RD4 cannot recognize 4R tau with deamidation at N279

Since N279 is located in the RD4 epitope, we next examined the effect of deamidation of N279 on immunoreactivity to RD4. Substitution of N279 to aspartic acid was introduced into 4R1N human tau isoform by site-directed mutagenesis, and immunoreactivity to mutant (N279D)-4R tau before and after phosphorylation by protein kinase A was compared with that of wild-type tau. As shown in Figure 2(a-d), RD4 was not able to recognize N279D-4R tau regardless of the phosphorylation state of Ser262, which is located near the RD4 epitope. To further investigate the deamidation of N279 of tau, we immunized a rabbit with a synthetic peptide, VQIIDKKLDLSNVQSKC, which is the RD4 antigen peptide with substitution of N279 to Asp. The antiserum anti-4R labeled both wild-type (WT) and N279D-4R tau equally (Figure 2c) and the immunoreactivity was unaffected by phosphorylation of Ser262/356 with PKA (Figure 2c). The anti-4R antibody specifically bound with recombinant human 4R tau isoforms, as did RD4, but did not react with the 3R tau isoforms (Figure 2e). The specificities of RD4 and anti-4R antibodies were further analyzed by means of ELISA assay using the antigen peptide of RD4 (L-Asn), the peptide with N279D substitution (L-Asp), the peptide with L-isoAsp substitution (L-isoAsp) and the peptide with N279 D-Asp substitution (D-Asp) (Figure 2f,g). These modifications are known to be related to deamidation of Asn residue (Figure 2h). RD4 failed to react with L-Asp (antigen peptide with N279D substitution), whereas anti-4R reacted almost equally with L-Asn (wild-type) and L-Asp peptide. Neither RD4 nor anti-4R reacted with D-Asp or L-isoAsp peptide (Figure 2f,g).

Open in a separate window

Figure 2

RD4 cannot recognize N279D-4R tau, but new anti-4R labels both WT and N279D-4R tau equally. Immunoblot analysis of wild-type (WT) and N279D mutant tau before (WT, N279D) and after (p-WT, p-N279D) phosphorylation with PKA, using T46 (a), RD4 (b), anti-4R (c) and 12E8 (d) antibodies. Immunoblot analysis of six recombinant human tau isoforms with T46, RD3, RD4 and anti-4R antibodies (e). Specificities of RD4 (1:1000 dilutions) and anti-4R (1:3000 dilutions) antibodies for synthetic peptides, L-Asn (wild-type), L-Asp, L-isoAsp and D-Asp peptides (0.625~10 μg/mL) tested by ELISA assay (f). Quantitation of the ELISA results (the mean of absorbance at 490 nm on 1.25 μg/mL peptide is shown (g). Pathways for deamidation of asparaginyl residues (h). L-Asn residue can be converted spontaneously via a succinimidyl intermediate to form L-Asp, D-Asp. L-isoAsp and D-isoAsp residue (modified from Ref. [9]).

Antiserum against peptide with deamidation of N279 strongly stained tau smears in AD

The immunoreactivity of anti-4R was compared to that of RD4, RD3 and T46 in immunoblotting of Sarkosyl-insoluble tau from tauopathy brains (Figure 3). Samples of insoluble tau from three AD (lane 1–3), two PSP (lane 3, 4) and two CBD (lane 6, 7) cases were examined (Figure 3a-d). RD4 faintly stained only two bands at 64 and 68 kDa in AD brains, whereas it stained several tau fragments in PSP and CBD, in addition to the two bands at 64 and 68 (Figure 3b). In contrast, anti-4R stained tau bands and smears in AD, like RD3, and this staining was much stronger than that of tau bands and smears in PSP and CBD (Figure 3d). These results strongly suggest that tau in AD brains is predominantly deamidated at N279, and that the levels of deamidation are much lower in tau from PSP and CBD brains.

