Spermidine/spermine-N1-acetyltransferase ablation impacts tauopathy-induced polyamine stress response

Abstract

Background: Tau stabilizes microtubules; however, in Alzheimer's disease (AD) and tauopathies, tau becomes hyperphosphorylated, aggregates, and results in neuronal death. Our group recently uncovered a unique interaction between polyamine metabolism and tau fate. Polyamines exert an array of physiological effects that support neuronal function and cognitive processing. Specific stimuli can elicit a polyamine stress response (PSR), resulting in altered central polyamine homeostasis. Evidence suggests that elevations in polyamines following a short-term stressor are beneficial; however, persistent stress and subsequent PSR activation may lead to maladaptive polyamine dysregulation, which is observed in AD, and may contribute to neuropathology and disease progression.

Methods: Male and female mice harboring tau P301L mutation (rTg4510) were examined for a tau-induced central polyamine stress response (tau-PSR). The direct effect of tau-PSR byproducts on tau fibrillization and oligomerization were measured using a thioflavin T assay and a N2a split superfolder GFP-Tau (N2a-ssGT) cell line, respectively. To therapeutically target the tau-PSR, we bilaterally injected caspase 3-cleaved tau truncated at aspartate 421 (AAV9 Tau ΔD421) into the hippocampus and cortex of spermidine/spermine-N¹-acetyltransferase (SSAT), a key regulator of the tau-PSR, knock out (SSAT-/-), and wild type littermates, and the effects on tau neuropathology, polyamine dysregulation, and behavior were measured. Lastly, cellular models were employed to further examine how SSAT repression impacted tau biology.

Results: Tau induced a unique tau-PSR signature in rTg4510 mice, notably in the accumulation of acetylated spermidine. In vitro, higher-order polyamines prevented tau fibrillization but acetylated spermidine failed to mimic this effect and even promoted fibrillization and oligomerization. AAV9 Tau ΔD421 also elicited a unique tau-PSR in vivo, and targeted disruption of SSAT prevented the accumulation of acetylated polyamines and impacted several tau phospho-epitopes. Interestingly, SSAT knockout mice presented with altered behavior in the rotarod task, the elevated plus maze, and marble burying task, thus highlighting the impact of polyamine homeostasis within the brain.

Conclusion: These data represent a novel paradigm linking tau pathology and polyamine dysfunction and that targeting specific arms within the polyamine pathway may serve as new targets to mitigate certain components of the tau phenotype.

Keywords: Alzheimer’s disease; Hippocampus; Polyamine dysregulation; Tau.

Fig. 1

Tau rTg4510 mice show significant polyamine dysregulation. a Simplified polyamine pathway. Ornithine decarboxylase antizyme 1 (OAZ1), ornithine decarboxylase (ODC), spermidine synthase (SRM), spermidine synthase (SMS), spermine oxidase (SMOX), spermidine/spermine-N¹-acetyltransferase (SSAT), polyamine oxidase (PAOX), polyamine-modulated factor 1 (PMF1), polyamine modulated factor binding protein 1 (PMFBP1). b, c Representative images and quantification of western blot analysis of 12-month-old nTg and rTg4510 (n = 5) hippocampal tau neuropathology (pSer199/202: t (4⁺) = − 7.490, p = .002; pSer356: t (4⁺) = − 4.588, p = .010; and total tau (H150): t (4⁺) = − 4.201, p = .014), followed by quantification (n = 5). Independent sample t test with ⁺Levene’s test for equality of variance correction, *p < .05. Data is represented by means ± S.E.M. d–h Representative images and quantification of immunohistochemical and western blot analysis of 14-month-old nTg and rTg4510 (n = 2–4) cortical and hippocampal C-terminally truncated tau neuropathology (tau ΔD421). Data is represented by means ± S.E.M. i–j Representative images and quantification of western blot analysis of 12-month-old nTg and rTg4510 (n = 4–5) polyamine dysregulation (ODC: t (4.843⁺) = 4.320, p = .008; SRM: t (5.347⁺) = − 3.756, p = .012; SMS: t(8) = 4.412, p = .003; SSAT: t(7) = − 2.849, p = .025; SMOX: t (4.399⁺) = − 2.958, p = .037; PMF1: t (4.113⁺) = − 2.690, p = .053; PMFBP1: t(8) = 3.850, p = .005). Independent sample t test with ⁺Levene’s test for equality of variance correction, *p < .05. Data is represented by means ± S.E.M.

