JOURNAL OF NEUROCHEMISTRY
| 2009 | 110 | 1400–1408
,1,2
doi: 10.1111/j.1471-4159.2009.06233.x
,1,2
,
,
,2
, ,2
,2
,2
,2
,2 ,
, ,
,2
,2
*Clinical Neurochemistry, National Parkinson Foundation Centre of Excellence Research Laboratory, Clinic and Policlinic for Psychiatry, Psychosomatic and Psychotherapy, University of Wu¨rzburg, Wu¨rzburg, Germany Department of Neuropathology, Institute of Pathology, University of Wu¨rzburg, Wu¨rzburg, Germany àDepartment of Psychiatry and Psychotherapy, RWTH Aachen University, Aachen, Germany §Department of Clinical Medicine, Kuopio University, Kuopio, Finland ¶Institut de Neuropathologia, Universitat de Barcelona, Barcelona, Spain **Department of Clinical Neuropathology, London Institute of Psychiatry, London, UK Institute of Neuropathology, University Hospital, Mu¨nster, Germany ààClinic of Psychiatry and Psychotherapy, Georg-August-University, Go¨ttingen, Germany §§Institute of Brain Research (Neuropathology), Eberhard Karls University of Tu¨bingen, Tu¨bingen, Germany ¶¶Department of Anatomical Pathology, Monash University, The Alfred Hospital, Prahran, Victoria, Australia ***National Neural Tissue Resource Centre, Australian Brain Bank Network, Melbourne, Victoria, Australia Prince of Wales Medical Research Institute and the University of New South Wales, Sydney, New South Wales, Australia àààNew South Wales Tissue Resource Centre, University of Sydney, Sydney, New South Wales, Asutralia §§§Department of Genetics and Pathology, Uppsala University, Uppsala, Sweden
Abstract Postmortem human brain tissue is widely used in neuroscience research, but use of tissue originating from different brain bank centers is considered inaccurate because of possible heterogeneity in sample quality. There is thus a need for well-characterized markers to assess the quality of postmortem brain tissue. Toward this aim, we determined tryptophan (TRP) concentrations, phosphofructokinase-1 and glutamate decarboxylase activities in 119 brain tissue samples. These neurochemical parameters were tested in samples from autopsied individuals, including control and pathological cases provided by 10 different brain bank cen-
ters. Parameters were assessed for correlation with agonal state, postmortem interval, age and gender, brain region, preservation and freezing methods, storage conditions and storage time, RNA integrity, and tissue pH value. TRP concentrations were elevated significantly (p = 0.045) with increased postmortem interval; which might indicate increased protein degradation. Therefore, TRP concentration might be one useful and convenient marker for estimating the quality of human postmortem brain tissue. Keywords: brain, glutamate decarboxylase, phosphofructokinase-1, postmortem, tryptophan. J. Neurochem. (2009) 110, 1400–1408.
Received May 12, 2009; accepted June 5, 2009. Address correspondence and reprint requests to Edna Gru¨nblatt, PhD, Neurochemistry Laboratory, Clinic and Policlinic for Psychiatry, Psychosomatic and Psychotherapy, University of Wu¨rzburg, Fu¨chsleinstr. 15, Wu¨rzburg D-97080, Germany. E-mail:
[email protected] 1 Both these authors contributed equally to this study.
2 All these authors are Member of BrainNet Europe II Consortium (BrainNet Europe; http://www.brainnet-europe.org). Abbreviations used: BBC, brain bank centers; GAD, glutamate decarboxylase; NP, neuropathology; PFK, phosphofructokinase-1; PMI, postmortem interval; PSY, psychiatry; RIN, RNA integrity number; TRP, tryptophan.
1400
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
Quality parameters in postmortem brain | 1401
Postmortem human brain tissue is being used for quantifying cellular and molecular markers of neural processes with the goal of better understanding the changes in the brain caused by CNS diseases (Gru¨nblatt et al. 2004, 2007; Reif et al. 2006; Bell et al. 2008; Schmitt et al. 2008). Because of the inaccessibility of the human brain and its poor unavailability for biopsy in non-malignant disorders (e.g. neurodegenerative diseases), fresh CNS tissue is rarely obtainable. Consequently, scientists studying human brain diseases rely heavily on the use of postmortem tissue. Although studies using human tissue have been performed for decades, today’s molecular assays are more sensitive and specific, requiring high-quality tissue samples (Rinne et al. 1974; Riederer and Wuketich 1976). Tissues must be carefully collected and handled; in addition, researchers need to establish standardized measures of tissue quality to ensure reliable research results (Bell et al. 2008; Kretzschmar 2009). Factors during the antemortem period, such as age, gender, medication, agonal state, and duration, as well as the tissue preparation methods and storage conditions, may affect postmortem brain tissue quality, e.g. in the case of neuropsychiatric disorders. New research techniques such as gene expression profiling and proteomics are providing exciting new avenues for research on human subjects (Gru¨nblatt et al. 2004; Clark et al. 2006; Sorolla et al. 2008). However, the extent to which the results of these techniques are affected by the postmortem interval (PMI; the interval between the death of a patient and the removal of the brain or other tissue prior to freezing or fixation), agonal state, refrigeration interval (the interval between the death of a patient and refrigeration of the body, typically to 4°C), preservation techniques, freezing techniques, packaging, and duration of storage until use has yet to be fully explored. Autopsy rates are declining worldwide, and it is increasingly difficult for scientists to collect the large number of postmortem brain tissue samples needed for statistical analysis. In addition to using samples from patients whose characteristics are already quite variable, it is sometimes necessary to use samples from different brain bank centers (BBCs) which may employ different isolation, preservation or storage standards (Bell et al. 2008; Sheedy et al. 2008). Therefore, in addition to standardizing methods for use in all BBCs, neuroscientists urgently need quality control parameters to assess postmortem brain tissue quality. This may help reduce the heterogeneity among samples that is currently a persistent problem when studying human postmortem brains. Therefore, we investigated whether glutamate decarboxylase (GAD; EC 4.1.1.15) and phosphofructokinase-1 (PFK; EC 2.7.1.11) enzyme activity and tryptophan [TRP; Chemical Abstracts Service (CAS) number 73-22-3] concentration, or tissue pH, could be used as additional quality control parameters for postmortem brain tissue. The activity of GAD, the enzyme catalyzing the formation of GABA
