Differential effects of 14-3-3 dimers on Tau phosphorylation, stability and toxicity in vivo

Differential effects of 14-3-3 dimers on Tau phosphorylation, stability and toxicity in vivo Abstract Neurodegenerative dementias collectively known as Tauopathies involve aberrant phosphorylation and aggregation of the neuronal protein Tau. The largely neuronal 14-3-3 proteins are also elevated in the central nervous system (CNS) and cerebrospinal fluid of Tauopathy patients, suggesting functional linkage. We use the simplicity and genetic facility of the Drosophila system to investigate in vivo whether 14-3-3s are causal or synergistic with Tau accumulation in precipitating pathogenesis. Proteomic, biochemical and genetic evidence demonstrate that both Drosophila 14-3-3 proteins interact with human wild-type and mutant Tau on multiple sites irrespective of their phosphorylation state. 14-3-3 dimers regulate steady-state phosphorylation of both wild-type and the R406W mutant Tau, but they are not essential for toxicity of either variant. Moreover, 14-3-3 elevation itself is not pathogenic, but recruitment of dimers on accumulating wild-type Tau increases its steady-state levels ostensibly by occluding access to proteases in a phosphorylation-dependent manner. In contrast, the R406W mutant, which lacks a putative 14-3-3 binding site, responds differentially to elevation of each 14-3-3 isoform. Although excess 14-3-3ζ stabilizes the mutant protein, elevated D14-3-3ɛ has a destabilizing effect probably because of altered 14-3-3 dimer composition. Our collective data demonstrate the complexity of 14-3-3/Tau interactions in vivo and suggest that 14-3-3 attenuation is not appropriate ameliorative treatment of Tauopathies. Finally, we suggest that ‘bystander’ 14-3-3s are recruited by accumulating Tau with the consequences depending on the composition of available dimers within particular neurons and the Tau variant. Introduction Tau is distributed primarily in central nervous system (CNS) axons (1,2), where among other functions mediates microtubule dynamics, axonal transport and neuronal morphology (3–5). In vertebrates including humans, alternative splicing of a single transcript yields multiple protein isoforms. They differ in their amino-termini by exclusion (0N), or inclusion of 1 (1N) or 2 (2N) sequences and inclusion of 3 (3R) or 4 (4R), of the evolutionarily conserved repeats (R) near the carboxy-terminus which are typically involved in microtubule binding (6,7). Tau is implicated in several degenerative dementias collectively termed Tauopathies and other neuropathologies (7–14). Alzheimer’s disease (AD) is the most widespread sporadic aging-related neurodegenerative Tauopathy globally (15). Frontotemporal dementia with Parkinsonism on chromosome 17 (FTDP-17) is a distinct dementia, typically affecting younger individuals and involves mutations in the tau gene (16,17). Significantly, both Tauopathies present distinct clinical profiles and different, but overlapping histopathology (10,13,18). Although, non-pathogenic Tau is phosphorylated on multiple sites, (19,20), it becomes further phosphorylated especially onto characteristic ‘pathology-associated sites’ to a state of ‘hyper-phosphorylation’, which has been associated with AD and other Tauopathies (13,21,22). The different pathophysiological and clinical profiles of the various Tauopathies suggest that in addition to common interactors such as microtubules, normal and mutant Tau isoforms may also interact differentially with distinct proteins within affected neurons. To address this hypothesis, we performed systematic proteomic screens in Drosophila expressing Tau variants (Papanikolopoulou, K., Samiotaki, M., Panayotou, G. and Skoulakis, E.M.C. manuscript in preparation). Herein we report on two proteins from that screen, Leonardo (Leo), and D14-3-3ɛ (Eps), which appeared to interact with human wild-type (WT) and the FTDP-17-linked R406W Tau mutant (RW) in Drosophila brains. Our results confirm and expand previous reports on genes upregulated in human AD brains, one of which was 14-3-3ζ (23). The 14-3-3s are a family of small acidic proteins present in all eukaryotes with roles in essential biological processes such as cell proliferation and differentiation, migration and survival, intracellular transport, neurite outgrowth and ion channel regulation (24–27). Multiple 14-3-3s exist in metazoans, with seven distinct isoforms (β/α, ɛ, ζ, γ, η, θ and σ) in vertebrates, which except for σ, are primarily expressed in the brain, but also present in most other tissues (24,27,28). In contrast, Drosophila contain only two 14-3-3 genes, leonardo (leo), encoding three closely related orthologs of the vertebrate 14-3-3ζ and an ortholog of the ɛ isoform, D14-3-3ɛ (29–31). 14-3-3s homo and heterodimerize, forming a cup-shaped negatively charged groove, which interacts with client proteins containing RS/xxpS/TxP, or RxY/FxpS/TxP (pS/T phosphoserine/threonine) motifs (25,32–35). Leo homo and heterodimers with D14-3-3ɛ have been reported in Drosophila (30). 14-3-3 binding has been reported to modulate the structure of client proteins, to mask or reveal functional motifs that regulate its localization, phosphorylation state and stability (28,36–39). Given that Tau contains multiple phosphorylated Serines and Threonines, 14-3-3 binding may serve to modulate its steady-state phosphorylation by protecting from phosphatases and bridging interactions with kinases (40–42). Tau and 14-3-3s have been reported as macromolecular complexes (43–45), with at least one ubiquitous kinase, GSK-3β (46) in bovine brains or culture cell systems. Moreover, 14-3-3s may modulate the local conformation of Tau domains (47) and/or its subcellular localization. Accumulating evidence links 14-3-3s and Tauopathies. 14-3-3s have been associated with hyper-phosphorylated Tau aggregates from AD brains (41,48) and Pick’s disease patients are reported immunoreactive for 14-3-3s (49). Consistent with these observations, 14-3-3s have been reported to promote Tau aggregation (50,51). Furthermore, independent studies suggest increased expression of 14-3-3s in AD patients (52) and elevation in their cerebrospinal fluid (CSF) (53,54). However, whether 14-3-3 elevation in AD neurons and CSF is a collateral consequence of the disease, or important for its pathogenesis is unclear. Since all 14-3-3s other than σ are expressed in the brain, it is essential to determine whether there are isoform-specific effects on Tau levels, its hyper-phosphorylation and toxicity. Based on the identification of 14-3-3s as Tau interactors on our proteomic screen, we use the advantages of Drosophila to ask whether 14-3-3s affect toxicity of WT and mutant Tau in the nervous system and whether any such effects are 14-3-3 isoform-specific in vivo. Such potential differential interactions may underlie the heterogeneous, but often overlapping clinical profiles of Tauopathies. Results 14-3-3s associate with Tau in vivo To verify and validate the proteomic results, we aimed to determine independently whether endogenous 14-3-3s interact with human Tau proteins in the fly CNS. To limit such potential interactions to the adult CNS, we used the TARGET system (55). Flies carrying the wild-type 0N4R-Tau (henceforth WT) and the FTDP-17-linked mutant 0N4R-R406W (henceforth RW) were raised at 18°C, the temperature non-permissive for Tau expression. Upon emergence, they were induced to express the human Tau proteins paneuronally for 10 days, while age-matched flies that remained at 18°C throughout the experiment were used as controls (Fig. 1A, upper panels). WT and RW Tau immunoprecipitated from fly head extracts with the biotinylated antibody HT7, co-precipitated both Leo and D14-3-3ɛ (EPS) (Fig. 1A). Figure 1. View largeDownload slide 14-3-3s interact with Tau in vivo. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts (TARGET system). (A) (upper panel) A western blot demonstrating accumulation of the WT and RW Tau isoforms in the heads of adults flies. Uninduced (U) refers to lysates from animals that remained at 18°C, whereas Induced (I), denotes lysates from animals expressing the respective Tau after shifting them to 30°C for 10 days. Syntaxin (Syx) was used as loading control, and Tau was revealed with the T46 non-phosphorylation-dependent antibody. The lower panels are representative of results of immunoprecipitations from head lysates of animals induced for 10 days and their respective controls using the HT7 biotinylated anti-Tau antibody. The resulting immune complexes were immunoblotted for the presence of 14-3-3 isoforms as indicated. (B) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with head lysates from WT-expressing flies and the complexes were immunoblotted with anti-LEO and anti-D14-3-3ε (EPS) antibodies as indicated. (C) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with fly head extracts expressing the indicated Tau protein and immunoblotted with the T46 anti-Tau antibody. (D) Head lysates from flies expressing pan-neuronally 2N4R-Flag (WT) and 2N4R-RW-Flag (RW) Tau proteins were either dephosphorylated (+λ) or treated with phosphatase inhibitors (−λ). Lysates were subsequently incubated with Glutathione-agarose beads coated with GST-difopein to collect the 14-3-3 proteins and then immunoblotted with T46 anti-Tau antibody. (E) Co-immunoprecipitation of Leo and D14-3-3ε (EPS) with Tau from animals expressing 2N4R-Flag (WT) and RW-2N4R-Flag (RW) as indicated by the bars in panel D, pan-neuronally. Prior to incubation with anti-Flag beads to collect the Tau proteins, the lysates were treated either with λ phosphatase (+λ) or with phosphatase inhibitors (–λ) as indicated. Figure 1. View largeDownload slide 14-3-3s interact with Tau in vivo. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts (TARGET system). (A) (upper panel) A western blot demonstrating accumulation of the WT and RW Tau isoforms in the heads of adults flies. Uninduced (U) refers to lysates from animals that remained at 18°C, whereas Induced (I), denotes lysates from animals expressing the respective Tau after shifting them to 30°C for 10 days. Syntaxin (Syx) was used as loading control, and Tau was revealed with the T46 non-phosphorylation-dependent antibody. The lower panels are representative of results of immunoprecipitations from head lysates of animals induced for 10 days and their respective controls using the HT7 biotinylated anti-Tau antibody. The resulting immune complexes were immunoblotted for the presence of 14-3-3 isoforms as indicated. (B) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with head lysates from WT-expressing flies and the complexes were immunoblotted with anti-LEO and anti-D14-3-3ε (EPS) antibodies as indicated. (C) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with fly head extracts expressing the indicated Tau protein and immunoblotted with the T46 anti-Tau antibody. (D) Head lysates from flies expressing pan-neuronally 2N4R-Flag (WT) and 2N4R-RW-Flag (RW) Tau proteins were either dephosphorylated (+λ) or treated with phosphatase inhibitors (−λ). Lysates were subsequently incubated with Glutathione-agarose beads coated with GST-difopein to collect the 14-3-3 proteins and then immunoblotted with T46 anti-Tau antibody. (E) Co-immunoprecipitation of Leo and D14-3-3ε (EPS) with Tau from animals expressing 2N4R-Flag (WT) and RW-2N4R-Flag (RW) as indicated by the bars in panel D, pan-neuronally. Prior to incubation with anti-Flag beads to collect the Tau proteins, the lysates were treated either with λ phosphatase (+λ) or with phosphatase inhibitors (–λ) as indicated. To perform the complementary experiment requiring purification of endogenous 14-3-3s, we used GST-difopein-conjugated beads because this synthetic peptide binds 14-3-3s with high affinity (26,56). As shown in Figure 1B, endogenous Leo and D14-3-3ɛ from fly head lysates were indeed bound on GST-difopein, but not on plain GST beads. Significantly, difopein-bound 14-3-3s from these lysates co-purified with either WT or RW (Fig. 1C). Collectively, these results validate the proteomic data and indicate that endogenous 14-3-3s directly or indirectly interact with Tau and are consistent with reports suggesting their co-localization in aggregates from Tauopathy patient brains (53,57). 14-3-3s bind phosphorylated Serines or Threonines (pSer/pThr) and are reported to modulate the phosphorylation state of their clients by occluding or facilitating access to kinases and phosphatases (38,39,58). Given the multiple pSer/pThr on Tau (59), we wondered whether its phosphorylation is required for the interaction with Leo and D14-3-3ɛ. Because the 0N4R isoforms were not available as Flag-Tagged constructs to allow the experiment we used the Flag-Tagged 2N4R and Flag-Tagged RW mutation in the 2N4R isoform. We reasoned that the aminoterminnal extension will not affect access of the phosphatase and in general the 0N4R and 2N4R isoforms behave very similarly in our assays (60,61). We treated head lysates from animals expressing the Flag-tagged 2N4R or Flag-tagged RW mutant in the 2N4R genetic background with λ-phosphatase. Although we cannot be certain that all phosphates were abolished, the majority of phosphates were removed as indicated by the shift in electrophoretic mobility (Fig. 1D), or with specific antibodies (Supplementary Material, Fig. S1). Treated and untreated lysates were mixed with GST-difopein-conjugated beads to collect 14-3-3s, the complexes resolved in acrylamide gels and probed for the presence of Tau in the pulled-down material. The results clearly demonstrate that difopein-bound 14-3-3s co-precipitated both phosphorylated and dephosphorylated Tau (Fig. 1D). Conversely, immunoprecipitation of transgenic Tau proteins via their Flag tag, confirmed co-precipitation of Leo and D14-3-3ɛ from treated and untreated lysates (Fig. 1E). Therefore, 14-3-3s bind dephosphorylated Tau indicating that they do not dock only on its phosphor-Serines and phosphor-Threonines. As predicted by these results, full-length Tau contains five putative 14-3-3 consensus binding sites involving potentially phosphorylated Ser and Thr (Fig. 2). However, as indicated by the results in Figure 1E, additional binding sites not involving phosphorylatable residues must exist and these were not apparent in our in silico search. The sites involving Ser198 and Ser214 are in the projection domain thought to interact with various cellular proteins other than microtubules (62). Ser214 is known to be phosphorylated (63) and has been reported critical for interaction with 14-3-3 and a mediator of Tau aggregation in vitro (64). A consensus binding site not reported before is anchored by Thr245 in the first repeat (R1), which has been reported constitutively phosphorylated on WT (20). An additional site in R4 precedes the Par1 kinase-targeted Ser356 (65). Finally, a putative 14-3-3 binding site is in the carboxy-terminal domain involving Ser413. Interestingly, Arg406 is the mandatory consensus-anchoring amino-acid for 14-3-3 binding to pSer413 and is mutated to Trp in R406W suggesting that this site is eliminated in the R406W mutant Tau. Figure 2. View largeDownload slide Proposed in silico 14-3-3 binding sites on WT Tau. The results of searching the full-length WT Tau sequence with all modes of the 14-3-3 binding consensus. Numbering of the amino-acids is based on the full-length 441 residue 2N4R WT Tau. Amino acids mandatory for 14-3-3 binding are shown in bold, while the potentially phosphorylated Ser or Thr are denoted by the numbered bold residues as reported previously [20]. Figure 2. View largeDownload slide Proposed in silico 14-3-3 binding sites on WT Tau. The results of searching the full-length WT Tau sequence with all modes of the 14-3-3 binding consensus. Numbering of the amino-acids is based on the full-length 441 residue 2N4R WT Tau. Amino acids mandatory for 14-3-3 binding are shown in bold, while the potentially phosphorylated Ser or Thr are denoted by the numbered bold residues as reported previously [20]. 14-3-3 dimers modulate Tau phosphorylation To determine whether 14-3-3s interactions may contribute to Tau phosphorylation, aggregation and toxicity, we attenuated by RNA-mediated interference (RNAi), each isoform in the CNS of flies also expressing WT. The available RNAi-mediating transgenes completely abrogated Leo levels (Fig. 3A), whereas D14-3-3ɛ was reduced by about 50% (Fig. 3B). Significantly, D14-3-3ɛ reduction was accompanied by a 50% Leo elevation (Fig. 3A) and a similar elevation of D14-3-3ɛ was observed in animals lacking Leo (Fig. 3B). This is congruent with prior reports of 14-3-3 homeostasis in D14-3-3ɛ mutant embryos (29), leo mutants and transgenic adults (30). Therefore, these genetic manipulations generate animals that lack Leo but harbor excess D14-3-3ɛ (Leo-), animals with nearly half of D14-3-3ɛ, but excess Leo (Eps-) and flies lacking Leo and half of D14-3-3ɛ (Leo-, Eps-). Figure 3. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on WT Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or least square means (LSM) contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing WT Tau probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (WTLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (WTEPS-) and also nearly eliminated (Dunnett’s, P < 0.0001, n = 3), when both isoforms were reduced concurrently (WTLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing WT Tau probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was significantly (Dunnett’s, P < 0.0001, n = 7) elevated (WTLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 10) upon D14-3-3ε attenuation and was also reduced (Dunnett’s, p = 0.0008, n = 5) when both isoforms were attenuated concurrently. (C) Total levels of WT in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences (ANOVA F(3, 27) = 0.2711, P = 0.8457) in WT levels (n > 6). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A reduction (Dunnett’s, P = 0.031, n = 10) in AT8 occupation was revealed only in animals with both isoforms reduced. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.029, n = 5) was observed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0017, n = 5) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0005, n = 6) in animals with both isoforms reduced. (G) Representative blots (n = 3) of 14-3-3 dimer formation investigation upon addition of cross-linker (+ linker) in adult head lysates of the indicated genotypes (bottom). Mirror blots were probed either with the anti-Leo (top) or the anti-D14-3-3ε (bottom) antibodies. The monomers (mono) and dimers are denoted. (H) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or WT Tau (WT) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing WT (solid bars) is shown. ANOVA indicated significant differences (F(6, 89) = 14.6695; P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus WT P < 0.0157; LEO- versus WT LEO- P < 0.0058; EPS- versus WT EPS- P <0.0389; LEO- EPS- versus WT LEO- EPS P < 0.0272. Figure 3. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on WT Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or least square means (LSM) contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing WT Tau probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (WTLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (WTEPS-) and also nearly eliminated (Dunnett’s, P < 0.0001, n = 3), when both isoforms were reduced concurrently (WTLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing WT Tau probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was significantly (Dunnett’s, P < 0.0001, n = 7) elevated (WTLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 10) upon D14-3-3ε attenuation and was also reduced (Dunnett’s, p = 0.0008, n = 5) when both isoforms were attenuated concurrently. (C) Total levels of WT in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences (ANOVA F(3, 27) = 0.2711, P = 0.8457) in WT levels (n > 6). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A reduction (Dunnett’s, P = 0.031, n = 10) in AT8 occupation was revealed only in animals with both isoforms reduced. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.029, n = 5) was observed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0017, n = 5) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0005, n = 6) in animals with both isoforms reduced. (G) Representative blots (n = 3) of 14-3-3 dimer formation investigation upon addition of cross-linker (+ linker) in adult head lysates of the indicated genotypes (bottom). Mirror blots were probed either with the anti-Leo (top) or the anti-D14-3-3ε (bottom) antibodies. The monomers (mono) and dimers are denoted. (H) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or WT Tau (WT) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing WT (solid bars) is shown. ANOVA indicated significant differences (F(6, 89) = 14.6695; P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus WT P < 0.0157; LEO- versus WT LEO- P < 0.0058; EPS- versus WT EPS- P <0.0389; LEO- EPS- versus WT LEO- EPS P < 0.0272. Because 14-3-3s are dimeric (25), we determined the effect of Leo and D14-3-3ɛ abrogation on dimer levels. Leo and D14-3-3ɛ dimers were apparent in control animals as expected (Fig. 3G). Leo-containing dimers were not apparent in Leo- animals despite the traces of monomer due to the large amounts of extract needed in cross-linking experiments. Surprisingly, abundant dimers were revealed with the D14-3-3ɛ antibody in these Leo- animals (Fig. 3G). This indicates that upon Leo loss D14-3-3ɛ homodimers form efficiently judging by their abundance. In confirmation, these dimers are also present in Leo-Eps- animals, albeit apparently reduced. In contrast, in flies with a 50% attenuation of D14-3-3ɛ (Eps-), we assume that the dimers are largely heterodimers with Leo or Leo homodimers (Fig. 3G). Interestingly, attenuation of either 14-3-3 alone or simultaneously, did not appear to affect WT levels (Fig. 3C). This indicates that either Leo nor D14-3-3ɛ are essential for steady-state WT levels in the fly CNS, or that the residual dimers in Leo-Eps- animals suffice to maintain Tau levels elevated. To determine whether 14-3-3 attenuation affects Tau phosphorylation, we assessed occupation of the pathology-associated sites AT8, AT100 and pSer262. Phosphorylation on Ser202/Thr205 recognized by the AT8 antibody appeared unchanged upon attenuation of either 14-3-3 protein alone, but was reduced upon concurrent attenuation of both Leo and D14-3-3ɛ (Fig. 3D). Because Leo reduction is accompanied by D14-3-3ɛ upregulation and vice versa (Fig. 3A and B) and in the doubly attenuated animals dimers levels are reduced, the results suggest that both isoforms contribute positively to AT8 occupation. In contrast, phosphorylation at the AT100 site, which includes the 14-3-3 binding pSer214 (Fig. 2), was elevated upon Leo abrogation (Fig. 3E). This suggests either that Leo homodimers or heterodimers occlude access to that site, or they promote its de-phosphorylation. Alternatively, D14-3-3ɛ excess or homodimers upon Leo loss may promote phosphorylation at AT100, or protect it from de-phosphorylation. Interestingly, AT100 occupation was not significantly altered upon D14-3-3ɛ attenuation and the concomitant Leo elevation, a situation that most likely enhances Leo heterodimer and homodimer levels. This indicates that Leo homodimer excess in Eps- flies does not occlude or inhibit AT100 occupation and therefore, increased AT100 phosphorylation in Leo- flies is likely the effect of excess D14-3-3ɛ. Accordingly, AT100 phosphorylation was significantly reduced (Fig. 3E) in Leo-, Eps- flies, ostensibly because the remaining D14-3-3ɛ is not sufficient to support AT100 occupation. Results were similar for pSer262 (Fig. 3F), which is not a 14-3-3 consensus site, but its occupation is regulated by the phosphorylation status of Ser238 and Thr245 (66), the latter of which appears to be (Fig. 2). Phosphorylation of Ser262 was also highly enhanced upon Leo abrogation, albeit even more than at AT100 and eliminated upon attenuation of both 14-3-3s (Fig. 3F). These results also indicate that Leo/D14-3-3ɛ heterodimers are essential for the steady-state phosphorylation of Ser262 and excess D14-3-3ɛ prevents dephosphorylation of the site. A parsimonious explanation of these results is a highly dynamic interaction of 14-3-3 homo and heterodimers modulating the balance between kinase and phosphatase access to the phosphosites. Leo homodimers appear to facilitate phosphatase access, which is likely suppressed by D14-3-3ɛ dimers in Leo- flies. In contrast, heterodimer binding in Eps- flies does not promote phosphatase access evidenced by the normal occupation levels. In Leo- Eps- flies where the stabilizing excess of D14-3-3ɛ is absent AT100 and pSer262 occupation is highly reduced. Do the effects of 14-3-3 attenuation on WT phosphorylation affect its toxicity? To address this, we quantified the pre-eclosion lethality associated with pan-neuronal WT expression alone or upon 14-3-3 attenuation. Indeed, as illustrated in Figure 3H, WT expression reduced the number of animals reaching adulthood relative to controls. Neither loss of Leo, nor D14-3-3ɛ attenuation alone affected viability (Fig. 3H, open bars) and they did not affect WT-dependent lethality (Fig. 3H, filled bars). However, reduction of both 14-3-3s (Leo- Eps-), decreased viability nearly 50% on its own, in accordance with prior reports (29) and this was further reduced upon WT expression (Fig. 3H, last two bars). Note that the difference in viability between animals with attenuated 14-3-3s alone and those also co-expressing WT remained relatively constant (the difference between open and filled bars in Fig. 3H). This indicates that attenuation of either 14-3-3 does not potentiate WT toxicity significantly. Rather the most significant effect on lethality was precipitated by the concomitant attenuation of both 14-3-3s, which decrease the total levels of 14-3-3 dimers. Interestingly the effects of 14-3-3 reduction (Fig. 4A and B) on the steady-state levels of RW and its phosphorylation at AT8, AT100 and pSer262 were similar with those for WT Tau (Fig. 4C–F). Lethality associated with RW expression was more pronounced than that of WT and reached nearly 50% of expected progeny (Fig. 4G). Again, attenuation of either 14-3-3 alone did not alter RW toxicity significantly, but concurrent reduction of both 14-3-3s resulted in 50% lethality on its own and this was mildly augmented by RW expression. Figure 4. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on R406W mutant Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing RW probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (RWLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (RWEPS-) and also eliminated (Dunnett’s, P < 0.0001, n = 6) when both isoforms were reduced concurrently (RWLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing RW probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was elevated (Dunnett’s, P < 0.001, n = 6) elevated (RWLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 6) upon D14-3-3ε attenuation and also reduced (Dunnett’s, P = 0.0022, n = 6) when both isoforms were reduced concurrently. (C) Total levels of RW in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences in Tau levels (n > 4). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. AT8 occupation was significantly reduced (Dunnett’s, P < 0.001, n = 9) only in animals with both isoforms attenuated. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.009, n = 4) was observed in RWLEO- animals and a significant decrease in (Dunnett’s, P = 0.0006, n = 4) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in RWLEO- animals and a decrease in (Dunnett’s, P = 0.0012, n = 6) in animals with both isoforms reduced. (G) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or RW Tau (RW) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing RW (solid bars) is shown. ANOVA indicated significant differences (F(6, 90) = 33.8151, P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus RW P < 0.0001; LEO- versus RW LEO- P < P < 0.0001; EPS- versus RW EPS- P < P < 0.0001; LEO- EPS- versus RW LEO- EPS- P < 0.0001. Figure 4. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on R406W mutant Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing RW probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (RWLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (RWEPS-) and also eliminated (Dunnett’s, P < 0.0001, n = 6) when both isoforms were reduced concurrently (RWLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing RW probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was elevated (Dunnett’s, P < 0.001, n = 6) elevated (RWLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 6) upon D14-3-3ε attenuation and also reduced (Dunnett’s, P = 0.0022, n = 6) when both isoforms were reduced concurrently. (C) Total levels of RW in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences in Tau levels (n > 4). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. AT8 occupation was significantly reduced (Dunnett’s, P < 0.001, n = 9) only in animals with both isoforms attenuated. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.009, n = 4) was observed in RWLEO- animals and a significant decrease in (Dunnett’s, P = 0.0006, n = 4) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in RWLEO- animals and a decrease in (Dunnett’s, P = 0.0012, n = 6) in animals with both isoforms reduced. (G) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or RW Tau (RW) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing RW (solid bars) is shown. ANOVA indicated significant differences (F(6, 90) = 33.8151, P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus RW P < 0.0001; LEO- versus RW LEO- P < P < 0.0001; EPS- versus RW EPS- P < P < 0.0001; LEO- EPS- versus RW LEO- EPS- P < 0.0001. Collectively, these results indicate dynamic interactions of 14-3-3 homo and heterodimers in the regulation of steady-state phosphorylation of both WT and RW and appear to have a minor effect on the toxicity of either Tau variant. Therefore, 14-3-3 attenuation does not synergize with Tau variants to augment their toxicity. Pharmacological inhibition of Tau/14-3-3 interactions 14-3-3 interactions with WT and RW could have been obscured because the genetic manipulations did not result in adult animals completely devoid of both 14-3-3s due to incomplete attenuation or homeostatic compensation (Figs 3A, B and 4A, B). This is further complicated by the partial functional overlap of 14-3-3s in vivo (29). To address and potentially augment this issue, we used BV02, a 14-3-3 antagonist reported to disrupt their biological activity in cultured cell systems (67). BV02 is a cell permeable non-peptide 14-3-3 antagonist which targets their amphipathic binding groove and disrupts interactions with client proteins (68). Because concurrent attenuation of both 14-3-3s resulted in elevated lethality, we titrated the concentration of the inhibitor such as to afford experimental resolution in combination with Tau expression. Control experiments indicated that this requirement is fulfilled at 0.15 nm throughout development, whereas at higher concentrations lethality was extensive. As suggested (67), at 0.15 nm and even at the lethal 0.30 nm concentration, the inhibitor did not affect the steady-state levels of 14-3-3 isoforms or Tau variants (Fig. 5A–D). Figure 5. View largeDownload slide Pharmacological inhibition of 14-3-3 dimer–client interaction. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) Representative blots probing the levels of LEO relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nM) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 36) = 0.6032, P = 0.5528). (B) Representative blots probing the levels of D14-3-3ɛ (EPS) relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nm) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 47) = 0.6833, P = 0.5101). (C) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 1.4411, P = 0.2552). (D) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 0.4593, P = 0.5120). (E) Representative blots probing the levels of AT100 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 10) = 21.5991, P < 0.0012). (F) Representative blots probing the levels of pS262 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 8) = 13.3574, P < 0.0081). (G) Representative blots probing the levels of AT100 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 11) = 45.7888, P < 0.0001). (H) Representative blots probing the levels of pS262 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 7) = 16.6753, P < 0.0065). (I) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CN) or WT (TAU) in media with and without 0.15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or TAU expressing animals (ANOVA F(1, 31) = 1.6096, P = 0.2143). (J) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with WT (TAU, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), or TAU, LEO- animals (ANOVA F(1, 15) = 0.2645, p = 0.6150). (K) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with WT (TAU, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for TAU, EPS- animals (ANOVA F(1, 17) = 27.0254, P < 0.0001). (L) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htau0N4R, leo RNAi recombinant (TAU, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of TAU, LEO-, EPS- flies (ANOVA F(1, 29) = 48.9572, P < 0.0001). (M) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or R406W (RW) in media with and without 15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or RW expressing animals (ANOVA F(1, 33) = 2.5017, P = 0.1236). (N) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with R406W (RW, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), but had a highly significant effects on teh viability of RW, LEO- animals (ANOVA F(1, 17) = 53.2494, P < 0.0001). (O) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with RW (RW, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for RW, EPS- animals (ANOVA F(1, 21) = 35.3185, P < 0.0001). (P) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htauR406W, leo RNAi recombinant (RW, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of RW, LEO-, EPS- flies (ANOVA F(1, 29) = 74.9826, P < 0.0001). Figure 5. View largeDownload slide Pharmacological inhibition of 14-3-3 dimer–client interaction. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) Representative blots probing the levels of LEO relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nM) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 36) = 0.6032, P = 0.5528). (B) Representative blots probing the levels of D14-3-3ɛ (EPS) relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nm) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 47) = 0.6833, P = 0.5101). (C) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 1.4411, P = 0.2552). (D) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 0.4593, P = 0.5120). (E) Representative blots probing the levels of AT100 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 10) = 21.5991, P < 0.0012). (F) Representative blots probing the levels of pS262 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 8) = 13.3574, P < 0.0081). (G) Representative blots probing the levels of AT100 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 11) = 45.7888, P < 0.0001). (H) Representative blots probing the levels of pS262 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 7) = 16.6753, P < 0.0065). (I) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CN) or WT (TAU) in media with and without 0.15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or TAU expressing animals (ANOVA F(1, 31) = 1.6096, P = 0.2143). (J) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with WT (TAU, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), or TAU, LEO- animals (ANOVA F(1, 15) = 0.2645, p = 0.6150). (K) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with WT (TAU, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for TAU, EPS- animals (ANOVA F(1, 17) = 27.0254, P < 0.0001). (L) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htau0N4R, leo RNAi recombinant (TAU, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of TAU, LEO-, EPS- flies (ANOVA F(1, 29) = 48.9572, P < 0.0001). (M) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or R406W (RW) in media with and without 15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or RW expressing animals (ANOVA F(1, 33) = 2.5017, P = 0.1236). (N) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with R406W (RW, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), but had a highly significant effects on teh viability of RW, LEO- animals (ANOVA F(1, 17) = 53.2494, P < 0.0001). (O) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with RW (RW, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for RW, EPS- animals (ANOVA F(1, 21) = 35.3185, P < 0.0001). (P) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htauR406W, leo RNAi recombinant (RW, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of RW, LEO-, EPS- flies (ANOVA F(1, 29) = 74.9826, P < 0.0001). We expected that the inhibitor may decrease phosphorylation of WT and RW at the sites assayed, essentially mimicking the observation in Leo-, Eps- animals (Figs 3D–F and 4D–F). However, BV02-mediated inhibition of 14-3-3s resulted in elevated AT100 and Ser262 phosphorylation on WT (Fig. 5E and F) and RW (Fig. 5G and H). These results confirm independently that Leo homodimers and heterodimers suppress occupation of the sites possibly by promoting phosphatase activity. This appears counteracted by 14-3-3ɛ dimers, which promote phosphorylation at these sites, evident in Leo- animals and upon inhibitor application. Furthermore, the results suggest that in Leo-Eps- animals, loss of phosphorylation at these sites may be a consequence of the large reduction in dimers (Fig. 3G), in addition to the large attenuation in D14-3-3 levels. Therefore, these data in agreement with those of genetic abrogation of 14-3-3s (Figs 3 and 4) strongly suggest complex regulation of Tau phosphorylation by 14-3-3 dimers. To determine the effects of pharmacological inhibition of 14-3-3 on WT and RW toxicity, we used 0.15 nm BV02 to avoid the lethality associated with higher concentrations. At 0.15 nm, the inhibitor did not significantly enhance the lethality of controls, WT-expressing (Fig. 5I), Leo- alone, or Leo-, WT-expressing flies (Fig. 5J). Because the inhibitor did not enhance the lethality of Leo- animals, genetic abrogation of Leo appears complete, in agreement with the western blot results (Figs 3A and 4A). This confirms independently that Leo is not required for WT toxicity. Inhibitor-mediated lethality was enhanced significantly if D14-3-3ɛ was attenuated in WT-expressing animals (Fig. 5K, TAU Eps-), consistent with the interpretation that D14-3-3ɛ heterodimers with the elevated Leo (Figs 3B and 4B) to account for the minor viability effects of WT, Eps- animals (Fig. 3H). However, inhibition of these heterodimers with BV02 precipitated the significant reduction in WT, Eps- viability (Fig. 5K). Consistent with this notion, treatment of already low viability Leo-Eps- flies with the inhibitor nearly eliminated them whether co-expressing WT, or not (Fig. 5L). Broadly, similar effects of the inhibitor were observed in flies expressing RW, except that as described earlier (Fig. 4G), the mutant protein was significantly more toxic on its own compared with WT (Fig. 5I versus Fig. 5M). Significantly however, in contrast to WT-expressing animals, BV02 treatment of RW-expressing Leo- flies precipitated nearly complete lethality (Fig. 5N). Since the inhibitor has no effect on Leo- animals, this is an effect precipitated specifically by the presence of RW. This suggests that D14-3-3ɛ dimers, abundant in Leo- animals, suppress RW toxicity and upon their inhibition with BV02 result in nearly complete lethality. It appears that Leo homodimers or heterodimers can compensate in part for this because although lethality of RW, Eps- was enhanced by the inhibitor (Fig. 5O), it was not as complete as in RW, Leo- (Fig. 5N). In agreement, strongly enhanced toxicity was apparent if both 14-3-3s were attenuated in RW-expressing animals and treated with BV02 (Fig. 5P). Therefore, D14-3-3ɛ differentially suppresses the toxicity of RW, an effect not apparent with WT. This differential interaction with the normal and mutant Tau may be mediated by the R406W mutation itself, possibly because it effects an apparent 14-3-3 binding site. 14-3-3 dosage affects WT-Tau levels Because 14-3-3s have been reported elevated in the brain of Tauopathy patients (53), we investigated whether increased Leo and D14-3-3ɛ levels affect phosphorylation, steady-state levels and the toxicity of WT (Fig. 6A). Of the three Leo isoforms, we used a transgene for the paneuronally expressed LeoII (31) throughout this work. Figure 6. View largeDownload slide 14-3-3 isoform-specific effects on WT Tau toxicity, but equivalent effects on its steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htau0N4R males (TAU). The difference is significantly different (ANOVA F(1, 21) = 163.16; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-WT expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared to the mean ratio of flies expressing both WT and either of the 14-3-3 transgenes (WTLEO, WTEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a significant enhancement of Leo-overexpression associated lethality by WT co-expression (ANOVA F(1, 20) = 69.12; P < 0.0001, n ≥ 10). Similarly, WT co-expression yielded significant lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 18) = 16.59; P < 0.0007, n > 9). (C–F) Lysates from the heads of flies expressing pan-neuronally WT alone, or also co-expressing the indicated Drosophila 14-3-3s were subjected to western blotting with the anti-total Tau (T14), or the indicated anti-phospho-Tau antibodies. The blots represent one of 3–4 independent experiments. For the quantification shown below each representative blot, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which was set to 1. The significance of potential differences was determined with Dunnett’s tests following ANOVA. The levels of WT Tau were significantly different (P < 0.01) than those in lysates also over-expressing Leo or D14-3-3ε (G) and remained significantly different (P < 0.05) when probed with the indicated phosphoantibodies (H–J). Figure 6. View largeDownload slide 14-3-3 isoform-specific effects on WT Tau toxicity, but equivalent effects on its steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htau0N4R males (TAU). The difference is significantly different (ANOVA F(1, 21) = 163.16; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-WT expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared to the mean ratio of flies expressing both WT and either of the 14-3-3 transgenes (WTLEO, WTEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a significant enhancement of Leo-overexpression associated lethality by WT co-expression (ANOVA F(1, 20) = 69.12; P < 0.0001, n ≥ 10). Similarly, WT co-expression yielded significant lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 18) = 16.59; P < 0.0007, n > 9). (C–F) Lysates from the heads of flies expressing pan-neuronally WT alone, or also co-expressing the indicated Drosophila 14-3-3s were subjected to western blotting with the anti-total Tau (T14), or the indicated anti-phospho-Tau antibodies. The blots represent one of 3–4 independent experiments. For the quantification shown below each representative blot, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which was set to 1. The significance of potential differences was determined with Dunnett’s tests following ANOVA. The levels of WT Tau were significantly different (P < 0.01) than those in lysates also over-expressing Leo or D14-3-3ε (G) and remained significantly different (P < 0.05) when probed with the indicated phosphoantibodies (H–J). Surprisingly, paneuronal overexpression of LeoII precipitated 30% lethality (Fig. 6B), but significantly, its co-expression with WT Tau increased lethality to nearly 80% (Fig. 6B). Hence excess Leo appears toxic on its own and this may contribute to pathology independently of that caused by Tau. Nevertheless, Leo elevation potentiates Tau toxicity and pathology. Interestingly, overexpression of a D14-3-3ɛ transgene did not precipitate significant lethality on its own and did not alter WT lethality as significantly as upon Leo elevation (Fig. 6B). Hence, 14-3-3 excess is not generally toxic, but rather it characterizes the Leo isoform. Interestingly, overexpression of either Leo or D14-3-3ɛ, did not result in significant compensatory elevation or decrease of the complementary isoform (Supplementary Material, Fig. S2A and B). Significantly, WT levels were nearly doubled upon co-expression with either Leo or D14-3-3ɛ (Fig. 6C). This was also apparent when the AT8, AT100 and pSer262 sites were surveyed (Fig. 6D–F). Because qualitatively and quantitatively similar results were obtained for all phosphorylation sites tested, elevated 14-3-3s do not appear to affect their occupation, but rather increase the steady-state levels of WT itself. However, although WT levels increased equivalently by over-expression of either Leo or D14-3-3ɛ (Fig. 6C), WT lethality was highly potentiated by Leo and much less by D14-3-3ɛ (Fig. 6B), indicating 14-3-3 isoform-specific effects on WT toxicity, but not on its levels. Differential effects of 14-3-3 excess on the R406W mutant Tau Do the effects of 14-3-3 elevation on WT generalize to the RW mutant? The nearly 50% lethality upon paneuronal expression of RW alone (Fig. 7A), was potentiated to over 95% upon co-expression with Leo (Fig. 7B). In contrast, D14-3-3ɛ elevation did not affect lethality on its own and its effects combined with R406W were marginal (Fig. 7B). Therefore, the effects of 14-3-3 excess on WT and RW toxicity are broadly similar, with Leo elevation enhancing it for both, but much more severely for the mutant Tau. Figure 7. View largeDownload slide 14-3-3 isoform-specific effects on R406W toxicity and steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htauR406W males (RW). The difference is significantly different (ANOVA F(1, 20) = 39.06; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-RW expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared with the mean ratio of flies expressing both RW and either of the 14-3-3 transgenes (RWLEO, RWEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a highly significant enhancement of Leo-overexpression associated lethality by RW co-expression (ANOVA F(1, 19) = 128.66, P < 0.0001, n ≥ 9). However, RW co-expression yielded marginal lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 16) = 4.87, P = 0.0422, n > 8). (C) Representative blot of lysates from the heads of pooled escaper flies expressing pan-neuronally under ElavGal4, RW alone or also co-expressing the indicated Drosophila 14-3-3s probed with anti-total Tau (T14). For the quantification shown below, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which is set to 1. RW levels in animals also co-expressing D14-3-3ε were significantly different than those expressing RW alone (Dunnett’s, P < 0.01, n = 4), but not those co-expressing Leo. (D–F) Representative blots and their respective quantifications from single flies expressing R406W alone, or co-overexpressing 14-3-3s. RW levels were quantified after normalization to Syntaxin (Syx) and significant differences (Dunnett’s, P < 0.01, n = 4 for all) in mean levels of RW alone which was set as 1, from its levels in the presence of excess Leo and D14-3-3ε. (G–I) Representative western blots and their respective quantifications from flies expressing RW alone, or also co-overexpressing 14-3-3s in adults only under Elav; Gal80ts at 25°C. RW levels were quantified after normalization to Syntaxin (Syx) and set to 1. Significant differences (Dunnett’s, P < 0.01, n = 3) in mean levels of RW alone versus its levels in the presence of excess Leo and D14-3-3ε. Figure 7. View largeDownload slide 14-3-3 isoform-specific effects on R406W toxicity and steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htauR406W males (RW). The difference is significantly different (ANOVA F(1, 20) = 39.06; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-RW expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared with the mean ratio of flies expressing both RW and either of the 14-3-3 transgenes (RWLEO, RWEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a highly significant enhancement of Leo-overexpression associated lethality by RW co-expression (ANOVA F(1, 19) = 128.66, P < 0.0001, n ≥ 9). However, RW co-expression yielded marginal lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 16) = 4.87, P = 0.0422, n > 8). (C) Representative blot of lysates from the heads of pooled escaper flies expressing pan-neuronally under ElavGal4, RW alone or also co-expressing the indicated Drosophila 14-3-3s probed with anti-total Tau (T14). For the quantification shown below, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which is set to 1. RW levels in animals also co-expressing D14-3-3ε were significantly different than those expressing RW alone (Dunnett’s, P < 0.01, n = 4), but not those co-expressing Leo. (D–F) Representative blots and their respective quantifications from single flies expressing R406W alone, or co-overexpressing 14-3-3s. RW levels were quantified after normalization to Syntaxin (Syx) and significant differences (Dunnett’s, P < 0.01, n = 4 for all) in mean levels of RW alone which was set as 1, from its levels in the presence of excess Leo and D14-3-3ε. (G–I) Representative western blots and their respective quantifications from flies expressing RW alone, or also co-overexpressing 14-3-3s in adults only under Elav; Gal80ts at 25°C. RW levels were quantified after normalization to Syntaxin (Syx) and set to 1. Significant differences (Dunnett’s, P < 0.01, n = 3) in mean levels of RW alone versus its levels in the presence of excess Leo and D14-3-3ε. Is the potentiation of RW toxicity reflected in the levels of the protein in the CNS? Because most animals with elevated Leo died as late pupae/pharate adults we pooled lysates from the few escapers, but surprisingly RW levels appeared unchanged upon Leo co-expression and equally surprisingly, significantly reduced upon elevation of D14-3-3ɛ (Fig. 7C). We noticed, however, that when using lysates from individual flies, in approximately 20% of Leo over-expressing animals, RW levels were nearly doubled (Fig. 7D), which agrees with the viability data. In contrast, RW levels were suppressed upon D14-3-3ɛ over-expression, whether pooled (Fig. 7C), or in single flies (Fig. 7D). Phosphorylation at AT8, AT100 were also correspondingly elevated or reduced (Fig. 7E and F) in these escapers. Brain lysates from dissected pharate adults that failed to eclose, indicated similarly elevated and reduced RW upon Leo and D14-3-3ɛ co-expression respectively (Supplementary Material, Fig. S3A). Therefore, RW levels in most animals failing to eclose are represented by those in 20% of the escapers (Fig. 7D). Collectively then, high levels of RW correlate well with the robust lethality of Leo co-expressing animals. In contrast, the drastically reduced RW levels upon D14-3-3ɛ elevation correlate well with the lack of significant lethality in these animals. To verify these results independently, increase experimental resolution and avoid the high mortality upon Leo/R406W co-expression, we used the TARGET system (55), to express the transgenes only in the adult CNS. Although RW under ElavGal4; Gal80ts at 25°C is expressed throughout development, its level is nearly 50% reduced relative to driving the same transgenes with ElavGal4 (Supplementary Material, Fig. S3B). Multiple independent experiments revealed that when restricted to the adult CNS, RW was significantly elevated upon Leo co-expression in all flies examined, but reduced by co-expression with D14-3-3ɛ (Fig. 7G). These results were corroborated with the phosphorylation-dependent antibodies AT8 and AT100 (Fig. 7H and I). R406W levels in D14-3-3ɛ-over-expressing animals (Fig. 7G–I) evaded detection with the AT100 antibody (Fig. 7I), unless large excess of lysate was loaded and exposure prolonged (not shown). Collectively, elevation of Leo results in increased steady-state levels of WT and RW Tau and potentiates their toxicity. In contrast, increased D14-3-3ɛ elevates WT, drastically decreases RWand has minor effects on their toxicity. These observations are in accordance with the abrogation results and support the notion that 14-3-3s have distinct functions in the regulation of Tau levels and its toxicity. Interestingly, although Leo elevation did not alter D14-3-3ɛ levels (Supplementary Material, Fig. S2C), over-expression of the later resulted in decreased levels of endogenous Leo (Supplementary Material, Fig. S2D). Leo reduction is expected to result in heterodimer attenuation and likely accounts for the reduced RW levels under these conditions (Fig. 7). 