Fig. 2

a, b Polyamine (putrescine (PTS: t(13) = − 7.687, p = .000), spermidine (SPD: t (13) = − .662, p = .520), spermine (SPM: t(13) = − 1.133, p = .278)) and acetylated polyamine (acetylputrescine (AcPTS: t(13) = 1.848, p = .087), acetylspermidine (AcSPD: t(13) = − 3.238, p = .006), acetylspermine (AcSPM: t(13) = − .052, p = .959)) quantification of brain homogenates from 8-month-old nTg and rTg4510 mice (n = 8). Independent samples t test *p < .05. Data is represented by means ± S.E.M.

Fig. 3

Polyamines and acetylated polyamines differentially impact tau fibrillization and acetylspermidine increases tau oligomerization. a–c Thioflavin T assay using recombinant 4R0N (P301L) tau and treatment of polyamines (putrescine, spermidine, spermine) or acetylated polyamines (acetylputrescine, acetylspermidine, acetylspermine) followed by quantification of area under the curve (AUC) compared to tau + vehicle controls. Data is represented as triplicate average and tables reflect AUC change from tau + vehicle control (% tau, % change). d, e Characterization and validation of split GFP-Tau individual plasmids (pmGFP10C-Tau, pmGFP11C-Tau) and monoclonal cell line (N2a-ssGT). f, g Representative images (at 24 h) and quantification of tau oligomerization following treatment with acetylspermidine. Simple main effects analysis showed that 30 μM acetylspermidine increased tau oligomerization (F(1, 78) = 7.140, p = .009. No effect of time (F(2,78) = .719, p = .490) or interaction of variables (F(2, 78) = .144, p = .866) was detected. 2 × 3 Factorial analysis of variance (ANOVA), followed by post hoc comparisons using Fishers PLSD. *p < .05, the asterisk inidcates the main effect of drug. Data is represented by means ± S.E.M.

Fig. 4

AAV9 Tau ΔD421 produces hippocampal inflammation and targeted disruption of SSAT increases NeuN expression and reduces tau neuropathology. a–d Representative images and quantification of immunohistochemical analysis of hippocampal and cortical inflammation (Iba1) and NeuN expression in response to 4-month incubation of either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 in 15-month-old nTg and SSAT-/- mice (n = 7–11). Iba1: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased inflammation (Iba1) in only the hippocampus (F(1, 36) = 6.823, p = .013). Further, within each genotype, pairwise comparisons revealed a significant difference in hippocampal Iba1 between only the SSAT-/- AAV9 empty capsid and SSAT-/- AAV9 Tau ΔD421 groups (p = .048). NeuN: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly decreased NeuN expression (NeuN) in only the hippocampus (F(1, 35) = 8.973, p = .005). Interestingly, simple main effects analysis also showed that SSAT-/- mice had significantly increased NeuN expression (NeuN) in both the hippocampus (F(1, 39) = 10.343, p = .003) and the cortex (F(1, 35) = 8.100, p = .007), relative to non-transgenic litter-mates. Further, within each genotype, pairwise comparisons revealed a significant difference in hippocampal NeuN between only the SSAT-/- AAV9 empty capsid and SSAT-/- AAV9 Tau ΔD421 groups (p = .013). e–h Representative images and quantification of immunohistochemical analysis of hippocampal (CA3) and anterior cortex (ACX) tau neuropathology in response to 4-month incubation of either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 in 15-month old nTg and SSAT-/- mice (n = 7–11). HT7: Simple main effect analysis showed that AAV9 Tau ΔD421 significantly increased total tau (HT7) in both the CA3 of the hippocampus (F(1, 34) = 55.939, p = .000) and the anterior cortex (ACX; F(1, 35) = 57.020, p = .000). Importantly, the level of total tau was not significantly different between nTg and SSAT-/- mice, ensuring treatment was equal across groups. pSer199/202: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased Tau pSer199/202 in the CA3 (F(1, 36) = 29.869, p = .000) and the ACX (F(1, 34) = 36.192, p = .000). Interestingly, in the CA3, there was also a main effect of genotype (F(1, 36) = 4.114, p = .05), and pairwise comparison revealed a significant difference between nTg AAV9 Tau ΔD421 and SSAT-/- AAV9 Tau ΔD421 mice (p = .010), identifying a protective effect of SSAT disruption in CA3 Tau pSer199/202. Further, in the ACX, there was a significant interaction of factors (F(1, 34) = 3.994, p = .05), and pairwise comparison again revealed a significant difference between nTg AAV9 Tau ΔD421 and SSAT-/- AAV9 Tau ΔD421 mice (p = 0.17), identifying a protective effect of SSAT disruption in ACX Tau pSer 199/202. AT8: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased phosphorylated paired helical filament tau (PHF; AT8) in both the CA3 (F(1, 36) = 32.677, p = .000) and ACX (F(1, 33) = 22.792, p = .000). No effect of genotype or interaction of factors was detected. 2 × 2 Factorial analysis of variance (ANOVA), followed by pairwise comparisons using Fisher’s PLSD. *p < .05; asterisks indicate the main effect of treatment and main effect of genotype, and ampersands indicate the interaction of genotype and treatment. Data is represented by means ± S.E.M.

