Tau is a neuronal protein present both in the cytoplasm, where it binds microtubules, and in the nucleus, where it exerts unconventional functions including the modulation of Alzheimer's disease-related genes. Here, we describe a method to investigate the function of nuclear Tau while excluding any interferences coming from cytoplasmic Tau.
Tau is a microtubule binding protein expressed in neurons and its main known function is related to the maintenance of cytoskeletal stability. However, recent evidence indicated that Tau is present also in other subcellular compartments including the nucleus where it is implicated in DNA protection, in rRNA transcription, in the mobility of retrotransposons and in the structural organization of the nucleolus. We have recently demonstrated that nuclear Tau is involved in the expression of the VGluT1 gene, suggesting a molecular mechanism that could explain the pathological increase of glutamate release in the early stages of Alzheimer's disease. Until recently, the involvement of nuclear Tau in modulating the expression of target genes has been relatively uncertain and ambiguous due to technical limitations that prevented the exclusion of the contribution of cytoplasmic Tau or the effect of other downstream factors not related to nuclear Tau. To overcome this uncertainty, we developed a method to study the expression of target genes specifically modulated by the nuclear Tau protein. We employed a protocol that couples the use of localization signals and the subcellular fractionation, allowing the exclusion of the interference from the cytoplasmic Tau molecules. Most notably, the protocol is easy and is composed of classic and reliable methods that are broadly applicable to study the nuclear function of Tau in other cell types and cellular conditions.
The functions of Tau protein in the nucleus have garnered significant interest in recent years, as it has been shown to be closely associated with nucleic acids1,2,3,4,5,6. Indeed, a recent genome-wide study demonstrated that Tau binds genic and intergenic DNA sequences in vivo7. A role in nucleolar organization has been suggested8,9,10,11. Moreover, Tau has been proposed to be involved in DNA protection from oxidative and hyperthermic stress5,10,12,13, whereas mutated Tau has been linked to chromosome instability and aneuploidy14,15,16.
Until now, the challenges in studying the functions of Tau in the nuclear compartment remained almost unsolved due to the difficulties in dissecting the specific contribution of nuclear Tau from the contribution of cytoplasmic Tau. Moreover, the functions attributed to Tau molecules in the nuclear compartment, up to now, are only correlative because they lack an unequivocal demonstration of the direct involvement of nuclear Tau proteins. Indeed, the involvement of Tau in the mobility of retrotransposons or in the rRNA transcription or in DNA protection11,12,17,18,19 might be also explained by the contribution of cytoplasmic Tau or by the effect of other downstream factors not related to nuclear Tau.
Here, we provide a method that can solve this issue by exploiting a classical procedure to isolate the nuclear compartment combined with the use of Tau constructs 0N4R tagged with nuclear localization (NLS) or nuclear export signals (NES). This approach eliminates the complex issues related to possible artefacts due to the spillover of Tau molecules from the cytoplasmic compartment. Moreover, Tau-NLS and Tau-NES constructs induce the enrichment or the exclusion of Tau molecules from the nuclear compartment, respectively, providing positive and negative controls for the involvement of nuclear Tau molecules in a specific function. The protocol is technically easy and it is composed of classic and reliable methods that are broadly applicable to study the nuclear function of Tau in other cell types, differentiated or not, such as cancer cells that reactivate Tau expression20,21. Moreover, it might be applied also to other proteins that are present in both the cytoplasm and the nucleus in order to dissect biological functions related to different compartments.
1. Cell Culture
2. Cell Differentiation
3. Chimeric Constructs Cloning
4. Cell Transfection
5. Immunofluorescence
6. Western Blot
The strategy used to dissect the impact of nuclear Tau in gene expression avoiding the contribution of cytoplasmic Tau proteins has been depicted in Figure 1. Briefly, Tau proteins tagged with NLS or NES are accumulated in or excluded from the nuclear compartment, respectively. The functional effect of this unbalance is the alteration of the gene expression measured as the product of the VGluT1 gene.
Following the protocol description, SH-SY5Y cells were treated with RA for 5 days and then with BDNF for 3 days in order to obtain post-mitotic neuron-like cells (Figure 2). In the absence of RA and BDNF, undifferentiated SH-SY5Y cells assume a rounder morphology and form cell clumps. As expected, starting the differentiation protocol, clumps unwind and cells spread out neurites; at the end of differentiation, cells are uniformly distributed and interconnected via a network of branched neurites.