14-3-3 levels affect Tau stability The WT and RW proteins are transgenic and therefore their transcription levels under UAS control should be relatively constant. Therefore, changes in their steady-state levels upon co-expression with Leo and D14-3-3ɛ must reflect alterations either in translation or their stability. Effects of 14-3-3 elevation on WT and RW translation cannot be totally excluded, but are improbable because both transgenes are controlled by the same vector-encoded regulatory regions. Hyperphosphorylation has been reported to decrease Tau ubiquitination and proteosomal degradation (69,70) and broad specificity phosphatases have been reported to promote Tau proteolysis (71). Given that 14-3-3s bind phosphorylated Ser and Thr (72), they could suppress Tau dephosphorylation by occluding phosphatases, thus enhancing its stability and steady-state levels. During these and previous experiments (60), we noticed that partially dephosphorylated WT Tau appeared unstable or expediently degraded by proteases endogenous to the brain lysate. This was modeled systematically by limited treatment of brain lysates with the broad specificity λ-phosphatase. After a 30 min incubation at 30°C with 50 units of λ-phosphatase, WT Tau became fully dephosphorylated, while 5 units appeared to have little if any effect under these conditions (Fig. 8A). An equal amount of lysate treated with 25 units λ-phosphatase was partially dephosphorylated indicated by its intermediate mobility and interestingly, it was significantly reduced quantitatively relative to the starting material. Confirmation that quantitatively comparable amounts of total lysate were present in all samples was provided by the Syntaxin loading controls. Additional experimental permutations such as limiting λ-phosphatase, longer or shorter incubations yielded qualitatively similar outcomes (not shown). These results suggest that Tau dephosphorylation, possibly at exposed or easily accessible sites, renders the protein susceptible to proteolytic degradation in vivo as previously suggested largely by in vitro results (73). Figure 8. View largeDownload slide 14-3-3 proteins affect Tau stability. WT, RW and STA Tau variants were expressed specifically in adult CNS under ElavGal4. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s was applied to reveal significant differences. (A) A western blot of head lysates from WT Tau-expressing animals incubated with the indicated units of λ-phosphatase for 30 min at 30°C probed with the T14 antibody. Partial dephosphorylation of WT with 25 units phosphatase reduced its levels compared with untreated, or expediently fully dephosphorylated protein. Note that complete dephosphorylation in the lysate is also confirmed by the increased electrophoretic mobility of the Syntaxin (Syx) loading control. (B) A representative blot and the quantification of three additional independent experiments of the levels of WT from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing WT and Leo or WT and D14-3-3ε retained significantly more Tau than lysates expressing WT alone (Dunnett’s, P < 0.001). (C) A representative blot and the quantification of additional independent experiments (n = 4) of the levels of RW from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (Input ratio-treated ratio). Treated lysates from samples co-overexpressing RW and D14-3-3ε retained significantly less Tau than lysates expressing RW alone or also co-overexpressing Leo (Dunnett’s P < 0.01). (D) A representative blot and the quantification of additional independent experiments (n = 3) of the levels of STA from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing STA and Leo or D14-3-3ε retained significantly more Tau than lysates expressing STA alone (Dunnett’s P < 0.0004). Figure 8. View largeDownload slide 14-3-3 proteins affect Tau stability. WT, RW and STA Tau variants were expressed specifically in adult CNS under ElavGal4. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s was applied to reveal significant differences. (A) A western blot of head lysates from WT Tau-expressing animals incubated with the indicated units of λ-phosphatase for 30 min at 30°C probed with the T14 antibody. Partial dephosphorylation of WT with 25 units phosphatase reduced its levels compared with untreated, or expediently fully dephosphorylated protein. Note that complete dephosphorylation in the lysate is also confirmed by the increased electrophoretic mobility of the Syntaxin (Syx) loading control. (B) A representative blot and the quantification of three additional independent experiments of the levels of WT from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing WT and Leo or WT and D14-3-3ε retained significantly more Tau than lysates expressing WT alone (Dunnett’s, P < 0.001). (C) A representative blot and the quantification of additional independent experiments (n = 4) of the levels of RW from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (Input ratio-treated ratio). Treated lysates from samples co-overexpressing RW and D14-3-3ε retained significantly less Tau than lysates expressing RW alone or also co-overexpressing Leo (Dunnett’s P < 0.01). (D) A representative blot and the quantification of additional independent experiments (n = 3) of the levels of STA from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing STA and Leo or D14-3-3ε retained significantly more Tau than lysates expressing STA alone (Dunnett’s P < 0.0004). Although over-expression did not alter WT and RW phosphoprofiles at the sites surveyed (Figs 6D–F and 7D–I), abrogation of 14-3-3s resulted in decreased phosphorylation (Figs 3D–E and 4D–E). Therefore, we investigated whether Tau elevation consequent to excess Leo or D14-3-3ɛ resulted from changes in phosphorylation state-dependent stability. To address this hypothesis, we incubated equal amounts of brain lysates from animals expressing WT and R406W with or without excess transgenic Leo and D14-3-3ɛ under conditions of partial dephosphorylation. As shown in Figure 8B, 60% of WT was degraded by endogenous proteases after λ-mediated limited dephosphorylation. However, co-overexpression of either 14-3-3 ameliorated this effect resulting in degradation of only 20–25% of the WT input. In contrast, the RW protein appeared less vulnerable to endogenous proteases since under the same conditions only 20% of the protein degraded (Fig. 8C). Interestingly however, excess D14-3-3ɛ resulted in significantly increased RW degradation with less than 40% of the starting material remaining at the end of incubation. In contrast, elevated Leo did not appear to affect RW levels. These results are consistent with the in vivo quantitation of WT and RW Tau steady-state levels upon excess 14-3-3s (Figs 6 and 7). To investigate whether RW destabilization upon D14-3-3ɛ excess is unique to this variant or it pertains to other mutants we subjected the STA variant in the same stability assay. This laboratory-generated mutant (61,66) contains Alanines in place of Ser238 and importantly replacing Thr245, which is proposed to bind 14-3-3s when phosphorylated (Fig. 2). As WT and RW, it binds both Leo and D14-3-3ɛ (Supplementary Material, Fig. S5A). Moreover, as for WT, co-expression with Leo or D14-3-3ɛ increases its levels in the adult CNS (Supplementary Material, Fig. S2B). Importantly, this appears to be the consequence of enhanced stability that appears to surpass even that of WT (Fig. 8D). Therefore, the destabilization of RW upon D14-3-3ɛ excess is specific to this particular variant and does not characterize all mutant Tau proteins. It follows then that 14-3-3 elevation has a stabilizing effect on Tau probably by suppressing access to phosphatases and/or proteases. Because 14-3-3 dimer composition is dynamic (30,74), particular homo and heterodimer distributions will likely favor distinct Tau conformations which potentially permit or occlude kinase and phosphatase access differentially. As D14-3-3ɛ overexpression results in decreased endogenous Leo (Supplementary Material, Fig. S2D), the relative pool of Leo homodimers is expected to decrease in response. This is likely exaggerated by D14-3-3ɛ excess, which may further decrease Leo homodimer levels. Under endogenous 14-3-3 levels and under Leo excess, WT appears stable. Therefore, its destabilization upon excessive D14-3-3ɛ is likely the consequence of decreased Leo homodimers. The RW mutation which encompasses a putative 14-3-3 binding site, appears to render the stability of the mutant protein particularly sensitive to Leo homodimer levels. Therefore, the results of 14-3-3 genetic and pharmacological attenuation and those of overexpression strongly indicate that the steady-state levels of WT and RW are modulated by the composition of 14-3-3 dimers. Discussion 14-3-3 interaction modes with Tau Our genetic and biochemical evidence in strong support of the proteomic results, demonstrate that 14-3-3s interact with human WT and the RW Tau variant in vivo, in agreement with in vitro evidence (44). 14-3-3s likely engage at least one or more of the five putative 14-3-3 consensus binding sites (Fig. 2), involving potentially phosphorylated Serines and Threonines. However, since Tau possesses over 80 potential phospho-Ser/phospho-Thr (75), additional phosphorylation-dependent, but non-consensus 14-3-3 docking sites may exist, as bona fide 14-3-3 clients with highly deviant binding motifs have been reported (76). Significantly, because both 14-3-3s bind dephosphorylated WT and RW (Fig. 1E), there are non-phosphorylation-dependent binding sites also. Accordingly, 14-3-3ζ binding on Ser262 (surrounding sequence KIGSTE) in R1 has been reported to be phosphorylation independent (41), but to promote its phosphorylation (41,44). Additional such sites could even involve interactions between the acidic binding groove of 14-3-3s with highly basic areas of Tau, which could emerge upon assumption of certain local conformations. Therefore, Tau presents 14-3-3 dimers with multiple sites and modes of intramolecular and intermolecular interactions. Because dimeric 14-3-3s can in principle bind in a bidentate mode with two client proteins concurrently (39), they may simultaneously bind phosphorylation-dependent and phosphorylation independent sites on the same Tau molecule, different Tau molecules, or Tau and interacting proteins. This in fact is reflected in the ability of the transgenic difopein to bind 14-3-3s and to co-immunoprecipitate Tau (Fig. 1B and C). The multiplicity of potential 14-3-3 binding sites on Tau and the potential for bidentate binding are expected to result in the observed complex interactions that largely depend on the available intracellular complement of 14-3-3s. Functional consequences of attenuated 14-3-3 interaction with Tau Collectively, our data address an important issue in 14-3-3 biology, namely that of their stoichiometry and the functional differences of dimeric states in vivo. Clearly the ratio of available 14-3-3 isoforms and consequently the composition of dimers and monomers influences significantly the steady-state phosphorylation status, at least of the pathology-relevant sites examined. However, 14-3-3s are not limiting for Tau stability and have marginal effects on its toxicity. Considering the genetic abrogation and pharmacological inhibition data, Leo-containing dimers promote low-level steady-state phosphorylation at the sites tested (Figs 3D–F, 4D–F and 5E–H). Importantly however, D14-3-3ɛ homodimers appear to promote phosphorylation of these sites because D14-3-3ɛ attenuation nearly abolishes their occupation (Figs 3D–F and 4D–F). 14-3-3s bind phosphorylated and un-phosphorylated Tau and in fact D14-3-3ɛ binding appears enhanced upon dephosphorylation (Fig. 1E). Therefore, we suggest that Leo-containing homo and heterodimers may use distinct phosphorylation-dependent and independent modes of interaction with Tau and this could stabilize its conformation locally in a manner that promotes or suppresses access to kinases or phosphatases. In addition, binding itself may occlude access of such enzymes to these sites. Such interactions are clearly sensitive to dimer composition, expected to be highly dynamic (77) and are consistent with the differential effects on occupation of the phospho-sites assessed when the intracellular complement of 14-3-3s is altered (Figs 3D–F and 4D–F). Therefore the differential distribution of 14-3-3 isoforms in different neuronal types (78), could be reflected in the reported tissue-specific Tau phosphorylation patterns (60,79–81). Interestingly, changes in the 14-3-3 complement in Leo-, Eps- and Leo- Eps- animals did not precipitate significant changes in WT or RW toxicity, unless functional inhibition was concurrently applied (Fig. 5K and O). This likely reflects both the complexity of Tau interactions with the distinct dimers and the partial functional redundancy of 14-3-3 proteins (29). Significantly, RW toxicity is differentially sensitive to D14-3-3ɛ levels (Fig. 5N), possibly resulting from the mutation on the putative binding site at aa406-413 (Fig. 2). It is highly interesting and significant that the partial suppression of RW toxicity is mediated by D14-3-3ɛ homodimers that appear to form when Leo is limiting (Fig. 3G). These results are consistent with the notion that 14-3-3s facilitate assumption of distinct local or global conformations for WT and RW with consequent changes in toxicity. Alternatively, distinct WT and RW conformations potentially mediated by phosphorylation, render them differentially accessible to 14-3-3s and other proteins mediating toxicity. Therefore, Tau expressing neurons do not tolerate well the functional loss of 14-3-3s. A potential reason for this is that 14-3-3 abrogation may induce ER (82), oxidative (83), or other type of stress in neurons already under oxidative stress because of the presence of pathogenic WT or RW Tau (84). The importance of 14-3-3s in cellular stress responses may be reflected by the strong lethality associated with genetic abrogation of both fly isoforms (Figs 3H and 4G), or their functional pharmacological inhibition (Fig. 5L and P). Therefore, our results strongly indicate that BV02-mediated inhibition of 14-3-3 binding is very likely to precipitate significant unwanted effects to be useful in treating Tauopathies. Functional consequences of 14-3-3 elevation Over-accumulation of Leo, but not D14-3-3ɛ precipitated significant lethality on its own (Fig. 6B). This combined with the drastic reduction in viability of animals with severely reduced functional 14-3-3s (Fig. 5L and P) and in agreement with prior results (29), demonstrate the importance of 14-3-3 homeostasis for survival. Because only Leo over-accumulation yielded lethality, it is likely that the effect is precipitated by excess Leo-containing dimers. The reason for this lethality is unclear to date, but it is under investigation and could be related to synaptic deficits (44) during development, or any of the multiple cellular processes requiring regulated 14-3-3 levels and activity (85,86). Interestingly over-accumulation of either 14-3-3 does not change the phosphorylation pattern of either WT or RW (Figs 6 and 7). Rather it affects their stability in an isoform-specific manner and differentially for WT and RW. Tau stability depends on its phosphorylation state [Fig. 8 and (70)] and hyper-phosphorylation protects Tau from proteolytic degradation (70). 14-3-3 over-accumulation did not change the phosphorylation state, at least of the sites tested, but rendered WT resistant to degradation. Therefore, the increase in WT levels is a likely result of excess 14-3-3-mediated conformational changes that render it resistant to endogenous phosphatases or proteases. In contrast, the effects of 14-3-3 elevation on the RW variant, which eliminates a single potential 14-3-3 binding site, are differential and isoform-specific. Whereas, Leo excess nearly doubles RW levels as for WT, over-accumulation of D14-3-3ɛ precipitates its drastic reduction (Fig. 4H–M). This appears to be a consequence of D14-3-3ɛ-dependent increase in susceptibility to degradation of under-phosphorylated RW (Fig. 7C). Hence, the D14-3-3ɛ excess and the resultant shift in heterodimer/homodimer balance appears to favor conformational state(s) that expose protease-sensitive sites on RW due to the mutation itself, which removes a 14-3-3 consensus binding site. Because RW degradation is not observed upon Leo excess, it is likely that over-accumulation of D14-3-3ɛ reduces the abundance of Leo/Leo homodimers, which appear to favor the degradation-resistant conformation. The differential effect is specific to RW, as the STA mutant affecting another putative 14-3-3 binding site, is completely protected from degradation by overexpression of either Leo or D14-3-3ɛ (Fig. 8D). Our results indicate that WT and STA in contrast to RW likely assume distinct conformations in the fly CNS, which render them differentially accessible to proteases and 14-3-3s are involved in these processes. The ability of 14-3-3 dimers for bidentate binding to different Tau client proteins could differentially promote stabilization of conformations that inhibit dephosphorylation and enhance stability. This may be further mediated by Tau interacting or regulating proteins whose identity is currently unknown, albeit under investigation. In summary, the data support the hypothesis that 14-3-3 excess stabilizes Tau except for RW, where removal of a binding site appears to trigger distinct conformational changes that render the variant protein particularly sensitive to the stoichiometry of 14-3-3 homo/heterodimers. Because 14-3-3s are differentially distributed in the mammalian CNS (78), it is possible that the expected distinct distribution of 14-3-3 homo and heterodimers, affects the pathogenicity of WT and mutant Tau proteins in a tissue-specific manner. This may underlie, at least in part, the neuronal type specificity of the various Tauopathies and the symptom variability and intensity that characterizes them (13). Are 14-3-3s Tauopathy mediators or innocent bystanders? Soon after their identification and characterization, 14-3-3 proteins became associated with diverse inflammatory, vascular, malignant and degenerative neuropathologies (53,87). Because in most of these disorders, elevation of 14-3-3 proteins was noted both in the CSF and in and around lesions, it remained unclear whether these proteins were incidental non-specific markers of neuronal stress and neurodegeneration, or involved in the pathogenesis or severity of each condition. Our data favor the first notion, because although overaccumulation of Leo resulted in lethality, this appears to be a consequence of altered 14-3-3 homeostasis. Pathogenic Tau phosphorylation increases potential 14-3-3 binding sites and probably recruits additional dimers. This may alter the intraneuronal distribution of 14-3-3s and may limit them for vital cellular processes. Given their homeostatic propensity, 14-3-3 levels may be elevated in compensation. Elevated 14-3-3s will favor stable Tau conformations and lead to its accumulation which increases neurotoxic consequences. In addition, the elevated 14-3-3 dimer load on hyper-phosphorylated Tau may stabilize transient conformations that promote additional phosphorylation at distant sites enhancing toxicity. This scenario of progressive 14-3-3 elevation and recruitment of an otherwise ‘bystander’ protein is a result of the large number of phopshorylation-dependent and non-phosphorylation-dependent 14-3-3 docking sites on Tau. Alternatively, accumulation of pathogenic Tau may induce cellular stress and 14-3-3s as stress-related proteins (23) could be elevated in response potentially initiating the progressive elevation scenario described above. In Drosophila however, 14-3-3 levels are not changed upon Tau accumulation in the CNS (Supplementary Material, Fig. S4). In vertebrates on the other hand, elevated 14-3-3s may be increasingly recruited to Tau molecules as they acquire pathogenic tags and either enhance or suppress conformations that favor their stability depending on the properties of the particular Tau variant and of the prevalent 14-3-3 dimers in that particular neuron. Finally, our results suggest that in vivo approaches in the CNS as the one described herein are invaluable in capturing the dynamic, adaptive and homeostatic interactions of 14-3-3 proteins with Tau and other client proteins with multiple modes of interaction and docking sites. Materials and Methods Drosophila culture and strains Flies were cultured in standard sugar-wheat flour food supplemented with soy flour and CaCl2. Pan-neuronal transgene expression was achieved using the driver ElavC155-Gal4 (ElavGal4) (88,89). Fly crosses and experiments were performed at 25°C unless noted otherwise. The ElavGal4; Gal80ts strain was constructed using standard methods. The Gal80ts strain (55) was obtained from the Bloomington Stock Centre. Conditional transgene expression under Elav-Gal4; Gal80ts was induced specifically in adult flies by incubation at 30°C for 10 days post-emergence. Transgenic fly lines carrying UAS-htau0N4R and UAS-htauR406W inserts were gifts of Dr M. Feany (Harvard Medical School) (90). The UAS-htauFLAG-2N4RSTA has been described in (61). The UAS-mycD14-3-3ɛ transgenic strains (91) and the D14-3-3ɛ RNAi-generating transgenes (92) have been described previously. The UAS-leo transgenes were generated by placing the entire leoII cDNA (93) in the pUAST vector (94), while Leo RNAi-generating transgenes were obtained from the Vienna Drosophila Resource Centre and were normalized to Cantonised w1188. Double transgenic strains were generated by standard crosses. The UAS-htauFLAG-2N4R strain has been described elsewhere (61). The UAS-htauFLAG-R406W-2N4R strain was generated by replacing Arg406 with Trp using the QuickChange XL site-directed mutagenesis kit according to the manufacturer’s instructions (Agilent Technologies). The mutagenic oligonucleotides 5′-GGGGACACGTCTCCATGGCATCTCAGCAATGTCTCC-3′ and 5′-GGAGACATTGCTGAGATGCCATGGAGACGTGTCCCC-3′ were annealed onto the UAS-htau2N4R-FLAG plasmid template and introduced an NcoI restriction site which was used for screening of positive clones. The sequence of the mutant was confirmed by dsDNA sequencing (Lark technologies). Transgenic flies were obtained with standard methods. To assess the effect of 14-3-3s on the levels and phosphorylation of Tau, we generated the double transgenic stocks UAS-leoII; UAS-htau0N4R, UAS-d14-3-3ɛ; UAS-htau0N4Rand UAS-leoII; UAS-htauR406W, UAS-d14-3-3ɛ; UAS-htauR406W which were crossed with ElavGal4 virgins as required. Conversely, to assess the effects of 14-3-3 attenuation, the following strains were generated ElavGal4; leoRNAi/CyO and ElavGal4; epsRNAi/TM3Sb and crossed to UAS-htau0N4R and UAS-htauR406W males as indicated. To generate flies lacking both 14-3-3 isoforms ElavGal4; epsRNAi/TM3Sb were crossed with UAS-leoRNAi; UAS-htau0N4R or with UAS-htauR406W. Viability assays To determine the effect of WT or RW expression on viability, 5 ElavGa4 females were crossed with 3 WT or RW homozygous males. After 24 h they transferred to new vials and allowed to lay eggs for three days and then discarded. The number of surviving progeny per live female was determined when adults emerged. Each assessment was performed at least in triplicate with five females each. Control progeny were obtained by concomitantly crossing the same number of ElavGal4 females to w1118 males and the mean number of their progeny per female was set as 100%. This number was used to normalize the numbers of Tau-expressing progeny expressed as a percent. 