Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Sep 1.
Published in final edited form as: J Neuropathol Exp Neurol. 2013 Sep;72(9):824–832. doi: 10.1097/NEN.0b013e3182a0a39f

Cerebrospinal Fluid Tau Cleaved by Caspase-6 Reflects Brain Levels and Cognition in Aging and Alzheimer Disease

Jasmine Ramcharitar 1,2, Steffen Albrecht 3, Veronica M Afonso 1,2, Vikas Kaushal 1,2, David A Bennett 4,5, Andrea C LeBlanc 1,2
PMCID: PMC3777607  NIHMSID: NIHMS506713  PMID: 23965742

Abstract

Caspase-6 (Casp6) activation in the brain is implicated early in the pathogenesis of Alzheimer disease (AD). In view of the need for early AD diagnosis, brain Casp6 activity was investigated by measuring Tau cleaved by Casp6 (TauΔCasp6) protein in postmortem cerebrospinal fluid (CSF) of 7 non-cognitively impaired, 5 mild cognitively impaired and 12 mild, moderate and severe AD patients. Levels of TauΔCasp6 in CSF accurately reflected the levels of active Casp6 and TauΔCasp6 detected using immunohistochemistry in hippocampal sections from the same individuals. Levels of CSF TauΔCasp6 significantly correlated with AD severity, and with lower global cognitive scores, mini mental state exam, and episodic, semantic, and working memory scores. Regression analyses suggested that the CSF TauΔCasp6 levels combined with TauΔCasp6 brain pathology predict cognitive performance. These results indicate that CSF TauΔCasp6 levels holds promise as a novel early biomarker of AD.

Keywords: Alzheimer disease, Caspase-6, Cognitive scores, Memory scores, Tau cleaved by Caspase-6, TauΔCasp6

INTRODUCTION

Earlier detection of Alzheimer disease (AD) will be essential to establishing efficient therapies against this disease. Cerebrospinal fluid (CSF) offers a window to the brain to measure and assess the predictive value of proteins involved in AD pathogenesis. CSF levels of amyloid β peptide 40 or 42 (Aβ40 or Aβ42), total Tau, and phosphorylated Tau (phospho-Tau) have been the most studied biomarkers of CSF to date (1, 2). Levels of CSF total Tau protein have a sensitivity and specificity of around 80% and 90%, respectively, for distinguishing AD from healthy controls (3). CSF phospho-Tau181 has a sensitivity and specificity of about 75% and 90%, respectively, for distinguishing AD from healthy controls (4). The combination of decreased Aβ42 and increased total Tau and phospho-Tau181 is currently the most accurate chemical biomarker for early AD (5) and can predict conversion from mild cognitively impaired (MCI) to AD (6). Compared to CSF Aβ42 and total Tau, phospho-Tauepitopes are most efficacious at distinguishing AD from non-AD and other neurodegenerative disorders (7). Whereas these markers are promising, there are problems associated with using them as biomarkers. Decreased levels of CSF Aβ42 are also found in other diseases exempt from amyloid pathology (8).

Increased CSF total Tau occurs following acute brain trauma or other neurodegenerative diseases, indicating that total Tau is increased in the CSF due to release from degenerating axons (9–11). Therefore, while Aβ, Tau and phospho-Tau have been widely tested as biomarkers of AD, there is as yet no conclusive biomarker that defines individuals at risk for AD.

Our group has identified high levels of Casp6 activity associated with neurofibrillary tangles (NFTs), neuritic plaques (NPs), and neuropil threads (NPTs) in both sporadic and familial AD (12–14). Casp6 activity is not limited to the brains of AD patients, but is also observed in some non-cognitively impaired (NCI) aged patient brains (13), where higher levels of Casp6 activity are associated with lower cognitive performance (15). By contrast, there is no active Casp6 detected in younger brains (14). In view of the role of Casp6 in cleaving several neuro-cytoskeletal or associated proteins and inducing axonal degeneration, the correlation between levels of Casp6 activity and cognition likely indicates causation (16–18). Therefore, detecting levels of Casp6 activity in aged brains might identify individuals at risk for developing AD. Because Tau is proteolytically cleaved by Casp6, and full-length Tau (FL Tau) and phospho-Tau are excreted in CSF, we hypothesized that Tau cleaved by Casp6 (TauΔCasp6) could be detected in CSF. We therefore developed an enzyme-linked immunosorbent assay (ELISA) to assess the levels of TauΔCasp6 in human CSF. Postmortem CSF was first investigated to determine if levels of TauΔCasp6 in CSF reflected levels assessed using immunohistochemistry in the brains from the same individuals. The results show that CSF TauΔCasp6 levels accurately reflect the levels of TauΔCasp6 and active Casp6 immunoreactivity in situ, and levels of CSF TauΔCasp6 correlate inversely with cognitive scores obtained within a year of death in these individuals.