The day after seeding, cells have been transfected with Tau-NLS or Tau-NES plasmids (section 4.2) with cationic lipids. For cells expressing Tau-NLS or Tau-NES constructs, Tau subcellular localization can be detected by immunofluorescence with anti-Tau antibodies. Depending on the efficiency of transfection, cells display a strong nuclear staining merging with the DAPI signal or a cytoplasmic staining with empty nuclei if they are successfully transfected with Tau-NLS or Tau-NES, respectively (Figure 3). The lack of these specific signals indicates an inefficient transfection.
To analyse the proteins enriched in different subcellular compartment, cells were collected and counted in order to process equal amounts of cells per sample. Any standard fractionation method that exploits increasing detergent and ionic strength and increasing centrifugation speed can be used to separate the cellular compartments from one another and thus isolate the cytosolic, the membrane-bound, the cytoskeletal and the nuclear fractions.
Once the nuclei have been isolated, the nuclear soluble fraction and the chromatin bound fraction were separated by adding 3 U/µL of micrococcal nuclease and 5 mM CaCl2.
For Western blot analysis, equal volumes of cytoplasmic and membrane fractions and half volumes of the other fractions have been loaded on a gradient precast acrylamide gel, to correct for the different amount of buffer added at each step.
To verify the efficient separation of different fractions, the Western blot exploiting the following antibodies has been performed: anti-GADPH (present in all fractions except the cytoskeleton and particularly enriched in the cytoplasmic fraction); anti-actin (particularly enriched in the cytoskeletal fraction); anti-tubulin (particularly enriched in Cytoplasmic and cytoskeletal fractions); anti-H2B (enriched in the nuclear fractions) (Figure 4).
An enrichment of these markers in different subcellular fractions indicates that the fractionation is not well performed. It must be noted that any protocol for subcellular fractionation might present a 10-15% of contamination between fractions.
Once verified the successful fractionation of the sample, the Western blot has been performed to check the signal of Tau in the nuclear compartment and the VGluT1 signal in the total extract (Figure 5). While untagged Tau is detectable in all fractions, Tau-NLS is strongly enriched in the nuclear compartment and it is poorly detectable in the cytoplasmic fraction. On the contrary, Tau-NES is enriched in the cytoplasmic fraction and it is less detectable in the nuclear fraction. The presence of a small amount of Tau-NES into the soluble nuclear fraction has to be expected since, like the endogenous Tau, it is translocated into the nucleus and once into the nuclear compartment the nuclear export signal allows its translocation to the cytoplasm. The detection of a different enrichment for these two fusion proteins might indicate a problem in the efficiency of transfection or in the cloning of constructs or in fractionation.
Quantitative analysis of Western blot can be done using ImageJ. Values are normalized for the housekeeping gene specific for each fraction (GAPDH for cytoplasmic fraction; histone H2B for soluble nuclear and chromatin-bound fractions).
The graph in Figure 5B reports the ratio of Tau in the soluble nuclear fraction and cytoplasmic fraction to highlight that Tau-NLS is highly enriched in the soluble nuclear fraction (SNF) while Tau-NES is decreased. Moreover, Tau-NLS is enriched in the chromatin-bound fraction (CBF) with respect to the cytoplasmic fraction (CF) while Tau-NES is decreased. SNF/CF = 1 and CBF/CF = 1 correspond to Tau ratio in control cells. The endogenous Tau is weakly detectable in all fractions as expected. The graph in Figure 5C reports the quantification of VGluT1 expression in the total extracts of samples expressing different amount of nuclear Tau. In cells expressing Tau-NES, VGluT1 expression is comparable to the baseline expression in control cells. On the contrary, in cells expressing untagged Tau or Tau-NLS, the expression of VGluT1 is more than doubled.
Figure 1: Graphical representation of the strategy used to allow a nuclear or a cytoplasmic accumulation of Tau. Tau-NLS is accumulated in the nuclear compartment while Tau-NES is excluded. The experimental readout is the modulation of the VGluT1 expression. Please click here to view a larger version of this figure.
Figure 2: Representative undifferentiated and differentiated cell culture. Image of undifferentiated SH-SY5Y (left), cells differentiated by RA (middle) and differentiated by RA and BDNF (right). Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 3: Representative image of Tau subcellular enrichment by immunofluorescence. Image of cells untransfected or expressing untagged Tau, Tau-NES or Tau-NLS constructs. Tau signal has been obtained by immunofluorescence (red), nuclei signal has been obtained by DAPI staining (blue), merged images are reported. Scale bar = 10 µm. Please click here to view a larger version of this figure.