from glutamate, has been reported to decrease as a consequence of prolonged agonal state and prolonged PMI (Perry et al. 1977; Spokes 1979; Martin et al. 2003). PFK plays a key role regulating carbohydrate metabolism, and its activity can be used as an indicator of glycolytic flux. PFK activity decreases with patient age as well as with prolonged agonal state (Iwangoff et al. 1980; Perry et al. 1982). TRP is one of the 20 standard amino acids, as well as an essential amino acid in the human diet. It is a both a building block in protein biosynthesis and a biochemical precursor for the biogenic amines, including serotonin, which is synthesized via tryptophan hydroxylase. Previous study of the influence of agonal state on all amino acids resulted in significant increase in only TRP with prolonged agonal state (Perry 1983). Studies using postmortem brains have shown that TRP decreased nominally in amygdala of schizophrenic subjects (Korpi et al. 1987), while TRP infusion in depressed subjects increased TRP levels in CSF and frontal cortex biopsies (Gillman et al. 1981). TRP levels were shown to increase as a result of PMI as well as with sample storage time (Perry et al. 1982; McIntyre and Stanley 1984; Ka¨rkela¨ and Scheinin 1992). Several reports indicate that pH (Hynd et al. 2003; Stan et al. 2006; Monoranu et al. 2009) and RNA quality (Barton et al. 1993; Bezchlibnyk et al. 2001; Hynd et al. 2003; Hunsucker et al. 2008) are indicators of postmortem brain tissue quality; here, we tested their influence on the other three parameters. For that purpose, we analyzed these three markers in 88 controls without any neurological disorders and additionally in pathological (e.g. infarct) postmortem brain tissue samples provided by 10 different BBCs. The three parameters were studied for their association with agonal state, PMI, age, and gender, as well as with the brain region from which the sample was taken, preservation and freezing method, sample storage conditions, and tissue pH value as well as RNA quality.
Materials and methods Brain tissue samples Postmortem human brain tissue was obtained from six European BBCs organized in BrainNet Europe II: Barcelona, London, Kuopio, Wu¨rzburg Neuropathology (NP) together with Wu¨rzburg Psychiatry (PSY) and Go¨ttingen (Bell et al. 2008). The NP departments of the Universities of Tu¨bingen and Mu¨nster (both members of BrainNet Germany) also provided cases for this study. The Australian BBCs in Sydney and Victoria contributed cases in which the agonal state has been well documented (Sarris et al. 2002). Whole human brains were obtained with the consent of next of kin according to the guidelines of the NIH Guide for the Care and Use of Laboratory human tissue and were approved by the national and local ethics committees. Cause of death was determined from clinical notes, the autopsy report, or the death certificate. In preparation for routine neuropathological examination, the brain was divided midsagittally, and one hemisphere was immersed in 4.5% p-formaldehyde (Fischar GmbH, Saarbru¨cken, Germany)
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
1402 | E. Gru¨nblatt et al.
for 3–4 weeks. The remaining half was cryopreserved following coronal slicing for cortical sampling, sagittal slicing for sampling the cerebellum and transverse sectioning of the brainstem. Tissue samples were snap frozen on brass plates, cooled in dry ice, and stored at )80°C until requested for experimental use. A neuropathological assessment was completed by macroscopic and histological examination. Because the tissue was obtained from different institutions, detailed information concerning tissue preparation and storage was collected. All samples described in this study were shipped on dry ice in frozen state without any defrost phase, and were processed and analyzed for all measurements in the neurochemistry laboratory of the Clinic for PSY in Wu¨rzburg. Eighty-eight control cases were available (Table S1), with ‘control’ defined as no history of any neurological or neuropsychiatric disease plus confirmation by histopathological examination. The gender distribution was balanced for control cases, with samples from 42 males and 46 females. Thirty-one pathological cases were available (19 males and 12 females); these included 14 cases with global brain ischemia (condition after resuscitation), four cases with regional ischemia (brain infarct), seven cases with Alzheimer’s disease, and six cases with schizophrenia. Both acute and subacute infarcts were included, as well as hemorrhagic infarcts; in some cases only tissue from the border area of the lesion (penumbra) was available (Table S1). For this study, frozen control samples from the frontal lobe (n = 65), cingulate gyrus (n = 61), striatum (n = 70), and cerebellum (n = 56) were available in most cases (hippocampus, n = 4; globus pallidus, n = 7). In this study, we did not differentiate between white and gray matter, as the aim of the study was finding any confounders for postmortem quality control. After determining that brain region had no effect on TRP levels or on GAD or PFK activity, the mean values for each case were calculated using measurements from the different brain regions. Detailed information about age (ranging from 0 to 98 years), PMI (range from 2 to 109 h) and storage time (range from 0 to 156 months) was provided for all cases. Sample characteristics are summarized in Table S2. Agonal state scoring A 3-point scale was used to classify the agonal state of each tissue donor as follows (adapted from Hardy et al. 1985). A score of 1: Rapid death because of violent or natural causes i.e. the sudden, unexpected death of a person who appeared to be reasonably healthy. The most frequent cause of death in this group was myocardial infarction. A score of 2: Intermediate agonal state, as when a person who was ill unexpectedly died. Deaths with a score of 2 were neither sudden nor slow. A score of 3: Slow death, such as a death after a long illness with a prolonged terminal phase. These patients typically died from cancer, cerebrovascular disease, or bronchopneumonia. A clinical record was available in most cases, and the agonal state was retrospectively graded. The agonal state was well documented in samples from the Australian BBCs. Evaluation of postmortem factors In addition to the clinical and neuropathological diagnosis, we assessed PMI data (the interval between death and freezing), which is frequently utilized as a measure of the quality of postmortem brain specimens. A related measure, storage time, is the interval from freezing the tissue at )80°C until use in the experiment. As we had