Fig. 5

AAV9 Tau ΔD421 produces robust hippocampal tau neuropathology; SSAT disruption prevents the accumulation of high molecular weight tau phospho-epitopes. a–i Representative images and quantification of western blot analysis of hippocampal tau neuropathology, exogenous Tau ΔD421, exogenous total tau (HT7), exogenous and endogenous total tau (Tau-5), pS199/202, tau-paired helical filament (Tau PHF; AT8), and pS396, in response to 4-month incubation of either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 in 15-month-old nTg and SSAT-/- mice (n = 3–5). Tau ΔD421: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased Tau ΔD421 (F(1, 16) = 24.925, p = .000). Importantly, the level of Tau ΔD421 was not significantly different between nTg and SSAT-/- mice, ensuring treatment was equal across groups. No main effect of genotype or interaction of factors was detected on Tau ΔD421. Total tau (HT7): Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased total tau (HT7) (F(1, 16) = 29.585, p = .000). Importantly, the level of total tau was not significantly different between nTg and SSAT-/- mice, ensuring treatment was equal across groups. No main effect of genotype or interaction of factors was detected on total tau (HT7). Tau-5: An interaction of factors was detected (F(1, 16) = 4.657, p = .046) on Tau-5, indicating the levels of Tau-5 are dependent on genotype, with targeted SSAT disruption decreasing Tau-5 levels. Tau pS199/202: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased monomeric (F(1, 16) = 26.621, p = .000) and high molecular weight (HMW; F(1, 16) = 5.143, p = .038) Tau pS199/202. No main effect of genotype or interaction of factors was detected on pS199/202. AT8: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased Tau PHF (AT8; F(1, 16) = 27.482, p = .000). No main effect of genotype or interaction of factors was detected on AT8. Tau pS396: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased monomeric (F(1, 16) = 15.084, p = .001) Tau pS396; however, pairwise comparisons revealed that AAV9 Tau ΔD421 only significantly increased monomeric pS396 in the nTg genotype (p = .003) and not in the SSAT-/- genotype (p = .073) indicating an impact of SSAT on monomeric Tau pS396. This notion is supported by the significant pairwise comparison between nTg AAV9 Tau ΔD421 and SSAT-/-0 AAV9 Tau ΔD421 groups (p = .048). Further, there was an interaction of factors on HMW Tau pS396 (F(1, 16) = 25.183, p = .000), indicating the levels of HMW Tau pS396 are dependent on genotype, with targeted SSAT disruption decreasing HMW pS396 levels. This is supported by the lack of significant pairwise comparison between SSAT-/- AAV9 empty capsid and SSAT-/- AAV9 Tau ΔD421 (p = .09), and the significant pairwise comparison between nTg AAV9 Tau ΔD421 and SSAT-/- AAV9 Tau ΔD421, again with SSAT-/- decreasing HMW pS396 levels. 2 × 2 Factorial analysis of variance (ANOVA), followed by pairwise comparisons using Fishers PLSD. *p < .05; asterisks indicate the main effect of treatment and main effect of genotype, ampersands indicate the interaction of genotype and treatment, and number signs indicate the pairwise comparison within genotype. Data is represented by means ± S.E.M.