14-3-3 Inhibitor experiments For the 14-3-3 inhibitor experiments, five ElavGal4 females were crossed to three relevant males from the abovementioned stocks and allowed to lay eggs for three days at 25°C on normal food supplemented with 0 (DMSO), 0.15 or 0.3 nm of the 14-3-3 antagonist II, BV02 (Calbiochem). Viability was monitored by counting the number of adult flies emerging per genotype with and without the antagonist. Control experiments with higher BV02 concentrations (0.35–0.5 nm) did not yield any progeny. Western blot analysis To determine total Tau levels and its phosphorylation status in different genotypes, adult female fly heads 1–3 days post-eclosion were homogenized in 1× Laemmli buffer (50 mm Tris pH 6.8, 100 mm DTT, 5% 2-mercaptoethanol, 2% SDS, 10% glycerol and 0.01% bromophenol blue). The lysates were boiled for 5 min at 95°C, centrifuged at 9000g for 5 min and separated by SDS-PAGE using a 10% separating gel. Proteins were transferred to a PVDF membrane at 100 V for 1 h and probed with the following monoclonal antibodies: T14 (Zymed laboratories) at 1: 1000 for 1 head/lane, T46 (Invitrogen) at 1: 2000 for 1 head/lane, AT100 (Pierce Endogen) at 1: 250 for 1 head/lane and the polyclonal antibodies pS262 (Biosource) at 1: 500 for 1 head/lane, anti-LEO (95) at 1: 20 000 for 0.2 head/lane and anti-D14-3-3ɛ (29) at 1: 2000 for 0.2 head/lane. The monoclonal antibody AT8 was kindly provided by A. Mudher and was used at 1: 200 dilution for 1 head/lane. To normalize for sample loading, the membranes were concurrently probed with an anti-syntaxin (Syn) monoclonal antibody (8C3, Developmental Studies Hybridoma Bank) at 1: 2000 for 1 head/lane and at 1: 500 for 0.2 heads/lane. Appropriate HRP-conjugated secondary antibodies were applied at 1: 5000. Proteins were visualized with chemiluminescence (ECL Plus, Amersham) and signals were quantified by densitometry with the ImageQ program. Results were plotted as means with S.E.M. from at least three independent experiments. The data were analyzed by standard parametric statistics as indicated in the figure legends. Pull-down assay Temporal regulation of UAS controlled transgenes under the ubiquitously expressed temperature sensitive Gal80ts has been described earlier (55). At the end of induction, flies were harvested, frozen at −72°C and then decapitated en masse by sieving in liquid nitrogen. Harvested heads were homogenized in PBST buffer (1× PBS pH 7.4 containing 0.01% Tween® 20) supplemented with protease inhibitors and lysates were incubated over-night with either magnetic beads coupled with HT7 anti-Tau antibody or with GST-difopein-conjugated beads. After washing, proteins were eluted with 0.1 m glycine-HCl pH 3.5 and subjected to SDS-PAGE. Immunoblotting was performed as described above. Biotinylated HT7 antibody (Thermo Scientific) was immobilized on Dynabeads M-280® Streptavidin (Invitrogen) according to the manufacturer’s instructions. The pGEX-6P1 vector encoding GST-tagged difopein was kindly provided by Dr Marco Lalle (56). To generate the protein, BL21(DE3) transformed cells were grown in LB medium (2% bacto tryptone, 0.5% yeast extract, 0.05% NaCl) at 37°C until OD600 0.5. Protein expression was induced by addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to 0.5 mm final concentration at 37°C for 3 h. GST-difopein was immobilized on glutathione-agarose beads (Sigma) according to the manufacturer’s instructions. Flag-tagged 2N4R-STA was immunoprecipated with anti-Flag M2 affinity resin (Sigma) according to manufacturer’s instructions. In vitro dephosphorylation Dephosphorylation prior to immunoprecipitation was achieved by treating the lysates with λ phosphatase (New England Biolabs). Briefly, head homogenates from adult flies expressing the indicated Flag-tagged transgenes pan-neuronally were prepared in lysis buffer (50 mm Tris pH 7.5, 150 mm NaCl, 1 mm MnCl2, 1% Triton) supplemented with Halt protease inhibitor cocktail (Pierce) EDTA-free. Lysates were split in two and they were either supplemented with phosphatase inhibitor cocktail (Sigma) or treated with λ phosphatase (2.5 mg of total protein were incubated with 1000 units of λ phosphatase at 30°C for 1 h). The reaction was stopped by addition of phosphatase inhibitors and lysates were incubated either with anti-Flag M2 affinity resin or with glutathione-agarose beads coated with GST-difopein over night at 4°C. The starting buffer of this protocol is different than the one below because immunoprecipitation has to follow dephosphorylation. Immunoprecipitation is not compatible with the standard de-phosphorylation buffer presented below. Homogenates from heads of adult flies were prepared in RIPA buffer (137 mm NaCl, 20 mm Tris pH 8.0, 10% glycerol, 0.1% SDS and 0.1% sodium deoxycholate). A total of 20 μg protein were incubated with the indicated units of λ phosphatase (New England Biolabs) at 30°C for 30 min, according to manufacturer’s instructions. As different batches of λ-phosphatase have different activities, we calibrated the units necessary to produce complete dephosphorylation per batch and then used the appropriate units to produce partial dephosphorylation as indicated. One-third of each sample and equivalent amount of untreated samples were resolved by SDS-PAGE and subjected to western blot analysis as described earlier. Cross-linking Eight heads were lysed in 70 μl of ice-cold buffer (20 mm sodium phosphate, pH 7.4, 150 mm NaCl supplemented with phosphatase and protease inhibitor mixture) and centrifuged for 15 min at 14 000g at 4°C. Head lysates were divided in two fractions of 30 μl. One was cross-linked by addition of 1.5 mm BS3 final concentration (Thermo Scientific) while the other remained untreated (lysis buffer was added instead of BS3). Samples were incubated for 2 h on ice and the reaction was stopped by the addition of 1 m Tris pH 7.4 buffer to a final concentration of 20 mm. Samples were analyzed by western blotting after the addition of 5× Laemmli Buffer. Statistical Analyses Untransformed (raw) data were analyzed parametrically with the JMP 7.1 statistical software package (SAS Institute Inc., Cary, NC). Supplementary Material Supplementary Material is available at HMG online. Acknowledgements The initial parts of this work were supported by grants 01ΕΔ207 (E.M.C.S.) and 04EP18 (K.P.) from the Hellenic General Secretariat for Research and Technology. We thank the Developmental Hybridoma Bank (DSHB) for antibodies, Drs Botas, Feany, Lu and Mudher for fly strains, Marco Lalle for GST-difopein reagents and George Panayotou for advise and support with the proteomic analyses. Conflict of Interest statement. None declared. References 1 Black M.M., Slaughter T., Moshiach S., Obrocka M., Fischer I. ( 1996) Tau is enriched on dynamic microtubules in the distal region of growing axons. J. Neurosci ., 16, 3601– 3619. Google Scholar CrossRef Search ADS PubMed  2 Kempf M., Clement A., Faissner A., Lee G., Brandt R. ( 1996) Tau binds to the distal axon early in development of polarity in a microtubule- and microfilament-dependent manner. J. Neurosci ., 16, 5583– 5592. Google Scholar CrossRef Search ADS PubMed  3 Avila J., Lucas J.J., Pérez M.A.R., Hernández F. ( 2004) Role of tau protein in both physiological and pathological conditions. Physiol. Rev ., 84, 361– 384. Google Scholar CrossRef Search ADS PubMed  4 Baas P.W., Qiang L. ( 2005) Neuronal microtubules: when the MAP is the roadblock. Trends Cell Biol ., 15, 183– 187. Google Scholar CrossRef Search ADS PubMed  5 Terwel D., Dewachter I., Van Leuven F. ( 2002) Axonal transport, tau protein, and neurodegeneration in Alzheimer's disease. Neuromol. Med ., 2, 151– 165. Google Scholar CrossRef Search ADS   6 Ballatore C., Lee V.M., Trojanowski J.Q. ( 2007) Tau-mediated neurodegeneration in Alzheimer's disease and related disorders. Nat. Rev. Neurosci ., 8, 663– 672. Google Scholar CrossRef Search ADS PubMed  7 Wang Y., Mandelkow E. ( 2016) Tau in physiology and pathology. Nat. Rev. Neurosci ., 17, 5– 21. Google Scholar CrossRef Search ADS PubMed  8 Avila J., Pérez M., Lim F., Gómez-Ramos A., Hernández F., Lucas J.J. ( 2004) Tau in neurodegenerative diseases: tau phosphorylation and assembly. Neurotox. Res ., 6, 477– 482. Google Scholar CrossRef Search ADS PubMed  9 Buée L., Bussière T., Buée-Scherrer V., Delacourte A., Hof P.R. ( 2000) Tau protein isoforms, phosphorylation and role in neurodegenerative disorders. Brain Res. Rev ., 33, 95– 130. Google Scholar CrossRef Search ADS   10 Buee L., Delacourte A. ( 1999) Comparative biochemistry of tau in progressive supranuclear palsy, corticobasal degeneration, FTDP-17 and Pick's disease. Brain Pathol ., 9, 681– 693. Google Scholar CrossRef Search ADS PubMed  11 Hardy J., Orr H. ( 2006) The genetics of neurodegenerative diseases. J. Neurochem ., 97, 1690– 1699. Google Scholar CrossRef Search ADS PubMed  12 Ingram E.M., Spillantini M.G. ( 2002) Tau gene mutations: dissecting the pathogenesis of FTDP-17. Trends Mol. Med ., 8, 555– 562. Google Scholar CrossRef Search ADS PubMed  13 Lee V.M.-Y., Goedert M., Trojanowski J.Q. ( 2001) Neurodegenerative tauopathies. Annu. Rev. Neurosci ., 24, 1121– 1159. Google Scholar CrossRef Search ADS PubMed  14 Spillantini M.G., Goedert M. ( 1998) Tau protein pathology in neurodegenerative diseases. Trends Neurosci ., 21, 428– 433. Google Scholar CrossRef Search ADS PubMed  15 Brookmeyer R., Johnson E., Ziegler-Graham K., Arrighi H.M. ( 2007) Forecasting the global burden of Alzheimer's disease. Alzheimers Dement ., 3, 186– 191. Google Scholar CrossRef Search ADS PubMed  16 Spillantini M.G., Van Swieten J.C., Goedert M. ( 2000) Tau mutations in frontotemporal dementia FTDP-17 and their relevance for Alzheimer's disease. Biochim. Biophys. Acta , 2, 0193– 0121. 17 Weder N.D., Aziz R., Wilkins K., Tampi R.R. ( 2007) Frontotemporal dementias: a review. Ann. Gen. Phychiatry , 6, 15. doi: 10.1186/1744-859X-6-15. Google Scholar CrossRef Search ADS   18 Sergeant N., Delacourte A., Buee L. ( 2005) Tau protein as a differential biomarker of tauopathies. Bioch. Biophys. Acta , 1739, 179– 197. 19 Johnson G.V., Stoothoff W.H. ( 2004) Tau phosphorylation in neuronal cell function and dysfunction. J. Cell Sci ., 117, 5721– 5729. Google Scholar CrossRef Search ADS PubMed  20 Martin L., Latypova X., Terro F. ( 2011) Post-translational modifications of tau protein: implications for Alzheimer's disease. Neurochem. Int ., 58, 458– 471. Google Scholar CrossRef Search ADS PubMed  21 Alonso A. d C., Li B., Grundke-Iqbal I., Iqbal K. ( 2008) Mechanism of tau-induced neurodegeneration in Alzheimer disease and related tauopathies. Curr. Alzheimer Res ., 5, 375– 384. Google Scholar CrossRef Search ADS PubMed  22 Gong C.-X., Liu F., Grundke-Iqbal I., Iqbal K. ( 2005) Post-translational modifications of tau protein in Alzheimer's disease. J. Neural Transm ., 112, 813– 838. Google Scholar CrossRef Search ADS   23 Soulié C., Nicole A., Delacourte A., Ceballos-Picot I. ( 2004) Examination of stress-related genes in human temporal versus occipital cortex in the course of neurodegeneration: involvement of 14-3-3 zeta in this dynamic process. Neurosci. Lett ., 365, 1– 5. Google Scholar CrossRef Search ADS PubMed  24 Aitken A. ( 2006) 14-3-3 proteins: a historic overview. Semin. Cancer Biol ., 16, 162– 172. Google Scholar CrossRef Search ADS PubMed  25 Gardino A.K., Smerdon S.J., Yaffe M.B. ( 2006) Structural determinants of 14-3-3 binding specificities and regulation of subcellular localization of 14-3-3-ligand complexes: a comparison of the X-ray crystal structures of all human 14-3-3 isoforms. Semin. Cancer Biol ., 16, 173– 182. Google Scholar CrossRef Search ADS PubMed  26 Masters S.C., Fu H. ( 2001) 14-3-3 proteins mediate an essential anti-apoptotic signal. J. Biol. Chem ., 276, 45193– 45200. Google Scholar CrossRef Search ADS PubMed  27 Aghazadeh Y., Papadopoulos V. ( 2016) The role of the 14-3-3 protein family in health, disease, and drug development. Drug Discov. Today , 21, 278– 287. Google Scholar CrossRef Search ADS PubMed  28 Aitken A., Baxter H., Dubois T., Clokie S., Mackie S., Mitchell K., Peden A., Zemlickova E. ( 2002) 14-3-3 proteins in cell regulation. Biochem. Soc. Trans ., 30, 351– 360. Google Scholar CrossRef Search ADS PubMed  29 Acevedo S.F., Tsigkari K.K., Grammenoudi S., Skoulakis E.M.C. ( 2007) In vivo functional specificity and homeostasis of Drosophila 14-3-3 proteins. Genetics , 177, 239– 253. Google Scholar CrossRef Search ADS PubMed  30 Messaritou G., Grammenoudi S., Skoulakis E. ( 2010) Dimerization is essential for 14-3-3zeta stability and function in vivo. J. Biol. Chem ., 285, 1692. Google Scholar CrossRef Search ADS PubMed  31 Messaritou G., Leptourgidou F., Franco M., Skoulakis E.M.C. ( 2009) A third functional isoform enriched in mushroom body neurons is encoded by the Drosophila 14-3-3zeta gene. FEBS Lett ., 583, 2934– 2938. Google Scholar CrossRef Search ADS PubMed  32 Muslin A.J., Tanner J.W., Allen P.M., Shaw A.S. ( 1996) Interaction of 14-3-3 with signaling proteins is mediated by the recognition of phosphoserine. Cell , 84, 889– 897. Google Scholar CrossRef Search ADS PubMed  33 Tzivion G., Avruch J. ( 2002) 14-3-3 Proteins: active cofactors in cellular regulation by Serine/Threonine phosphorylation. J. Biol. Chem ., 277, 3061– 3064. Google Scholar CrossRef Search ADS PubMed  34 Yaffe M.B., Elia A.E.H. ( 2001) Phosphoserine/threonine-binding domains. Curr. Opin. Cell. Biol ., 13, 131– 138. Google Scholar CrossRef Search ADS PubMed  35 Kaplan A., Ottmann C., Fournier A. ( 2017) 14-3-3 adaptor protein-protein interactions as therapeutic targets for CNS diseases. Pharmacol. Res ., 125, 114. Google Scholar CrossRef Search ADS PubMed  36 Bridges D., Moorhead G.B.G. ( 2005) 14-3-3 proteins: a number of functions for a numbered protein. Sci. STKE , 2005, re10. Google Scholar PubMed  37 van Hemert M.J., Steensma H.Y., van Heusden G.P. ( 2001) 14-3-3 proteins: key regulators of cell division, signalling and apoptosis. BioEssays , 23, 936– 946. Google Scholar CrossRef Search ADS PubMed  38 van Heusden G. ( 2005) 14-3-3 proteins: regulators of numerous eukaryotic proteins. IUBMB Life , 57, 623– 629. Google Scholar CrossRef Search ADS PubMed  39 Yaffe M.B. ( 2002) How do 14-3-3 proteins work?-Gatekeeper phosphorylation and the molecular anvil hypothesis. FEBS Lett ., 513, 53– 57. Google Scholar CrossRef Search ADS PubMed  40 Chun J. ( 2004) 14-3-3 Protein mediates phosphorylation of microtubule-associated protein tau by serum- and glucocorticoid-induced protein kinase 1. Mol. Cells , 18, 360– 368. Google Scholar PubMed  41 Hashiguchi M., Sobue K., Paudel H.K. ( 2000) 14-3-3 z is an effector of tau protein phosphorylation. J. Biol. Chem ., 275, 25247– 25254. Google Scholar CrossRef Search ADS PubMed  42 Yuan Z., Agarwal-Mawal A., Paudel H.K. ( 2004) 14-3-3 binds to and mediates phosphorylation of microtubule-associated tau protein by Ser9-phosphorylated glycogen synthase kinase 3beta in the brain. J. Biol. Chem ., 279, 26105– 26114. Google Scholar CrossRef Search ADS PubMed  43 Li T., Paudel H.K. ( 2007) 14-3-3zeta facilitates GSK3beta-catalyzed tau phosphorylation in HEK-293 cells by a mechanism that requires phosphorylation of GSK3beta on Ser9. Neurosci. Lett ., 414, 203– 208. Google Scholar CrossRef Search ADS PubMed  44 Qureshi H.Y. ( 2013) Overexpression of 14-3-3z promotes tau phosphorylation at Ser262 and accelerates proteosomal degradation of synaptophysin in rat primary hippocampal neurons. PLoS One , 8, e84615. Google Scholar CrossRef Search ADS PubMed  45 Qureshi H.Y., Li T., MacDonald R., Cho C.M., Leclerc N., Paudel H.K. ( 2013) Interaction of 14-3-3ζ with microtubule-associated protein tau within Alzheimer's disease neurofibrillary tangles. Biochemistry , 52, 6445– 6455. Google Scholar CrossRef Search ADS PubMed  46 Agarwal-Mawal A., Qureshi H.Y., Cafferty P.W., Yuan Z., Han D., Lin R., Paudel H.K. ( 2003) 14-3-3 connects glycogen synthase kinase-3 beta to tau within a brain microtubule-associated tau phosphorylation complex. J. Biol. Chem ., 278, 12722– 12728. Google Scholar CrossRef Search ADS PubMed  47 Bustos D. ( 2012) The role of protein disorder in the 14-3-3 interaction network. Mol. Biosyst ., 8, 178– 184. Google Scholar CrossRef Search ADS PubMed  48 Umahara T., Uchihara T., Tsuchiya K., Nakamura A., Iwamoto T., Ikeda K., Takasaki M. ( 2004) 14-3-3 proteins and zeta isoform containing neurofibrillary tangles in patients with Alzheimer's disease. Acta Neuropathol ., 108, 279– 286. Google Scholar CrossRef Search ADS PubMed  49 Umahara T., Uchihara T., Tsuchiya K., Nakamura A., Ikeda K., Iwamoto T., Takasaki M. ( 2004) Immunolocalization of 14-3-3 isoforms in brains with Pick body disease. Neurosci. Lett ., 371, 215– 219. Google Scholar CrossRef Search ADS PubMed  50 Hernandez F., Cuadros R., Avila J. ( 2004) Zeta 14-3-3 protein favours the formation of human tau fibrillar polymers. Neurosci. Lett ., 357, 143– 146. Google Scholar CrossRef Search ADS PubMed  51 Sadik G., Tanaka T., Kato K., Yanagi K., Kudo T., Takeda M. ( 2009) Differential interaction and aggregation of 3-repeat and 4-repeat tau isoforms with 14-3-3zeta protein. Biochem. Biophys. Res. Commun ., 383, 37– 41. Google Scholar CrossRef Search ADS PubMed  52 Fountoulakis M., Cairns N., Lubec G. ( 1999) Increased levels of 14-3-3 gamma and epsilon proteins in brain of patients with Alzheimer's disease and Down syndrome. J. Neural. Transm. Suppl ., 57, 323– 335. Google Scholar PubMed  53 Berg D., Holzmann C., Riess O. ( 2003) 14-3-3 proteins in the nervous system. Nature , 4, 752– 762. 54 Jeganathan S., Hascher A., Chinnathambi S., Biernat J., Mandelkow E.-M., Mandelkow E. ( 2008) Proline-directed pseudo-phosphorylation at AT8 and PHF1 epitopes induces a compaction of the paperclip folding of Tau and generates a pathological (MC-1) conformation. J. Biol. Chem ., 283, 32066– 32076. Google Scholar CrossRef Search ADS PubMed  55 McGuire S.E., Mao Z., Davis R.L. ( 2004) Spatiotemporal gene expression targeting with the TARGET and gene-switch systems in Drosophila. Sci. STKE , 2004, pl6. Google Scholar PubMed  56 Lalle M., Salzano A.M., Crescenzi M., Pozio E. ( 2006) The Giardia duodenalis 14-3-3 protein is post-translationally modified by phosphorylation and polyglycylation of the C-terminal tail. J. Biol. Chem ., 281, 5137– 5148. Google Scholar CrossRef Search ADS PubMed  57 Sluchanko N., Gusev N. ( 2011) Probable participation of 14-3-3 in tau protein oligomerization and aggregation. J. Alzheimers Dis ., 27, 467– 476. Google Scholar PubMed  58 Obsilova V., Nedbalkova E., Silhan J., Boura E., Herman P., Vecer J., Sulc M., Teisinger J., Dyda F., Obsil T. ( 2008) The 14-3-3 protein affects the conformation of the regulatory domain of human tyrosine hydroxylase. Biochemistry , 47, 1768– 1777. Google Scholar CrossRef Search ADS PubMed  59 Papanikolopoulou K., Skoulakis E.M. ( 2011) The power and richness of modelling tauopathies in Drosophila. Mol. Neurobiol ., 44, 122– 133. Google Scholar CrossRef Search ADS PubMed  60 Grammenoudi S., Kosmidis S., Skoulakis E.M. ( 2006) Cell type-specific processing of human Tau proteins in Drosophila. FEBS Lett ., 580, 4602– 4606. Google Scholar CrossRef Search ADS PubMed  61 Kosmidis S., Grammenoudi S., Papanikolopoulou K., Skoulakis E.M.C. ( 2010) Differential effects of Tau on the integrity and function of neurons essential for learning in Drosophila. J. Neurosci ., 30, 464– 477. Google Scholar CrossRef Search ADS PubMed  62 Gustke N., Trinczek B., Biernat J., Mandelkow E.M., Mandelkow E. ( 1994) Domains of tau protein and interactions with microtubules. Biochemistry , 33, 9511– 9522. Google Scholar CrossRef Search ADS PubMed  63 Zheng-Fischhofer Q., Biernat J., Mandelkow E.-M., Illenberger S., Godemann R., Mandelkow E. ( 1998) Sequential phosphorylation of Tau by glycogen synthase kinase-3beta and protein kinase A at Thr212 and Ser214 generates the Alzheimer-specific epitope of antibody AT100 and requires a paired-helical-filament-like conformation. Eur. J. Biochem ., 252, 542– 552. Google Scholar CrossRef Search ADS PubMed  64 Sadik G., Tanaka T., Kato K., Yamamori H., Nessa B.N., Morihara T., Takeda M. ( 2009) Phosphorylation of tau at Ser214 mediates its interaction with 14-3-3 protein: implications for the mechanism of tau aggregation. J. Neurochem ., 108, 33– 43. Google Scholar CrossRef Search ADS PubMed  65 Nishimura I., Yang Y., Lu B. ( 2004) PAR-1 kinase plays an initiator role in a temporally ordered phosphorylation process that confers tau toxicity in Drosophila. Cell , 116, 671– 682. Google Scholar CrossRef Search ADS PubMed  66 Papanikolopoulou K., Skoulakis E. ( 2015) Temporally distinct phosphorylations differentiate Tau-dependent learning deficits and premature mortality in Drosophila. Hum. Mol. Genet ., 24, 2065– 2077. Google Scholar CrossRef Search ADS PubMed  67 Mancini M., Corradi V., Petta S., Barbieri E., Manetti F., Botta M., Santucci M.A. ( 2011) A new nonpeptidic inhibitor of 14-3-3 induces apoptotic cell death in chronic myeloid leukemia sensitive or resistant to imatinib. J. Pharmacol. Exp. Ther ., 336, 596– 604. Google Scholar CrossRef Search ADS PubMed  68 Corradi V., Mancini M., Manetti F., Petta S., Santucci M.A., Botta M. ( 2010) Identification of the first non-peptidic small molecule inhibitor of the c-Abl/14-3-3 protein-protein interactions able to drive sensitive and Imatinib-resistant leukemia cells to apoptosis. Bioorg. Med. Chem. Lett ., 20, 6133– 6137. Google Scholar CrossRef Search ADS PubMed  69 Iqbal K., Liu F., Gong C.-X., Alonso A.D.C., Grundke-Iqbal I. ( 2009) Mechanisms of tau-induced neurodegeneration. Acta Neuropathol ., 118, 53– 69. Google Scholar CrossRef Search ADS PubMed  70 Sluchanko N.N., Sudnitsyna M.V., Chernik I.S., Seit-Nebi A.S., Gusev N.B. ( 2011) Phosphomimicking mutations of human 14-3-3ζ affect its interaction with tau protein and small heat shock protein HspB6. Arch. Biochem. Biophys ., 506, 24– 34. Google Scholar CrossRef Search ADS PubMed  71 Lee S., Shea T. ( 2013) Regulation of tau proteolysis by phosphatases. Brain Res ., 1495, 30– 36. Google Scholar CrossRef Search ADS PubMed  72 Obsilova V. ( 2008) 14-3-3 proteins: a family of versatile molecular regulators. Physiol. Res ., 57(Suppl 3), S11– S21. Google Scholar PubMed  73 Wang Y., Garg S., Mandelkow E.-M., Mandelkow E. ( 2010) Proteolytic processing of tau. Biochem. Soc. Trans ., 38, 955– 961. Google Scholar CrossRef Search ADS PubMed  74 Ghorbani S., Fossbakk A., Jorge-Finnigan A., Flydal M.I., Haavik J., Kleppe R. ( 2016) Regulation of tyrosine hydroxylase is preserved across different homo- and heterodimeric 14-3-3 proteins. Amino Acids , 48, 1221– 1229. Google Scholar CrossRef Search ADS PubMed  75 Alonso A., Corbo C., Beharry C. ( 2013) Abnormal hyperphosphorylation of tau: sites, regulation, and molecular mechanism of neurofibrillary degeneration. J. Alzheimers Dis ., 9, P673– P139. 76 Coblitz B., Wu M., Shikano S., Li M. ( 2006) C-terminal binding: an expanded repertoire and function of 14-3-3proteins. FEBS Lett ., 580, 1531– 1535. Google Scholar CrossRef Search ADS PubMed  77 Woodcock J. ( 2017) Role of salt bridges in the dimer interface of 14-3-3ζ in dimer dynamics, N-terminal α-helical order and molecular chaperone activity. J. Biol. Chem ., doi: 10.1074/jbc.M117.801019. 78 Martin H., Rostas J., Patel Y., Aitken A. ( 2002) Subcellular localisation of 14-3-3 isoforms in rat brain using specific antibodies. J. Neurochem ., 63, 2259– 2265. Google Scholar CrossRef Search ADS   79 Götz J., Nitsch R. ( 2001) Compartmentalized tau hyperphosphorylation and increased levels of kinases in transgenic mice. Neuroreport , 3, 2007– 2016. 80 Holzer M., Holzapfel H.-P., Zedlick D., Brückner M.K., Arendt T. ( 1994) Abnormally phosphorylated tau protein in Alzheimer's disease: heterogeneity of individual regional distribution and relationship to clinical severity. Neuroscience , 63, 499– 516. Google Scholar CrossRef Search ADS PubMed  81 Lauckner J., Frey P., Geula C. ( 2003) Comparative distribution of tau phosphorylated at Ser262 in pre-tangles and tangles. Neurobiol. Aging , 24, 767– 776. Google Scholar CrossRef Search ADS PubMed  82 Cao L., Lei H., Chang M.-Z., Liu Z.-Q., Bie X.-H. ( 2015) Down-regulation of 14-3-3β exerts anti-cancer effects through inducing ER stress in human glioma U87 cells: involvement of CHOP-Wnt pathway. Biochem. Biophys. Res. Commun ., 462, 389– 395. Google Scholar CrossRef Search ADS PubMed  83 Thandavarayan R.A., Giridharan V.V., Sari F.R., Arumugam S., Veeraveedu P.T., Pandian G.N., Palaniyandi S.S., Ma M., Suzuki K., Gurusamy N., Watanabe K. ( 2011) Depletion of 14-3-3 protein exacerbates cardiac oxidative stress, inflammation and remodeling process via modulation of MAPK/NF-ĸB signaling pathways after streptozotocin-induced diabetes mellitus. Cell Physiol. Biochem ., 28, 911– 922. Google Scholar CrossRef Search ADS PubMed  84 Dias-Santagata D., Fulga T.A., Duttaroy A., Feany M.B. ( 2007) Oxidative stress mediates tau-induced neurodegeneration in Drosophila. J. Clin. Invest ., 117, 236– 245. Google Scholar CrossRef Search ADS PubMed  85 Mackintosh C. ( 2004) Dynamic interactions between 14-3-3 proteins and phosphoproteins regulate diverse cellular processes. Biochem. J ., 381, 329– 342. Google Scholar CrossRef Search ADS PubMed  86 Sluchanko N., Gusev N. ( 2017) Moonlighting chaperone-like activity of the universal regulatory 14-3-3 proteins. FEBS J ., 284, 1279– 1295. Google Scholar CrossRef Search ADS PubMed  87 Boston P.F., Jackson P., Thompson R.J. ( 1982) Human 14-3-3 protein: radioimmunoassay, tissue distribution, and cerebrospinal fluid levels in patients with neurological disorders. J. Neurochem ., 38, 1475– 1482. Google Scholar CrossRef Search ADS PubMed  88 Lin D.M., Goodman C.S. ( 1994) Ectopic and increased expression of Fasciclin II alters motoneuron growth cone guidance. Neuron , 13, 507– 523. Google Scholar CrossRef Search ADS PubMed  89 Robinow S., White K. ( 1988) The locus elav of Drosophila melanogaster is expressed in neurons at all developmental stages. Dev. Biol ., 126, 294– 303. Google Scholar CrossRef Search ADS PubMed  90 Wittmann C.W. ( 2001) Tauopathy in Drosophila: neurodegeneration without neurofibrillary tangles. Science , 293, 711– 714. Google Scholar CrossRef Search ADS PubMed  91 Chen H.-K., Fernandez-Funez P., Acevedo S.F., Lam Y.C., Kaytor M.D., Fernandez M.H., Aitken A., Skoulakis E.M.C., Orr H.T., Botas J., Zoghbi H.Y. ( 2003) Interaction of Akt-phosphorylated ataxin-1 with 14-3-3 mediates neurodegeneration in spinocerebellar ataxia type 1. Cell , 113, 457– 468. Google Scholar CrossRef Search ADS PubMed  92 Tsigkari K., Acevedo S., Skoulakis E. ( 2012) 14-3-3ɛ is required for germ cell migration in Drosophila. PLoS One , 7, e36702. Google Scholar CrossRef Search ADS PubMed  93 Philip N., Acevedo S., Skoulakis E.M.C. ( 2001) Conditional rescue of olfactory learning and memory defects in mutants of the 14-3-3z gene leonardo. J. Neurosci ., 21, 8417– 8425. Google Scholar CrossRef Search ADS PubMed  94 Brand A.H., Perrimon N. ( 1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development , 118, 401– 415. Google Scholar PubMed  95 Skoulakis E.M.C., Davis R.L. ( 1996) Olfactory learning deficits in mutants for leonardo, a Drosophila gene encoding a 14-3-3 protein. Neuron , 17, 931– 944. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Human Molecular Genetics Oxford University Press

Differential effects of 14-3-3 dimers on Tau phosphorylation, stability and toxicity in vivo

Loading next page...
 
/lp/ou_press/differential-effects-of-14-3-3-dimers-on-tau-phosphorylation-stability-5RK0o0eLtw
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0964-6906
eISSN
1460-2083
D.O.I.