Open in a separate window

Figure 3

New anti-4R strongly stained tau bands and smears in AD. Immunoblot analyses of Sarkosyl-insoluble tau from three AD (1–3), two PSP(4, 5) and two CBD (6, 7) cases with T46 (a, 1:2000 dilution), RD4 (b, 1:1000 dilution), RD3 (c, 1:1000 dilution) and anti-4R antibodies (d, 1:2000 dilution).

Next, we compared the immunostaining of tau pathologies with RD3, RD4 and anti-4R on formalin-fixed brain sections of AD (Figure 4). None of the three antibodies stained tau on formalin-fixed sections in the absence of formic acid treatment or autoclaving (not shown), suggesting that the epitopes of these antibodies are masked. After autoclaving or formic acid treatment, neurofibrillary tangles (NFTs) and neuropil threads (NTs) were detected with these antibodies, and dual treatment with both autoclaving and formic acid strongly enhanced the staining (not shown). The new anti-4R antibody stained intracellular NFTs and NTs more extensively than did RD4 (Figure 4). This was most evident in the CA1 region, where anti-4R strongly stained RD3+/RD4− NFTs (Figure 4d,f). This result indicates that the RD4 epitope is deamidated in pathological tau from AD brain, especially in RD3+/RD4− NFTs. RD3 stained abundant ghost tangles in entorhinal cortex and NFTs in CA1, but failed to stain fine processes of NFTs and NTs (Figure 4), as previously reported [10]. Anti-4R also failed to detect ghost tangles in entorhinal cortex.

Open in a separate window

Figure 4

New anti-4R antibody stained intracellular NFTs more extensively than did RD4. Immunostaining of entorhinal cortex (EC) and CA1 sections of AD brain after autoclaving and formic acid treatments, using RD3 (a, b), RD4 (c, d) and anti-4R (e, f) antibodies. NFTs with higher magnification are shown in insets. Bar=100 μm (25 μm in insets).

Preparation of sarkosyl-insoluble fractions

Brain samples (0.5 g) from patients with AD, PSP and CBD were each homogenized in 10 ml of homogenization buffer (HB: 10 mM Tris–HCl, pH 7.5 containing 0.8 M NaCl, 1 mM EGTA, 1 mM dithiothreitol). Sarkosyl was added to the lysates (final concentration: 2%), which were then incubated for 30 min at 37°C and centrifuged at 20,000 g for 10 min at 25°C. The supernatant was divided into eight tubes (each 1.3 mL) and centrifuged at 100,000 g for 20 min at 25°C. The pellets were further washed with sterile saline (0.5 mL/tube) and centrifuged at 100,000 g for 20 min. The resulting pellets were used as Sarkosyl-insoluble fraction (ppt).

LC/MS/MS analysis of sarkosyl-insoluble tau

Sarkosyl-insoluble tau from AD brains was subjected to SDS–PAGE using 4–20% polyacrylamide gel (PAGE mini, Daiichi, Tokyo). After staining with Coomassie brilliant blue R-250 (CBB), the bands corresponding to the phosphorylated tau (64 and 68 kDa) were cut out. In-gel digestion of proteins with 1 μg/ml trypsin was carried out as described previously and the resulting peptides were analyzed by an ion-trap spectrometry (Velos Pro; Thermo Fisher Scientific Inc. Waltham, MA). The MS/MS data files were searched and analyzed using the Mascot Server (Matrix Science Inc., Boston, MA).

Recombinant tau proteins

Expression constructs for six human tau isoforms in plasmid pRK172 were kindly provided by Dr. Goedert. Site-directed mutagenesis was used to change N279 to Asp (numbering refers to the 441-amino-acid isoform of human brain tau) in the four-repeat 412-amino-acid isoform (expressed from cDNA clone htau46). Wild-type and mutated tau proteins were expressed in Escherichia coli BL21(DE3) and purified as described previously[22]. For in vitro phosphorylation, purified tau (10 μg/ml) was incubated with PKA (10,000 U/ml; New England Biolabs, Beverly, MA) in 30 mM Tris–HCl buffer (pH 7.5) containing 0.1 mM EGTA, 10 mM MgSO₄, 0.8 mM PMSF and 2 mM ATP, at 30°C for 1 hr.