MATERIALS AND METHODS

Cloning of TauΔ Casp6 and FL Tau Proteins

TauΔCasp6 and Tau full-length cDNA were PCR amplified from a red fluorescent protein-Tau cDNA construct (kind gift from Dr. Yasuo Ihara, University of Tokyo, Japan) using forward primer 5′ TTCAGGATCCGCTGAGCCCCGCCAGGAG 3′ and reverse primers 5′ ACCGCTCGAGTTAGTCCCCAGACACCACTGG 3′ for TauΔCasp6 and 5′ ACACCGCTCGAGTTACAAACCCTGCTTGGCCAG 3′ for FL Tau (Life Technologies Inc. Burlington, Canada). The start codon of the cloning vector, pET28a+ (Novagen, Millipore, Billerica, MA) was used so that the vector’s N-terminal His tag would be incorporated into both FL Tau and TauΔCasp6 proteins upon protein translation. The PCR reactions were performed using 500 ng of RFP-Tau cDNA, 20 μM of each primer, 10 mM dNTP, 0.025 units/μl of FideliTaq deoxyribonucleic acid (DNA) polymerase (Affymetrix, CA), and 1x FideliTaq buffer. The annealing temperature was 52.8°C. The amplified DNA was cloned into the BamH1 and Xho1 sites of pET28a+ (Novagen) and positive clones were confirmed by sequencing at Genome Quebec.

Recombinant Protein Expression and Purification of TauΔCasp6 and FL Tau

FL Tau and TauΔCasp6 transformed Rosettacells (EMD Millipore) were induced at OD600 ~0.55–0.65 with 0.2 mM IPTG for 4 hours shaking at 37°C. Cells were collected by centrifugation at 6000 g for 15 minutes and lysed in 25 mM phosphate buffer (0.2 M sodium phosphate, mono-sodium salt, 0.2 M sodium phosphate, di-sodium salt), 300 mM NaCl, 1 mg/mL lysozyme and fresh protease inhibitors (38 mg/mL 4-(2-aminoethyl)-benzenesulfonyl fluoride, 0.1 μg/mL Nα-tosyl-L-lysine chloromethyl ketone, 0.1 μg/mL pepstatin A and 0.5 μg/mL) (all from Sigma, ON, Canada). Lysates were cleared by centrifugation at 45,000 g for 45 min at 4° C and filtered through a 0.22 μm pore (Millipore, ON, Canada). The recombinant proteins were purified on Nickel Sepharose 6 Fast Flow Beads (GE Healthcare) and stored at −80°C.

Collection of Brain Tissues, Fixation, and Slide Preparation

Brain tissue samples were obtained from subjects who participated in the Religious Orders Study (19). The Religious Orders Study includes older nuns, priests and brothers who have agreed to yearly clinical evaluations and brain donation at time of death. The clinical evaluations include a medical history, neurologic examination, and 21 cognitive function tests including the mini-mental state examination (MMSE) to characterize the cohort. Nineteen tests assess episodic, semantic, and working memory, perceptual speed, and visuospatial ability, as previously described (20). A global cognitive score was generated by converting the values from 19 different cognitive tests into Z scores (using the mean and standard deviation at baseline) and averaged. Participants were diagnostically classified by a clinician following the NINCDS-ADRDA criteria (21). An AD diagnosis was assigned to persons with a history of cognitive decline and evidence of impairment in memory and other cognitive domains. MCI referred to individuals who displayed cognitive impairment upon neuropsychological evaluations but were not clinically diagnosed with dementia. NCI referred to persons without dementia or MCI. Scores were disclosed once the TauΔCasp6 levels were obtained.

Immunohistochemistry

Four-μm-thick formalin-fixed, paraffin-embedded hippocampal tissue sections were deparaffinized, rehydrated and treated with antigen retrieval buffer (10 mM Tris Base, 1 mM EDTA, 0.05% Tween 20, pH 9) for 20 minutes at 97°C in the Dako Cytomation Pascal (Dako, Burlington, ON, Canada). Sections were immunostained using the Dako Autostainer Plus and EnVisionTM Flex Peroxidase kits (Dako). Sections were blocked for 30 minutes, treated with primary antibodies for 30 minutes. The 10630 rabbit anti-active Casp6 (p20Csp6) antiserum (1:1000) was developed against the p20 subunit C-terminal PLDVVD sequence of Casp6 (14). The 10635 TauΔCasp6 antiserum (1:25000) was developed against the C-terminal KSPVVSGD epitope of Tau generated by cleavage with Casp6 (14). Staining was visualized by incubating the slides for 10 minutes with diaminobenzidine (Dako); the slides were then counterstained with Dako hematoxylin.

Assessment of Immunohistochemical Staining

The MIRAX Digital Slide Scanner (Zeiss, Canada) was used to scan tissue sections and generate high-resolution digital images that were analyzed using the MIRAX Viewer Program (Zeiss, Germany). The Atlas of the Human Brain was used as a reference to identify the hippocampus proper (CA1–CA4 and subiculum) and the entorhinal cortex (ERC), trans-ERC and temporal cortices in each tissue section (22). Some areas were not present in all sections and this is reflected by different numbers of data points in the figures. Scoring was done in collaboration with neuropathologist, Dr. Steffen Albrecht, in a blinded manner. The densities of NFTs, NPs and NPTs were scored semiquantitatively as absent (0), absent to low (0.5), low/mild (1), mild to moderate (1–2), moderate (2), moderate to severe (2–3) or high/severe (3). Scoring diagrams developed by the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) for assessing AD pathology in autopsy brains were used as guides (23). Alzheimer-type Tau pathology (NFTs, NPTs, and NPs) was assessed using conventional well-established neuropathological diagnostic criteria (23).