10.1093/hmg/ddy129
Publisher site
See Article on Publisher Site

Abstract

Abstract Neurodegenerative dementias collectively known as Tauopathies involve aberrant phosphorylation and aggregation of the neuronal protein Tau. The largely neuronal 14-3-3 proteins are also elevated in the central nervous system (CNS) and cerebrospinal fluid of Tauopathy patients, suggesting functional linkage. We use the simplicity and genetic facility of the Drosophila system to investigate in vivo whether 14-3-3s are causal or synergistic with Tau accumulation in precipitating pathogenesis. Proteomic, biochemical and genetic evidence demonstrate that both Drosophila 14-3-3 proteins interact with human wild-type and mutant Tau on multiple sites irrespective of their phosphorylation state. 14-3-3 dimers regulate steady-state phosphorylation of both wild-type and the R406W mutant Tau, but they are not essential for toxicity of either variant. Moreover, 14-3-3 elevation itself is not pathogenic, but recruitment of dimers on accumulating wild-type Tau increases its steady-state levels ostensibly by occluding access to proteases in a phosphorylation-dependent manner. In contrast, the R406W mutant, which lacks a putative 14-3-3 binding site, responds differentially to elevation of each 14-3-3 isoform. Although excess 14-3-3ζ stabilizes the mutant protein, elevated D14-3-3ɛ has a destabilizing effect probably because of altered 14-3-3 dimer composition. Our collective data demonstrate the complexity of 14-3-3/Tau interactions in vivo and suggest that 14-3-3 attenuation is not appropriate ameliorative treatment of Tauopathies. Finally, we suggest that ‘bystander’ 14-3-3s are recruited by accumulating Tau with the consequences depending on the composition of available dimers within particular neurons and the Tau variant. Introduction Tau is distributed primarily in central nervous system (CNS) axons (1,2), where among other functions mediates microtubule dynamics, axonal transport and neuronal morphology (3–5). In vertebrates including humans, alternative splicing of a single transcript yields multiple protein isoforms. They differ in their amino-termini by exclusion (0N), or inclusion of 1 (1N) or 2 (2N) sequences and inclusion of 3 (3R) or 4 (4R), of the evolutionarily conserved repeats (R) near the carboxy-terminus which are typically involved in microtubule binding (6,7). Tau is implicated in several degenerative dementias collectively termed Tauopathies and other neuropathologies (7–14). Alzheimer’s disease (AD) is the most widespread sporadic aging-related neurodegenerative Tauopathy globally (15). Frontotemporal dementia with Parkinsonism on chromosome 17 (FTDP-17) is a distinct dementia, typically affecting younger individuals and involves mutations in the tau gene (16,17). Significantly, both Tauopathies present distinct clinical profiles and different, but overlapping histopathology (10,13,18). Although, non-pathogenic Tau is phosphorylated on multiple sites, (19,20), it becomes further phosphorylated especially onto characteristic ‘pathology-associated sites’ to a state of ‘hyper-phosphorylation’, which has been associated with AD and other Tauopathies (13,21,22). The different pathophysiological and clinical profiles of the various Tauopathies suggest that in addition to common interactors such as microtubules, normal and mutant Tau isoforms may also interact differentially with distinct proteins within affected neurons. To address this hypothesis, we performed systematic proteomic screens in Drosophila expressing Tau variants (Papanikolopoulou, K., Samiotaki, M., Panayotou, G. and Skoulakis, E.M.C. manuscript in preparation). Herein we report on two proteins from that screen, Leonardo (Leo), and D14-3-3ɛ (Eps), which appeared to interact with human wild-type (WT) and the FTDP-17-linked R406W Tau mutant (RW) in Drosophila brains. Our results confirm and expand previous reports on genes upregulated in human AD brains, one of which was 14-3-3ζ (23). The 14-3-3s are a family of small acidic proteins present in all eukaryotes with roles in essential biological processes such as cell proliferation and differentiation, migration and survival, intracellular transport, neurite outgrowth and ion channel regulation (24–27). Multiple 14-3-3s exist in metazoans, with seven distinct isoforms (β/α, ɛ, ζ, γ, η, θ and σ) in vertebrates, which except for σ, are primarily expressed in the brain, but also present in most other tissues (24,27,28). In contrast, Drosophila contain only two 14-3-3 genes, leonardo (leo), encoding three closely related orthologs of the vertebrate 14-3-3ζ and an ortholog of the ɛ isoform, D14-3-3ɛ (29–31). 14-3-3s homo and heterodimerize, forming a cup-shaped negatively charged groove, which interacts with client proteins containing RS/xxpS/TxP, or RxY/FxpS/TxP (pS/T phosphoserine/threonine) motifs (25,32–35). Leo homo and heterodimers with D14-3-3ɛ have been reported in Drosophila (30). 14-3-3 binding has been reported to modulate the structure of client proteins, to mask or reveal functional motifs that regulate its localization, phosphorylation state and stability (28,36–39). Given that Tau contains multiple phosphorylated Serines and Threonines, 14-3-3 binding may serve to modulate its steady-state phosphorylation by protecting from phosphatases and bridging interactions with kinases (40–42). Tau and 14-3-3s have been reported as macromolecular complexes (43–45), with at least one ubiquitous kinase, GSK-3β (46) in bovine brains or culture cell systems. Moreover, 14-3-3s may modulate the local conformation of Tau domains (47) and/or its subcellular localization. Accumulating evidence links 14-3-3s and Tauopathies. 14-3-3s have been associated with hyper-phosphorylated Tau aggregates from AD brains (41,48) and Pick’s disease patients are reported immunoreactive for 14-3-3s (49). Consistent with these observations, 14-3-3s have been reported to promote Tau aggregation (50,51). Furthermore, independent studies suggest increased expression of 14-3-3s in AD patients (52) and elevation in their cerebrospinal fluid (CSF) (53,54). However, whether 14-3-3 elevation in AD neurons and CSF is a collateral consequence of the disease, or important for its pathogenesis is unclear. Since all 14-3-3s other than σ are expressed in the brain, it is essential to determine whether there are isoform-specific effects on Tau levels, its hyper-phosphorylation and toxicity. Based on the identification of 14-3-3s as Tau interactors on our proteomic screen, we use the advantages of Drosophila to ask whether 14-3-3s affect toxicity of WT and mutant Tau in the nervous system and whether any such effects are 14-3-3 isoform-specific in vivo. Such potential differential interactions may underlie the heterogeneous, but often overlapping clinical profiles of Tauopathies. Results 14-3-3s associate with Tau in vivo To verify and validate the proteomic results, we aimed to determine independently whether endogenous 14-3-3s interact with human Tau proteins in the fly CNS. To limit such potential interactions to the adult CNS, we used the TARGET system (55). Flies carrying the wild-type 0N4R-Tau (henceforth WT) and the FTDP-17-linked mutant 0N4R-R406W (henceforth RW) were raised at 18°C, the temperature non-permissive for Tau expression. Upon emergence, they were induced to express the human Tau proteins paneuronally for 10 days, while age-matched flies that remained at 18°C throughout the experiment were used as controls (Fig. 1A, upper panels). WT and RW Tau immunoprecipitated from fly head extracts with the biotinylated antibody HT7, co-precipitated both Leo and D14-3-3ɛ (EPS) (Fig. 1A). Figure 1. View largeDownload slide 14-3-3s interact with Tau in vivo. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts (TARGET system). (A) (upper panel) A western blot demonstrating accumulation of the WT and RW Tau isoforms in the heads of adults flies. Uninduced (U) refers to lysates from animals that remained at 18°C, whereas Induced (I), denotes lysates from animals expressing the respective Tau after shifting them to 30°C for 10 days. Syntaxin (Syx) was used as loading control, and Tau was revealed with the T46 non-phosphorylation-dependent antibody. The lower panels are representative of results of immunoprecipitations from head lysates of animals induced for 10 days and their respective controls using the HT7 biotinylated anti-Tau antibody. The resulting immune complexes were immunoblotted for the presence of 14-3-3 isoforms as indicated. (B) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with head lysates from WT-expressing flies and the complexes were immunoblotted with anti-LEO and anti-D14-3-3ε (EPS) antibodies as indicated. (C) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with fly head extracts expressing the indicated Tau protein and immunoblotted with the T46 anti-Tau antibody. (D) Head lysates from flies expressing pan-neuronally 2N4R-Flag (WT) and 2N4R-RW-Flag (RW) Tau proteins were either dephosphorylated (+λ) or treated with phosphatase inhibitors (−λ). Lysates were subsequently incubated with Glutathione-agarose beads coated with GST-difopein to collect the 14-3-3 proteins and then immunoblotted with T46 anti-Tau antibody. (E) Co-immunoprecipitation of Leo and D14-3-3ε (EPS) with Tau from animals expressing 2N4R-Flag (WT) and RW-2N4R-Flag (RW) as indicated by the bars in panel D, pan-neuronally. Prior to incubation with anti-Flag beads to collect the Tau proteins, the lysates were treated either with λ phosphatase (+λ) or with phosphatase inhibitors (–λ) as indicated. Figure 1. View largeDownload slide 14-3-3s interact with Tau in vivo. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts (TARGET system). (A) (upper panel) A western blot demonstrating accumulation of the WT and RW Tau isoforms in the heads of adults flies. Uninduced (U) refers to lysates from animals that remained at 18°C, whereas Induced (I), denotes lysates from animals expressing the respective Tau after shifting them to 30°C for 10 days. Syntaxin (Syx) was used as loading control, and Tau was revealed with the T46 non-phosphorylation-dependent antibody. The lower panels are representative of results of immunoprecipitations from head lysates of animals induced for 10 days and their respective controls using the HT7 biotinylated anti-Tau antibody. The resulting immune complexes were immunoblotted for the presence of 14-3-3 isoforms as indicated. (B) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with head lysates from WT-expressing flies and the complexes were immunoblotted with anti-LEO and anti-D14-3-3ε (EPS) antibodies as indicated. (C) Glutathione-agarose beads coated with either GST or GST-difopein (GST-dif) were incubated with fly head extracts expressing the indicated Tau protein and immunoblotted with the T46 anti-Tau antibody. (D) Head lysates from flies expressing pan-neuronally 2N4R-Flag (WT) and 2N4R-RW-Flag (RW) Tau proteins were either dephosphorylated (+λ) or treated with phosphatase inhibitors (−λ). Lysates were subsequently incubated with Glutathione-agarose beads coated with GST-difopein to collect the 14-3-3 proteins and then immunoblotted with T46 anti-Tau antibody. (E) Co-immunoprecipitation of Leo and D14-3-3ε (EPS) with Tau from animals expressing 2N4R-Flag (WT) and RW-2N4R-Flag (RW) as indicated by the bars in panel D, pan-neuronally. Prior to incubation with anti-Flag beads to collect the Tau proteins, the lysates were treated either with λ phosphatase (+λ) or with phosphatase inhibitors (–λ) as indicated. To perform the complementary experiment requiring purification of endogenous 14-3-3s, we used GST-difopein-conjugated beads because this synthetic peptide binds 14-3-3s with high affinity (26,56). As shown in Figure 1B, endogenous Leo and D14-3-3ɛ from fly head lysates were indeed bound on GST-difopein, but not on plain GST beads. Significantly, difopein-bound 14-3-3s from these lysates co-purified with either WT or RW (Fig. 1C). Collectively, these results validate the proteomic data and indicate that endogenous 14-3-3s directly or indirectly interact with Tau and are consistent with reports suggesting their co-localization in aggregates from Tauopathy patient brains (53,57). 14-3-3s bind phosphorylated Serines or Threonines (pSer/pThr) and are reported to modulate the phosphorylation state of their clients by occluding or facilitating access to kinases and phosphatases (38,39,58). Given the multiple pSer/pThr on Tau (59), we wondered whether its phosphorylation is required for the interaction with Leo and D14-3-3ɛ. Because the 0N4R isoforms were not available as Flag-Tagged constructs to allow the experiment we used the Flag-Tagged 2N4R and Flag-Tagged RW mutation in the 2N4R isoform. We reasoned that the aminoterminnal extension will not affect access of the phosphatase and in general the 0N4R and 2N4R isoforms behave very similarly in our assays (60,61). We treated head lysates from animals expressing the Flag-tagged 2N4R or Flag-tagged RW mutant in the 2N4R genetic background with λ-phosphatase. Although we cannot be certain that all phosphates were abolished, the majority of phosphates were removed as indicated by the shift in electrophoretic mobility (Fig. 1D), or with specific antibodies (Supplementary Material, Fig. S1). Treated and untreated lysates were mixed with GST-difopein-conjugated beads to collect 14-3-3s, the complexes resolved in acrylamide gels and probed for the presence of Tau in the pulled-down material. The results clearly demonstrate that difopein-bound 14-3-3s co-precipitated both phosphorylated and dephosphorylated Tau (Fig. 1D). Conversely, immunoprecipitation of transgenic Tau proteins via their Flag tag, confirmed co-precipitation of Leo and D14-3-3ɛ from treated and untreated lysates (Fig. 1E). Therefore, 14-3-3s bind dephosphorylated Tau indicating that they do not dock only on its phosphor-Serines and phosphor-Threonines. As predicted by these results, full-length Tau contains five putative 14-3-3 consensus binding sites involving potentially phosphorylated Ser and Thr (Fig. 2). However, as indicated by the results in Figure 1E, additional binding sites not involving phosphorylatable residues must exist and these were not apparent in our in silico search. The sites involving Ser198 and Ser214 are in the projection domain thought to interact with various cellular proteins other than microtubules (62). Ser214 is known to be phosphorylated (63) and has been reported critical for interaction with 14-3-3 and a mediator of Tau aggregation in vitro (64). A consensus binding site not reported before is anchored by Thr245 in the first repeat (R1), which has been reported constitutively phosphorylated on WT (20). An additional site in R4 precedes the Par1 kinase-targeted Ser356 (65). Finally, a putative 14-3-3 binding site is in the carboxy-terminal domain involving Ser413. Interestingly, Arg406 is the mandatory consensus-anchoring amino-acid for 14-3-3 binding to pSer413 and is mutated to Trp in R406W suggesting that this site is eliminated in the R406W mutant Tau. Figure 2. View largeDownload slide Proposed in silico 14-3-3 binding sites on WT Tau. The results of searching the full-length WT Tau sequence with all modes of the 14-3-3 binding consensus. Numbering of the amino-acids is based on the full-length 441 residue 2N4R WT Tau. Amino acids mandatory for 14-3-3 binding are shown in bold, while the potentially phosphorylated Ser or Thr are denoted by the numbered bold residues as reported previously [20]. Figure 2. View largeDownload slide Proposed in silico 14-3-3 binding sites on WT Tau. The results of searching the full-length WT Tau sequence with all modes of the 14-3-3 binding consensus. Numbering of the amino-acids is based on the full-length 441 residue 2N4R WT Tau. Amino acids mandatory for 14-3-3 binding are shown in bold, while the potentially phosphorylated Ser or Thr are denoted by the numbered bold residues as reported previously [20]. 14-3-3 dimers modulate Tau phosphorylation To determine whether 14-3-3s interactions may contribute to Tau phosphorylation, aggregation and toxicity, we attenuated by RNA-mediated interference (RNAi), each isoform in the CNS of flies also expressing WT. The available RNAi-mediating transgenes completely abrogated Leo levels (Fig. 3A), whereas D14-3-3ɛ was reduced by about 50% (Fig. 3B). Significantly, D14-3-3ɛ reduction was accompanied by a 50% Leo elevation (Fig. 3A) and a similar elevation of D14-3-3ɛ was observed in animals lacking Leo (Fig. 3B). This is congruent with prior reports of 14-3-3 homeostasis in D14-3-3ɛ mutant embryos (29), leo mutants and transgenic adults (30). Therefore, these genetic manipulations generate animals that lack Leo but harbor excess D14-3-3ɛ (Leo-), animals with nearly half of D14-3-3ɛ, but excess Leo (Eps-) and flies lacking Leo and half of D14-3-3ɛ (Leo-, Eps-). Figure 3. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on WT Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or least square means (LSM) contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing WT Tau probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (WTLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (WTEPS-) and also nearly eliminated (Dunnett’s, P < 0.0001, n = 3), when both isoforms were reduced concurrently (WTLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing WT Tau probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was significantly (Dunnett’s, P < 0.0001, n = 7) elevated (WTLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 10) upon D14-3-3ε attenuation and was also reduced (Dunnett’s, p = 0.0008, n = 5) when both isoforms were attenuated concurrently. (C) Total levels of WT in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences (ANOVA F(3, 27) = 0.2711, P = 0.8457) in WT levels (n > 6). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A reduction (Dunnett’s, P = 0.031, n = 10) in AT8 occupation was revealed only in animals with both isoforms reduced. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.029, n = 5) was observed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0017, n = 5) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0005, n = 6) in animals with both isoforms reduced. (G) Representative blots (n = 3) of 14-3-3 dimer formation investigation upon addition of cross-linker (+ linker) in adult head lysates of the indicated genotypes (bottom). Mirror blots were probed either with the anti-Leo (top) or the anti-D14-3-3ε (bottom) antibodies. The monomers (mono) and dimers are denoted. (H) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or WT Tau (WT) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing WT (solid bars) is shown. ANOVA indicated significant differences (F(6, 89) = 14.6695; P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus WT P < 0.0157; LEO- versus WT LEO- P < 0.0058; EPS- versus WT EPS- P <0.0389; LEO- EPS- versus WT LEO- EPS P < 0.0272. Figure 3. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on WT Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or least square means (LSM) contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing WT Tau probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (WTLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (WTEPS-) and also nearly eliminated (Dunnett’s, P < 0.0001, n = 3), when both isoforms were reduced concurrently (WTLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing WT Tau probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was significantly (Dunnett’s, P < 0.0001, n = 7) elevated (WTLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 10) upon D14-3-3ε attenuation and was also reduced (Dunnett’s, p = 0.0008, n = 5) when both isoforms were attenuated concurrently. (C) Total levels of WT in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences (ANOVA F(3, 27) = 0.2711, P = 0.8457) in WT levels (n > 6). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A reduction (Dunnett’s, P = 0.031, n = 10) in AT8 occupation was revealed only in animals with both isoforms reduced. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.029, n = 5) was observed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0017, n = 5) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in WTLEO- animals and a significant decrease in (Dunnett’s, P = 0.0005, n = 6) in animals with both isoforms reduced. (G) Representative blots (n = 3) of 14-3-3 dimer formation investigation upon addition of cross-linker (+ linker) in adult head lysates of the indicated genotypes (bottom). Mirror blots were probed either with the anti-Leo (top) or the anti-D14-3-3ε (bottom) antibodies. The monomers (mono) and dimers are denoted. (H) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or WT Tau (WT) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing WT (solid bars) is shown. ANOVA indicated significant differences (F(6, 89) = 14.6695; P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus WT P < 0.0157; LEO- versus WT LEO- P < 0.0058; EPS- versus WT EPS- P <0.0389; LEO- EPS- versus WT LEO- EPS P < 0.0272. Because 14-3-3s are dimeric (25), we determined the effect of Leo and D14-3-3ɛ abrogation on dimer levels. Leo and D14-3-3ɛ dimers were apparent in control animals as expected (Fig. 3G). Leo-containing dimers were not apparent in Leo- animals despite the traces of monomer due to the large amounts of extract needed in cross-linking experiments. Surprisingly, abundant dimers were revealed with the D14-3-3ɛ antibody in these Leo- animals (Fig. 3G). This indicates that upon Leo loss D14-3-3ɛ homodimers form efficiently judging by their abundance. In confirmation, these dimers are also present in Leo-Eps- animals, albeit apparently reduced. In contrast, in flies with a 50% attenuation of D14-3-3ɛ (Eps-), we assume that the dimers are largely heterodimers with Leo or Leo homodimers (Fig. 3G). Interestingly, attenuation of either 14-3-3 alone or simultaneously, did not appear to affect WT levels (Fig. 3C). This indicates that either Leo nor D14-3-3ɛ are essential for steady-state WT levels in the fly CNS, or that the residual dimers in Leo-Eps- animals suffice to maintain Tau levels elevated. To determine whether 14-3-3 attenuation affects Tau phosphorylation, we assessed occupation of the pathology-associated sites AT8, AT100 and pSer262. Phosphorylation on Ser202/Thr205 recognized by the AT8 antibody appeared unchanged upon attenuation of either 14-3-3 protein alone, but was reduced upon concurrent attenuation of both Leo and D14-3-3ɛ (Fig. 3D). Because Leo reduction is accompanied by D14-3-3ɛ upregulation and vice versa (Fig. 3A and B) and in the doubly attenuated animals dimers levels are reduced, the results suggest that both isoforms contribute positively to AT8 occupation. In contrast, phosphorylation at the AT100 site, which includes the 14-3-3 binding pSer214 (Fig. 2), was elevated upon Leo abrogation (Fig. 3E). This suggests either that Leo homodimers or heterodimers occlude access to that site, or they promote its de-phosphorylation. Alternatively, D14-3-3ɛ excess or homodimers upon Leo loss may promote phosphorylation at AT100, or protect it from de-phosphorylation. Interestingly, AT100 occupation was not significantly altered upon D14-3-3ɛ attenuation and the concomitant Leo elevation, a situation that most likely enhances Leo heterodimer and homodimer levels. This indicates that Leo homodimer excess in Eps- flies does not occlude or inhibit AT100 occupation and therefore, increased AT100 phosphorylation in Leo- flies is likely the effect of excess D14-3-3ɛ. Accordingly, AT100 phosphorylation was significantly reduced (Fig. 3E) in Leo-, Eps- flies, ostensibly because the remaining D14-3-3ɛ is not sufficient to support AT100 occupation. Results were similar for pSer262 (Fig. 3F), which is not a 14-3-3 consensus site, but its occupation is regulated by the phosphorylation status of Ser238 and Thr245 (66), the latter of which appears to be (Fig. 2). Phosphorylation of Ser262 was also highly enhanced upon Leo abrogation, albeit even more than at AT100 and eliminated upon attenuation of both 14-3-3s (Fig. 3F). These results also indicate that Leo/D14-3-3ɛ heterodimers are essential for the steady-state phosphorylation of Ser262 and excess D14-3-3ɛ prevents dephosphorylation of the site. A parsimonious explanation of these results is a highly dynamic interaction of 14-3-3 homo and heterodimers modulating the balance between kinase and phosphatase access to the phosphosites. Leo homodimers appear to facilitate phosphatase access, which is likely suppressed by D14-3-3ɛ dimers in Leo- flies. In contrast, heterodimer binding in Eps- flies does not promote phosphatase access evidenced by the normal occupation levels. In Leo- Eps- flies where the stabilizing excess of D14-3-3ɛ is absent AT100 and pSer262 occupation is highly reduced. Do the effects of 14-3-3 attenuation on WT phosphorylation affect its toxicity? To address this, we quantified the pre-eclosion lethality associated with pan-neuronal WT expression alone or upon 14-3-3 attenuation. Indeed, as illustrated in Figure 3H, WT expression reduced the number of animals reaching adulthood relative to controls. Neither loss of Leo, nor D14-3-3ɛ attenuation alone affected viability (Fig. 3H, open bars) and they did not affect WT-dependent lethality (Fig. 3H, filled bars). However, reduction of both 14-3-3s (Leo- Eps-), decreased viability nearly 50% on its own, in accordance with prior reports (29) and this was further reduced upon WT expression (Fig. 3H, last two bars). Note that the difference in viability between animals with attenuated 14-3-3s alone and those also co-expressing WT remained relatively constant (the difference between open and filled bars in Fig. 3H). This indicates that attenuation of either 14-3-3 does not potentiate WT toxicity significantly. Rather the most significant effect on lethality was precipitated by the concomitant attenuation of both 14-3-3s, which decrease the total levels of 14-3-3 dimers. Interestingly the effects of 14-3-3 reduction (Fig. 4A and B) on the steady-state levels of RW and its phosphorylation at AT8, AT100 and pSer262 were similar with those for WT Tau (Fig. 4C–F). Lethality associated with RW expression was more pronounced than that of WT and reached nearly 50% of expected progeny (Fig. 4G). Again, attenuation of either 14-3-3 alone did not alter RW toxicity significantly, but concurrent reduction of both 14-3-3s resulted in 50% lethality on its own and this was mildly augmented by RW expression. Figure 4. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on R406W mutant Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing RW probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (RWLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (RWEPS-) and also eliminated (Dunnett’s, P < 0.0001, n = 6) when both isoforms were reduced concurrently (RWLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing RW probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was elevated (Dunnett’s, P < 0.001, n = 6) elevated (RWLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 6) upon D14-3-3ε attenuation and also reduced (Dunnett’s, P = 0.0022, n = 6) when both isoforms were reduced concurrently. (C) Total levels of RW in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences in Tau levels (n > 4). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. AT8 occupation was significantly reduced (Dunnett’s, P < 0.001, n = 9) only in animals with both isoforms attenuated. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.009, n = 4) was observed in RWLEO- animals and a significant decrease in (Dunnett’s, P = 0.0006, n = 4) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in RWLEO- animals and a decrease in (Dunnett’s, P = 0.0012, n = 6) in animals with both isoforms reduced. (G) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or RW Tau (RW) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing RW (solid bars) is shown. ANOVA indicated significant differences (F(6, 90) = 33.8151, P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus RW P < 0.0001; LEO- versus RW LEO- P < P < 0.0001; EPS- versus RW EPS- P < P < 0.0001; LEO- EPS- versus RW LEO- EPS- P < 0.0001. Figure 4. View largeDownload slide 14-3-3 attenuation affects steady-state phosphorylation of selected sites on R406W mutant Tau. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. Representative blots of head lysates are shown probed with the indicated antibodies. For the quantifications below, levels of the protein or phosphorylation site occupation indicated were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing WT alone, which is set to 1. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences from the controls (open bars) indicated by stars. (A) Leo, D14-3-3ε and both isoform attenuation in animals expressing RW probed for Leo. Compared to its levels in control animals Leo was significantly (Dunnett’s, P < 0.0001, n = 6) abrogated (RWLEO-), significantly (Dunnett’s, P < 0.0001, n = 6) elevated upon D14-3-3ε attenuation (RWEPS-) and also eliminated (Dunnett’s, P < 0.0001, n = 6) when both isoforms were reduced concurrently (RWLEO-EPS-). (B) D14-3-3ε, Leo and both isoform attenuation in animals expressing RW probed for D14-3-3ε. Compared to its levels in control animals D14-3-3ε was elevated (Dunnett’s, P < 0.001, n = 6) elevated (RWLEO-), significantly reduced (Dunnett’s, P < 0.0001, n = 6) upon D14-3-3ε attenuation and also reduced (Dunnett’s, P = 0.0022, n = 6) when both isoforms were reduced concurrently. (C) Total levels of RW in the indicated genotypes revealed with the T46 antibody. Quantification did not reveal significant differences in Tau levels (n > 4). (D) AT8 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. AT8 occupation was significantly reduced (Dunnett’s, P < 0.001, n = 9) only in animals with both isoforms attenuated. (E) AT100 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. An increase (Dunnett’s, P = 0.009, n = 4) was observed in RWLEO- animals and a significant decrease in (Dunnett’s, P = 0.0006, n = 4) in animals with both isoforms reduced. (F) pSer262 occupation levels in animals with attenuated Leo, D14-3-3ε and both isoforms. A significant elevation (Dunnett’s, P < 0.0001, n = 6) was revealed in RWLEO- animals and a decrease in (Dunnett’s, P = 0.0012, n = 6) in animals with both isoforms reduced. (G) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or RW Tau (RW) and flies bearing Leo (LEO-) or D14-3-3ε (EPS-) or both (LEO-, EPS-) abrogating transgenes alone (open bars) or co-expressing RW (solid bars) is shown. ANOVA indicated significant differences (F(6, 90) = 33.8151, P < 0.0001), which were resolved by LSM contrast analysis as follows: CN versus RW P < 0.0001; LEO- versus RW LEO- P < P < 0.0001; EPS- versus RW EPS- P < P < 0.0001; LEO- EPS- versus RW LEO- EPS- P < 0.0001. Collectively, these results indicate dynamic interactions of 14-3-3 homo and heterodimers in the regulation of steady-state phosphorylation of both WT and RW and appear to have a minor effect on the toxicity of either Tau variant. Therefore, 14-3-3 attenuation does not synergize with Tau variants to augment their toxicity. Pharmacological inhibition of Tau/14-3-3 interactions 14-3-3 interactions with WT and RW could have been obscured because the genetic manipulations did not result in adult animals completely devoid of both 14-3-3s due to incomplete attenuation or homeostatic compensation (Figs 3A, B and 4A, B). This is further complicated by the partial functional overlap of 14-3-3s in vivo (29). To address and potentially augment this issue, we used BV02, a 14-3-3 antagonist reported to disrupt their biological activity in cultured cell systems (67). BV02 is a cell permeable non-peptide 14-3-3 antagonist which targets their amphipathic binding groove and disrupts interactions with client proteins (68). Because concurrent attenuation of both 14-3-3s resulted in elevated lethality, we titrated the concentration of the inhibitor such as to afford experimental resolution in combination with Tau expression. Control experiments indicated that this requirement is fulfilled at 0.15 nm throughout development, whereas at higher concentrations lethality was extensive. As suggested (67), at 0.15 nm and even at the lethal 0.30 nm concentration, the inhibitor did not affect the steady-state levels of 14-3-3 isoforms or Tau variants (Fig. 5A–D). Figure 5. View largeDownload slide Pharmacological inhibition of 14-3-3 dimer–client interaction. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) Representative blots probing the levels of LEO relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nM) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 36) = 0.6032, P = 0.5528). (B) Representative blots probing the levels of D14-3-3ɛ (EPS) relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nm) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 47) = 0.6833, P = 0.5101). (C) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 1.4411, P = 0.2552). (D) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 0.4593, P = 0.5120). (E) Representative blots probing the levels of AT100 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 10) = 21.5991, P < 0.0012). (F) Representative blots probing the levels of pS262 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 8) = 13.3574, P < 0.0081). (G) Representative blots probing the levels of AT100 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 11) = 45.7888, P < 0.0001). (H) Representative blots probing the levels of pS262 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 7) = 16.6753, P < 0.0065). (I) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CN) or WT (TAU) in media with and without 0.15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or TAU expressing animals (ANOVA F(1, 31) = 1.6096, P = 0.2143). (J) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with WT (TAU, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), or TAU, LEO- animals (ANOVA F(1, 15) = 0.2645, p = 0.6150). (K) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with WT (TAU, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for TAU, EPS- animals (ANOVA F(1, 17) = 27.0254, P < 0.0001). (L) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htau0N4R, leo RNAi recombinant (TAU, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of TAU, LEO-, EPS- flies (ANOVA F(1, 29) = 48.9572, P < 0.0001). (M) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or R406W (RW) in media with and without 15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or RW expressing animals (ANOVA F(1, 33) = 2.5017, P = 0.1236). (N) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with R406W (RW, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), but had a highly significant effects on teh viability of RW, LEO- animals (ANOVA F(1, 17) = 53.2494, P < 0.0001). (O) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with RW (RW, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for RW, EPS- animals (ANOVA F(1, 21) = 35.3185, P < 0.0001). (P) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htauR406W, leo RNAi recombinant (RW, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of RW, LEO-, EPS- flies (ANOVA F(1, 29) = 74.9826, P < 0.0001). Figure 5. View largeDownload slide Pharmacological inhibition of 14-3-3 dimer–client interaction. Human WT and RW Tau variants were expressed specifically in adult CNS under ElavGal4; Gal80ts. All quantifications were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) Representative blots probing the levels of LEO relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nM) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 36) = 0.6032, P = 0.5528). (B) Representative blots probing the levels of D14-3-3ɛ (EPS) relative to Syntaxin used as a loading control in flies raised in media with and without the indicated concentration (in nm) of BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(2, 47) = 0.6833, P = 0.5101). (C) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 1.4411, P = 0.2552). (D) Representative blots probing the levels of WT relative to Syntaxin used as a loading control in flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). Significant differences were not unveiled (ANOVA F(1, 12) = 0.4593, P = 0.5120). (E) Representative blots probing the levels of AT100 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 10) = 21.5991, P < 0.0012). (F) Representative blots probing the levels of pS262 relative to Syntaxin in WT-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 8) = 13.3574, P < 0.0081). (G) Representative blots probing the levels of AT100 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 11) = 45.7888, P < 0.0001). (H) Representative blots probing the levels of pS262 relative to Syntaxin in RW-expressing flies raised in media with and without 0.15 nm BV02. For the quantifications below, levels of the two proteins in untreated animals were set to 1 (open bars). The difference is significant (ANOVA F(1, 7) = 16.6753, P < 0.0065). (I) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CN) or WT (TAU) in media with and without 0.15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or TAU expressing animals (ANOVA F(1, 31) = 1.6096, P = 0.2143). (J) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with WT (TAU, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), or TAU, LEO- animals (ANOVA F(1, 15) = 0.2645, p = 0.6150). (K) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with WT (TAU, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for TAU, EPS- animals (ANOVA F(1, 17) = 27.0254, P < 0.0001). (L) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htau0N4R, leo RNAi recombinant (TAU, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of TAU, LEO-, EPS- flies (ANOVA F(1, 29) = 48.9572, P < 0.0001). (M) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) or R406W (RW) in media with and without 15 nm BV02. The drug did not affect viability of controls (ANOVA F(1, 33) = 0.0011, P = 0.9735), or RW expressing animals (ANOVA F(1, 33) = 2.5017, P = 0.1236). (N) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; leo RNAi/CyO (LEO-), or after crossing them with R406W (RW, LEO-) in media with and without 0.15 nm BV02. The drug did not affect viability of LEO- (ANOVA F(1, 19) = 0.0492, P = 0.8270), but had a highly significant effects on teh viability of RW, LEO- animals (ANOVA F(1, 17) = 53.2494, P < 0.0001). (O) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb (EPS-), or after crossing them with RW (RW, EPS-) in media with and without 0.15 nm BV02. The drug did not affect viability of EPS- (ANOVA F(1, 37) = 2.0118, P = 0.1647), but it did for RW, EPS- animals (ANOVA F(1, 21) = 35.3185, P < 0.0001). (P) The mean number of non-balancer-bearing progeny per female ± S.E.M. from ElavGal4; epsRNAi/TM3Sb crossed to leo RNAi/CyO (LEO-, EPS-), or to a UAS-htauR406W, leo RNAi recombinant (RW, LEO-, EPS-) in media with and without 0.15 nm BV02. The drug had a significant effect on LEO-, EPS- (ANOVA F(1, 25) = 49.6475, P < 0.0001), and on the viability of RW, LEO-, EPS- flies (ANOVA F(1, 29) = 74.9826, P < 0.0001). We expected that the inhibitor may decrease phosphorylation of WT and RW at the sites assayed, essentially mimicking the observation in Leo-, Eps- animals (Figs 3D–F and 4D–F). However, BV02-mediated inhibition of 14-3-3s resulted in elevated AT100 and Ser262 phosphorylation on WT (Fig. 5E and F) and RW (Fig. 5G and H). These results confirm independently that Leo homodimers and heterodimers suppress occupation of the sites possibly by promoting phosphatase activity. This appears counteracted by 14-3-3ɛ dimers, which promote phosphorylation at these sites, evident in Leo- animals and upon inhibitor application. Furthermore, the results suggest that in Leo-Eps- animals, loss of phosphorylation at these sites may be a consequence of the large reduction in dimers (Fig. 3G), in addition to the large attenuation in D14-3-3 levels. Therefore, these data in agreement with those of genetic abrogation of 14-3-3s (Figs 3 and 4) strongly suggest complex regulation of Tau phosphorylation by 14-3-3 dimers. To determine the effects of pharmacological inhibition of 14-3-3 on WT and RW toxicity, we used 0.15 nm BV02 to avoid the lethality associated with higher concentrations. At 0.15 nm, the inhibitor did not significantly enhance the lethality of controls, WT-expressing (Fig. 5I), Leo- alone, or Leo-, WT-expressing flies (Fig. 5J). Because the inhibitor did not enhance the lethality of Leo- animals, genetic abrogation of Leo appears complete, in agreement with the western blot results (Figs 3A and 4A). This confirms independently that Leo is not required for WT toxicity. Inhibitor-mediated lethality was enhanced significantly if D14-3-3ɛ was attenuated in WT-expressing animals (Fig. 5K, TAU Eps-), consistent with the interpretation that D14-3-3ɛ heterodimers with the elevated Leo (Figs 3B and 4B) to account for the minor viability effects of WT, Eps- animals (Fig. 3H). However, inhibition of these heterodimers with BV02 precipitated the significant reduction in WT, Eps- viability (Fig. 5K). Consistent with this notion, treatment of already low viability Leo-Eps- flies with the inhibitor nearly eliminated them whether co-expressing WT, or not (Fig. 5L). Broadly, similar effects of the inhibitor were observed in flies expressing RW, except that as described earlier (Fig. 4G), the mutant protein was significantly more toxic on its own compared with WT (Fig. 5I versus Fig. 5M). Significantly however, in contrast to WT-expressing animals, BV02 treatment of RW-expressing Leo- flies precipitated nearly complete lethality (Fig. 5N). Since the inhibitor has no effect on Leo- animals, this is an effect precipitated specifically by the presence of RW. This suggests that D14-3-3ɛ dimers, abundant in Leo- animals, suppress RW toxicity and upon their inhibition with BV02 result in nearly complete lethality. It appears that Leo homodimers or heterodimers can compensate in part for this because although lethality of RW, Eps- was enhanced by the inhibitor (Fig. 5O), it was not as complete as in RW, Leo- (Fig. 5N). In agreement, strongly enhanced toxicity was apparent if both 14-3-3s were attenuated in RW-expressing animals and treated with BV02 (Fig. 5P). Therefore, D14-3-3ɛ differentially suppresses the toxicity of RW, an effect not apparent with WT. This differential interaction with the normal and mutant Tau may be mediated by the R406W mutation itself, possibly because it effects an apparent 14-3-3 binding site. 14-3-3 dosage affects WT-Tau levels Because 14-3-3s have been reported elevated in the brain of Tauopathy patients (53), we investigated whether increased Leo and D14-3-3ɛ levels affect phosphorylation, steady-state levels and the toxicity of WT (Fig. 6A). Of the three Leo isoforms, we used a transgene for the paneuronally expressed LeoII (31) throughout this work. Figure 6. View largeDownload slide 14-3-3 isoform-specific effects on WT Tau toxicity, but equivalent effects on its steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htau0N4R males (TAU). The difference is significantly different (ANOVA F(1, 21) = 163.16; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-WT expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared to the mean ratio of flies expressing both WT and either of the 14-3-3 transgenes (WTLEO, WTEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a significant enhancement of Leo-overexpression associated lethality by WT co-expression (ANOVA F(1, 20) = 69.12; P < 0.0001, n ≥ 10). Similarly, WT co-expression yielded significant lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 18) = 16.59; P < 0.0007, n > 9). (C–F) Lysates from the heads of flies expressing pan-neuronally WT alone, or also co-expressing the indicated Drosophila 14-3-3s were subjected to western blotting with the anti-total Tau (T14), or the indicated anti-phospho-Tau antibodies. The blots represent one of 3–4 independent experiments. For the quantification shown below each representative blot, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which was set to 1. The significance of potential differences was determined with Dunnett’s tests following ANOVA. The levels of WT Tau were significantly different (P < 0.01) than those in lysates also over-expressing Leo or D14-3-3ε (G) and remained significantly different (P < 0.05) when probed with the indicated phosphoantibodies (H–J). Figure 6. View largeDownload slide 14-3-3 isoform-specific effects on WT Tau toxicity, but equivalent effects on its steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s or LSM contrast analysis were applied to reveal significant differences. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htau0N4R males (TAU). The difference is significantly different (ANOVA F(1, 21) = 163.16; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-WT expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared to the mean ratio of flies expressing both WT and either of the 14-3-3 transgenes (WTLEO, WTEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a significant enhancement of Leo-overexpression associated lethality by WT co-expression (ANOVA F(1, 20) = 69.12; P < 0.0001, n ≥ 10). Similarly, WT co-expression yielded significant lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 18) = 16.59; P < 0.0007, n > 9). (C–F) Lysates from the heads of flies expressing pan-neuronally WT alone, or also co-expressing the indicated Drosophila 14-3-3s were subjected to western blotting with the anti-total Tau (T14), or the indicated anti-phospho-Tau antibodies. The blots represent one of 3–4 independent experiments. For the quantification shown below each representative blot, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which was set to 1. The significance of potential differences was determined with Dunnett’s tests following ANOVA. The levels of WT Tau were significantly different (P < 0.01) than those in lysates also over-expressing Leo or D14-3-3ε (G) and remained significantly different (P < 0.05) when probed with the indicated phosphoantibodies (H–J). Surprisingly, paneuronal overexpression of LeoII precipitated 30% lethality (Fig. 6B), but significantly, its co-expression with WT Tau increased lethality to nearly 80% (Fig. 6B). Hence excess Leo appears toxic on its own and this may contribute to pathology independently of that caused by Tau. Nevertheless, Leo elevation potentiates Tau toxicity and pathology. Interestingly, overexpression of a D14-3-3ɛ transgene did not precipitate significant lethality on its own and did not alter WT lethality as significantly as upon Leo elevation (Fig. 6B). Hence, 14-3-3 excess is not generally toxic, but rather it characterizes the Leo isoform. Interestingly, overexpression of either Leo or D14-3-3ɛ, did not result in significant compensatory elevation or decrease of the complementary isoform (Supplementary Material, Fig. S2A and B). Significantly, WT levels were nearly doubled upon co-expression with either Leo or D14-3-3ɛ (Fig. 6C). This was also apparent when the AT8, AT100 and pSer262 sites were surveyed (Fig. 6D–F). Because qualitatively and quantitatively similar results were obtained for all phosphorylation sites tested, elevated 14-3-3s do not appear to affect their occupation, but rather increase the steady-state levels of WT itself. However, although WT levels increased equivalently by over-expression of either Leo or D14-3-3ɛ (Fig. 6C), WT lethality was highly potentiated by Leo and much less by D14-3-3ɛ (Fig. 6B), indicating 14-3-3 isoform-specific effects on WT toxicity, but not on its levels. Differential effects of 14-3-3 excess on the R406W mutant Tau Do the effects of 14-3-3 elevation on WT generalize to the RW mutant? The nearly 50% lethality upon paneuronal expression of RW alone (Fig. 7A), was potentiated to over 95% upon co-expression with Leo (Fig. 7B). In contrast, D14-3-3ɛ elevation did not affect lethality on its own and its effects combined with R406W were marginal (Fig. 7B). Therefore, the effects of 14-3-3 excess on WT and RW toxicity are broadly similar, with Leo elevation enhancing it for both, but much more severely for the mutant Tau. Figure 7. View largeDownload slide 14-3-3 isoform-specific effects on R406W toxicity and steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htauR406W males (RW). The difference is significantly different (ANOVA F(1, 20) = 39.06; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-RW expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared with the mean ratio of flies expressing both RW and either of the 14-3-3 transgenes (RWLEO, RWEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a highly significant enhancement of Leo-overexpression associated lethality by RW co-expression (ANOVA F(1, 19) = 128.66, P < 0.0001, n ≥ 9). However, RW co-expression yielded marginal lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 16) = 4.87, P = 0.0422, n > 8). (C) Representative blot of lysates from the heads of pooled escaper flies expressing pan-neuronally under ElavGal4, RW alone or also co-expressing the indicated Drosophila 14-3-3s probed with anti-total Tau (T14). For the quantification shown below, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which is set to 1. RW levels in animals also co-expressing D14-3-3ε were significantly different than those expressing RW alone (Dunnett’s, P < 0.01, n = 4), but not those co-expressing Leo. (D–F) Representative blots and their respective quantifications from single flies expressing R406W alone, or co-overexpressing 14-3-3s. RW levels were quantified after normalization to Syntaxin (Syx) and significant differences (Dunnett’s, P < 0.01, n = 4 for all) in mean levels of RW alone which was set as 1, from its levels in the presence of excess Leo and D14-3-3ε. (G–I) Representative western blots and their respective quantifications from flies expressing RW alone, or also co-overexpressing 14-3-3s in adults only under Elav; Gal80ts at 25°C. RW levels were quantified after normalization to Syntaxin (Syx) and set to 1. Significant differences (Dunnett’s, P < 0.01, n = 3) in mean levels of RW alone versus its levels in the presence of excess Leo and D14-3-3ε. Figure 7. View largeDownload slide 14-3-3 isoform-specific effects on R406W toxicity and steady-state levels. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. (A) The mean number of progeny per female ± S.E.M. from crossing ElavGal4 to w1118 males (CNT) is shown, relative to the mean number of progeny per female ± S.E.M. of crossing ElavGal4 to UAS-htauR406W males (RW). The difference is significantly different (ANOVA F(1, 20) = 39.06; P < 0.0001, n > 10). (B) The mean number of observed/expected progeny ± S.E.M. for non-RW expressing Leo (LEO), or D14-3-3ε (EPS) overexpressing animals is compared with the mean ratio of flies expressing both RW and either of the 14-3-3 transgenes (RWLEO, RWEPS). Note that the means are not normalized to 100% to accommodate the lethality associated with overexpression of the 14-3-3 transgenes alone. There is a highly significant enhancement of Leo-overexpression associated lethality by RW co-expression (ANOVA F(1, 19) = 128.66, P < 0.0001, n ≥ 9). However, RW co-expression yielded marginal lethality over that of animals expressing D14-3-3ε alone (ANOVA F(1, 16) = 4.87, P = 0.0422, n > 8). (C) Representative blot of lysates from the heads of pooled escaper flies expressing pan-neuronally under ElavGal4, RW alone or also co-expressing the indicated Drosophila 14-3-3s probed with anti-total Tau (T14). For the quantification shown below, Tau levels were normalized using the Syntaxin (Syx) loading control and shown as a ratio of their means ± S.E.M. relative to their respective levels in fly heads expressing Tau alone, which is set to 1. RW levels in animals also co-expressing D14-3-3ε were significantly different than those expressing RW alone (Dunnett’s, P < 0.01, n = 4), but not those co-expressing Leo. (D–F) Representative blots and their respective quantifications from single flies expressing R406W alone, or co-overexpressing 14-3-3s. RW levels were quantified after normalization to Syntaxin (Syx) and significant differences (Dunnett’s, P < 0.01, n = 4 for all) in mean levels of RW alone which was set as 1, from its levels in the presence of excess Leo and D14-3-3ε. (G–I) Representative western blots and their respective quantifications from flies expressing RW alone, or also co-overexpressing 14-3-3s in adults only under Elav; Gal80ts at 25°C. RW levels were quantified after normalization to Syntaxin (Syx) and set to 1. Significant differences (Dunnett’s, P < 0.01, n = 3) in mean levels of RW alone versus its levels in the presence of excess Leo and D14-3-3ε. Is the potentiation of RW toxicity reflected in the levels of the protein in the CNS? Because most animals with elevated Leo died as late pupae/pharate adults we pooled lysates from the few escapers, but surprisingly RW levels appeared unchanged upon Leo co-expression and equally surprisingly, significantly reduced upon elevation of D14-3-3ɛ (Fig. 7C). We noticed, however, that when using lysates from individual flies, in approximately 20% of Leo over-expressing animals, RW levels were nearly doubled (Fig. 7D), which agrees with the viability data. In contrast, RW levels were suppressed upon D14-3-3ɛ over-expression, whether pooled (Fig. 7C), or in single flies (Fig. 7D). Phosphorylation at AT8, AT100 were also correspondingly elevated or reduced (Fig. 7E and F) in these escapers. Brain lysates from dissected pharate adults that failed to eclose, indicated similarly elevated and reduced RW upon Leo and D14-3-3ɛ co-expression respectively (Supplementary Material, Fig. S3A). Therefore, RW levels in most animals failing to eclose are represented by those in 20% of the escapers (Fig. 7D). Collectively then, high levels of RW correlate well with the robust lethality of Leo co-expressing animals. In contrast, the drastically reduced RW levels upon D14-3-3ɛ elevation correlate well with the lack of significant lethality in these animals. To verify these results independently, increase experimental resolution and avoid the high mortality upon Leo/R406W co-expression, we used the TARGET system (55), to express the transgenes only in the adult CNS. Although RW under ElavGal4; Gal80ts at 25°C is expressed throughout development, its level is nearly 50% reduced relative to driving the same transgenes with ElavGal4 (Supplementary Material, Fig. S3B). Multiple independent experiments revealed that when restricted to the adult CNS, RW was significantly elevated upon Leo co-expression in all flies examined, but reduced by co-expression with D14-3-3ɛ (Fig. 7G). These results were corroborated with the phosphorylation-dependent antibodies AT8 and AT100 (Fig. 7H and I). R406W levels in D14-3-3ɛ-over-expressing animals (Fig. 7G–I) evaded detection with the AT100 antibody (Fig. 7I), unless large excess of lysate was loaded and exposure prolonged (not shown). Collectively, elevation of Leo results in increased steady-state levels of WT and RW Tau and potentiates their toxicity. In contrast, increased D14-3-3ɛ elevates WT, drastically decreases RWand has minor effects on their toxicity. These observations are in accordance with the abrogation results and support the notion that 14-3-3s have distinct functions in the regulation of Tau levels and its toxicity. Interestingly, although Leo elevation did not alter D14-3-3ɛ levels (Supplementary Material, Fig. S2C), over-expression of the later resulted in decreased levels of endogenous Leo (Supplementary Material, Fig. S2D). Leo reduction is expected to result in heterodimer attenuation and likely accounts for the reduced RW levels under these conditions (Fig. 7). 