Screening of tau protein kinase inhibitors in a tauopathy-relevant cell-based model of tau hyperphosphorylation and oligomerization

Tauopathies are a class of neurodegenerative disorders characterized by abnormal deposition of post-translationally modified tau protein in the human brain. Tauopathies are associated with Alzheimer’s disease (AD), chronic traumatic encephalopathy (CTE), and other diseases. Hyperphosphorylation increases tau tendency to aggregate and form neurofibrillary tangles (NFT), a pathological hallmark of AD. In this study, okadaic acid (OA, 100 nM), a protein phosphatase 1/2A inhibitor, was treated for 24h in mouse neuroblastoma (N2a) and differentiated rat primary neuronal cortical cell cultures (CTX) to induce tau-hyperphosphorylation and oligomerization as a cell-based tauopathy model. Following the treatments, the effectiveness of different kinase inhibitors was assessed using the tauopathy-relevant tau antibodies through tau-immunoblotting, including the sites: pSer202/pThr205 (AT8), pThr181 (AT270), pSer202 (CP13), pSer396/pSer404 (PHF-1), and pThr231 (RZ3). OA-treated samples induced tau phosphorylation and oligomerization at all tested epitopes, forming a monomeric band (46–67 kDa) and oligomeric bands (170 kDa and 240 kDa). We found that TBB (a casein kinase II inhibitor), AR and LiCl (GSK-3 inhibitors), cyclosporin A (calcineurin inhibitor), and Saracatinib (Fyn kinase inhibitor) caused robust inhibition of OA-induced monomeric and oligomeric p-tau in both N2a and CTX culture. Additionally, a cyclin-dependent kinase 5 inhibitor (Roscovitine) and a calcium chelator (EGTA) showed contrasting results between the two neuronal cultures. This study provides a comprehensive view of potential drug candidates (TBB, CsA, AR, and Saracatinib), and their efficacy against tau hyperphosphorylation and oligomerization processes. These findings warrant further experimentation, possibly including animal models of tauopathies, which may provide a putative Neurotherapy for AD, CTE, and other forms of tauopathy-induced neurodegenerative diseases.


165
The culture lysate harvesting for N2a cells and CTX culture were identical. After the treatment, 166 conditioned media were collected from each well and added into separate tubes on ice and 8 167 centrifuged at 10,000 x g for 10 min at 4 o C. Lysis buffer was added to the attached cells on the 12-168 well plates (100 µl per well). The Triton-X lysis buffer included: 1mM DTT, 1% phosphatase 169 inhibitors (Sigma), 1% Mini-Complete protease inhibitor cocktail tablet (Roche Biochemicals), 170 and 1% Triton X-100. The attached cells were then scraped down into the lysis buffer and collected 171 into separate 1.5 ml Eppendorf tubes. The insoluble pellets from the conditioned culture media 172 were combined with the lysed cells in the lysis buffer. The cell lysates were incubated for 90 173 minutes at 4 º C and then centrifuged at 15,000 rpm for 15 minutes to remove cell debris. hyperphosphorylation and oligomerization for 6h and 24h (Figure 1a,  tau protein bands at 46 kDa and 48 kDa at 6h and 24h (Figure 1a). The intensity of the band at 46 217 kDa was detected at higher levels compared to the band at 48 kDa in control samples. (Figure 1a).

218
Treatment with OA (100 nM) for 6h and 24h showed a dramatic decrease in levels of the  (Figure 1a, 1b). 234 On the other hand, in our cell culture experimental conditions, treatment with OA (100 235 nM) for less than 6h did not show any detectable tau bands with CP13 and PHF-1 (data not shown).
It is well-known that OA induces apoptosis in human neuroblastoma cells, mouse neuroblastoma, 237 and rat cerebellum neurons (47). Thus, the time points were not increased beyond 24h of treatment 238 to avoid tau phosphorylation modifications resulting from proteolysis and neural death.

239
Probing with CP13 (pSer202) antibody did not show any detectable bands of tau protein in 240 control samples (Figure 1a). This result indicates that endogenous phosphorylation of tau at 241 Ser202 site is low under normal growth conditions. However, with OA treatment, CP13 showed 242 HMW band formed at 110 kDa (x2 size of monomeric tau) with 6h and 24h (Figure 1a, 1b).

247
Low molecular weight monomeric tau (LMW-MT) bands were not detected with either 248 CP13 or PHF-1. It should be noted that the DA9 antibody recognizes total tau epitopes from aa.

255
Taken together, these data strongly suggest that OA treatment caused protein phosphatase   the intact αII-spectrin band was detected at 240 kDa, and no SBDP150/145 or SBDP120 was 293 observed with the TBB treated conditions suggesting a healthy culture.