Figure 4: Representative detection of proteins enriched in subcellular fractions by Western blot. Western blot of subcellular fractions from SH-SY5Y cells. CF = cytoplasmic fraction; MF = membrane fraction; SNF = soluble nuclear fraction; CBF = chromatin bound fraction; CKF = cytoskeletal fraction. Please click here to view a larger version of this figure.
Figure 5: Representative detection of nuclear Tau and VGluT1 proteins. (A) Western blot of Tau protein detected in the nuclear and cytoplasmic fractions. (B) The graph reports the ratio of Tau in the nuclear fractions and cytoplasmic fraction. The values have been normalized on the endogenous Tau. SNF/CF = 1 and CBF/CF = 1 corresponds to endogenous Tau ratio in control cells. (C) Western blot of VGluT1 protein. (D) The graph reports the quantification of VGluT1 expression in the total extracts of samples expressing different amount of nuclear Tau. Kruskal-Wallis ANOVA and Mann-Whitney test; *** p < 0.001, **** p < 0.0001, n.s. p > 0.05. All results are shown as mean ± SEM from at least three independent experiments. This representative figure has been modified from Siano et al.31. Please click here to view a larger version of this figure.
We describe a method to measure the impact of nuclear Tau protein on gene expression. With this protocol the contribution of cytoplasmic Tau is strongly limited. Critical steps of this protocol are the following: the differentiation of human neuroblastoma SH-SY5Y cells, the subcellular fractionation and the localization of Tau protein in the nuclear compartment.
First, as shown in the representative results section, the differentiation of SH-SY5Y cells by adding RA and BDNF is crucial to obtain a good preparation of neuron-like cells in culture. The density of cells seeded is particularly important since a lower density might impact cell proliferation. Moreover, for experiments that need a high number of cells, like cellular fractionation and Western blot, it is important to note that the BDNF differentiation step blocks the cellular proliferation to allow the terminal differentiation, thus limiting the number of cells in culture. Alternative differentiation protocols use only RA or NGF instead of BDNF. However, while adding BDNF after RA allows to reach a better morphological differentiation32,33, NGF induces a weaker neurite outgrowth in SH-SY5Y cells34. Moreover, it has been extensively demonstrated that the combination of RA and BDNF allows to obtain a homogeneous neuronal population with expression of neuronal markers and decreased proliferation35. For this reason, the differentiation protocol exploited here combines RA and BDNF.
However, the procedure reported to dissect the role of Tau in different subcellular compartments can be used also for undifferentiated cells or for different cell types.
The subcellular fractionation is a very critical step andit is crucial to have enough starting material: a commercial kit requires only 1 x 106 cells, whereas other procedures may need a much higher starting quantity. Moreover, the use of a kit with standard buffers and steps guarantees the reproducibility of the experiment that is unavoidable and essential. However, since the composition of buffers is often proprietary, they might contain detergents which may alter the function of the protein of interest and it might be difficult to optimize the isolation of the fractions. Moreover, even in the best condition, there might be a 10-15% of contamination between fractions. A poor yield from each fraction could be overcome by increasing the incubation time in extraction buffers of specific fractions.
Since the functions of nuclear Tau have gained significant interest in recent years, it is particularly important to provide a reliable method to dissect the function of Tau in different cellular compartments. Coupling the subcellular fractionation, with the expression of Tau constructs specifically directed or excluded from the nucleus, allows one to finely tune the amount of Tau in different compartments.
A critical step in this part of the protocol is the cloning of Tau tagged with nuclear localization signal or with the nuclear export signal. The efficiency of the NLS is guaranteed by the presence of a 3XNLS consensus sequence from the SV40 virus. The nuclear translocation of the protein can be easily checked by immunofluorescence, and the lack of signal into the nucleus might be due to an incorrect cloning or to an inefficient transfection. On the contrary, the nuclear export is guaranteed by the NES consensus sequence. In this case, the immunofluorescence allows checking of the export of Tau from the nucleus. However, a weak nuclear signal is not to be excluded since Tau-NES protein enters the nucleus and then, due to the NES sequence, it is exported into the cytosol.
Up to now, the function of nuclear Tau has been studied only by correlative approaches that do not assure its direct involvement. The protocol here described, provide the first approach allowing to clearly discriminate the specific function of Tau into the nuclear compartment. As previously demonstrated, the endogenous Tau does not affect the results obtained by this protocol. Indeed, the same experiment performed in non-neuronal cells that do not express endogenous Tau, leads to VGluT1 altered expression. We applied this protocol to study the expression of disease-related genes31. Anyhow, it could be exploited also to investigate other nuclear Tau functions, such as the involvement on DNA damage, the interaction with nuclear cofactors or with the chromatin.