Assessment of TauΔCasp6 Protein Levels in the CSF Using a Sandwich ELISA

ELISA experiments were conducted simultaneously on all AD and control samples using several plates that were coated at the same time. Subsequent experiments on other neurodegenerative diseases were run in the same ELISA plate and at the same time. The postmortem CSF from AD and controls was provided by the Religious Orders Study. The multiple sclerosis (MS) and Parkinson disease (PD) CSF was obtained from Precision Med, Inc. (San Diego, CA). Multiple sclerosis and PD CSF was obtained premortem. Although the ages of the MS group (62.2 ± 6.7 years) and PD group (73.6 ± 4.3 years) were significantly younger than the AD group (89.6 ± 4.4 years; p < 0.01), the PD age did not differ significantly from the NCI group used in this study. The 10635 TauΔCasp6 antiserum (14) was diluted 1:125 in 1x phosphate-buffered saline (diluted from 10x concentrate: 80 g NaCl, 2 g KCl, 11.5 g Na2HPO4, 2 g KH2PO4, pH 7.4). One hundred μl of TauΔCasp6 antiserum solution was added to each well of a 96-well clear, flat-bottom plate (Nunc Maxisorp, Sigma-Aldrich, St. Louis, MO) and incubated overnight at 4°C. Each well was washed with 300 μL of Tris-buffered saline with Tween ([TBST]; 50 mM Tris, 150 mM NaCl, 0.05% Tween 20, pH 7.4) 4 × 2 minutes, shaking at a speed setting of 3 (Barnstead/Lab-line Rotator Model 1314). All subsequent washes and incubations were also done while shaking as described above. Wells were blocked with 300 μL of 2.5% bovine serum albumin in TBST ([BSA-TBST] pH 7.4) for 2 hours at room temperature (RT), and washed with 300 μL of TBST 4 × 2 minutes. Control samples were prepared in Tris-buffered saline (TBS; 50 mM Tris, 150 mM NaCl, pH 7.4). One hundred μL of undiluted control/CSF samples in triplicates were incubated for 3 hours at RT and washed with 300 μL of TBST 4 × 2 minutes. Anti-human Tau HT7 monoclonal antibody (1:1000 dilution in 1% BSA-TBS) generated to the N-terminal peptide PPGQK of Tau (200 μg/mL) (Pierce Endogen, ON, Canada) was used as the detection antibody. One hundred μL of the HT7 antibody solution was incubated overnight at 4°C and washed with 300 μL of TBST 4 × 2 minutes. HRP anti-mouse IgG Fc antibody (0.8 mg/mL; 1:100,000 dilution in 1% BSA-TBS) (Jackson ImmunoResearch, ON, Canada) was used as the developing antibody. One hundred μL of the HRP anti-mouse antibody solution was incubated for 1 hour at RT, and washed with 300 μL of TBST 4 × 2 minutes. The ELAST ELISA Amplification System was used to increase assay sensitivity (PerkinElmer, USA). Biotinyl tyramide and HRP avidin solutions were prepared according to the manufacturer’s manual. One hundred μL of the biotinyl tyramide solution was added to each well and incubated protected from light for 15 minutes at RT and washed with 300 μL of TBST 4 × 2 minutes. One hundred μL of the HRP avidin solution was incubated protected from light for 15 minutes at RT. One hundred μl of the Ready-to-use Enhancer K-Blue TMB HRP substrate (Neogen, Lansing, MI) was added. Plates were read at a wavelength of 650 nm, every 45 seconds for 2 hours at 32°C, using a H4 Synergy Plate reader (Bio-Tek, Winooski, VT). Recombinant TauΔCasp6 protein standard curve was quantified in pg/mL at each time point. The standard curve with the highest correlation coefficient value was chosen to assess the amount of TauΔCasp6 in each CSF sample. Using this technique, we observed acceptable intra-assay variability in the triplicates (10% ± 7%). Inter-assay variability (30%–70%) was larger, although the rank amount was retained, i.e. low samples remained lowest and high samples remained highest.

Correlations, Linear Regressions and Statistical Analyses

InStat 3 statistical software (GraphPad Software Inc., San Diego, CA) was used to evaluate group differences by one-way Analysis of Variance and post-hoc Bonferroni or Tukey-Kramer multiple comparisons test. Correlations were done by Spearman Rank correlations (two-tailed p < 0.05). To identify the most important TauΔCasp6 measurement that predicts cognition, separate stepwise linear regression analyses (using SPSS 17 software) were performed for each cognitive score entering the levels of CSF TauΔCasp6, and the pathological levels of TauΔCasp6 in the ERC, subiculum, and CA1–4. In addition, the ages of the individuals were always included as a potential variable in predicting cognitive scores.

RESULTS

Demographics of NCI, MCI and AD Patients

A total of 24 cases were used in this study: 7 NCI, 5 MCI and 12 AD patients (Table 1). There were no significant differences in age, education, or postmortem interval among the groups. Although the number of females and males is not identical within groups, overall there were 11 males and 13 females. The average MMSE score of the NCI and MCI groups were significantly higher (p < 0.01) vs. the AD group. The average global cognitive, episodic memory, semantic memory, working memory and visuospatial ability scores were also higher in the NCI and MCI groups than in the AD group.

Table 1.