samples from several BBCs, we were interested in knowing precisely how tissue was prepared and stored. We gave a questionnaire to each BBC to obtain this information. Concerning the storage temperature until dissection, the Mu¨nster BBC and the Tu¨bingen BBC kept the body at 18°C; the remaining eight BBCs stored it at 4°C. The Mannheim BBC and the Mu¨nster BBC used liquid nitrogen for freezing the tissue, while the others used dry ice. Packaging during storage could also influence tissue quality. Only the Wu¨rzburg BBC and the Barcelona BBC used aluminum foil; in contrast, the other eight BBCs used plastic bags. Tissue pH measurements Tissue pH values were determined for fresh dissected tissue or for thawed frozen blocks of tissue using a pH meter as previously described (Monoranu et al. 2009). Tryptophan concentration determination Deep-frozen tissue was weighed and homogenized with 0.1 N HCl (1 : 19 w/v) on ice using sonifier (Branson-Digital Sonifier, G. Heinemann, Schwa¨bisch-Gmu¨nd, Germany). The homogenate was centrifuged for 20 min at 36 721 g at 4°C. TRP content was determined in the supernatant (10 lL) using HPLC Agilent 1100Serie (Waldbronn, Germany). The TRP measurements were conducted according to the method described previously for tissue-free amino acid determination (Graser et al. 1985) with slight modifications. For the fluorometric measurements, a 3-lm 3-EC 150/4.6 Nucleodur C18 Gravity column (Machery-Nagel, Du¨ren, Germany) was used, with 330 nm excitation and 450 nm emission. The gradient mobile phases consisted of mobile phase 1 (25 mM phosphate buffer, pH 6.8, and 0.5% tetrahydrofurane) and mobile phase 2 (55% 40 mM phosphate buffer, pH 6.8, 27% methanol, and 18% acetonitrile). The flow was set to 0.90 mL/min at 30°C. TRP tissue levels were calculated for all samples in micromolar concentration (all samples were of 52.63 mg tissue/mL). Glutamate decarboxylase activity assay Glutamate decarboxylase activity was determined by measuring the production of radiolabeled carbon dioxide from 14C-glutamate as described previously (Chalmers et al. 1970; Geigerseder et al. 2003). In brief, deep-frozen brain tissue was weighed and homogenized on ice with buffer (1 mg: 4 lL buffer) containing 0.1 M potassium phosphate buffer, pH 7.1, 0.5% Triton X-100, 200 mM 2-(2-aminoethyl)-2-thiopseudourea dihydrobromide and CompleteÔ Mini protease inhibitors (Roche Applied Science, Roche Diagnostics GmbH, Mannheim, Germany) using a glassTeflon potter. The homogenate was then centrifuged at 2543 g for 2 min at 4°C and the supernatant was assayed. The total protein content was determined using the Bradford assay (Bradford 1976). All samples were pre-adjusted to 3–4 lg protein/lL using homogenization buffer. The assay was performed in a total reaction volume of 60 lL, with 20 lL of sample plus 30 lL of 2x incubation mix (0.5 M EDTA, 0.5% Triton X-100, 1 M dithiothreitol, 40 mM L-glutamate, 60 mM potassium phosphate buffer, pH 7.1, and 8.1 mM pyridoxalphosphate) plus 10 lL neutralized 14C-L-glutamate mix (50 lCi 14C-L-glutamate in 0.1 N HCl and 0.1 M potassium hydroxide). The reaction mix was incubated for 1 h at 37°C and stopped by the addition of 200 lL of 10% trichloroacetic acid. Released CO2 was absorbed on benzethonium
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
Quality parameters in postmortem brain | 1403
hydroxide-drenched filter disks, and filter-bound radioactivity was determined using a LS-6500 Multi-purpose liquid scintillation counter (Beckman CoulterÔ GmbH, Krefeld, Germany). Control reactions lacking radioactive glutamate were performed in parallel. For measurement of total radioactivity, the reaction consisted of 10 lL neutralized 14C-L-glutamate mix put directly into scintillation tube. Assays were conducted in triplicate for each sample, and results were normalized for the total protein content measured via Bradford. Enzyme activity was expressed as nmol/mg protein/h. Phosphofructokinase-1 activity assay Determination of PFK activity by measuring the production of radiolabeled D-fructose-1,6-biphosphate from [c-32P]-ATP was performed as described previously (Sola-Penna et al. 2002). In brief, deep-frozen brain tissue was weighed and homogenized on ice (1 mg: 10 lL) with 50 mM Tris–HCl buffer, pH 7.4, 5 mM magnesium chloride, and 5 mM ammonium sulfate containing CompleteÔ Mini protease inhibitors (Roche Applied Science, Roche Diagnostics GmbH) using a 5-mL glass-Teflon potter. The homogenate was then centrifuged at 36 721 g for 20 min at 4°C. Total protein content in the supernatant was determined using the Bradford assay (Bradford 1976). All samples were adjusted to 100 lg protein/320 lL using homogenization buffer. Each assay was performed in a total volume of 400 lL containing 50 mM Tris– HCl, pH 7.4, 5 mM magnesium chloride, 5 mM ammonium sulfate, 1 mM fructose-6-phosphate, and 0.1 mM [c-32P]-ATP (4 lCi/lmol) plus 320 lL supernatant (100 lg protein). The reaction was incubated for 3 min at 37°C and then stopped by the addition of 1 mL activated charcoal suspended in 0.1 N HCl (25 g activated charcoal per 100 mL 0.1 N HCl). The suspension was centrifuged at 4465 g for 10 min at 4°C. Supernatant (400 lL) was added to 4 mL scintillation fluid and counted in a LS-6500 Multi-purpose liquid scintillation counter (Beckman CoulterÔ GmbH). Blank reactions including radioactivity were performed in parallel by adding directly to the reaction activated charcoal. For measurement of total radioactivity, the reaction without activated charcoal was inserted into a scintillation tube. Assays were conducted in triplicate for each sample, and results were normalized for total protein content measured via Bradford. Enzyme activity was expressed as nmol fructose-1,6-bisphosphate/mg protein/min. RNA quality analysis Total RNA from samples with high/low TRP (n = 6 and 6, respectively) levels was isolated using RNeasy Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer’s instructions. RNA purity was assessed using the A260/A280 ratio (A260 = absorbance at 260 nm and A280 = absorbance at 280 nm). RNA integrity was analyzed on an Agilent 2100 Bioanalyzer, which also automatically calculates the sample concentration, the 28S/18S ribosomal ratio, and the RNA integrity number (RIN). RIN values lower than 7 are usually regarded as degraded RNA/low RNA quality, while values over 7 are regarded as good RNA quality. Statistical methods After finding that brain region had no effect on TRP levels or GAD or PFK activity, the mean values for these parameters were calculated for each case using the measurements from all six brain regions. Further statistical analysis was conducted using these new
mean values for each case (data not shown). Log transformation was applied to GAD and PFK values, as they were approximately lognormally distributed. Data were reported as the mean ± SD. The t-test was used to compare two groups with normal distribution. Otherwise, the Mann–Whitney test was used. For comparison of more than two groups, the non-parametrical Kruskal– Wallis test was used. Correlations were calculated using Spearman’s correlation coefficient. Values of p < 0.05 were considered statistically significant. Variance analysis and multivariate analysis were not possible as the data were not normally distributed or the sample size was too small. The analysis was conducted using SPSS 15.0 software (SPSS Inc. Headquarters, Chicago, IL, USA).
