Cell culture

Human SK-N-SH neuroblastoma cells (European Collection of Cell Cultures #86012802, Salisbury, UK) were cultured in Dulbecco’s modified Eagle’s medium with GlutaMax (Gibco BRL, Carlsbad, CA) supplemented with 10% fetal calf serum (Lonza, Basel, Switzerland) and 100 U/ml penicillin (Yamanouchi Pharma BV, Leiderdorp, the Netherlands). Cells were differentiated in culture medium supplemented with all trans-retinoic acid (Sigma, St Louis, MO, USA) in a final concentration of 10 µM for 5 days. Differentiated cells were subsequently treated with tunicamycin(TM) or 2-deoxyglucose (2-DG) at indicated concentrations for 16 h.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Ubiquilin 2 immunoreactivity in tauopathies.

Representative immunohistochemical staining of ubiquilin 2 in the CA1 region of the hippocampus in a control (A) and an AD case (B). Detailed images showing plaque-like structures (C) and perisomatic granules (D) in an AD case. Ubiquilin 2 immunoreactivity in the subiculum of a sporadic PiD case (E), a G272V MAPT mutation carrier (F), and a PSP case (G). AD, Alzheimer’s disease; CA, cornu ammonis, FTD, frontotemporal dementia; MAPT, microtubule associated protein tau.

https://doi.org/10.1371/journal.pone.0076598.g001

Postmortem material

Post-mortem brain material was obtained from the Netherlands Brain Bank (Amsterdam, The Netherlands). Patients or their next of kin gave informed consent for brain autopsy and use of tissue and medical records for research purposes. Staging of AD pathology was evaluated according to Braak and Braak [19]. Age, gender, clinical diagnosis, Braak score for NFTs and post-mortem interval of the cases used in this study are listed in Tables S1, S2, S3 and S4.

Immunohistochemistry

Formalin-fixed, paraffin-embedded tissue was selected from age-matched non-neurological controls, AD and tau-FTD cases (Table 1). Sections (5 µm thick) were mounted on Superfrost plus tissue slides (Menzel-Glaser, Germany) and dried overnight at 37°C. After deparaffinizing, endogenous peroxidase activity was quenched using 0.3% H₂O₂ in Phosphate-buffered saline (PBS) for 30 min. Sections were treated in 10 mM pH 6.0 sodium citrate buffer heated by autoclave during 10 min for antigen retrieval. Sections were washed with PBS and incubated with ubiquilin 2 antibody (Novus Biologicals, Littleton, CO, USA) diluted in Dako antibody diluent (1:10 000, Dako, Hamburg, Germany) for 1 h at room temperature. Negative controls were generated by omission of primary antibodies. After washing with PBS sections were incubated with horseradish peroxidase-labelled mouse/rabbit secondary antibody (REAL EnVision/HRP Rabbit/Mouse, Dako) for 30 min at room temperature. After another washing step color was developed using 3,3′-diaminobenzidine (DAB, Sigma) as chromogen for 10 min. Sections were counterstained with hematoxylin and mounted using Depex (BDH Laboratories Supplies, East Grinstead, UK). For double staining with AT8 (tau pSer202/Thr205, Thermo, Fisher Scientific, Rockford, IL) sections were treated as described above for ubiquilin 2 staining. After developing with 3,3′-diaminobenzidine sections were washed and subsequently treated with 10 mmol/l pH 6.0 sodium citrate buffer heated by autoclave for 10min to denature bound antibodies. After short washing in PBS, sections were blocked with 10% normal goat serum (Dako) for 10 min at room temperature. Incubation with AT8 antibody (1:100 in antibody diluent) was performed overnight at 4 °C. After washing sections were incubated with HRP-conjugated goat anti-mouse IgG1 (1:100 in antibody diluent) for 1 h. Color was developed using Liquid Permanent Red (LPR; DAKO). Sections were counterstained with hematoxylin and mounted using aquamount (BDH Laboratories Supplies). Images were taken with the same microscope settings using a 20 x objective. At least two different fields of view were analyzed per patient. Ubiquilin 2 and AT8 signals were unmixed using Nuance spectral imaging system (CRi, Woburn, MA, USA) as described previously [20]. AT8 and ubiquilin 2 double stained sections of 7 AD cases (Table 1) were used to compare ubiquilin 2 signals between AT8 positive and AT8 negative cells. Numbers of AT8 positive and AT8 negative cells, respectively, were determined and assigned to the respective group. To determine neurons displaying a high ubiquilin 2 signal in these groups, a threshold for ubiquilin 2 intensity was set using ImageJ software. The threshold was changed for two sections (case 4, 7) because of a variation in staining intensity. Neurons, under and above the threshold, respectively, were counted and assigned to ubiquilin 2 low or ubiquilin 2 high group. Subsequently, frequency distributions were compared between AT8 positive and AT8 negative groups using the chi-square test. A p value of <0.05 was taken as significant.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Tau inclusions are not immunoreactive for ubiquilin 2.