Demographics and Cognitive Performance

NCI MCI AD
Number of cases 7 5 12
Age at death (y) 83.35 ± 8.36 84.09 ± 9.77 89.94 ± 4.04
Education (y) 19.43 ± 3.60 17.00 ± 4.00 18.92 ± 2.23
Postmortem Interval (h) 6.03 ± 2.47 6.91 ± 2.04 12.03 ± 7.95
Gender 5 M, 2 F 3 M, 2 F 3 M, 9 F
Global Cognitive Score −0.11 ± 0.55** 0.22 ± 0.24** −2.07 ± 1.09
MMSE 27.71 ± 1.80** 28.19 ± 1.29** 13.35 ± 9.13
Episodic Memory 0.06 ± 0.45** 0.56 ± 0.29** −2.37 ± 1.18
Semantic Memory 0.30 ± 0.48* 0.18 ± 0.47* −1.80 ± 1.33
Working Memory −0.43 ± 0.78* −0.34 ± 0.78* −1.56 ± 0.90
Visuospatial Ability −0.36 ± 1.00 0.03 ± 0.48 −1.32 ± 0.84
Perceptual Speed −0.56 ± 1.04* 0.34 ± 0.49* −1.85 ± 0.98
Open in a new tab

Data represent mean ± SD.

F, female; M, male; NCI, non-cognitive impairment; MCI, mild-cognitive impairment; AD, Alzheimer disease; MMSE, Mini-mental state examination.

ANOVA p < 0.0001, Bonferroni Multiple Comparisons post Test

*

p < 0.01 and

**

p < 0.001 vs. AD.

CSF Levels of TauΔCasp6 Correlate with Protein Immunoreactivity in the Brain

TauΔCasp6 and active Casp6-immunopositive NFTs, NPs and NPTs were semiquantitatively determined in the trans-ERC, ERC, subiculum, CA1, CA2, CA3, CA4 and temporal cortex regions of 24 cases based on scoring diagrams developed by the CERAD (Fig. 1A). ELISA of TauΔCasp6 vs. FL Tau was validated with purified recombinant proteins (Fig. 1B). The ELISA detected recombinant TauΔCasp6 protein within a linear range (r2 = 0.999) between 15.6 to 1000 pg/mL and had a detection limit of 3.9 pg/mL. The TauΔCasp6 sandwich ELISA showed little detection of the recombinant FL Tau protein even at 1000 pg/mL protein concentrations (Fig. 1B). The TauΔCasp6 sandwich ELISA was thus established as a suitable method for specifically and quantitatively measuring TauΔCasp6 in biological samples. Because of acceptable intra-assay variability, but not inter-assay variability, all samples were assessed simultaneously on a single batch of plates in triplicate.

Figure 1.

Figure 1

Open in a new tab

CSF Tau cleaved by caspase-6 (TauΔCasp6) levels correlate with TauΔCasp6 levels in the brain. (A) Micrographs of hippocampi immunostained with anti-TauΔCasp6 (upper panels) or anti-active Casp6 (lower panels) demonstrate variable abundance of pathological Tau immunoreactivity on which pathological scores of 0 to 3 were assigned. (B) Example of 2 standard curves obtained by ELISA with recombinant TauΔCasp6 and full-length (FL) Tau confirm the specificity of the assay for TauΔCasp6. (C, D) Spearman rank correlation (r) between the levels of TauΔCasp6 in CSF and TauΔCasp6 (C) or active caspase-6 (D) immunostaining scores in samples from the indicated brain regions. *p < 0.05, **p < 0.01. Trans-ERC, trans-entorhinal cortex; ERC, entorhinal cortex.

Given that protein concentration in the CSF varies with age (4) and postmortem collection interval (5), Pearson correlations were performed between TauΔCasp6 CSF levels and age or postmortem interval. CSF TauΔCasp6 did not correlate to either of these variables (p > 0.168). To determine if the level of TauΔCasp6 in the CSF reflected the levels of active Casp6 or TauΔCasp6 in NFTs, NPTs and NPs in brains, Spearman rank correlations were performed between the amount of TauΔCasp6 in CSF and the pathological scores for Casp6 in the medial temporal cortex (Fig. 1C, D). The results show that the levels of CSF TauΔCasp6 correlated positively with the amount of immunoreactivity for TauΔCasp6 (Fig 1C) or active Casp6 (Fig 1D) in the brain. The correlation analyses yielded significant r-values for TauΔCasp6 in the ERC, subiculum, CA1, CA3 and CA4 areas of the hippocampus as well as the temporal cortex but they were not significant in the trans-ERC and CA2 areas. Significant r-values for active Casp6 were found in the trans-ERC, ERC, subiculum, CA2, CA3 and CA4 areas of the hippocampus as well as the temporal cortex but did not reach significance in the CA1. The lack of significant correlation in a few areas are possibly due to a higher turnover of the active Casp6 or TauΔCasp6 in certain areas but since at least 1 of the markers of Casp6 activity is present in each area of the hippocampus, these results indicate that the level of TauΔCasp6 detected in the CSF accurately reflects Casp6 activity in the brains of these individuals.

CSF TauΔCasp6 Levels Correlate with Severity of Disease

To determine if CSF TauΔCasp6 levels reflected the diagnostic status of these individuals, we compared the levels of TauΔCasp6 in CSF in each diagnostic group (Fig. 2A). There was a significant correlation between the levels of CSF TauΔCasp6 and the stage of disease. The levels of TauΔCasp6 distinguished the NCI, MCI and mild AD cases from severe AD cases but did not distinguish the NCI cases from MCI or milder cases of AD. Nevertheless, when comparing averaged CSF TauΔCasp6 levels per group with averaged global cognitive scores obtained within a year of death, a strong correlation (r2 = 0.9469) was obtained (Fig. 2B). These results indicate that levels of CSF TauΔCasp6 increase with lower cognitive performance.

Figure 2.

