and proteins in plasma and cerebrospinal fluid as biomarkers for the prediction, diagnosis, and monitoring of therapeutic efficacy of Alzheimer’s disease. Biochm Biophys Acta 1782:549–558 Le Bastard N, Aerts L, Leurs J et al (2009) No correlation between time-linked plasma and CSF Ab levels. Neurochem Int 55:820–825 Le Bastard N, Leurs J, Blomme W et al (2010) Plasma amyloidbeta forms in Alzheimer’s disease and non-Alzheimer’s disease patients. J Alzheimers Dis 21:291–301 Schneider P, Hampel H, Buerger K (2009) Biological marker candidates of Alzheimer’s disease in blood, plasma, and serum. CNS Neurosci Ther 15:358–374 McKhann G, Drachman D, Folstein M et al (1984) Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA work group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s disease. Neurology 34:939–944 American Psychiatric Association (1994) DSM-IV: diagnostic and statistical manual of mental disorders, 4th edn. American Psychiatric Association, Washington Roma´n GC, Tatemichi TK, Erkinjuntti T et al (1993) Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN international workshop. Neurology 43:250–260 Folstein M, Folstein S, McHugh PR (1975) Mini-Mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189–196 Engelborghs S, Sleegers K, Cras P et al (2007) No association of CSF biomarkers with APOEe4, plaque and tangle burden in definite Alzheimer’s disease. Brain 130:2320–2326 Andreasen N, Blennow K (2005) CSF biomarkers for mild cognitive impairment and early Alzheimer’s disease. Clin Neurol Neurosurg 107:165–173 Fei M, Jianghua W, Rujuan M et al. (2011) The relationship of plasma Ab levels to dementia in aging individuals with mild cognitive impairment. J Neurol Sci doi:10.1016/j.jns.2011.03.005 Danh C, Benedetti S, Dostert P (1985) Age-related changes in glutamine synthetase activity of rat brain, liver and heart. Gerontology 31:95–100 Launer LJ, Andersen K, Dewey ME et al (1999) Rates and risk factors for dementia and Alzheimer’s disease: results from EURODEM pooled analyses. EURODEM incidence research group and work groups. European studies of dementia. Neurology 52:78–84