14-3-3 levels affect Tau stability The WT and RW proteins are transgenic and therefore their transcription levels under UAS control should be relatively constant. Therefore, changes in their steady-state levels upon co-expression with Leo and D14-3-3ɛ must reflect alterations either in translation or their stability. Effects of 14-3-3 elevation on WT and RW translation cannot be totally excluded, but are improbable because both transgenes are controlled by the same vector-encoded regulatory regions. Hyperphosphorylation has been reported to decrease Tau ubiquitination and proteosomal degradation (69,70) and broad specificity phosphatases have been reported to promote Tau proteolysis (71). Given that 14-3-3s bind phosphorylated Ser and Thr (72), they could suppress Tau dephosphorylation by occluding phosphatases, thus enhancing its stability and steady-state levels. During these and previous experiments (60), we noticed that partially dephosphorylated WT Tau appeared unstable or expediently degraded by proteases endogenous to the brain lysate. This was modeled systematically by limited treatment of brain lysates with the broad specificity λ-phosphatase. After a 30 min incubation at 30°C with 50 units of λ-phosphatase, WT Tau became fully dephosphorylated, while 5 units appeared to have little if any effect under these conditions (Fig. 8A). An equal amount of lysate treated with 25 units λ-phosphatase was partially dephosphorylated indicated by its intermediate mobility and interestingly, it was significantly reduced quantitatively relative to the starting material. Confirmation that quantitatively comparable amounts of total lysate were present in all samples was provided by the Syntaxin loading controls. Additional experimental permutations such as limiting λ-phosphatase, longer or shorter incubations yielded qualitatively similar outcomes (not shown). These results suggest that Tau dephosphorylation, possibly at exposed or easily accessible sites, renders the protein susceptible to proteolytic degradation in vivo as previously suggested largely by in vitro results (73). Figure 8. View largeDownload slide 14-3-3 proteins affect Tau stability. WT, RW and STA Tau variants were expressed specifically in adult CNS under ElavGal4. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s was applied to reveal significant differences. (A) A western blot of head lysates from WT Tau-expressing animals incubated with the indicated units of λ-phosphatase for 30 min at 30°C probed with the T14 antibody. Partial dephosphorylation of WT with 25 units phosphatase reduced its levels compared with untreated, or expediently fully dephosphorylated protein. Note that complete dephosphorylation in the lysate is also confirmed by the increased electrophoretic mobility of the Syntaxin (Syx) loading control. (B) A representative blot and the quantification of three additional independent experiments of the levels of WT from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing WT and Leo or WT and D14-3-3ε retained significantly more Tau than lysates expressing WT alone (Dunnett’s, P < 0.001). (C) A representative blot and the quantification of additional independent experiments (n = 4) of the levels of RW from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (Input ratio-treated ratio). Treated lysates from samples co-overexpressing RW and D14-3-3ε retained significantly less Tau than lysates expressing RW alone or also co-overexpressing Leo (Dunnett’s P < 0.01). (D) A representative blot and the quantification of additional independent experiments (n = 3) of the levels of STA from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing STA and Leo or D14-3-3ε retained significantly more Tau than lysates expressing STA alone (Dunnett’s P < 0.0004). Figure 8. View largeDownload slide 14-3-3 proteins affect Tau stability. WT, RW and STA Tau variants were expressed specifically in adult CNS under ElavGal4. Stars indicate significant differences from controls (open bars) per the statistical treatments listed below. All quantification were analyzed by initial ANOVA, and if positive Dunnett’s was applied to reveal significant differences. (A) A western blot of head lysates from WT Tau-expressing animals incubated with the indicated units of λ-phosphatase for 30 min at 30°C probed with the T14 antibody. Partial dephosphorylation of WT with 25 units phosphatase reduced its levels compared with untreated, or expediently fully dephosphorylated protein. Note that complete dephosphorylation in the lysate is also confirmed by the increased electrophoretic mobility of the Syntaxin (Syx) loading control. (B) A representative blot and the quantification of three additional independent experiments of the levels of WT from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing WT and Leo or WT and D14-3-3ε retained significantly more Tau than lysates expressing WT alone (Dunnett’s, P < 0.001). (C) A representative blot and the quantification of additional independent experiments (n = 4) of the levels of RW from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (Input ratio-treated ratio). Treated lysates from samples co-overexpressing RW and D14-3-3ε retained significantly less Tau than lysates expressing RW alone or also co-overexpressing Leo (Dunnett’s P < 0.01). (D) A representative blot and the quantification of additional independent experiments (n = 3) of the levels of STA from the indicated genotypes either prior (input), or after incubation with 20 units of λ phosphatase at 30°C for 30 min (treated). Bars represent mean protein levels ± S.E.M.s of difference in the Tau/Syx ratios of protein remaining in the lysate after phosphatase treatment from those on the respective untreated sample (input ratio-treated ratio). Treated lysates from samples co-overexpressing STA and Leo or D14-3-3ε retained significantly more Tau than lysates expressing STA alone (Dunnett’s P < 0.0004). Although over-expression did not alter WT and RW phosphoprofiles at the sites surveyed (Figs 6D–F and 7D–I), abrogation of 14-3-3s resulted in decreased phosphorylation (Figs 3D–E and 4D–E). Therefore, we investigated whether Tau elevation consequent to excess Leo or D14-3-3ɛ resulted from changes in phosphorylation state-dependent stability. To address this hypothesis, we incubated equal amounts of brain lysates from animals expressing WT and R406W with or without excess transgenic Leo and D14-3-3ɛ under conditions of partial dephosphorylation. As shown in Figure 8B, 60% of WT was degraded by endogenous proteases after λ-mediated limited dephosphorylation. However, co-overexpression of either 14-3-3 ameliorated this effect resulting in degradation of only 20–25% of the WT input. In contrast, the RW protein appeared less vulnerable to endogenous proteases since under the same conditions only 20% of the protein degraded (Fig. 8C). Interestingly however, excess D14-3-3ɛ resulted in significantly increased RW degradation with less than 40% of the starting material remaining at the end of incubation. In contrast, elevated Leo did not appear to affect RW levels. These results are consistent with the in vivo quantitation of WT and RW Tau steady-state levels upon excess 14-3-3s (Figs 6 and 7). To investigate whether RW destabilization upon D14-3-3ɛ excess is unique to this variant or it pertains to other mutants we subjected the STA variant in the same stability assay. This laboratory-generated mutant (61,66) contains Alanines in place of Ser238 and importantly replacing Thr245, which is proposed to bind 14-3-3s when phosphorylated (Fig. 2). As WT and RW, it binds both Leo and D14-3-3ɛ (Supplementary Material, Fig. S5A). Moreover, as for WT, co-expression with Leo or D14-3-3ɛ increases its levels in the adult CNS (Supplementary Material, Fig. S2B). Importantly, this appears to be the consequence of enhanced stability that appears to surpass even that of WT (Fig. 8D). Therefore, the destabilization of RW upon D14-3-3ɛ excess is specific to this particular variant and does not characterize all mutant Tau proteins. It follows then that 14-3-3 elevation has a stabilizing effect on Tau probably by suppressing access to phosphatases and/or proteases. Because 14-3-3 dimer composition is dynamic (30,74), particular homo and heterodimer distributions will likely favor distinct Tau conformations which potentially permit or occlude kinase and phosphatase access differentially. As D14-3-3ɛ overexpression results in decreased endogenous Leo (Supplementary Material, Fig. S2D), the relative pool of Leo homodimers is expected to decrease in response. This is likely exaggerated by D14-3-3ɛ excess, which may further decrease Leo homodimer levels. Under endogenous 14-3-3 levels and under Leo excess, WT appears stable. Therefore, its destabilization upon excessive D14-3-3ɛ is likely the consequence of decreased Leo homodimers. The RW mutation which encompasses a putative 14-3-3 binding site, appears to render the stability of the mutant protein particularly sensitive to Leo homodimer levels. Therefore, the results of 14-3-3 genetic and pharmacological attenuation and those of overexpression strongly indicate that the steady-state levels of WT and RW are modulated by the composition of 14-3-3 dimers. Discussion 14-3-3 interaction modes with Tau Our genetic and biochemical evidence in strong support of the proteomic results, demonstrate that 14-3-3s interact with human WT and the RW Tau variant in vivo, in agreement with in vitro evidence (44). 14-3-3s likely engage at least one or more of the five putative 14-3-3 consensus binding sites (Fig. 2), involving potentially phosphorylated Serines and Threonines. However, since Tau possesses over 80 potential phospho-Ser/phospho-Thr (75), additional phosphorylation-dependent, but non-consensus 14-3-3 docking sites may exist, as bona fide 14-3-3 clients with highly deviant binding motifs have been reported (76). Significantly, because both 14-3-3s bind dephosphorylated WT and RW (Fig. 1E), there are non-phosphorylation-dependent binding sites also. Accordingly, 14-3-3ζ binding on Ser262 (surrounding sequence KIGSTE) in R1 has been reported to be phosphorylation independent (41), but to promote its phosphorylation (41,44). Additional such sites could even involve interactions between the acidic binding groove of 14-3-3s with highly basic areas of Tau, which could emerge upon assumption of certain local conformations. Therefore, Tau presents 14-3-3 dimers with multiple sites and modes of intramolecular and intermolecular interactions. Because dimeric 14-3-3s can in principle bind in a bidentate mode with two client proteins concurrently (39), they may simultaneously bind phosphorylation-dependent and phosphorylation independent sites on the same Tau molecule, different Tau molecules, or Tau and interacting proteins. This in fact is reflected in the ability of the transgenic difopein to bind 14-3-3s and to co-immunoprecipitate Tau (Fig. 1B and C). The multiplicity of potential 14-3-3 binding sites on Tau and the potential for bidentate binding are expected to result in the observed complex interactions that largely depend on the available intracellular complement of 14-3-3s. Functional consequences of attenuated 14-3-3 interaction with Tau Collectively, our data address an important issue in 14-3-3 biology, namely that of their stoichiometry and the functional differences of dimeric states in vivo. Clearly the ratio of available 14-3-3 isoforms and consequently the composition of dimers and monomers influences significantly the steady-state phosphorylation status, at least of the pathology-relevant sites examined. However, 14-3-3s are not limiting for Tau stability and have marginal effects on its toxicity. Considering the genetic abrogation and pharmacological inhibition data, Leo-containing dimers promote low-level steady-state phosphorylation at the sites tested (Figs 3D–F, 4D–F and 5E–H). Importantly however, D14-3-3ɛ homodimers appear to promote phosphorylation of these sites because D14-3-3ɛ attenuation nearly abolishes their occupation (Figs 3D–F and 4D–F). 14-3-3s bind phosphorylated and un-phosphorylated Tau and in fact D14-3-3ɛ binding appears enhanced upon dephosphorylation (Fig. 1E). Therefore, we suggest that Leo-containing homo and heterodimers may use distinct phosphorylation-dependent and independent modes of interaction with Tau and this could stabilize its conformation locally in a manner that promotes or suppresses access to kinases or phosphatases. In addition, binding itself may occlude access of such enzymes to these sites. Such interactions are clearly sensitive to dimer composition, expected to be highly dynamic (77) and are consistent with the differential effects on occupation of the phospho-sites assessed when the intracellular complement of 14-3-3s is altered (Figs 3D–F and 4D–F). Therefore the differential distribution of 14-3-3 isoforms in different neuronal types (78), could be reflected in the reported tissue-specific Tau phosphorylation patterns (60,79–81). Interestingly, changes in the 14-3-3 complement in Leo-, Eps- and Leo- Eps- animals did not precipitate significant changes in WT or RW toxicity, unless functional inhibition was concurrently applied (Fig. 5K and O). This likely reflects both the complexity of Tau interactions with the distinct dimers and the partial functional redundancy of 14-3-3 proteins (29). Significantly, RW toxicity is differentially sensitive to D14-3-3ɛ levels (Fig. 5N), possibly resulting from the mutation on the putative binding site at aa406-413 (Fig. 2). It is highly interesting and significant that the partial suppression of RW toxicity is mediated by D14-3-3ɛ homodimers that appear to form when Leo is limiting (Fig. 3G). These results are consistent with the notion that 14-3-3s facilitate assumption of distinct local or global conformations for WT and RW with consequent changes in toxicity. Alternatively, distinct WT and RW conformations potentially mediated by phosphorylation, render them differentially accessible to 14-3-3s and other proteins mediating toxicity. Therefore, Tau expressing neurons do not tolerate well the functional loss of 14-3-3s. A potential reason for this is that 14-3-3 abrogation may induce ER (82), oxidative (83), or other type of stress in neurons already under oxidative stress because of the presence of pathogenic WT or RW Tau (84). The importance of 14-3-3s in cellular stress responses may be reflected by the strong lethality associated with genetic abrogation of both fly isoforms (Figs 3H and 4G), or their functional pharmacological inhibition (Fig. 5L and P). Therefore, our results strongly indicate that BV02-mediated inhibition of 14-3-3 binding is very likely to precipitate significant unwanted effects to be useful in treating Tauopathies. Functional consequences of 14-3-3 elevation Over-accumulation of Leo, but not D14-3-3ɛ precipitated significant lethality on its own (Fig. 6B). This combined with the drastic reduction in viability of animals with severely reduced functional 14-3-3s (Fig. 5L and P) and in agreement with prior results (29), demonstrate the importance of 14-3-3 homeostasis for survival. Because only Leo over-accumulation yielded lethality, it is likely that the effect is precipitated by excess Leo-containing dimers. The reason for this lethality is unclear to date, but it is under investigation and could be related to synaptic deficits (44) during development, or any of the multiple cellular processes requiring regulated 14-3-3 levels and activity (85,86). Interestingly over-accumulation of either 14-3-3 does not change the phosphorylation pattern of either WT or RW (Figs 6 and 7). Rather it affects their stability in an isoform-specific manner and differentially for WT and RW. Tau stability depends on its phosphorylation state [Fig. 8 and (70)] and hyper-phosphorylation protects Tau from proteolytic degradation (70). 14-3-3 over-accumulation did not change the phosphorylation state, at least of the sites tested, but rendered WT resistant to degradation. Therefore, the increase in WT levels is a likely result of excess 14-3-3-mediated conformational changes that render it resistant to endogenous phosphatases or proteases. In contrast, the effects of 14-3-3 elevation on the RW variant, which eliminates a single potential 14-3-3 binding site, are differential and isoform-specific. Whereas, Leo excess nearly doubles RW levels as for WT, over-accumulation of D14-3-3ɛ precipitates its drastic reduction (Fig. 4H–M). This appears to be a consequence of D14-3-3ɛ-dependent increase in susceptibility to degradation of under-phosphorylated RW (Fig. 7C). Hence, the D14-3-3ɛ excess and the resultant shift in heterodimer/homodimer balance appears to favor conformational state(s) that expose protease-sensitive sites on RW due to the mutation itself, which removes a 14-3-3 consensus binding site. Because RW degradation is not observed upon Leo excess, it is likely that over-accumulation of D14-3-3ɛ reduces the abundance of Leo/Leo homodimers, which appear to favor the degradation-resistant conformation. The differential effect is specific to RW, as the STA mutant affecting another putative 14-3-3 binding site, is completely protected from degradation by overexpression of either Leo or D14-3-3ɛ (Fig. 8D). Our results indicate that WT and STA in contrast to RW likely assume distinct conformations in the fly CNS, which render them differentially accessible to proteases and 14-3-3s are involved in these processes. The ability of 14-3-3 dimers for bidentate binding to different Tau client proteins could differentially promote stabilization of conformations that inhibit dephosphorylation and enhance stability. This may be further mediated by Tau interacting or regulating proteins whose identity is currently unknown, albeit under investigation. In summary, the data support the hypothesis that 14-3-3 excess stabilizes Tau except for RW, where removal of a binding site appears to trigger distinct conformational changes that render the variant protein particularly sensitive to the stoichiometry of 14-3-3 homo/heterodimers. Because 14-3-3s are differentially distributed in the mammalian CNS (78), it is possible that the expected distinct distribution of 14-3-3 homo and heterodimers, affects the pathogenicity of WT and mutant Tau proteins in a tissue-specific manner. This may underlie, at least in part, the neuronal type specificity of the various Tauopathies and the symptom variability and intensity that characterizes them (13). Are 14-3-3s Tauopathy mediators or innocent bystanders? Soon after their identification and characterization, 14-3-3 proteins became associated with diverse inflammatory, vascular, malignant and degenerative neuropathologies (53,87). Because in most of these disorders, elevation of 14-3-3 proteins was noted both in the CSF and in and around lesions, it remained unclear whether these proteins were incidental non-specific markers of neuronal stress and neurodegeneration, or involved in the pathogenesis or severity of each condition. Our data favor the first notion, because although overaccumulation of Leo resulted in lethality, this appears to be a consequence of altered 14-3-3 homeostasis. Pathogenic Tau phosphorylation increases potential 14-3-3 binding sites and probably recruits additional dimers. This may alter the intraneuronal distribution of 14-3-3s and may limit them for vital cellular processes. Given their homeostatic propensity, 14-3-3 levels may be elevated in compensation. Elevated 14-3-3s will favor stable Tau conformations and lead to its accumulation which increases neurotoxic consequences. In addition, the elevated 14-3-3 dimer load on hyper-phosphorylated Tau may stabilize transient conformations that promote additional phosphorylation at distant sites enhancing toxicity. This scenario of progressive 14-3-3 elevation and recruitment of an otherwise ‘bystander’ protein is a result of the large number of phopshorylation-dependent and non-phosphorylation-dependent 14-3-3 docking sites on Tau. Alternatively, accumulation of pathogenic Tau may induce cellular stress and 14-3-3s as stress-related proteins (23) could be elevated in response potentially initiating the progressive elevation scenario described above. In Drosophila however, 14-3-3 levels are not changed upon Tau accumulation in the CNS (Supplementary Material, Fig. S4). In vertebrates on the other hand, elevated 14-3-3s may be increasingly recruited to Tau molecules as they acquire pathogenic tags and either enhance or suppress conformations that favor their stability depending on the properties of the particular Tau variant and of the prevalent 14-3-3 dimers in that particular neuron. Finally, our results suggest that in vivo approaches in the CNS as the one described herein are invaluable in capturing the dynamic, adaptive and homeostatic interactions of 14-3-3 proteins with Tau and other client proteins with multiple modes of interaction and docking sites. Materials and Methods Drosophila culture and strains Flies were cultured in standard sugar-wheat flour food supplemented with soy flour and CaCl2. Pan-neuronal transgene expression was achieved using the driver ElavC155-Gal4 (ElavGal4) (88,89). Fly crosses and experiments were performed at 25°C unless noted otherwise. The ElavGal4; Gal80ts strain was constructed using standard methods. The Gal80ts strain (55) was obtained from the Bloomington Stock Centre. Conditional transgene expression under Elav-Gal4; Gal80ts was induced specifically in adult flies by incubation at 30°C for 10 days post-emergence. Transgenic fly lines carrying UAS-htau0N4R and UAS-htauR406W inserts were gifts of Dr M. Feany (Harvard Medical School) (90). The UAS-htauFLAG-2N4RSTA has been described in (61). The UAS-mycD14-3-3ɛ transgenic strains (91) and the D14-3-3ɛ RNAi-generating transgenes (92) have been described previously. The UAS-leo transgenes were generated by placing the entire leoII cDNA (93) in the pUAST vector (94), while Leo RNAi-generating transgenes were obtained from the Vienna Drosophila Resource Centre and were normalized to Cantonised w1188. Double transgenic strains were generated by standard crosses. The UAS-htauFLAG-2N4R strain has been described elsewhere (61). The UAS-htauFLAG-R406W-2N4R strain was generated by replacing Arg406 with Trp using the QuickChange XL site-directed mutagenesis kit according to the manufacturer’s instructions (Agilent Technologies). The mutagenic oligonucleotides 5′-GGGGACACGTCTCCATGGCATCTCAGCAATGTCTCC-3′ and 5′-GGAGACATTGCTGAGATGCCATGGAGACGTGTCCCC-3′ were annealed onto the UAS-htau2N4R-FLAG plasmid template and introduced an NcoI restriction site which was used for screening of positive clones. The sequence of the mutant was confirmed by dsDNA sequencing (Lark technologies). Transgenic flies were obtained with standard methods. To assess the effect of 14-3-3s on the levels and phosphorylation of Tau, we generated the double transgenic stocks UAS-leoII; UAS-htau0N4R, UAS-d14-3-3ɛ; UAS-htau0N4Rand UAS-leoII; UAS-htauR406W, UAS-d14-3-3ɛ; UAS-htauR406W which were crossed with ElavGal4 virgins as required. Conversely, to assess the effects of 14-3-3 attenuation, the following strains were generated ElavGal4; leoRNAi/CyO and ElavGal4; epsRNAi/TM3Sb and crossed to UAS-htau0N4R and UAS-htauR406W males as indicated. To generate flies lacking both 14-3-3 isoforms ElavGal4; epsRNAi/TM3Sb were crossed with UAS-leoRNAi; UAS-htau0N4R or with UAS-htauR406W. Viability assays To determine the effect of WT or RW expression on viability, 5 ElavGa4 females were crossed with 3 WT or RW homozygous males. After 24 h they transferred to new vials and allowed to lay eggs for three days and then discarded. The number of surviving progeny per live female was determined when adults emerged. Each assessment was performed at least in triplicate with five females each. Control progeny were obtained by concomitantly crossing the same number of ElavGal4 females to w1118 males and the mean number of their progeny per female was set as 100%. This number was used to normalize the numbers of Tau-expressing progeny expressed as a percent. 