Double immunostaining for ubiquilin 2 and AT8 in an AD case (A), a G272V MAPT mutation carrier (B) and a sporadic PiD case (C). Detailed images of double labeling show neurons with AT8-positive tau inclusions (D-O). Images were spectrally unmixed, and shown separately and merged in artificial colors (ubiquilin 2: green; AT8: red). AD, Alzheimer’s disease; CA, cornu ammonis FTD, frontotemporal dementia; MAPT, microtubule-associated protein tau, Ub2, ubiquilin 2.

https://doi.org/10.1371/journal.pone.0076598.g002

Western Blot

Brain material (Table 1) was homogenized in lysis buffer (150 mM NaCl, 50 mM Tris HCl pH 7.6, 1% NP-40, 0.5% Sodium Deoxycholate, 0.1% SDS, 2 mM EDTA) supplemented with protease inhibitors (Complete protease inhibitor cocktail from Roche, Penzberg, Germany) and phosphatase inhibitors (PhosSTOP, Roche). Lysates were incubated for 30 min at 4°C and centrifuged for 20 min at 20.000 × g. For analysis of insoluble proteins, pellets were resuspended in 70% formic acid, incubated for 30 min at room temperature and lyophilized using a SpeedVac concentrator. Proteins were resolved in PBS and loaded in a 10% SDS gel. Protein concentrations of the supernatants were determined using Bicinchoninic acid (BCA) assay reagent (Thermo, Fisher Scientific, Rockford, IL, USA). Equal amounts of protein were loaded on a 10% SDS gel and analysis was performed as described previously [20]. Ubiquilin 2 antibody (Novus Biologicals) and GAPDH (Millipore Corporation, Billerica, MA, USA) antibody, respectively, were employed in a 1:1000 dilution. GraphPad Prism Sotfware was used for statistical analysis of the data. Mann-Whitney test was used to test for differences between groups. A p value of <0.05 was taken as significant.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Ubiquilin 2 levels are decreased in AD patients.

(A) Western blot analysis of brain lysates from AD (Braak stage 5-6) and control (Braak stage 1) brain material using an ubiquilin 2 antibody and a GAPDH antibody as loading control. (B) Quantification of Western blot signals in (A). Ubiquilin 2 signals were normalized to GAPDH and statistically analysed (Mann–Whitney test, * p <0.05). (C) Supernatant and pellet lysate fractions of brain material from 3 AD patients (Braak stage 6) and 3 controls (Braak stage 1) were analysed by Western blotting. Pellets were treated with formic acid to dissolve aggregates prior to Western blotting. GAPDH was used as loading control (supernatant); tau was used as a control for successful extraction of the aggregates in the AD samples (pellet). AD, Alzheimer’s disease.

https://doi.org/10.1371/journal.pone.0076598.g003

RNA isolation and cDNA synthesis

For RNA isolation, brain material (Table 1) or cells were homogenized in TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and total RNA was isolated using the QIAcube (QIAGEN, Venlo, The Netherlands). The protocol was used according to the manufacturer’s specifications. RNA concentrations and purity were assessed by OD measurements at 260 and 280nm on a NanoDrop spectrophotometer (Thermo Scientific). For cDNA synthesis, 1μg of RNA and 125pmol OligodT12 primer were dissolved in a total of 10μl H2O and incubated at 72°C for 10min. Reverse transcriptase mix was added, consisting of 5μl 5 × first-strand buffer (Invitrogen), 0.5μl SuperScript II RNA polymerase (Invitrogen), 10mM dNTPs and 25mM MgCl2 in a total of 15μl H20. The mixture was incubated at 42°C for 1h, followed by 15min at 70°C. cDNA quality was assessed on 0.8% agarose gel.

Real-time qPCR

Real-time qPCR was performed using the Light Cycler 480 system (Roche Applied Science, Indianapolis, IN, USA). Oligonucleotide primers (Sigma) used for qPCR are listed in table 2. Reaction volumes of 5μl contained cDNA, 0.1μM Universal Probe Library probe (Roche Applied Science), also listed in table 2, 0.4μM forward primer, 0.4μ reverse primer and 2.5μl 2 × LightCycler 480 Probes Master (Roche Applied Science). After denaturation for 10min at 95°C, amplification was performed using 35 cycles of denaturation (95°C for 10s), followed by annealing (58°C for 15s) and elongation (72°C for 15s). Results were analyzed using the LightCycler 480 software (Roche Applied Science) version 1.5. GraphPad Prism Software was used for statistical analysis of the data. Mann-Whitney test was used to test for differences between groups. A p value of <0.05 was taken as significant.