Results Postmortem tissue characteristics: tissue source and diagnosis We analyzed tissue samples from 119 individuals aged 0–98 years (mean age, 64.40 ± 23.49 years). The characteristics of the cases and parameters used to assess tissue quality (e.g. storage time and PMI) are shown in Table S2, with additional characteristics shown in Table S1. The PMI ranged from 2.5 to 109 h (mean PMI, 28.58 ± 21.83 h) and storage time ranged from 0 to 156 months (mean storage time, 59.57 ± 43.98 months). Tryptophan, PFK, and GAD in different brain regions No difference was found in TRP levels or PFK or GAD activity in the different brain regions in control cases (Table S3). Tryptophan, PFK, and GAD in postmortem brains from different BBCs In control postmortem brain samples, the mean TRP values from Wu¨rzburg Department of NP (n = 16, 46.1 ± 12.2 lM) were significantly lower (Kruskal–Wallis test, p < 0.001) than those from other BBCs (Table S4). PFK activity (log value) was lower in both the Wu¨rzburg NP BBC (n = 16, 4.3 ± 0.3 Log(nmol fructose-1,6-bis-P/mg protein/min)) and the Wu¨rzburg Clinic for PSY BBC (n = 5, 3.8 ± 0.5) compared with the other BBCs; this decrease was significant only for the Wu¨rzburg NP BBC (Kruskal–Wallis test, p < 0.01). No difference was found in GAD activity in the different BBCs. Correlation of gender and age with TRP, PFK, and GAD Glutamate decarboxylase activity in the control samples differed significantly in men and women (Mann–Whitney test, p = 0.002): GAD activity was higher in men (0.87 ± 0.86 Log(nmol/mg protein/h)) than in women (0.33 ± 0.74; Table S5). No difference was observed in the two genders for TRP levels or PFK activity. Regarding age as an influencing factor, there was a tendency toward a negative correlation
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
1404 | E. Gru¨nblatt et al.
between GAD activity and age, with GAD activity decreasing with age (Spearman’s correlation coefficient rho = )0.253, p = 0.018). There was no correlation between TRP levels or PFK activity and age in the control samples (Fig. S1). Correlation of tissue handling and storage with TRP, PFK, and GAD We found that the body storage temperature had no effect on these parameters (Table S5). However, freezing tissue using dry ice resulted in significantly lower TRP levels (Mann– Whitney test, p = 0.003; n = 65, 86.7 ± 34.6 lM) and significantly higher GAD activity (Mann–Whitney test, p = 0.020; n = 65, 0.70 ± 0.87 Log(nmol/mg protein/h)) compared with freezing with liquid nitrogen (Fig. S2a,b and Table S5). The packaging used to store the frozen tissue influenced both TRP levels and PFK activity. TRP levels were significantly lower (Mann–Whitney test, p < 0.001) when aluminum foil was used (n = 32, 70.61 ± 32.91 lM) compared with plastic bag use (n = 56, 106.5 ± 32.1 lM), and PFK activity was also significantly lower (t-test, p < 0.001) using aluminum foil [n = 32, 4.36 ± 0.47 Log(nmol fructose-1,6-bis-P/mg protein/min)] compared with plastic bags [n = 56, 4.81 ± 0.53 Log(nmol fructose-1,6-bis-P/mg protein/min)]. In contrast, GAD activity was not affected by the packaging material (Fig. S2c,d and Table S5). Correlation of PMI and storage time with TRP, PFK, and GAD Tryptophan levels and PFK and GAD activity showed no correlation with storage time, although we observed a tendency for TRP levels to increase with increased storage time (Fig. 1a–c). In contrast, TRP levels and PMI were significantly correlated (Spearman’s correlation coefficient rho = 0.216, p = 0.045), with TRP levels increasing with increasing PMI (Fig. 1d). The two other parameters, PFK and GAD activity, did not correlate with PMI (Fig. 1e and f). Correlation of agonal state with TRP, PFK, and GAD No significant difference was observed in TRP levels or PFK and GAD activity in the three agonal state groups in control samples (Table 1). Correlation of pathology with TRP, PFK, and GAD No significant difference was observed in TRP levels or PFK and GAD activity in postmortem brain tissue from individuals with global ischemia versus controls (Table 2). Correlation of TRP, PFK, and GAD with pH and with each other We performed a correlation analysis of pH levels with the three biochemical parameters. Both PFK and GAD activities were positively correlated with pH (rho = 0.327, p = 0.002
and rho = 0.537, p < 0.001, respectively) (Fig. S1). Thus, increased enzyme activity was observed with increased pH values (less acidic, pH 6.8). TRP levels did not correlate with pH values (rho = 0.076, p = 0.490) (Fig. S3). We also tested whether our measured neurochemical parameters correlated with each other. We found a slight but significant correlation between PFK activity and GAD activity (Spearman’s correlation coefficient rho = 0.249, p = 0.019), showing increased PFK activity with increased GAD activity (Fig. S3). TRP showed no correlation whatsoever with PFK or GAD activity (Fig. S3). Correlation of TRP and pH with RNA integrity Correlation analysis of TRP and pH with RIN values and the ratio of 28S/18S ribosomal were conducted in brain tissues of high and low levels of TRP samples. In addition, the RIN and 28S/18S ribosomal ratio were tested for their correlation with storage time and PMI. No correlation, whatsoever, was found between TRP or pH and the integrity of the RNA isolated from these samples (Table S6). Neither the RNA integrity showed any correlation to storage time nor PMI (Table S6).