14-3-3 Inhibitor experiments For the 14-3-3 inhibitor experiments, five ElavGal4 females were crossed to three relevant males from the abovementioned stocks and allowed to lay eggs for three days at 25°C on normal food supplemented with 0 (DMSO), 0.15 or 0.3 nm of the 14-3-3 antagonist II, BV02 (Calbiochem). Viability was monitored by counting the number of adult flies emerging per genotype with and without the antagonist. Control experiments with higher BV02 concentrations (0.35–0.5 nm) did not yield any progeny. Western blot analysis To determine total Tau levels and its phosphorylation status in different genotypes, adult female fly heads 1–3 days post-eclosion were homogenized in 1× Laemmli buffer (50 mm Tris pH 6.8, 100 mm DTT, 5% 2-mercaptoethanol, 2% SDS, 10% glycerol and 0.01% bromophenol blue). The lysates were boiled for 5 min at 95°C, centrifuged at 9000g for 5 min and separated by SDS-PAGE using a 10% separating gel. Proteins were transferred to a PVDF membrane at 100 V for 1 h and probed with the following monoclonal antibodies: T14 (Zymed laboratories) at 1: 1000 for 1 head/lane, T46 (Invitrogen) at 1: 2000 for 1 head/lane, AT100 (Pierce Endogen) at 1: 250 for 1 head/lane and the polyclonal antibodies pS262 (Biosource) at 1: 500 for 1 head/lane, anti-LEO (95) at 1: 20 000 for 0.2 head/lane and anti-D14-3-3ɛ (29) at 1: 2000 for 0.2 head/lane. The monoclonal antibody AT8 was kindly provided by A. Mudher and was used at 1: 200 dilution for 1 head/lane. To normalize for sample loading, the membranes were concurrently probed with an anti-syntaxin (Syn) monoclonal antibody (8C3, Developmental Studies Hybridoma Bank) at 1: 2000 for 1 head/lane and at 1: 500 for 0.2 heads/lane. Appropriate HRP-conjugated secondary antibodies were applied at 1: 5000. Proteins were visualized with chemiluminescence (ECL Plus, Amersham) and signals were quantified by densitometry with the ImageQ program. Results were plotted as means with S.E.M. from at least three independent experiments. The data were analyzed by standard parametric statistics as indicated in the figure legends. Pull-down assay Temporal regulation of UAS controlled transgenes under the ubiquitously expressed temperature sensitive Gal80ts has been described earlier (55). At the end of induction, flies were harvested, frozen at −72°C and then decapitated en masse by sieving in liquid nitrogen. Harvested heads were homogenized in PBST buffer (1× PBS pH 7.4 containing 0.01% Tween® 20) supplemented with protease inhibitors and lysates were incubated over-night with either magnetic beads coupled with HT7 anti-Tau antibody or with GST-difopein-conjugated beads. After washing, proteins were eluted with 0.1 m glycine-HCl pH 3.5 and subjected to SDS-PAGE. Immunoblotting was performed as described above. Biotinylated HT7 antibody (Thermo Scientific) was immobilized on Dynabeads M-280® Streptavidin (Invitrogen) according to the manufacturer’s instructions. The pGEX-6P1 vector encoding GST-tagged difopein was kindly provided by Dr Marco Lalle (56). To generate the protein, BL21(DE3) transformed cells were grown in LB medium (2% bacto tryptone, 0.5% yeast extract, 0.05% NaCl) at 37°C until OD600 0.5. Protein expression was induced by addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to 0.5 mm final concentration at 37°C for 3 h. GST-difopein was immobilized on glutathione-agarose beads (Sigma) according to the manufacturer’s instructions. Flag-tagged 2N4R-STA was immunoprecipated with anti-Flag M2 affinity resin (Sigma) according to manufacturer’s instructions. In vitro dephosphorylation Dephosphorylation prior to immunoprecipitation was achieved by treating the lysates with λ phosphatase (New England Biolabs). Briefly, head homogenates from adult flies expressing the indicated Flag-tagged transgenes pan-neuronally were prepared in lysis buffer (50 mm Tris pH 7.5, 150 mm NaCl, 1 mm MnCl2, 1% Triton) supplemented with Halt protease inhibitor cocktail (Pierce) EDTA-free. Lysates were split in two and they were either supplemented with phosphatase inhibitor cocktail (Sigma) or treated with λ phosphatase (2.5 mg of total protein were incubated with 1000 units of λ phosphatase at 30°C for 1 h). The reaction was stopped by addition of phosphatase inhibitors and lysates were incubated either with anti-Flag M2 affinity resin or with glutathione-agarose beads coated with GST-difopein over night at 4°C. The starting buffer of this protocol is different than the one below because immunoprecipitation has to follow dephosphorylation. Immunoprecipitation is not compatible with the standard de-phosphorylation buffer presented below. Homogenates from heads of adult flies were prepared in RIPA buffer (137 mm NaCl, 20 mm Tris pH 8.0, 10% glycerol, 0.1% SDS and 0.1% sodium deoxycholate). A total of 20 μg protein were incubated with the indicated units of λ phosphatase (New England Biolabs) at 30°C for 30 min, according to manufacturer’s instructions. As different batches of λ-phosphatase have different activities, we calibrated the units necessary to produce complete dephosphorylation per batch and then used the appropriate units to produce partial dephosphorylation as indicated. One-third of each sample and equivalent amount of untreated samples were resolved by SDS-PAGE and subjected to western blot analysis as described earlier. Cross-linking Eight heads were lysed in 70 μl of ice-cold buffer (20 mm sodium phosphate, pH 7.4, 150 mm NaCl supplemented with phosphatase and protease inhibitor mixture) and centrifuged for 15 min at 14 000g at 4°C. Head lysates were divided in two fractions of 30 μl. One was cross-linked by addition of 1.5 mm BS3 final concentration (Thermo Scientific) while the other remained untreated (lysis buffer was added instead of BS3). Samples were incubated for 2 h on ice and the reaction was stopped by the addition of 1 m Tris pH 7.4 buffer to a final concentration of 20 mm. Samples were analyzed by western blotting after the addition of 5× Laemmli Buffer. Statistical Analyses Untransformed (raw) data were analyzed parametrically with the JMP 7.1 statistical software package (SAS Institute Inc., Cary, NC). Supplementary Material Supplementary Material is available at HMG online. Acknowledgements The initial parts of this work were supported by grants 01ΕΔ207 (E.M.C.S.) and 04EP18 (K.P.) from the Hellenic General Secretariat for Research and Technology. We thank the Developmental Hybridoma Bank (DSHB) for antibodies, Drs Botas, Feany, Lu and Mudher for fly strains, Marco Lalle for GST-difopein reagents and George Panayotou for advise and support with the proteomic analyses. Conflict of Interest statement. None declared. References 1 Black M.M., Slaughter T., Moshiach S., Obrocka M., Fischer I. ( 1996) Tau is enriched on dynamic microtubules in the distal region of growing axons. J. Neurosci ., 16, 3601– 3619. Google Scholar CrossRef Search ADS PubMed  2 Kempf M., Clement A., Faissner A., Lee G., Brandt R. ( 1996) Tau binds to the distal axon early in development of polarity in a microtubule- and microfilament-dependent manner. J. Neurosci ., 16, 5583– 5592. Google Scholar CrossRef Search ADS PubMed  3 Avila J., Lucas J.J., Pérez M.A.R., Hernández F. ( 2004) Role of tau protein in both physiological and pathological conditions. Physiol. Rev ., 84, 361– 384. Google Scholar CrossRef Search ADS PubMed  4 Baas P.W., Qiang L. ( 2005) Neuronal microtubules: when the MAP is the roadblock. Trends Cell Biol ., 15, 183– 187. Google Scholar CrossRef Search ADS PubMed  5 Terwel D., Dewachter I., Van Leuven F. ( 2002) Axonal transport, tau protein, and neurodegeneration in Alzheimer's disease. Neuromol. Med ., 2, 151– 165. Google Scholar CrossRef Search ADS   6 Ballatore C., Lee V.M., Trojanowski J.Q. ( 2007) Tau-mediated neurodegeneration in Alzheimer's disease and related disorders. Nat. Rev. Neurosci ., 8, 663– 672. Google Scholar CrossRef Search ADS PubMed  7 Wang Y., Mandelkow E. ( 2016) Tau in physiology and pathology. Nat. Rev. Neurosci ., 17, 5– 21. Google Scholar CrossRef Search ADS PubMed  8 Avila J., Pérez M., Lim F., Gómez-Ramos A., Hernández F., Lucas J.J. ( 2004) Tau in neurodegenerative diseases: tau phosphorylation and assembly. Neurotox. Res ., 6, 477– 482. Google Scholar CrossRef Search ADS PubMed  9 Buée L., Bussière T., Buée-Scherrer V., Delacourte A., Hof P.R. ( 2000) Tau protein isoforms, phosphorylation and role in neurodegenerative disorders. Brain Res. Rev ., 33, 95– 130. Google Scholar CrossRef Search ADS   10 Buee L., Delacourte A. ( 1999) Comparative biochemistry of tau in progressive supranuclear palsy, corticobasal degeneration, FTDP-17 and Pick's disease. Brain Pathol ., 9, 681– 693. Google Scholar CrossRef Search ADS PubMed  11 Hardy J., Orr H. ( 2006) The genetics of neurodegenerative diseases. J. Neurochem ., 97, 1690– 1699. Google Scholar CrossRef Search ADS PubMed  12 Ingram E.M., Spillantini M.G. ( 2002) Tau gene mutations: dissecting the pathogenesis of FTDP-17. Trends Mol. Med ., 8, 555– 562. Google Scholar CrossRef Search ADS PubMed  13 Lee V.M.-Y., Goedert M., Trojanowski J.Q. ( 2001) Neurodegenerative tauopathies. Annu. Rev. Neurosci ., 24, 1121– 1159. Google Scholar CrossRef Search ADS PubMed  14 Spillantini M.G., Goedert M. ( 1998) Tau protein pathology in neurodegenerative diseases. Trends Neurosci ., 21, 428– 433. Google Scholar CrossRef Search ADS PubMed  15 Brookmeyer R., Johnson E., Ziegler-Graham K., Arrighi H.M. ( 2007) Forecasting the global burden of Alzheimer's disease. Alzheimers Dement ., 3, 186– 191. Google Scholar CrossRef Search ADS PubMed  16 Spillantini M.G., Van Swieten J.C., Goedert M. ( 2000) Tau mutations in frontotemporal dementia FTDP-17 and their relevance for Alzheimer's disease. Biochim. Biophys. Acta , 2, 0193– 0121. 17 Weder N.D., Aziz R., Wilkins K., Tampi R.R. ( 2007) Frontotemporal dementias: a review. Ann. Gen. Phychiatry , 6, 15. doi: 10.1186/1744-859X-6-15. Google Scholar CrossRef Search ADS   18 Sergeant N., Delacourte A., Buee L. ( 2005) Tau protein as a differential biomarker of tauopathies. Bioch. Biophys. Acta , 1739, 179– 197. 19 Johnson G.V., Stoothoff W.H. ( 2004) Tau phosphorylation in neuronal cell function and dysfunction. J. Cell Sci ., 117, 5721– 5729. Google Scholar CrossRef Search ADS PubMed  20 Martin L., Latypova X., Terro F. ( 2011) Post-translational modifications of tau protein: implications for Alzheimer's disease. Neurochem. Int ., 58, 458– 471. Google Scholar CrossRef Search ADS PubMed  21 Alonso A. d C., Li B., Grundke-Iqbal I., Iqbal K. ( 2008) Mechanism of tau-induced neurodegeneration in Alzheimer disease and related tauopathies. Curr. Alzheimer Res ., 5, 375– 384. Google Scholar CrossRef Search ADS PubMed  22 Gong C.-X., Liu F., Grundke-Iqbal I., Iqbal K. ( 2005) Post-translational modifications of tau protein in Alzheimer's disease. J. Neural Transm ., 112, 813– 838. Google Scholar CrossRef Search ADS   23 Soulié C., Nicole A., Delacourte A., Ceballos-Picot I. ( 2004) Examination of stress-related genes in human temporal versus occipital cortex in the course of neurodegeneration: involvement of 14-3-3 zeta in this dynamic process. Neurosci. Lett ., 365, 1– 5. Google Scholar CrossRef Search ADS PubMed  24 Aitken A. ( 2006) 14-3-3 proteins: a historic overview. Semin. Cancer Biol ., 16, 162– 172. Google Scholar CrossRef Search ADS PubMed  25 Gardino A.K., Smerdon S.J., Yaffe M.B. ( 2006) Structural determinants of 14-3-3 binding specificities and regulation of subcellular localization of 14-3-3-ligand complexes: a comparison of the X-ray crystal structures of all human 14-3-3 isoforms. Semin. Cancer Biol ., 16, 173– 182. Google Scholar CrossRef Search ADS PubMed  26 Masters S.C., Fu H. ( 2001) 14-3-3 proteins mediate an essential anti-apoptotic signal. J. Biol. Chem ., 276, 45193– 45200. Google Scholar CrossRef Search ADS PubMed  27 Aghazadeh Y., Papadopoulos V. ( 2016) The role of the 14-3-3 protein family in health, disease, and drug development. Drug Discov. Today , 21, 278– 287. Google Scholar CrossRef Search ADS PubMed  28 Aitken A., Baxter H., Dubois T., Clokie S., Mackie S., Mitchell K., Peden A., Zemlickova E. ( 2002) 14-3-3 proteins in cell regulation. Biochem. Soc. Trans ., 30, 351– 360. Google Scholar CrossRef Search ADS PubMed  29 Acevedo S.F., Tsigkari K.K., Grammenoudi S., Skoulakis E.M.C. ( 2007) In vivo functional specificity and homeostasis of Drosophila 14-3-3 proteins. Genetics , 177, 239– 253. Google Scholar CrossRef Search ADS PubMed  30 Messaritou G., Grammenoudi S., Skoulakis E. ( 2010) Dimerization is essential for 14-3-3zeta stability and function in vivo. J. Biol. Chem ., 285, 1692. Google Scholar CrossRef Search ADS PubMed  31 Messaritou G., Leptourgidou F., Franco M., Skoulakis E.M.C. ( 2009) A third functional isoform enriched in mushroom body neurons is encoded by the Drosophila 14-3-3zeta gene. FEBS Lett ., 583, 2934– 2938. Google Scholar CrossRef Search ADS PubMed  32 Muslin A.J., Tanner J.W., Allen P.M., Shaw A.S. ( 1996) Interaction of 14-3-3 with signaling proteins is mediated by the recognition of phosphoserine. Cell , 84, 889– 897. Google Scholar CrossRef Search ADS PubMed  33 Tzivion G., Avruch J. ( 2002) 14-3-3 Proteins: active cofactors in cellular regulation by Serine/Threonine phosphorylation. J. Biol. Chem ., 277, 3061– 3064. Google Scholar CrossRef Search ADS PubMed  34 Yaffe M.B., Elia A.E.H. ( 2001) Phosphoserine/threonine-binding domains. Curr. Opin. Cell. Biol ., 13, 131– 138. Google Scholar CrossRef Search ADS PubMed  35 Kaplan A., Ottmann C., Fournier A. ( 2017) 14-3-3 adaptor protein-protein interactions as therapeutic targets for CNS diseases. Pharmacol. Res ., 125, 114. Google Scholar CrossRef Search ADS PubMed  36 Bridges D., Moorhead G.B.G. ( 2005) 14-3-3 proteins: a number of functions for a numbered protein. Sci. STKE , 2005, re10. Google Scholar PubMed  37 van Hemert M.J., Steensma H.Y., van Heusden G.P. ( 2001) 14-3-3 proteins: key regulators of cell division, signalling and apoptosis. BioEssays , 23, 936– 946. Google Scholar CrossRef Search ADS PubMed  38 van Heusden G. ( 2005) 14-3-3 proteins: regulators of numerous eukaryotic proteins. IUBMB Life , 57, 623– 629. Google Scholar CrossRef Search ADS PubMed  39 Yaffe M.B. ( 2002) How do 14-3-3 proteins work?-Gatekeeper phosphorylation and the molecular anvil hypothesis. FEBS Lett ., 513, 53– 57. Google Scholar CrossRef Search ADS PubMed  40 Chun J. ( 2004) 14-3-3 Protein mediates phosphorylation of microtubule-associated protein tau by serum- and glucocorticoid-induced protein kinase 1. Mol. Cells , 18, 360– 368. Google Scholar PubMed  41 Hashiguchi M., Sobue K., Paudel H.K. ( 2000) 14-3-3 z is an effector of tau protein phosphorylation. J. Biol. Chem ., 275, 25247– 25254. Google Scholar CrossRef Search ADS PubMed  42 Yuan Z., Agarwal-Mawal A., Paudel H.K. ( 2004) 14-3-3 binds to and mediates phosphorylation of microtubule-associated tau protein by Ser9-phosphorylated glycogen synthase kinase 3beta in the brain. J. Biol. Chem ., 279, 26105– 26114. Google Scholar CrossRef Search ADS PubMed  43 Li T., Paudel H.K. ( 2007) 14-3-3zeta facilitates GSK3beta-catalyzed tau phosphorylation in HEK-293 cells by a mechanism that requires phosphorylation of GSK3beta on Ser9. Neurosci. Lett ., 414, 203– 208. Google Scholar CrossRef Search ADS PubMed  44 Qureshi H.Y. ( 2013) Overexpression of 14-3-3z promotes tau phosphorylation at Ser262 and accelerates proteosomal degradation of synaptophysin in rat primary hippocampal neurons. PLoS One , 8, e84615. Google Scholar CrossRef Search ADS PubMed  45 Qureshi H.Y., Li T., MacDonald R., Cho C.M., Leclerc N., Paudel H.K. ( 2013) Interaction of 14-3-3ζ with microtubule-associated protein tau within Alzheimer's disease neurofibrillary tangles. Biochemistry , 52, 6445– 6455. Google Scholar CrossRef Search ADS PubMed  46 Agarwal-Mawal A., Qureshi H.Y., Cafferty P.W., Yuan Z., Han D., Lin R., Paudel H.K. ( 2003) 14-3-3 connects glycogen synthase kinase-3 beta to tau within a brain microtubule-associated tau phosphorylation complex. J. Biol. Chem ., 278, 12722– 12728. Google Scholar CrossRef Search ADS PubMed  47 Bustos D. ( 2012) The role of protein disorder in the 14-3-3 interaction network. Mol. Biosyst ., 8, 178– 184. Google Scholar CrossRef Search ADS PubMed  48 Umahara T., Uchihara T., Tsuchiya K., Nakamura A., Iwamoto T., Ikeda K., Takasaki M. ( 2004) 14-3-3 proteins and zeta isoform containing neurofibrillary tangles in patients with Alzheimer's disease. Acta Neuropathol ., 108, 279– 286. Google Scholar CrossRef Search ADS PubMed  49 Umahara T., Uchihara T., Tsuchiya K., Nakamura A., Ikeda K., Iwamoto T., Takasaki M. ( 2004) Immunolocalization of 14-3-3 isoforms in brains with Pick body disease. Neurosci. Lett ., 371, 215– 219. Google Scholar CrossRef Search ADS PubMed  50 Hernandez F., Cuadros R., Avila J. ( 2004) Zeta 14-3-3 protein favours the formation of human tau fibrillar polymers. Neurosci. Lett ., 357, 143– 146. Google Scholar CrossRef Search ADS PubMed  51 Sadik G., Tanaka T., Kato K., Yanagi K., Kudo T., Takeda M. ( 2009) Differential interaction and aggregation of 3-repeat and 4-repeat tau isoforms with 14-3-3zeta protein. Biochem. Biophys. Res. Commun ., 383, 37– 41. Google Scholar CrossRef Search ADS PubMed  52 Fountoulakis M., Cairns N., Lubec G. ( 1999) Increased levels of 14-3-3 gamma and epsilon proteins in brain of patients with Alzheimer's disease and Down syndrome. J. Neural. Transm. Suppl ., 57, 323– 335. Google Scholar PubMed  53 Berg D., Holzmann C., Riess O. ( 2003) 14-3-3 proteins in the nervous system. Nature , 4, 752– 762. 54 Jeganathan S., Hascher A., Chinnathambi S., Biernat J., Mandelkow E.-M., Mandelkow E. ( 2008) Proline-directed pseudo-phosphorylation at AT8 and PHF1 epitopes induces a compaction of the paperclip folding of Tau and generates a pathological (MC-1) conformation. J. Biol. Chem ., 283, 32066– 32076. Google Scholar CrossRef Search ADS PubMed  55 McGuire S.E., Mao Z., Davis R.L. ( 2004) Spatiotemporal gene expression targeting with the TARGET and gene-switch systems in Drosophila. Sci. STKE , 2004, pl6. Google Scholar PubMed  56 Lalle M., Salzano A.M., Crescenzi M., Pozio E. ( 2006) The Giardia duodenalis 14-3-3 protein is post-translationally modified by phosphorylation and polyglycylation of the C-terminal tail. J. Biol. Chem ., 281, 5137– 5148. Google Scholar CrossRef Search ADS PubMed  57 Sluchanko N., Gusev N. ( 2011) Probable participation of 14-3-3 in tau protein oligomerization and aggregation. J. Alzheimers Dis ., 27, 467– 476. Google Scholar PubMed  58 Obsilova V., Nedbalkova E., Silhan J., Boura E., Herman P., Vecer J., Sulc M., Teisinger J., Dyda F., Obsil T. ( 2008) The 14-3-3 protein affects the conformation of the regulatory domain of human tyrosine hydroxylase. Biochemistry , 47, 1768– 1777. Google Scholar CrossRef Search ADS PubMed  59 Papanikolopoulou K., Skoulakis E.M. ( 2011) The power and richness of modelling tauopathies in Drosophila. Mol. Neurobiol ., 44, 122– 133. Google Scholar CrossRef Search ADS PubMed  60 Grammenoudi S., Kosmidis S., Skoulakis E.M. ( 2006) Cell type-specific processing of human Tau proteins in Drosophila. FEBS Lett ., 580, 4602– 4606. Google Scholar CrossRef Search ADS PubMed  61 Kosmidis S., Grammenoudi S., Papanikolopoulou K., Skoulakis E.M.C. ( 2010) Differential effects of Tau on the integrity and function of neurons essential for learning in Drosophila. J. Neurosci ., 30, 464– 477. Google Scholar CrossRef Search ADS PubMed  62 Gustke N., Trinczek B., Biernat J., Mandelkow E.M., Mandelkow E. ( 1994) Domains of tau protein and interactions with microtubules. Biochemistry , 33, 9511– 9522. Google Scholar CrossRef Search ADS PubMed  63 Zheng-Fischhofer Q., Biernat J., Mandelkow E.-M., Illenberger S., Godemann R., Mandelkow E. ( 1998) Sequential phosphorylation of Tau by glycogen synthase kinase-3beta and protein kinase A at Thr212 and Ser214 generates the Alzheimer-specific epitope of antibody AT100 and requires a paired-helical-filament-like conformation. Eur. J. Biochem ., 252, 542– 552. Google Scholar CrossRef Search ADS PubMed  64 Sadik G., Tanaka T., Kato K., Yamamori H., Nessa B.N., Morihara T., Takeda M. ( 2009) Phosphorylation of tau at Ser214 mediates its interaction with 14-3-3 protein: implications for the mechanism of tau aggregation. J. Neurochem ., 108, 33– 43. Google Scholar CrossRef Search ADS PubMed  65 Nishimura I., Yang Y., Lu B. ( 2004) PAR-1 kinase plays an initiator role in a temporally ordered phosphorylation process that confers tau toxicity in Drosophila. Cell , 116, 671– 682. Google Scholar CrossRef Search ADS PubMed  66 Papanikolopoulou K., Skoulakis E. ( 2015) Temporally distinct phosphorylations differentiate Tau-dependent learning deficits and premature mortality in Drosophila. Hum. Mol. Genet ., 24, 2065– 2077. Google Scholar CrossRef Search ADS PubMed  67 Mancini M., Corradi V., Petta S., Barbieri E., Manetti F., Botta M., Santucci M.A. ( 2011) A new nonpeptidic inhibitor of 14-3-3 induces apoptotic cell death in chronic myeloid leukemia sensitive or resistant to imatinib. J. Pharmacol. Exp. Ther ., 336, 596– 604. Google Scholar CrossRef Search ADS PubMed  68 Corradi V., Mancini M., Manetti F., Petta S., Santucci M.A., Botta M. ( 2010) Identification of the first non-peptidic small molecule inhibitor of the c-Abl/14-3-3 protein-protein interactions able to drive sensitive and Imatinib-resistant leukemia cells to apoptosis. Bioorg. Med. Chem. Lett ., 20, 6133– 6137. Google Scholar CrossRef Search ADS PubMed  69 Iqbal K., Liu F., Gong C.-X., Alonso A.D.C., Grundke-Iqbal I. ( 2009) Mechanisms of tau-induced neurodegeneration. Acta Neuropathol ., 118, 53– 69. Google Scholar CrossRef Search ADS PubMed  70 Sluchanko N.N., Sudnitsyna M.V., Chernik I.S., Seit-Nebi A.S., Gusev N.B. ( 2011) Phosphomimicking mutations of human 14-3-3ζ affect its interaction with tau protein and small heat shock protein HspB6. Arch. Biochem. Biophys ., 506, 24– 34. Google Scholar CrossRef Search ADS PubMed  71 Lee S., Shea T. ( 2013) Regulation of tau proteolysis by phosphatases. Brain Res ., 1495, 30– 36. Google Scholar CrossRef Search ADS PubMed  72 Obsilova V. ( 2008) 14-3-3 proteins: a family of versatile molecular regulators. Physiol. Res ., 57(Suppl 3), S11– S21. Google Scholar PubMed  73 Wang Y., Garg S., Mandelkow E.-M., Mandelkow E. ( 2010) Proteolytic processing of tau. Biochem. Soc. Trans ., 38, 955– 961. Google Scholar CrossRef Search ADS PubMed  74 Ghorbani S., Fossbakk A., Jorge-Finnigan A., Flydal M.I., Haavik J., Kleppe R. ( 2016) Regulation of tyrosine hydroxylase is preserved across different homo- and heterodimeric 14-3-3 proteins. Amino Acids , 48, 1221– 1229. Google Scholar CrossRef Search ADS PubMed  75 Alonso A., Corbo C., Beharry C. ( 2013) Abnormal hyperphosphorylation of tau: sites, regulation, and molecular mechanism of neurofibrillary degeneration. J. Alzheimers Dis ., 9, P673– P139. 76 Coblitz B., Wu M., Shikano S., Li M. ( 2006) C-terminal binding: an expanded repertoire and function of 14-3-3proteins. FEBS Lett ., 580, 1531– 1535. Google Scholar CrossRef Search ADS PubMed  77 Woodcock J. ( 2017) Role of salt bridges in the dimer interface of 14-3-3ζ in dimer dynamics, N-terminal α-helical order and molecular chaperone activity. J. Biol. Chem ., doi: 10.1074/jbc.M117.801019. 78 Martin H., Rostas J., Patel Y., Aitken A. ( 2002) Subcellular localisation of 14-3-3 isoforms in rat brain using specific antibodies. J. Neurochem ., 63, 2259– 2265. Google Scholar CrossRef Search ADS   79 Götz J., Nitsch R. ( 2001) Compartmentalized tau hyperphosphorylation and increased levels of kinases in transgenic mice. Neuroreport , 3, 2007– 2016. 80 Holzer M., Holzapfel H.-P., Zedlick D., Brückner M.K., Arendt T. ( 1994) Abnormally phosphorylated tau protein in Alzheimer's disease: heterogeneity of individual regional distribution and relationship to clinical severity. Neuroscience , 63, 499– 516. Google Scholar CrossRef Search ADS PubMed  81 Lauckner J., Frey P., Geula C. ( 2003) Comparative distribution of tau phosphorylated at Ser262 in pre-tangles and tangles. Neurobiol. Aging , 24, 767– 776. Google Scholar CrossRef Search ADS PubMed  82 Cao L., Lei H., Chang M.-Z., Liu Z.-Q., Bie X.-H. ( 2015) Down-regulation of 14-3-3β exerts anti-cancer effects through inducing ER stress in human glioma U87 cells: involvement of CHOP-Wnt pathway. Biochem. Biophys. Res. Commun ., 462, 389– 395. Google Scholar CrossRef Search ADS PubMed  83 Thandavarayan R.A., Giridharan V.V., Sari F.R., Arumugam S., Veeraveedu P.T., Pandian G.N., Palaniyandi S.S., Ma M., Suzuki K., Gurusamy N., Watanabe K. ( 2011) Depletion of 14-3-3 protein exacerbates cardiac oxidative stress, inflammation and remodeling process via modulation of MAPK/NF-ĸB signaling pathways after streptozotocin-induced diabetes mellitus. Cell Physiol. Biochem ., 28, 911– 922. Google Scholar CrossRef Search ADS PubMed  84 Dias-Santagata D., Fulga T.A., Duttaroy A., Feany M.B. ( 2007) Oxidative stress mediates tau-induced neurodegeneration in Drosophila. J. Clin. Invest ., 117, 236– 245. Google Scholar CrossRef Search ADS PubMed  85 Mackintosh C. ( 2004) Dynamic interactions between 14-3-3 proteins and phosphoproteins regulate diverse cellular processes. Biochem. J ., 381, 329– 342. Google Scholar CrossRef Search ADS PubMed  86 Sluchanko N., Gusev N. ( 2017) Moonlighting chaperone-like activity of the universal regulatory 14-3-3 proteins. FEBS J ., 284, 1279– 1295. Google Scholar CrossRef Search ADS PubMed  87 Boston P.F., Jackson P., Thompson R.J. ( 1982) Human 14-3-3 protein: radioimmunoassay, tissue distribution, and cerebrospinal fluid levels in patients with neurological disorders. J. Neurochem ., 38, 1475– 1482. Google Scholar CrossRef Search ADS PubMed  88 Lin D.M., Goodman C.S. ( 1994) Ectopic and increased expression of Fasciclin II alters motoneuron growth cone guidance. Neuron , 13, 507– 523. Google Scholar CrossRef Search ADS PubMed  89 Robinow S., White K. ( 1988) The locus elav of Drosophila melanogaster is expressed in neurons at all developmental stages. Dev. Biol ., 126, 294– 303. Google Scholar CrossRef Search ADS PubMed  90 Wittmann C.W. ( 2001) Tauopathy in Drosophila: neurodegeneration without neurofibrillary tangles. Science , 293, 711– 714. Google Scholar CrossRef Search ADS PubMed  91 Chen H.-K., Fernandez-Funez P., Acevedo S.F., Lam Y.C., Kaytor M.D., Fernandez M.H., Aitken A., Skoulakis E.M.C., Orr H.T., Botas J., Zoghbi H.Y. ( 2003) Interaction of Akt-phosphorylated ataxin-1 with 14-3-3 mediates neurodegeneration in spinocerebellar ataxia type 1. Cell , 113, 457– 468. Google Scholar CrossRef Search ADS PubMed  92 Tsigkari K., Acevedo S., Skoulakis E. ( 2012) 14-3-3ɛ is required for germ cell migration in Drosophila. PLoS One , 7, e36702. Google Scholar CrossRef Search ADS PubMed  93 Philip N., Acevedo S., Skoulakis E.M.C. ( 2001) Conditional rescue of olfactory learning and memory defects in mutants of the 14-3-3z gene leonardo. J. Neurosci ., 21, 8417– 8425. Google Scholar CrossRef Search ADS PubMed  94 Brand A.H., Perrimon N. ( 1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development , 118, 401– 415. Google Scholar PubMed  95 Skoulakis E.M.C., Davis R.L. ( 1996) Olfactory learning deficits in mutants for leonardo, a Drosophila gene encoding a 14-3-3 protein. Neuron , 17, 931– 944. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Human Molecular GeneticsOxford University Press

Published: Apr 12, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off