The authors have nothing to disclose.
This work was supported by grants from Scuola Normale Superiore (SNS14_B_DIPRIMIO; SNS16_B_DIPRIMIO).
Alexa Fluor 633 goat anti-mouse IgG | Life Technologies | A21050 | IF 1:500 |
anti Actin Antibody | BETHYL LABORATORIE | A300-485A | anti-rabbit WB 1:10000 |
anti GAPDH Antibody | Fitzgerald Industries International | 10R-G109a | anti-mouse WB 1:10000 |
anti H2B Antibody | Abcam | ab1790 | anti-rabbit WB 1:15000 |
anti Tau-13 Antibody | Santa Cruz Biotechnology | sc-21796 | anti-mouse WB 1:1000; IF 1:500 |
anti Tubulin alpha Antibody | Thermo Fisher Scientific | PA5-16891 | anti-mouse WB 1:5000 |
anti VGluT1 Antibody | Sigma-Aldrich | AMAb91041 | anti-mouse WB 1:500 |
BCA Protein Assay Kit | Euroclone | EMPO14500 | |
BDNF | Alomone Labs | B-250 | |
Blotting-Grade Blocker | Biorad | 1706404 | Non-fat dry milk |
BOVIN SERUM ALBUMIN | Sigma-Aldrich | A4503-50g | |
cOmplete Mini | Roche | 11836170001 | protease inhibitor |
Criterion TGX 4-20% Stain Free, 10 well | Biorad | 5678093 | |
DAPI | Thermo Fisher Scientific | 62247 | |
DMEM/F-12 | GIBCO | 21331-020 | |
Dulbecco's Modified Eagle's Medium Low Glucose | Euroclone | ECM0060L | |
EDTA | Sigma-Aldrich | 0390-100ml | pH=8 0.5M |
Foetal Bovine Serum | Euroclone | EC50182L | |
Glycerol | Sigma-Aldrich | G5516-500ml | |
Goat anti-mouse IgG-HPR | Santa Cruz Biotechnology | sc-2005 | WB 1:1000 |
Goat anti-rabbit IgG-HPR | Santa Cruz Biotechnology | sc-2004 | WB 1:1000 |
IGEPAL CA-630 | Sigma-Aldrich | I8896-50ml | Octylphenoxy poly(ethyleneoxy)ethanol |
Immobilon Western | MERCK | WBKLS0500 | |
Lab-Tech Chamber slide 8 well glass slide | nunc | 177402 | |
L-glutamine | Euroclone | ECB3000D | 100X |
Lipofectamine 2000 transfection reagent | Thermo Fisher Scientific | 12566014 | cationic lipid |
Methanol | Sigma-Aldrich | 322415-6X1L | |
MgCl2 | Sigma-Aldrich | M8266-100G | |
NaCl | Sigma-Aldrich | S3014-1kg | |
Opti-MEM reduced serum medium | Gibco | 31985070 | |
PEI | Sigma-Aldrich | 40,872-7 | |
Penicillin/Streptomycin | Thermo Fisher Scientific | 15140122 | 10,000 U/ml, 100ml |
Phosphate Buffered Saline (Dulbecco A) | OXOID | BR0014G | |
PhosStop | Roche | 4906837001 | phosphatase inhibitor |
QIAGEN Plasmid Maxi Kit | Qiagen | 12163 | Step 3.10 |
Retinoic acid | Sigma-Aldrich | R2625-100mg | |
Subcellular Protein Fractionation Kit for cultured cells | Thermo Fisher Scientific | 78840 | |
Supported Nitrocellulose membrane | Biorad | 1620097 | |
TC-Plate 6well | SARSTEDT | 833,920 | |
TCS SP2 laser scanning confocal microscope | Leica | N/A | |
Triton x-100 | Sigma-Aldrich | X100-500ml | Non-ionic surfactant |
Trypsin-EDTA | Thermo Fisher Scientific | 15400054 | 0.50% |
Tween-20 | Sigma-Aldrich | P9416-100ml | |
VECTASHIELD antifade mounting medium | Vector Laboratories | H-1000 | |
Wizard Plus SV Minipreps DNA Purification Systems | Promega | A1330 | Step 3.5 |