Fig. 6

AAV9 AAV9 Tau ΔD421 induces tau-PSR; SSAT disruption prevents polyamine dysregulation. a–h Representative images and quantification of western blot analysis of hippocampal polyamine dysregulation (ODC, OAZ1, SRM, SMS, SMOX, PMF1, and PMFBP1) in response to 4-month incubation of either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 in 15-month-old nTg and SSAT-/- mice (n = 4–5). ODC: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly decreased ornithine decarboxylase (ODC) (F(1, 16) = 13.164, p = .002). Further, within each genotype, pairwise comparisons revealed a significant reduction of ODC by AAV9 Tau ΔD421 in SSAT-/- mice (p = .003); an effect that was absent in non-transgenic mice. No main effect of genotype (F(1, 16) = 2.358, p = .144) or interaction of factors (F(1, 16) = 1.511, p = .237) was detected on ODC. OAZ1: No main effect of genotype (F(1, 16) = .040, p = .844), AAV9 Tau ΔD421 (F(1, 16) = 3.430, p = .083), or interaction of factors (F(1, 16) = .388, p = .542) was detected in OAZ1. SRM: No main effect of genotype (F(1, 16) = 4.035, p = .062), AAV9 Tau ΔD421 (F(1, 16) = .064, p = .804), or interaction of factors (F(1, 16) = .300, p = .592) was detected in SRM. SMS: Simple main effects analysis showed that AAV9 Tau ΔD421 significantly decreased spermine synthase (SMS) (F(1, 15) = 23.066, p = .000). No effect of genotype (F(1, 15) = 1.918, p = .186) or interaction of factors (F(1, 15) = .850, p = .371) was detected on SMS. SMOX: No main effect of genotype (F(1, 16) = .519, p = .482), AAV9 Tau ΔD421 (F(1, 16) = 2.604, p = .126), or interaction of factors (F(1, 16) = .121, p = .733) was detected in SMOX. PMF1: Simple main effects analysis showed that SSAT disruption significantly increased polyamine-modulated factor 1 (PMF1) (F(1, 16) = 4.522, p = .049). No main effect of AAV9 Tau ΔD421 (F(1, 16) = .114, p = .740) or interaction of factors (F(1, 16) = .000, p = .993) was detected in PMF1. PMFBP1: Simple main effects analysis showed that SSAT disruption significantly increased polyamine-modulated factor binding protein 1 (PMFBP1) (F(1, 16) = 6.499, p = .021). Further, within each genotype, while there was no simple main effect of AAV9 Tau ΔD421 (F (1,16) = 3.248, p = .090), pairwise comparison revealed a significant difference in PMFBP1 between only the SSAT-/- AAV9 empty capsid and SSAT-/- AAV9 Tau ΔD421 groups (p = .017), identifying a SSAT-dependent effect of AAV9 Tau ΔD421 on PMFBP1. There was no significant interaction of factors (F(1, 16) = 3.786, p = .069) on PMFBP1. 2 × 2 Factorial analysis of variance (ANOVA), followed by pairwise comparisons using Fishers PLSD. *p < .05; asterisks indicate the main effect of treatment and main effect of genotype, number signs indicate the pairwise comparison within genotype. Data is represented by means ± S.E.M.