Antibodies

RD3 (directed to residues 209~224: Millipore), RD4 (residues 275~291: Millipore), T46 (residues 404~441: Invitrogen) and pS396 (phospho-Ser396: Calbiochem) were purchased. Antiserum anti-4R was raised against a synthetic peptide VQIIDKKLDLSNVQSKC which corresponds to residues 275~291 of human tau (441 residues), with substitution of N279 to Asp (Sigma Aldrich Japan). The peptide was conjugated to m-maleimidobenzoyl- N-hydrosuccinimide ester-activated keyhole limpet hemocyanin (KLH). The KLH-peptide complex (1 mg of each immunogen) emulsified in Freund’s complete adjuvant was injected subcutaneously into a New Zealand White rabbit, followed by 5 weekly subcutaneous injections of 150 μg KLH-peptide complex emulsified in Freund’s incomplete adjuvant, starting 3 weeks after the first immunization.

ELISA assay

Each synthetic peptide consisting of residues 275~291 (VQIINKKLDLSNVQSKC) with the fifth position being replaced by L-Asp, L-isoAsp, or D-Asp was synthesized by the solid-phase method (Sigma Aldrich Japan) These peptides, L-Asn (wild-type), L-Asp, L-isoAsp, D-Asp (0.625~10 μg/ml in 50 mM Tris–HCl, pH 8.8) were coated onto microtitre plates (SUMILON) at 4°C for 16 h. The plates were blocked with 10% fetal bovine serum (FBS) in PBS, incubated with the first antibodies (RD4, 1:1000; anti-4R, 1:3000) diluted in 10% FBS/PBS at room temperature for 1.5 h, followed by incubation with HRP-goat anti-rabbit IgG (Bio-Rad) at 1:1000 dilution, and reacted with the substrate, 0.4 mg/ml o-phenylendiamine, in citrate buffer (24 mM citric acid, 51 mM Na₂HPO₄), The absorbance at 490 nm was measured using Plate Chameleon (HIDEX) as described [23].

Microtubule assembly and tau binding

Purified recombinant wild-type and mutant tau (htau46) proteins (0.1 mg/ml, 2.3 μM) were incubated with bovine brain tubulin (1 mg/ml, 20 μM, cytoskeleton) in assembly buffer at 37°C, as described [22]. The assembly of tubulin was monitored in terms of the change in turbidity at 350 nm. The binding assay was performed as described [24]. Briefly, purified tubulin was incubated at 37°C in the presence of 1 mM GTP and 20 μM taxol. Tau protein was added at various concentrations and each mixture was incubated for 10 min. The suspensions were centrifuged for 100,000 g at 37°C. The resulting pellets were resuspended in 50 mM PIPES pH 6.9, 1 mM EGTA, 0.2 mM MgCl₂, 5 mM DTT, 0.5 M NaCl. The pellets and supernatants (containing bound and free tau, respectively) were subjected to SDS-PAGE and stained with Coomassie brilliant blue R250. The gels were scanned at 400 dpi on a gel scanner and evaluated using the software provided.

Gel electrophoresis and immunoblotting

Samples were run on gradient 4-20% or 10% polyacrylamide gels and electrophoretically transferred to PVDF membranes. Residual protein-binding sites were blocked by incubation with 3% gelatin (Wako) for 10 min at 37°C, followed by overnight incubation at room temperature with the primary antibody. The membrane was then incubated for 1 hr at room temperature with anti-rabbit IgG (BA-1000, Vector Lab) or anti-mouse IgG (BA-2000, Vector lab), then incubated for 30 min with avidin-horseradish peroxidase (Vector Lab), and the reaction product was visualized by using 0.1% 3,3-diaminobenzidine (DAB) and 0.2 mg/ml NiCl₂ as the chromogen.