Discussion The only parameter that showed promise as a neurochemical marker of postmortem tissue quality was the level of TRP. PMI seems to play a crucial role in many types of protein preservation, as TRP levels increased significantly with increased PMI (Fig. 1). This finding was in agreement with previous report (Ka¨rkela¨ and Scheinin 1992), who found an increase in TRP levels in postmortem human cisternal fluid with increased PMI. Increase TRP levels in postmortem tissue might indicate protein degradation and enzyme activity alterations, as previously reported (Birkmayer and Riederer 1980; Ferrer et al. 2007). Still not all proteins degrade in similar manner (Omalu et al. 2005), therefore scientist should test such factors ahead. TRP levels were shown to alter in infarct, hepatic failure, and schizophrenia (Curzon et al. 1973; Riederer et al. 1981; Jellinger and Riederer 1983; Korpi et al. 1987). Although we found no influence of pathological factors on TRP, any study of protein/enzymes should be critical. Storage duration, age, gender, pathological state, or agonal state did not affect TRP levels in this study, in contrast to some previous reports (Perry et al. 1982; Ka¨rkela and Scheinin 1992; Myint et al. 2007; Girela et al. 2008). One possibility to the inconsistency of results with other reports is the use of a sensitive method for the TRP determination (see Materials and methods). TRP levels were significantly influenced by the freezing and packaging methods: lower TRP levels were found using dry ice to freeze the sample and aluminum foil for storage which were used in some BBC in this study. We suggest the use of dry ice and aluminum to preserve the quality of the tissue and for protein preservation in particular, but these points must be
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
Quality parameters in postmortem brain | 1405
(b) 7
(c) 3
6
2
PFK (log)
200 150 100 50 0
5 4 3
0 20 40 60 80 100 120 140
GAD (log)
Tryptophan (µM)
(a) 250
(e)
200
PFK (log)
Tryptophan (µM)
250
150 100
Storage time (months)
(f)
7
3
6
2
5 4
50 0 0
20
40
60
80 100 120
0 20 40 60 80 100 120 140
Storage time (months)
GAD (log)
(d)
0 –1
0 20 40 60 80 100 120 140
Storage time (months)
1
3
0
20
40
PMI (h)
60
80 100 120
PMI (h)
1 0 –1
0
20
40
60
80 100 120
PMI (h)
Fig. 1 Correlation analysis of TRP levels, PFK activity, and GAD activity with storage time and PMI in control postmortem brain tissue samples. Spearman-Rank correlation coefficient analysis was conducted for all parameters. (a) rho = 0.120, p = 0.264 for TRP and storage time; (b) rho = )0.164, p = 0.127 for PFK and storage time;
(c) rho = )0.047, p = 0.663 for GAD and storage time, (d) rho = 0.216, p = 0.045 for TRP and PMI; (e) rho = )0.051, p = 0.641 for PFK and PMI; and (f) rho = )0.010, p = 0.929 for GAD and PMI. PFK, phosphofructokinase-1; GAD, glutamate decarboxylase; PMI, postmortem interval.
further investigated. TRP measurements can thus be used as one quality control parameter for postmortem brain tissue. Additionally, assessment of TRP levels for the use of quality assessment of the tissue in regard to PMI is independent of all other parameters (Fig. S3). In contrast to previous reports concerning PFK activity in postmortem brains (Perry et al. 1982; Bigl et al. 1996, 2000), we did not find any pre- or postmortem factors that caused major alterations in PFK enzyme activity. PFK activity was significantly lower when aluminum foil rather than a plastic bag was used to package the frozen tissue (Fig. S2 and Table S5). In addition, it seems that PFK activity is influenced by tissue pH (Fig. S3), which was previously found to be influenced by the agonal state (Monoranu et al. 2009). Besides, the level of PFK activity correlated significantly with GAD activity; this, too, might be because of the influence of tissue pH.
Many studies have identified GAD activity as important indicator of postmortem brain tissue quality (Perry et al. 1982; Martin et al. 2003; Girela et al. 2008). GAD is involved in the synthesis of GABA and is therefore involved in controlling neurotransmitter levels. This study did not confirm previous findings that alterations in GAD activity correlated with PMI (Ritchie et al. 1986; Dodd et al. 1988; Martin et al. 2003), storage duration and temperature (Puymirat et al. 1979), agonal state (Bowen et al. 1976; Spokes 1979; Maker et al. 1981; Perry et al. 1982; Martin 1986; Dodd et al. 1988), or tissue pathology (Lloyd et al.
Table 1 TRP levels, PFK activity, and GAD activity in control postmortem brain tissue samples from donors with different agonal states
TRP (lM)
PFK activity (Log)
GAD activity (Log)
Agonal state (n)
Mean ± SD
Mean ± SD
Mean ± SD
Rapid (42) Intermediate (18) Prolonged (20)
87.41 ± 38.73 104.18 ± 28.83 98.52 ± 37.52
4.80 ± 0.56 4.60 ± 0.50 4.57 ± 0.43
0.74 ± 0.92 0.59 ± 0.69 0.37 ± 0.80
Log transformation was applied to GAD and PFK values as they were approximately lognormal distributed. As the sample number was low, the non-parametrical Kruskal–Wallis test was used. TRP, tryptophan; GAD, glutamate decarboxylase; PFK, phosphofructokinase-1.