Ubiquilin 2 is not associated with inclusions in tauopathies

Deng et al. showed that wt ubiquilin 2 accumulates in pathological inclusions in the spinal cord and brain of ALS/dementia patients [2]. Because ubiquilin 2 associates with different types of protein inclusions, this could indicate a general role for ubiquilin in protein aggregates associated with neurodegeneration. To investigate whether ubiquilin 2 is involved in tau pathology, immunohistochemical analyses were performed on paraffin-embedded brain tissue from tauopathy patients and non-demented controls (Table 1). Hippocampus and temporal cortex are areas with prominent tau pathology in these diseases and were thus selected for the analysis. In all cases, ubiquilin 2 reactivity was diffusely distributed throughout the cytoplasm of granular and pyramidal cells in the hippocampus (Figure 1). A number of neurons displayed a strong nuclear ubiquilin 2 signal. The intensity of ubiquilin 2 immunoreactivity varied among subfields of the hippocampus and areas that displayed a strong ubiquilin 2 signal differed between cases. However, these differences in ubiquilin 2 staining patterns were also observed in control cases and showed no consistent pattern correlating with disease. Ubiquilin 2 reactivity was also found associated with extracellular plaque-like structures and granules resembling perisomatic granules described by Probst et al. [21] and this was predominantly found in AD cases (Figure 1C, D). No intracellular ubiquilin 2 positive inclusions were observed in hippocampal neurons in any of the tauopathy cases. In line with the findings in the hippocampus, ubiquilin 2 was diffusely distributed in neurons in the temporal cortex (Figure S1).

To further investigate the potential role of ubiquilin 2 in tau pathology, hippocampal tissue of eight AD and six FTD patients (Table 1) was double-stained with the AT8 antibody, which specifically detects hyperphosphorylated tau (Figure 2). Consistent with literature, pronounced AT8 immunoreactivity was detected in Cornu ammonis (CA) region 1, CA region 3 and the subiculum in the cohort [19,22]. In Pick’s disease, tau inclusions known as Pick bodies were also present in the granular cells of the dentate gyrus [22]. Spectral unmixing demonstrated that ubiquilin 2 does not co-localize with tau-positive inclusions (Figure 2, D-O) in any of the cases. These results indicate that direct association of ubiquilin 2 with pathological inclusions as described for ALS/dementia using the same antibody is not detected in tauopathies.

Ubiquilin 2 levels are decreased in AD

Lower levels of ubiquilin 1 were previously reported for AD brain [15]. Therefore, ubiquilin 2 protein levels in cortical tissue of four AD patients and controls were determined by Western blotting. The result showed that ubiquilin 2 levels were significantly decreased in AD (Figure 3) The ubiquilin 2 antibody we used detected a weak band below the ubiquilin 2 band, which suggests minor crossreactivity with the 66 kD ubiquilin 1. Since the soluble protein fraction was used in this analysis, we wanted to exclude the possibility that ubiquilin 2 protein was retained by aggregates present in the insoluble protein fraction. Thus, supernatant and pellet fraction from whole cell lysates were examined from three AD and three control cases. Pellets were treated with formic acid to dissolve aggregated proteins prior to Western blot analysis. Consistent with our first analysis ubiquilin 2 was reduced in supernatant fraction of the AD samples (data not shown). A smear of tau positive material was detected in the pellet fraction of AD samples indicating that extraction of proteins from the pellet was successful (Figure 3 C). The ubiquilin 2 levels in the pellet fraction varied between patients, but showed no consistent difference between AD patients and controls. This indicates that ubiquilin 2 is not sequestered in the insoluble fraction, which is supported by the lack of co-localization of ubiquilin 2 immunoreactivity with tau inclusions (Figure 2, D-O). Therefore, our data indicate that the expression of ubiquilin 2 protein is decreased in the brains of AD patients.

To investigate the cause of the decrease in protein levels, we analyzed whether the ubiquilin 2 mRNA levels are decreased as well and found no difference between AD and controls, neither in the hippocampus nor in the cortex (Figure 4). This suggests that a post-transcriptional mechanism (decreased translation or increased degradation) is involved in the downregulation of the ubiquilin 2 levels in AD.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. Ubiquilin 2 mRNA levels are not changed in AD.

Ubiquilin 2 mRNA levels were assessed by qPCR in hippocampus and temporal cortex from AD and control brain. GAPDH was used as a reference gene (Mann–Whitney test, non-significant).

https://doi.org/10.1371/journal.pone.0076598.g004

Low ubiquilin 2 levels are not linked to tau pathology

To determine whether ubiquilin 2 levels are selectively lower in cells showing tau pathology, a semi-quantitative analysis of AD cases double-stained with ubiquilin 2 and AT8 was performed (described in detail in material and methods). Briefly, number of neurons with and without AT8 staining, respectively, were determined in the CA1 area of the hippocampus in 7 AD patients and assigned to AT8 positive or AT8 negative group. Subsequently, for each group numbers of cells showing no or low immunoreactivity for ubiquilin 2 (ubiquilin 2 low) and with high immunoreactivity (ubiquilin 2 high), respectively, were counted and frequency distributions were compared between the two groups using a chi-square test (Table 3). Analysis of the total cell numbers from 7 patients showed no association of lower ubiquilin 2 levels with AT8 staining. In contrast, we found that ubiquilin 2 signals are higher in AT8 positive cells. Since the cell numbers counted were different for each patient, we analysed the cases separately and found a high variation in frequency distributions. However, only case 6 showed lower ubiquilin 2 levels in AT8 positive group (Table 3). Clearly, our analysis showed that low ubiquilin 2 levels in AD detected in the Western blot analysis are not associated with the occurrence of tau pathology.