Figure 2

Open in a new tab

CSF Tau cleaved by caspase 6 (TauΔCasp6) levels in groups of non-cognitively impaired (NCI), mild cognitively impaired (MCI), mild, moderate and severe Alzheimer disease (AD). (A) Linear regression of levels of CSF TauΔCasp6 vs. each subgroup of disease. ANOVA between groups p < 0.0001 and Bonferroni multiple comparison post hoc test indicates significant difference between NCI, MCI or mild AD and severe AD. (B) Linear regression (r2) between the averaged levels of TauΔCasp6 in CSF vs. the averaged global cognitive score of each subgroup. Bars indicate SD. ANOVA for the linear regression was significant [F(1,3) = 53.506, p = 0.0053].

CSF TauΔCasp6 Levels Predict Cognitive Performance

To determine if levels of CSF TauΔCasp6 reflected a specific type of cognitive ability, the levels of CSF TauΔCasp6 was correlated to global cognition and to separate measures of episodic, working, and semantic memory and to perceptual speed and visuospatial ability (Fig. 3). Spearman Rank correlations yielded a statistically significant r-value of −0.547 and −0.555 for GCS (p < 0.01) and MMSE (p<0.01), respectively (Fig. 3A, B). Furthermore, the level of CSF TauΔCasp6 increased as episodic, semantic, and working memory decreased (Fig. 3C–E). In contrast, no significant correlation was obtained between CSF TauΔCasp6 levels and visuospatial abilities and perceptual speed scores (Fig. 3F, G).

Figure 3.

Figure 3

Open in a new tab

CSF Tau cleaved by caspase 6 (TauΔCasp6) levels correlates with cognitive decline. A–G: Spearman rank correlations (r) of TauΔCasp6 in CSF vs. global cognitive (GC) scores (A), mini mental status exam (MMSE) (B), episodic (C), semantic (D), working (E) memory and visuospatial ability (F), or perceptual speed (G). *p < 0.05, **p < 0.01

To identify the most important TauΔCasp6 measurement that predicts cognition, separate stepwise linear regression analyses were performed for each cognitive score entering the levels of CSF and pathological TauΔCasp6 in the trans-ERC, ERC, subiculum, CA1–4, and temporal cortex. The results indicate that the CSF TauΔCasp6 levels combined with TauΔCasp6 pathology in the ERC, CA2 and subiculum were significant predictors for global cognitive scores, MMSE and semantic memory, respectively (Table 2). Episodic memory was best predicted by ERC TauΔCasp6 pathology, whereas age of death best predicted working memory, consistent with the known decline in working memory with age (26). These results suggest again that both TauΔCasp6 CSF and pathology combine to predict specific types of cognitive performance. The prediction of episodic memory by TauΔCasp6 pathology in ERC is consistent with data previously obtained from aged NCI individuals (15).

Table 2.

Regression analysis predicting cognition from Tau cleaved by caspase 6 cerebrospinal fluid levels and brain pathology

Model Estimate (SE) p value Beta Constant (SE) Adjusted R2
Global Cognitive Score ERC + −0.638 (0.469) 0.006 −0.579
CSF −0.007 (0.003) 0.041 −0.400 1.519 (0.512) 0.670
MMSE CSF + −0.072 (0.021) 0.004 −0.556
CA2 −4.190 (1.368) 0.010 −0.484 37.974 (3.673) 0.698
Episodic Memory ERC −1.113 (0.219) <0.0001 −0.815 1.368 (0.528) 0.639
Semantic Memory SUB + −0.521 (0.194) 0.020 −0.500
CSF −0.008 (0.003) 0.030 −0.459 1.324 (0.537) 0.620
Working Memory Age −0.096 (0.028) 0.004 −0.696 7.333 (2.398) 0.445
Open in a new tab

In the model column, the CSF Tau cleaved by Casp6 (TauΔCasp6) is indicated by “CSF”; the pathological TauΔCasp6 score is indicated by brain area as follows: ERC, entorhinal cortex; SUB, subiculum. CSF, cerebrospinal fluid; MMSE, Mini-mental state examination; SE, standard error.

Data represent statistic parameters for predictor variables of the various models.

Levels of TauΔCasp6 in AD Patient CSF Are Significantly Higher than in MS and PD Patient CSF

To assess if TauΔCasp6 is present in the CSF of individuals with other neurodegenerative diseases, we compared CSF of AD patients with those of patients with MS and PD, 2 progressive neurodegenerative diseases. The level of CSF TauΔCasp6 for both the MS and PD did not differ significantly from each other but were approximately 55% that of the AD CSF TauΔCasp6 levels (Fig. 4). This closely resembled the levels of CSF TauΔCasp6 in the NCI group (60% ± 20% of AD levels). Unfortunately, no cognitive scores were available from the MS and PD cases; nevertheless, these results indicate that the presence of TauΔCasp6 does not increase in the CSF in all neurodegenerative diseases.

Figure 4.

Figure 4

Open in a new tab

CSF Tau cleaved by caspase 6 (TauΔCasp6) levels of Parkinson disease (PD) and multiple sclerosis (MS) patients are significantly lower than those of Alzheimer disease (AD) patients. Levels of CSF TauΔCasp6 in AD (n = 5) vs. MS (n = 5) and PD (n = 5) cases. ***p < 0.001 vs. AD (one-way ANOVA followed by a Tukey-Kramer post hoc analysis).