Fig. 7

Polyamine quantification (putrescine, spermidine, acetylputrescine, acetylspermidine) of the brain homogenates in response to 4-month incubation of either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 in 15-month-old nTg and SSAT-/- mice (n = 9–11). a Putrescine: Simple main effects analysis showed that SSAT disruption significantly reduced putrescine (F(1, 35) = 332.618, p = .000). Further, within each genotype, pairwise comparison revealed a trend in difference in putrescine between only the nTg AAV9 empty capsid and nTg AAV9 Tau ΔD421 groups (p = .061). No main effect of AAV9 Tau ΔD421 (F(1, 35) = 1.698, p = .201) or interaction of factors (F(1, 35) = 1.960, p = .170) was detected on putrescine. b Acetylputrescine: Simple main effects analysis showed that SSAT disruption significantly reduced acetylputrescine (F(1, 37) = 57.551, p = .000). No main effect of AAV9 Tau ΔD421 (F(1, 37) = .075, p = .785) or interaction of factors (F(1, 37) = .008, p = .928) was detected on acetylputrescine. c Spermidine: Simple main effects analysis showed that SSAT disruption significantly reduced spermidine (F(1, 37) = 4.736, p = .036). No main effect of AAV9 Tau ΔD421 (F(1, 37) = .001, p = .970) or interaction of factors (F(1, 37) = 2.009, p = .165) was detected on spermidine. d Acetylspermidine: Simple main effects of genotype (F(1, 37) = 27.723, p = .000), AAV9 Tau ΔD421 (F(1, 37) = 11.661, p = .002) and interaction of factors (F(1, 37) = 6.539, p = .015) was detected on acetylspermidine. e Spermine: No main effect of genotype (F(1, 35) = 2.012, p = .165), AAV9 Tau ΔD421 (F(1, 35) = .236, p = .630), or interaction of factors (F(1, 35) = .296, p = .590) was detected on Spermine. f Acetylspermine: Simple main effects analysis showed that SSAT disruption significantly increased acetylspermine (F(1, 38) = 9.360, p = .004). No main effect of AAV9 Tau ΔD421 (F(1, 38) = .004, p = .949) or interaction of factors (F(1, 38) = 2.439, p = .127) was detected on acetylspermine. g SSAT mRNA: mRNA expression of SSAT in response to 4-month incubation of either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 in 15-month old nTg and SSAT-/- mice (n = 3–5). Simple main effects showed that SSAT disruption significantly reduced SSAT expression, normalized to GAPDH, (F(1, 12) = 20.660, p = .001. No effect of AAV9 Tau ΔD421 (F(1, 12) = .687, p = .423) or interaction of factors (F(1, 12) = .694, p = .421) was detected on SSAT expression. 2 × 2 Factorial analysis of variance (ANOVA), followed by pairwise comparisons using Fishers PLSD. *p < .05; asterisks indicate the main effect of treatment and main effect of genotype, ampersands indicate the interaction of genotype and treatment, and number signs indicate the pairwise comparison within genotype. Data is represented by means ± S.E.M.