296
CaMKII by blocking the Ca++ mitochondrial permeability (54) ( Table 1). Thus, CsA was selected 297 in this study as a calcium-dependent kinase inhibitor to assess its effect on OA-induced tau  Table 3).

302
Phosphorylation sites at human tau threonine 181 and 231 have been shown to differentiate AD 303 patient from a control subject (35, 55). Hence, RZ3 (pT231) and AT270 (pT181) antibodies were 304 used to study these sites.

305
RZ3 antibody showed a complete reduction of the 48 kDa band (monomeric p-tau) when cells 306 were treated with CsA ( Supplementary Figure 1a, Table 3). As for apoptotic pathway activation, αII-spectrin antibody did not show any effect 324 on the 240 kDa band (intact form), and the SBDP150/145 or SBDP120 bands were not detected 325 with EGTA treatment.

327
Tau is a substrate of Glycogen synthase kinase-3 (GSK-3)(57), and p-tau phosphorylation 328 and oligomerization could be inhibited by GSK-3 inhibition. To test this hypothesis, the effects of  Table 3).

347
Probing with total tau DA9 showed with A-107 treatment, a 13% reduction of 170 kDa band (DA9) 348 and did not show a statistically significant effect on the 48 kDa band (Supplementary Figure 2a, b, Table 3). As for caspase-3, calpain, and cell injury activation, αII-spectrin did not show SBDP 350 150/145 or SBDP120 post-treatment, indicative of a healthy neuronal culture.

352
Saracatinib is an inhibitor of the Src/abl kinase family, developed initially for several types of 353 cancer but withdrawn for the lack of effectiveness (60). However, Saracatinib is also a potent 354 inhibitor of Fyn kinase, which is linked to tau (16). Fyn has been reported to phosphorylate 355 dendritic tau, which allows Fyn to localize to the post-synaptic density (16). In the current study,   (Figure 2a, 2b, and Table 3). K252a treatment caused 40% reduction of 170 kDa (DA9; 375 oligomeric p-tau) compared to OA treatment alone (average of n=3) (Figure 2a, 2b, and Table   376 3   Table 4).

482
Unexpectedly, EGTA caused an adverse effect in CTX culture by further enhancing physiological In the present study, OA was used to induce tau hyperphosphorylation and oligomerization 500 in mouse neuroblastoma N2a culture and rat primary cerebrocortical neuronal cultures (CTX) to 501 screen for various tau kinase inhibitors as potential drug candidates. The N2a neuronal cultures 502 have been widely used to study mechanisms of neurodegeneration because they are a homogenous  It is widely established that PP2A is the primary enzyme responsible for dephosphorylation 529 of tau protein throughout the brain, controlling all tau phosphorylation sites. PP2A activity is 530 decreased in AD and TBI brains (12, 82). Therefore, the OA-induced inhibition of PP2A is a highly relevant model to study various tau protein kinase inhibitors as modulators of tau 532 hyperphosphorylation and oligomerization targeting tau pathology ( Figure 6, Table 2).

533
In contrast to N2a cell culture, the monomeric form of tau (ranging from 63 kDa -67 kDa) 534 was only observed in CTX culture. OA might be able to cause tau oligomerization inducing  Table 4) and was less 557 effective in N2a cells (Figure 2a, 2b, and Table 3). It was also observed that the effect of AR is 558 more prominent compared to another GSK3 inhibitor, A-107 in CTX primary culture. This effect 559 could be attributed, in part, to the high selectivity and specificity of AR to GSK3β (89) compared 560 to A-107. A-107 display selectivity for both GSK3α and GSK3β (K i = 0.6 nM for both) (90) thereby 561 might dilute the effect of inhibition of GSK3β, which is regarded as the critical kinase in AD (88).

562
Similarly, a study has shown that hypothermia-induced tau hyperphosphorylation was 563 reduced with AR treatment in human neuroblastoma SH-SY5Y 3R-Tau (76). In another study, AR 564 protected N2a cell culture against apoptosis by inhibition of the phosphatidylinositol-3 565 kinase/protein kinase B pathway and showed neuroprotective properties against neurotoxicity 566 caused by the β-amyloid peptide in hippocampal slices (91). The lack of AR effect on N2a culture 567 might be attributed to differences in cellular mechanisms from CTX culture, mediating OA-568 induced tau phosphorylation at multiple levels and different sites.

569
LiCl is well-known to inhibit GSK3 and other kinases (76). In CTX culture, LiCl caused 570 dramatic inhibition of basal and OA-induced tau hyperphosphorylation at all tested tau epitopes.

571
Consistent with previous reports, LiCl was shown to reduce tau phosphorylation in cultured cells,