DISCUSSION

In this study, we show that the levels of CSF TauΔCasp6 of aged individuals reflect accurately the immunoreactivity of TauΔCasp6 and active Casp6 enzyme in the brain. Except for TauΔCasp6 in trans-ERC and CA2 and active Casp6 in CA1, immunostaining scores in all areas of the hippocampal formation and the temporal cortex correlated with CSF TauΔCasp6 levels. The lack of significant correlations for some brain regions may have been due to a higher turnover of either Casp6 or TauΔCasp6 in certain brain areas. Despite the lack of CSF correlations with Casp6 activity in some brain regions, the TauΔCasp6 in CSF always correlated with at least one of the Casp6 biomarkers of brain pathology. Recently, Casp6 activity has been associated with axonal degeneration of neurons (17, 18, 27). Therefore, the CSF TauΔCasp6 levels may become an important biomarker of ongoing neurodegeneration. Clearly in AD, neuronal Casp6 activity is associated with the NFT, NPT, and NP pathological lesions that define the disease. Nevertheless, determining TauΔCasp6 levels with the phospho-Tau and Aβ42 AD CSF levels might increase the specificity and sensitivity of the analyses for the diagnosis of AD.

The second major finding of this study is that the levels of TauΔCasp6 in CSF increase with disease severity based on the clinical diagnosis of AD dementia whereby subjects were determined to have NCI, MCI, mild, moderate, and severe AD. When global cognitive scores and TauΔCasp6 levels are averaged within these categories, a strong linear regression was achieved (r2 = 0.94). Although the NCI, MCI and mild AD groups could be distinguished significantly from severe AD, levels of CSF TauΔCasp6 did not differentiate NCI from MCI or mild AD. There are 2 possible interpretations of this result. First, this could reflect the heterogeneous nature of the NCI individuals, some of whom are likely to have eventually progressed to AD if they had lived longer. This result could thus be predicted if TauΔCasp6 is a very early marker of cognitive decline. Consequently, in future studies it would be essential to include CSF from young subjects, and from living subjects who are followed longitudinally for the development of cognitive impairment. Second, the clinical diagnoses of NCI, MCI and AD are a more subjective assessment whereas cognitive scores are collected objectively using a quantitative scale. Therefore, better correlations may be obtained between CSF TauΔCasp6 levels and cognitive status than with diagnosis status.

In addition to the association between CSF TauΔCasp6 levels and cognitive categories, CSF TauΔCasp6 levels also correlated significantly with an overall measure of global cognition and to specific measures of episodic memory, semantic memory, and working memory scores. Furthermore, CSF TauΔCasp6 combined with the TauΔCasp6 pathology in various brain structures to predict cognitive performance. That CSF TauΔCasp6 levels correlate so well with cognitive impairment is consistent with the known effects of active Casp6 on axonal degeneration (17, 18, 27), the presence of active Casp6 and TauΔCasp6 in aged NCI individuals (13), and the association of higher levels of Casp6 activity in ERC and CA1 regions of the hippocampus with lower memory performance in NCI individuals (15). Interestingly, no significant correlation was achieved between CSF TauΔCasp6 levels and with visuospatial abilities or perceptual speed. This is consistent with data indicating that perceptual speed and visuospatial abilities have been associated with vascular disease, which often overlaps AD (28, 29). The fact that CSF and brain TauΔCasp6 levels were both related to cognitive performance in the same models provides additional evidence for the idea that CSF TauΔCasp6 levels are a biomarker of ongoing neurodegeneration and provide information complementary to the pathological findings in the brain.

There are a number of limitations to this study. First, we have not yet conclusively evaluated whether Casp6 activation is specific to AD. Clearly, in AD, the activity of Casp6, based on immunodetection of the active subunit of Casp6 and TauΔCasp6, is intimately associated with the NFTs, NPTs, and NPs defining AD (12–14). There is no Casp6 activity in younger brains or in AD cerebellum (14). Furthermore, some aged brains do not have any active Casp6 or TauΔCasp6 immunoreactivity (15). Nevertheless, other neurodegenerative disease could involve the activation of Casp6 (30–35). Indeed, Casp6 has been suggested to impart neurotoxicity in a mouse model of Huntington disease (30), although recent studies crossing the mutant with Casp6 null mice could not find evidence for a Casp6-mediated neurotoxic Htt fragment (36). Casp6 also has been implicated in Parkinson disease (34). We show here that levels of CSF TauΔCasp6 in MS and PD, 2 progressive neurodegenerative diseases, are similar to those found in NCI individuals rather than those in AD. While these results suggest some specificity for TauΔCasp6 in AD, the brains and CSF of other patients with neurodegenerative diseases will need to be evaluated in future studies. For an accurate diagnostic test to be developed for AD, it is likely that both clinical and biomarker status will need to be considered. Second, our study is based only on postmortem CSF and the ELISA needs to be applied to premortem CSF for clinical application. Because no correlation was obtained between levels of CSF TauΔCasp6 and time of postmortem interval, it is unlikely that the postmortem interval affected protein levels in our study. This is consistent with observed preservation of the blood-brain barrier in the first 24 hours after death (25). Furthermore, it is unlikely that the CSF TauΔCasp6 is only the result of blood-brain barrier breakdown after death because total Tau and phospho-Tau efficiently cross the blood-brain barrier in premortem CSF (37). Third, longitudinal studies will be required to establish if TauΔCasp6 in CSF has a predictive value for AD. Fourth, as with all ELISA-based experiments, there is variability between different experiments despite low intra-experiment variability. The inter-experiment variability is likely due to the coating of the plates with the capture antisera. We are presently exploring alternative technologies that can reduce this variability and provide reproducible measures of the absolute amount of TauΔCasp6 in CSF.