Table 2 TRP concentration, PFK activity, and GAD activity in postmortem brain control and pathological tissue samples
TRP (lM)
PFK activity (Log)
GAD activity (Log)
Group (n)
Mean ± SD
Mean ± SD
Mean ± SD
Control (88) Infarct (4) Global ischemia (14) Schizophrenia (6) AD (7)
93.48 ± 36.57 104.58 ± 39.09 101.48 ± 31.92
0.59 ± 0.84 0.45 ± 0.70 0.23 ± 0.62
80.63 ± 38.92
4.65 ± 0.55 4.71 ± 0.47 4.48 ± 0.77 (n = 13) 4.40 ± 0.25
71.91 ± 35.06
3.89 ± 1.46
)0.15 ± 0.89
0.78 ± 0.28
Log transformation was applied to the GAD and PFK values, as they were approximately log-normally distributed. As the sample number was low, the non-parametrical Mann–Whitney test was used to compare the control and global ischemia groups. BBC, Brain Bank Center; TRP, tryptophan; GAD, glutamate decarboxylase; PFK, phosphofructokinase-1; AD, Alzheimer’s disease.
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
1406 | E. Gru¨nblatt et al.
1976; Martin 1986). Similar to our results, Spokes (1979) also found a weak correlation between GAD activity and donor age in which GAD activity tended to decrease with increasing age. We found that GAD activity in all regions of the brain was significantly higher in tissues from males than from females (Table S5). The freezing method significantly influenced GAD activity (Fig. S2 and Table S5): GAD activity was higher when dry ice rather than liquid nitrogen was used. In addition, we found a significant correlation between GAD activity and tissue pH (Fig. S3). As with PFK activity, this may bias the study. Many studies discuss the importance of RNA integrity in postmortem brains and its influence on gene expression studies (Barton et al. 1993; Hynd et al. 2003; Tomita et al. 2004; Barrachina et al. 2006; Ervin et al. 2007; Chevyreva et al. 2008; Ferrer et al. 2008; Popova et al. 2008). Although RNA integrity was not the focus of this study, we could not find any correlation between RNA quality measured by RIN values and TRP levels. This finding, points to the fact that not only molecular parameters are important, but also biochemical parameters such as TRP changes indicating postmortem tissue quality are of importance. In addition, although the sample size was rather small, we found no correlation between RNA quality and pH, PMI, and storage time. This finding is in accordance to several reports indicating no correlation to these parameters (Stan et al. 2006; Ervin et al. 2007; Ferrer et al. 2008; Popova et al. 2008). In conclusion, our study reveals that TRP levels in postmortem brain tissue depend on factors associated with tissue handling and PMI and depend less on factors such as agonal state or pathological conditions. We therefore conclude that TRP levels may be useful as additional indicator that BBCs can use to estimate the quality of postmortem brain tissue they are banking. In turn, this will help neuroscience researchers produce reliable and reproducible research results.
Acknowledgements We thank all tissue donors and their families as well as to the BrainNet Europe II Consortium and the German BrainNet. In Australia, tissue samples were received from the Australian Brain Bank Network. The Network is supported by the National Health and Medical Research Council of Australia and Neurosciences Australia. We thank Prof. Dr. med. Hans A. Kretzschmar the coordinator of the BrainNet Europe II Consortium for his enormous work. We thank Prof. Gerhard Ransmayr, Prof. Kurt Jellinger, Prof. Fritz Leblhuber, Dr. Thomas Arzberger, Dr Rene Silye, Dr Susanne Kneitz, Hannelore Schraut, Rainer Burger, Monika Siemer, and Carola Gagel for technical assistance.
Funding This study was supported by the European Commission’s Sixth Framework Program (BrainNet Europe II, LSHM-CT-2004503039).
Supporting information Additional Supporting information may be found in the online version of this article: Fig. S1 Scatter plot diagram of TRP levels and PFK and GAD activity distribution according to donor age in control postmortem brain tissue samples. Fig. S2 Differences in TRP levels, PFK activity, and GAD activity in control postmortem brain tissue samples according to freezing method (a and b) and tissue packaging material (c and d). Fig. S3 Correlation analysis of TRP levels, PFK activity, and GAD activity with pH values and with each other for control postmortem brain tissue samples. Table S1 Demographic characteristics for control and pathological cases Table S2 Characteristics of the control and pathological postmortem brain tissue samples from 10 different brain bank centers. Table S3 TRP concentration, PFK activity, and GAD activity in control postmortem brain tissue samples from different brain regions.
Table S4 TRP levels, PFK activity, and GAD activity in control postmortem brain tissue samples from different BBCs. Table S5 TRP concentration, PFK activity, and GAD activity in control postmortem brain tissue samples according to gender, body storage temperature, freezing method, and tissue packaging material. Table S6 TRP concentration, pH value correlation to RNA integrity values as well as to storage time and PMI in samples with high/low levels of TRP As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
References Barrachina M., Castano E. and Ferrer I. (2006) TaqMan PCR assay in the control of RNA normalization in human post-mortem brain tissue. Neurochem. Int. 49, 276–284. Barton A. J., Pearson R. C., Najlerahim A. and Harrison P. J. (1993) Preand postmortem influences on brain RNA. J. Neurochem. 61, 1–11. Bell J. E., Alafuzoff I., Al-Sarraj S. et al. (2008) Management of a twenty-first century brain bank: experience in the BrainNet Europe consortium. Acta Neuropathol. 115, 497–507. Bezchlibnyk Y. B., Wang J. F., McQueen G. M. and Young L. T. (2001) Gene expression differences in bipolar disorder revealed by cDNA array analysis of post-mortem frontal cortex. J. Neurochem. 79, 826–834. Bigl M., Bleyl A. D., Zedlick D., Arendt T., Bigl V. and Eschrich K. (1996) Changes of activity and isozyme pattern of phosphofructokinase in the brains of patients with Alzheimer’s disease. J. Neurochem. 67, 1164–1171. Bigl M., Beck M., Bleyl A. D., Bigl V. and Eschrich K. (2000) Altered phosphofructokinase mRNA levels but unchanged isoenzyme pattern in brains from patients with Alzheimer’s disease. Brain Res. Mol. Brain Res. 76, 411–414. Birkmayer W. and Riederer P. (1980) Die Parkinson-Krankheit: Biochemie, Klinik, Therapie. Springer-Verlag, Vienna. Bowen D. M., Smith C. B., White P. and Davison A. N. (1976) Neurotransmitter-related enzymes and indices of hypoxia in senile dementia and other abiotrophies. Brain 99, 459–496.