Fig. 8

AAV9 Tau ΔD421 produces cognitive impairment; SSAT disruption produces behavioral phenotype. a Radial arm water maze errors for per block and per day of 15-month-old nTg and SSAT-/- mice administered either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 (n = 9–11). Simple main effects analysis showed that AAV9 Tau ΔD421 significantly increased errors (F(1, 36) = 5.204, p = .029) on day 2 of the radial arm water maze (RAWM). No effect of genotype (F(1, 36) = 1.521, p = .225), AAV9 Tau ΔD421 (F(1, 36) = 3.196, p = .082), or interaction of factors (F(1, 36) = .392, p = .535) was found on errors on day 1 of RAWM. No effect of genotype (F(1, 36) = .051, p = .822) or interaction of factors (F(1, 36) = .525, p = .473) was found on errors on day 2 of RAWM. Repeated measures mixed analysis of variance (ANOVA), *p < .05. Data is represented by means ± S.E.M. Asterisks indicate the main effect of treatment. b Rotarod performance latency per trial and per day of 15-month-old nTg and SSAT-/- mice administered either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 (n = 9–11). Simple main effects analysis showed that SSAT disruption significantly increased latency to fall on day 1 (F(1, 37) = 14.164, p = .001) and day 2 (F(1, 37) = 19.626, p = .000) of the rotarod task. No effect of AAV9 Tau ΔD421 on day 1 or day 2 (F(1, 37) = .779, p = .383; F(1, 37) = 2.985, p = .092), respectively), or interaction of factors on day 1 or day 2 (F(1, 37) = .080, p = .779; F(1, 37) = .149, p = .702, respectively) was detected on rotarod latency to fall. Repeated measures mixed analysis of variance (ANOVA), *p < .05. Data is represented by means ± S.E.M. Asterisks indicate the main effect of genotype. c–h Behavioral assessment in the open field, elevated plus maze, and marble burying tasks of 15-month-old nTg and SSAT-/- mice administered either AAV9 empty capsid (EC) or AAV9 Tau ΔD421 (n = 9–11). 2 × 2 Factorial analysis of variance (ANOVA), followed by pairwise comparisons using Fishers PLSD. *p < .05; Asterisks indicate the main effect of genotype and number signs indicate the pairwise comparison within genotype. Data is represented by means ± S.E.M. c–e Open field: Simple main effects analysis showed no effect of SSAT disruption on total distance traveled (F(1, 38) = 1.488, p = .230), number of entries to the center zone (F(1, 38) = 1.825, p = .185), or time in the center zone (F(1, 38) = .001, p = .970). While there was no effect of AAV9 Tau ΔD421 on total distance traveled (F(1, 38) = .844, p = .364) or number of entries to the center zone (F(1, 38) = .007, p = .935), there was a main effect of treatment on time in the center zone in that AAV9 Tau ΔD421 significantly decreased time in the center zone (F(1, 38) = 5.065, p = .030); however, pairwise comparisons revealed this effect was only present in SSAT-/- mice (p = .01), suggesting a modest interaction of genotype and treatment on anxiety-like behavior as measured by this task. f, g EPM: Simple main effects analysis showed that SSAT disruption significantly increased entries to the open arm (F(1, 38) = 11.894, p = .001) and time in the open arm (F(1, 38) = 4.892, p = .033). No main effect of AAV9 Tau ΔD421 (F(1, 3) = .014, p = .908; F(1, 38) = .001, p = .982, respectively) or interaction of factors (F(1, 38) = .006, p = .937; F(1, 38) = 1.070, p = .307, respectively) was detected on EPM open arm entries and open arm time. h Marble burying: Simple main effects analysis showed that SSAT disruption significantly increased percent of marbles buried (F(1, 37) = 20.576, p = .000). No main effect of AAV9 Tau ΔD421 (F(1, 37) = 3.034, p = .090) or interaction of factors (F(1, 37) = .029, p = .866) was detected on percent marbles buried. i–k Comparing nTg and SSAT-/- mice (n = 6), normal input/output curve at Schaffer collateral/CA1 synapses was observed. Nonlinear regression, line fits were compared using the extra sum-of-squares F test for the Fiber volley (p = 0.4907) and fEPSP (p = 0.1019), respectively. No difference was seen in short-term presynaptic plasticity measured by paired-pulse facilitation (PPF), (F(1, 48) = 0.010, p = 0.920.) LTP was induced in nTg and SSAT-/- slices using a theta-burst stimulation protocol and no significant differences were seen in the induction or maintenance of LTP in SSAT-/- compared to nTg slices (F(1, 12) = 0.642, p = 0.439)

Fig. 9

siRNA repression of SAT1 reduces tau pathology in vitro. a Representative images and quantification of western blot analysis of C3 HeLa cells stably overexpressing tau (4R0N) following transfection with siRNA scramble negative control or siRNA SAT1 (n = 3). b One way analysis of variance (ANOVA) revealed an effect of group on monomeric Tau pSer356 (F(3, 8) = 4.743, p = .035), and post hoc comparisons revealed a significant decrease in response to siRNA SAT1 with and without DSPM (p = .029, p = .010, respectively), compared to siRNA scrambled control. c One way analysis of variance (ANOVA) revealed an effect of group on monomeric Tau pSer396 (F(3, 8) = 15.829, p = .001), and post hoc comparisons revealed a significant decrease in response to siRNA SAT1 with and without DSPM (p = .003, p = .001, respectively), compared to siRNA scrambled control. d One way analysis of variance (ANOVA) revealed no effect of group on monomeric total tau (F(3, 8) = 2.236, p = .161); however, post hoc comparisons revealed a significant decrease in response to siRNA SAT1 without DSPM (p = .05), compared to siRNA scrambled control. One way analysis of variance (ANOVA), followed by post hoc comparisons using Fishers PLSD, *p < .05. Data is represented by means ± S.E.M.