The β-amyloid peptide (Aβ) and tau are key molecules in Alzheimer’s disease causing neuronal dysfunction and cell death. Evidence exists that tau mediates the pathology downstream of Aβ. N-methyl-D-aspartate receptors (NMDARs) are supposed to play an essential role in the pathophysiology of Aβ and tau. However, the exact mechanisms deciphering how Aβ could induce tau-dependent neuronal dysfunctions are still unclear.

Here we show that virus-mediated expression of human tau causes neuronal degeneration in organotypic hippocampal slice cultures from APP transgenic mice but not in non-transgenic cultures. Treatment with therapeutic concentrations of NMDAR open channel blocker memantine completely abolished tau-dependent cell death in APP transgenic cultures. Removing extracellular calcium with calcium chelator BAPTA also prevented tau toxicity. Our data indicate that human tau-dependent neuronal cell death in APP transgenic slice cultures is mediated by calcium influx through NMDARs.

Alzheimer’s disease (AD) is characterized by aggregates of Aβ and tau. Evidence exists that tau mediates pathological events downstream of Aβ^[1][2][3]. NMDARs play an essential role in Aβ-induced neuronal dysfunction, and inhibition of NMDARs protects against Aβ-induced synaptic loss^[4][5]. Furthermore, the effect of Aβ on tau is mediated by NMDARs as blocking glutamate-binding to NMDARs prevents the induction of a tau-dependent neuronal cell death in APP transgenic cultures^[6]. NMDARs can modulate intracellular signaling cascades via calcium influx into the cell. However, metabotropic NMDAR signaling, i.e., independent of ion flux, has also been described^[7][8]. It has been recently shown that oligomeric Aβ caused synaptic loss in hippocampal slice cultures via NMDAR signaling but independently of calcium influx^[9]. In the present study, we investigated whether the tau-dependent neuronal cell death in APP transgenic cultures depends on ionotropic (calcium influx) or metabotropic NMDAR functions.

It has been shown that synaptic loss in APP transgenic slice cultures occurred via NMDA receptor signaling but independent of calcium flux. Here, we aim to determine whether neuronal tau-dependent cell death in the same cultures also depends on calcium flux-independent NMDA receptor activity.

We prepared organotypic hippocampal slices from APP transgenic mice and non-transgenic littermates and expressed EGFP (control) or EGFP-coupled human 441 tau selectively in neurons using neurotropic Sindbis virus. After infection, EGFP fluorescence was observed in all hippocampal subregions, i.e., dentate gyrus (DG) as well as in the cornu ammonis areas with the strongest signals in CA3 (Fig. 1A).

In organotypic slices from APP transgenic mice, Aβ induced a human tau-dependent cell death^[6]. Here, we first confirmed this data by showing that EGFP-tau expression increased cytotoxicity assay luminescence in APP transgenic compared to non-transgenic cultures (Fig. 1B; n=5, p=0.019, two-tailed paired student’s t-test). Considering slight variations in infection efficiencies between different cultures and experiments, we normalized cytotoxicity assay luminescence from each culture to the EGFP/EGFP-tau signal, measured in the lysate from the respective culture.

In contrast to EGFP-tau expressing neurons, no toxicity was observed in APP transgenic cultures upon expression of EGFP alone (Fig. 1B). This suggests that transgenic expression of APP and its cleavage products, such as Aβ, do not cause cell death per se but require the presence of human tau. To determine whether tau toxicity in APP transgenic cultures was mediated by ion flux through NMDA receptors, we first treated cultures with clinically relevant concentrations (1 and 10 µM, according to^[10]) of the NMDAR open channel blocker memantine. 1 µM memantine has been shown to block synaptic transmission in organotypic slices^[9]. At both concentrations, memantine treatment completely abolished tau-dependent cell death in APP transgenic cultures (Fig. 1B). To corroborate that the influx of calcium through the NMDAR is an essential process in this event, we treated slices with 2 mM non-cell-permeable calcium chelator BAPTA or its solvent NaHCO3 as control. BAPTA treatment completely abolished tau toxicity in APP transgenic cultures indicating that tau-dependent cell death requires calcium flux into the neuron.