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
Quality parameters in postmortem brain | 1407
Bradford M. M. (1976) Rapid and sensitive method for quantitation of microgram quantities of proteins utilizing principle of protein dye binding. Anal. Biochem. 72, 248–254. Chalmers A., McGeer E. G., Wickson V. and McGeer P. L. (1970) Distribution of glutamic acid decarboxylase in the brains of various mammalian species. Comp. Gen. Pharmacol. 1, 385–390. Chevyreva I., Faull R. L., Green C. R. and Nicholson L. F. (2008) Assessing RNA quality in postmortem human brain tissue. Exp. Mol. Pathol. 84, 71–77. Clark D., Dedova I., Cordwell S. and Matsumoto I. (2006) A proteome analysis of the anterior cingulate cortex gray matter in schizophrenia. Mol. Psychiatry 11, 423, 459–470. Curzon G., Kantamaneni B. D., Winch J., Rojas-Bueno A., Murray-Lyon I. M. and Williams R. (1973) Plasma and brain tryptophan changes in experimental acute hepatic failure. J. Neurochem. 21, 137–145. Dodd P. R., Hambley J. W., Cowburn R. F. and Hardy J. A. (1988) A comparison of methodologies for the study of functional transmitter neurochemistry in human brain. J. Neurochem. 50, 1333– 1345. Ervin J. F., Heinzen E. L., Cronin K. D., Goldstein D., Szymanski M. H., Burke J. R., Welsh-Bohmer K. A. and Hulette C. M. (2007) Postmortem delay has minimal effect on brain RNA integrity. J. Neuropathol. Exp. Neurol. 66, 1093–1099. Ferrer I., Santpere G., Arzberger T. et al. (2007) Brain protein preservation largely depends on the postmortem storage temperature: implications for study of proteins in human neurologic diseases and management of brain banks: a BrainNet Europe Study. J. Neuropathol. Exp. Neurol. 66, 35–46. Ferrer I., Martinez A., Boluda S., Parchi P. and Barrachina M. (2008) Brain banks: benefits, limitations and cautions concerning the use of post-mortem brain tissue for molecular studies. Cell Tissue Bank 9, 181–194. Geigerseder C., Doepner R., Thalhammer A., Frungieri M. B., GamelDidelon K., Calandra R. S., Kohn F. M. and Mayerhofer A. (2003) Evidence for a GABAergic system in rodent and human testis: local GABA production and GABA receptors. Neuroendocrinology 77, 314–323. Gillman P. K., Bartlett J. R., Bridges P. K., Hunt A., Patel A. J., Kantamaneni B. D. and Curzon G. (1981) Indolic substances in plasma, cerebrospinal fluid, and frontal cortex of human subjects infused with saline or tryptophan. J. Neurochem. 37, 410–417. Girela E., Villanueva E., Irigoyen P., Girela V., Hernandez-Cueto C. and Peinado J. M. (2008) Free amino acid concentrations in vitreous humor and cerebrospinal fluid in relation to the cause of death and postmortem interval. J. Forensic Sci. 53, 730–733. Graser T. A., Godel H. G., Albers S., Foldi P. and Furst P. (1985) An ultra rapid and sensitive high-performance liquid chromatographic method for determination of tissue and plasma free amino acids. Anal. Biochem. 151, 142–152. Gru¨nblatt E., Mandel S., Jacob-Hirsch J. et al. (2004) Gene expression profiling of parkinsonian substantia nigra pars compacta; alterations in ubiquitin-proteasome, heat shock protein, iron and oxidative stress regulated proteins, cell adhesion/cellular matrix and vesicle trafficking genes. J. Neural Transm. 111, 1543– 1573. Gru¨nblatt E., Zander N., Bartl J. et al. (2007) Comparison analysis of gene expression patterns between sporadic Alzheimer’s and Parkinson’s disease. J. Alzheimers Dis. 12, 291–311. Hardy J. A., Wester P., Winblad B., Gezelius C., Bring G. and Eriksson A. (1985) The patients dying after long terminal phase have acidotic brains; implications for biochemical measurements on autopsy tissue. J. Neural Transm. 61, 253–264. Hunsucker S. W., Solomon B., Gawryluk J., Geiger J. D., Vacano G. N., Duncan M. W. and Patterson D. (2008) Assessment of post-mor-
tem-induced changes to the mouse brain proteome. J. Neurochem. 105, 725–737. Hynd M. R., Lewohl J. M., Scott H. L. and Dodd P. R. (2003) Biochemical and molecular studies using human autopsy brain tissue. J. Neurochem. 85, 543–562. Iwangoff P., Enz A., Armbruster R., Emmenegger H., Pataki A. and Sandoz P. (1980) The influence of aging, post-mortem delay until isolation of the tissue and duration of agony on some glycolytic enzymes in human autoptic brain tissue [author’s translation]. Aktuelle Gerontol. 10, 203. Jellinger K. and Riederer P. (1983) Central neurotransmitters in cerebral ischemia and stroke [author’s translation]. Fortschr. Neurol. Psychiatr. 51, 91–123. Ka¨rkela¨ J. and Scheinin M. (1992) Tryptophan and biogenic amine metabolites in post-mortem human cisternal fluid: effects of post-mortem interval and agonal time. J. Neurol. Sci. 107, 239– 245. Korpi E. R., Kleinman J. E., Goodman S. I. and Wyatt R. J. (1987) Neurotransmitter amino acids in post-mortem brains of chronic schizophrenic patients. Psychiatry Res. 22, 291–301. Kretzschmar H. (2009) Brain banking: opportunities, challenges and meaning for the future. Nat. Rev. Neurosci. 