In conclusion, we have established an ELISA-based test that accurately measures the level of TauΔCasp6 in CSF. Analyses of postmortem CSF shows that the levels of TauΔCasp6 in the CSF reflect the levels of TauΔCasp6 immunoreactivity in the brain and that the levels of brain and CSF TauΔCasp6 increase with cognitive decline in aged individuals.

Acknowledgments

This work was supported by the Canadian Foundation of Innovation, CIHR IAP-102238, and 2011 MOP-243413-BCA-CGAG-45097 to ALB, P30AG10161 and R01AG15819 to DAB.

The authors gratefully acknowledge the gift of PHF-1 antibody from Dr. Peter Davies (Albert Einstein College of Medicine, NY).

References

  • 1.Hampel H, Frank R, Broich K, et al. Biomarkers for Alzheimer’s disease: academic, industry and regulatory perspectives. Nat Rev Drug Discov. 2010;9:560–74. doi: 10.1038/nrd3115. [DOI] [PubMed] [Google Scholar]
  • 2.Humpel C. Identifying and validating biomarkers for Alzheimer’s disease. Trends Biotechnol. 2011;29:26–32. doi: 10.1016/j.tibtech.2010.09.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Blennow K. Cerebrospinal fluid protein biomarkers for Alzheimer’s disease. NeuroRx. 2004;1:213–25. doi: 10.1602/neurorx.1.2.213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Blennow K. CSF biomarkers for mild cognitive impairment. J Intern Med. 2004;256:224–34. doi: 10.1111/j.1365-2796.2004.01368.x. [DOI] [PubMed] [Google Scholar]
  • 5.Mattsson N, Blennow K, Zetterberg H. CSF biomarkers: pinpointing Alzheimer pathogenesis. Ann N Y Acad Sci. 2009;1180:28–35. doi: 10.1111/j.1749-6632.2009.04944.x. [DOI] [PubMed] [Google Scholar]
  • 6.Diniz BS, Pinto JA, Júnior, Forlenza OV. Do CSF total tau, phosphorylated tau, and beta-amyloid 42 help to predict progression of mild cognitive impairment to Alzheimer’s disease? A systematic review and meta-analysis of the literature. World J Biol Psychiatry. 2008;9:172–82. doi: 10.1080/15622970701535502. [DOI] [PubMed] [Google Scholar]
  • 7.Tapiola T, Alafuzoff I, Herukka SK, et al. Cerebrospinal fluid {beta}-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol. 2009;66:382–9. doi: 10.1001/archneurol.2008.596. [DOI] [PubMed] [Google Scholar]
  • 8.Blennow K, Hampel H. CSF markers for incipient Alzheimer’s disease. Lancet Neurol. 2003;2:605–13. doi: 10.1016/s1474-4422(03)00530-1. [DOI] [PubMed] [Google Scholar]
  • 9.Hesse C, Rosengren L, Andreasen N, et al. Transient increase in total tau but not phospho-tau in human cerebrospinal fluid after acute stroke. Neurosci Lett. 2001;297:187–90. doi: 10.1016/s0304-3940(00)01697-9. [DOI] [PubMed] [Google Scholar]
  • 10.Ost M, Nylen K, Csajbok L, et al. Initial CSF total tau correlates with 1-year outcome in patients with traumatic brain injury. Neurology. 2006;67:1600–4. doi: 10.1212/01.wnl.0000242732.06714.0f. [DOI] [PubMed] [Google Scholar]
  • 11.Otto M, Wiltfang J, Tumani H, et al. Elevated levels of tau-protein in cerebrospinal fluid of patients with Creutzfeldt-Jakob disease. Neurosci Lett. 1997;225:210–2. doi: 10.1016/s0304-3940(97)00215-2. [DOI] [PubMed] [Google Scholar]
  • 12.Albrecht S, Bogdanovic N, Ghetti B, et al. Caspase-6 activation in familial Alzheimer disease brains carrying amyloid precursor protein or presenilin i or presenilin II mutations. J Neuropathol Exp Neurol. 2009;68:1282–93. doi: 10.1097/NEN.0b013e3181c1da10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Albrecht S, Bourdeau M, Bennett D, et al. Activation of caspase-6 in aging and mild cognitive impairment. Am J Pathol. 2007;170:1200–9. doi: 10.2353/ajpath.2007.060974. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Guo H, Albrecht S, Bourdeau M, et al. Active Caspase-6 and Caspase-6 cleaved Tau in neuropil threads, neuritic plaques and neurofibrillary tangles of Alzheimer’s Disease. Am J Pathol. 2004;165:523–31. doi: 10.1016/S0002-9440(10)63317-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ramcharitar J, Afonso VM, Albrecht S, et al. Caspase-6 activity predicts lower episodic memory ability in aged individuals. Neurobiol Aging. 2013;34:1815–24. doi: 10.1016/j.neurobiolaging.2013.01.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Klaiman G, Petzke TL, Hammond J, et al. Targets of caspase-6 activity in human neurons and Alzheimer disease. Mol Cell Proteomics. 2008;7:1541–55. doi: 10.1074/mcp.M800007-MCP200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Nikolaev A, McLaughlin T, O’Leary DD, et al. APP binds DR6 to trigger axon pruning and neuron death via distinct caspases. Nature. 2009;457:981–9. doi: 10.1038/nature07767. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  • 18.Sivananthan SN, Lee AW, Goodyer CG, et al. Familial amyloid precursor protein mutants cause caspase-6-dependent but amyloid beta-peptide-independent neuronal degeneration in primary human neuron cultures. Cell Death Dis. 2010;1:e100. doi: 10.1038/cddis.2010.74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bennett DA, Schneider JA, Buchman AS, et al. Overview and findings from the rush memory and aging project. Curr Alzheimer Res. 2012;9:646–63. doi: 10.2174/156720512801322663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Wilson RS, Schneider JA, Arnold SE, et al. Conscientiousness and the incidence of Alzheimer disease and mild cognitive impairment. Arch Gen Psychiatry. 2007;64:1204–12. doi: 10.1001/archpsyc.64.10.1204. [DOI] [PubMed] [Google Scholar]
  • 21.Bennett DA, Schneider JA, Bienias JL, et al. Mild cognitive impairment is related to Alzheimer disease pathology and cerebral infarctions. Neurology. 2005;64:834–41. doi: 10.1212/01.WNL.0000152982.47274.9E. [DOI] [PubMed] [Google Scholar]
  • 22.Mai JK, Paxinos G, Voss T. Atlas of the Human Brain. 3. Amsterdam: Elsevier; 2008. [Google Scholar]
  • 23.Mirra SS, Hart MN, Terry RD. Making the diagnosis of Alzheimer’s disease: a primer for practicing pathologists. Arch Pathol Lab Med. 1993;117:132–44. [PubMed] [Google Scholar]
  • 24.Platt MS, McClure S, Clarke R, et al. Postmortem cerebrospinal fluid pleocytosis. Am J Forensic Med Pathol. 1989;10:209–12. doi: 10.1097/00000433-198909000-00007. [DOI] [PubMed] [Google Scholar]
  • 25.Wyler D, Marty W, Bar W. Correlation between the post-mortem cell content of cerebrospinal fluid and time of death. Int J Legal Med. 1994;106:194–9. doi: 10.1007/BF01371336. [DOI] [PubMed] [Google Scholar]
  • 26.Wang M, Gamo NJ, Yang Y, et al. Neuronal basis of age-related working memory decline. Nature. 2011;476:210–3. doi: 10.1038/nature10243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Cusack CL, Swahari V, Hampton Henley W, et al. Distinct pathways mediate axon degeneration during apoptosis and axon-specific pruning. Nature communications. 2013;4:1876. doi: 10.1038/ncomms2910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Graham NL, Emery T, Hodges JR. Distinctive cognitive profiles in Alzheimer’s disease and subcortical vascular dementia. J Neurol Neurosurg Psychiatry. 2004;75:61–71. [PMC free article] [PubMed] [Google Scholar]
  • 29.Schneider JA, Wilson RS, Cochran EJ, et al. Relation of cerebral infarctions to dementia and cognitive function in older persons. Neurology. 2003;60:1082–8. doi: 10.1212/01.wnl.0000055863.87435.b2. [DOI] [PubMed] [Google Scholar]
  • 30.Graham RK, Deng Y, Slow EJ, et al. Cleavage at the caspase-6 site is required for neuronal dysfunction and degeneration due to mutant huntingtin. Cell. 2006;125:1179–91. doi: 10.1016/j.cell.2006.04.026. [DOI] [PubMed] [Google Scholar]
  • 31.Uribe V, Wong BK, Graham RK, et al. Rescue from excitotoxicity and axonal degeneration accompanied by age-dependent behavioral and neuroanatomical alterations in caspase-6-deficient mice. Hum Mol Genet. 2012;21:1954–67. doi: 10.1093/hmg/dds005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Waldron-Roby E, Ratovitski T, Wang X, et al. Transgenic mouse model expressing the caspase 6 fragment of mutant huntingtin. J Neurosci. 2012;32:183–93. doi: 10.1523/JNEUROSCI.1305-11.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Warby SC, Doty CN, Graham RK, et al. Activated caspase-6 and caspase-6-cleaved fragments of huntingtin specifically colocalize in the nucleus. Hum Mol Genet. 2008;17:2390–404. doi: 10.1093/hmg/ddn139. [DOI] [PubMed] [Google Scholar]
  • 34.Giaime E, Sunyach C, Druon C, et al. Loss of function of DJ-1 triggered by Parkinson’s disease-associated mutation is due to proteolytic resistance to caspase-6. Cell Death Differ. 2010;17:158–69. doi: 10.1038/cdd.2009.116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Robert G, Puissant A, Dufies M, et al. The caspase 6 derived N-terminal fragment of DJ-1 promotes apoptosis via increased ROS production. Cell Death Differ. 2012;19:1769–78. doi: 10.1038/cdd.2012.55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Gafni J, Papanikolaou T, Degiacomo F, et al. Caspase-6 activity in a BACHD mouse modulates steady-state levels of mutant huntingtin protein but is not necessary for production of a 586 amino acid proteolytic fragment. J Neurosci. 2012;32:7454–65. doi: 10.1523/JNEUROSCI.6379-11.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Blennow K, Vanmechelen E, Hampel H. CSF total tau, Abeta42 and phosphorylated tau protein as biomarkers for Alzheimer’s disease. Mol Neurobiol. 2001;24:87–97. doi: 10.1385/MN:24:1-3:087. [DOI] [PubMed] [Google Scholar]

RESOURCES