We showed that human tau expression in APP transgenic but not in non-transgenic cultures causes neuronal cell death. The induction of tau toxicity in these cultures can be blocked by applying anti-Aβ antibodies^[6], indicating that Aβ and not APP or any other APP cleavage product confers toxicity to human tau. We further provide evidence that calcium influx through NMDARs is required for this process. In contrast, previous studies showed that Aβ-induced synaptic loss did not depend on calcium influx^[9] and did not depend on the presence of tau^[6]. This indicates that Aβ-induced loss of synapses and neuronal cell death are two independent events, induced by distinct pathways. Although our data show that tau-dependent toxicity in APP transgenic cultures requires NMDAR-mediated calcium flux, it remains to be shown whether calcium influx is an event upstream or downstream of tau toxicity. Ittner and colleagues proposed a model in which the presence of tau is required for Aβ-induced excitotoxicity, which supports a role of calcium flux downstream of tau^[3]. In contrast, pharmacological activation of extrasynaptic NMDARs- even in the absence of Aβ- increased phosphorylation and toxicity of tau^[6]. This suggests a role of calcium influx upstream of tau. Thus, it will be important to analyze whether the treatment with memantine or BAPTA prevents aberrant posttranslational modifications of tau, such as increased phosphorylation. This would confirm a role of calcium flux upstream of tau. These experiments could be performed in a follow-up study.

Organotypic hippocampal slice cultures

6 to 7 day old APP transgenic (ArcAβ mice^[11]) and non-transgenic littermates were decapitated, brains were removed, hippocampi were isolated and cut into 400 μm thick slices. Three slices were placed per cell culture insert (Millipore). Slices were placed in a culture medium (minimum essential medium Eagle with HEPES modification, 25% basal medium with Earle’s modification, 25% heat-inactivated horse serum, 2 mM glutamine, 50 units per ml penicillin, 50 μg/ml streptomycin, 0.6% glucose, pH 7.2). Culture medium was exchanged every second or third day. On days in vitro (DIV) 11, culture medium was replaced by low-serum Nb-N2 medium (Neurobasal medium, 0.5% heat-inactivated horse serum, 2 mM glutamine, 50 units/ml penicillin, 50 μg/ml streptomycin, 0.6% glucose, 1×N2 supplement, pH 7.2) to ensure more defined condition during analysis. On DIV12, slices were infected with sindbis virus.

Sindbis virus

The following viral constructs were used for experiments: pSinRep5-EGFP, pSinRep5-EGFP-441wt tau. Infection of slices causes neuron-specific expression of EGFP or EGFP-coupled tau.

Pharmacological treatments

BAPTA (1,2-Bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid) and memantine (3,5-Dimethyl-tricyclo[3.3.1.13,7]decan-1-amine hydrochloride) were purchased from Tocris. Cultures were treated in parallel to EGFP/EGFP-tau expression from DIV12-16.

Memantine and BAPTA (each prepared from the same batch as in the present study) are pharmacologically active at the used concentrations^[9].

Cytotoxicity analysis

On DIV 16, cell culture supernatant from each well was collected for cytotoxicity analysis. In addition, slices were lysed (all three slices per well were pooled) in RIPA buffer (50 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 1% NP-40, 0.5% deoxycholate and 0.1% SDS, pH 8.0) containing phosphatase inhibitor cocktail (Sigma) and protease inhibitor cocktail (Roche) and centrifuged at 5,000 g for 10 min at 4°C. Lysate and supernatant were frozen in liquid nitrogen and stored at -80°C until further use. Cytotoxicity was measured using CytotoxGlo assay (Promega), according to the manufacturer’s recommendations. EGFP and EGFP-tau signals were determined using microplate reader (Synergy HT; BioTek, Germany). Cytotoxicity assay luminescence was divided by EGFP or EGFP-tau signal to normalize toxicity to EGFP or EGFP-tau viral infection efficiencies.

The authors declare no conflicts of interest.

All animal experiments were performed in accordance with the guidelines of the Swiss veterinary cantonal office.

No fraudulence is committed in performing these experiments or during processing of the data. We understand that in the case of fraudulence, the study can be retracted by ScienceMatters.