10, 70–78. Lloyd K., Mo¨hler H., Bartholini G. and Hornykiewicz O. (1976) Pathological alterations in glutamic acid decarboxylase activity in Parkinson’s disease, in Advances in Parkinsonism, Vol. 1 (Birkmayer W. and Hornykiewicz O., eds), pp. 186–192. F. HoffmannLa Roche, Basel, Switzerland. Maker H. S., Weiss C., Weissbarth S., Silides D. J. and Whetsell W. (1981) Regional activities of metabolic enzymes and glutamate decarboxylase in human brain. Ann. Neurol. 10, 377–383. Martin D. (1986) Brain glutamate decarboxylase, in Neuromethods: Neurotransmitter Enzymes (Boulton A., Baker G. and Yu P., eds), pp. 361–388. Human Press, Clifton, NJ. Martin S. B., Waniewski R. A., Battaglioli G. and Martin D. L. (2003) Post-mortem degradation of brain glutamate decarboxylase. Neurochem. Int. 42, 549–554. McIntyre I. M. and Stanley M. (1984) Postmortem and regional changes of serotonin, 5-hydroxyindoleacetic acid, and tryptophan in brain. J. Neurochem. 42, 1588–1592. Monoranu C. M., Apfelbacher M., Gru¨nblatt E. et al. (2009) pH measurement as quality control on human postmortem brain tissue: a Study of the BrainNet Europe Consortium. Neuropathol. Appl. Neurobiol. 35, 329–337. Myint A. M., Kim Y. K., Verkerk R., Park S. H., Scharpe S., Steinbusch H. W. and Leonard B. E. (2007) Tryptophan breakdown pathway in bipolar mania. J. Affect. Disord. 102, 65–72. Omalu B. I., Mancuso J. A., Cho P. and Wecht C. H. (2005) Diagnosis of Alzheimer’s disease in an exhumed decomposed brain after twenty months of burial in a deep grave. J. Forensic Sci. 50, 1453– 1458. Perry E. (1983) Neurochemical markers in the assessment of postmortem neurobiological status, in Human Brain Dissection (Pope A., ed.), pp. 59–72. US Department of Health and Human Services, Washington, DC. Perry E. K., Gibson P. H., Blessed G., Perry R. H. and Tomlinson B. E. (1977) Neurotransmitter enzyme abnormalities in senile dementia. Choline acetyltransferase and glutamic acid decarboxylase activities in necropsy brain tissue. J. Neurol. Sci. 34, 247–265. Perry E. K., Perry R. H. and Tomlinson B. E. (1982) The influence of agonal status on some neurochemical activities of postmortem human brain tissue. Neurosci. Lett. 29, 303–307. Popova T., Mennerich D., Weith A. and Quast K. (2008) Effect of RNA quality on transcript intensity levels in microarray analysis of human post-mortem brain tissues. BMC Genomics 9, 91.
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
1408 | E. Gru¨nblatt et al.
Puymirat J., Javoy-Agid F., Gaspar P., Ploska A., Prochiantz A. and Agid Y. (1979) Post mortem stability and storage in the cold of brain enzymes. J. Neurochem. 32, 449–454. Reif A., Fritzen S., Finger M., Strobel A., Lauer M., Schmitt A. and Lesch K. P. (2006) Neural stem cell proliferation is decreased in schizophrenia, but not in depression. Mol. Psychiatry 11, 514–522. Riederer P. and Wuketich S. (1976) Time course of nigrostriatal degeneration in Parkinson’s disease. A detailed study of influential factors in human brain amine analysis. J. Neural Transm. 38, 277–301. Riederer P., Kruzik P., Kienzl E., Kleinberger G., Jellinger K. and Wesemann W. (1981) Central aminergic function and its disturbance by hepatic disease: the current status of L-valine pharmacotherapy in metabolic coma, in Transmitter Biochemistry of Human Brain Tissue (Riederer P. and Usdin E., eds), pp. 143–182. Macmillan Publisher Ltd, London. Rinne U. K., Laaksonen H., Riekkinen P. and Sonninen V. (1974) Brain glutamic acid decarboxylase activity in Parkinson’s disease. Eur. Neurol. 12, 13–19. Ritchie T., Scully S. A., de Vellis J. and Noble E. P. (1986) Stability of neuronal and glial marker enzymes in post-mortem rat brain. Neurochem. Res. 11, 383–392. Sarris M., Garrick T. M., Sheedy D. and Harper C. G. (2002) Banking for the future: an Australian experience in brain banking. Pathology 34, 225–229.
Schmitt A., Parlapani E., Bauer M., Heinsen H. and Falkai P. (2008) Is brain banking of psychiatric cases valuable for neurobiological research? Clinics 63, 255–266. Sheedy D., Garrick T., Dedova I., Hunt C., Miller R., Sundqvist N. and Harper C. (2008) An Australian Brain Bank: a critical investment with a high return! Cell Tissue Bank 9, 205–216. Sola-Penna M., dos Santos A. C., Alves G. G., El-Bacha T., FaberBarata J., Pereira M. F., Serejo F. C., Da Poian A. T. and Sorenson M. (2002) A radioassay for phosphofructokinase-1 activity in cell extracts and purified enzyme. J. Biochem. Biophys. Methods 50, 129–140. Sorolla M. A., Reverter-Branchat G., Tamarit J., Ferrer I., Ros J. and Cabiscol E. (2008) Proteomic and oxidative stress analysis in human brain samples of Huntington disease. Free Radic. Biol. Med. 45, 667–678. Spokes E. G. (1979) An analysis of factors influencing measurements of dopamine, noradrenaline, glutamate decarboxylase and choline acetylase in human post-mortem brain tissue. Brain 102, 333–346. Stan A. D., Ghose S., Gao X. M., Roberts R. C., Lewis-Amezcua K., Hatanpaa K. J. and Tamminga C. A. (2006) Human postmortem tissue: what quality markers matter? Brain Res. 1123, 1–11. Tomita H., Vawter M. P., Walsh D. M. et al. (2004) Effect of agonal and postmortem factors on gene expression profile: quality control in microarray analyses of postmortem human brain. Biol. Psychiatry 55, 346–352.
Ó 2009 The Authors Journal Compilation Ó 2009 International Society for Neurochemistry, J. Neurochem. (2009) 110, 1400–1408
