Simufilam suppresses overactive mTOR and restores its sensitivity to insulin in Alzheimer's disease patient lymphocytes

Abstract

Introduction: Implicated in both aging and Alzheimer's disease (AD), mammalian target of rapamycin (mTOR) is overactive in AD brain and lymphocytes. Stimulated by growth factors such as insulin, mTOR monitors cell health and nutrient needs. A small molecule oral drug candidate for AD, simufilam targets an altered conformation of the scaffolding protein filamin A (FLNA) found in AD brain and lymphocytes that induces aberrant FLNA interactions leading to AD neuropathology. Simufilam restores FLNA's normal shape to disrupt its AD-associated protein interactions. Methods: We measured mTOR and its response to insulin in lymphocytes of AD patients before and after oral simufilam compared to healthy control lymphocytes. Results: mTOR was overactive and its response to insulin reduced in lymphocytes from AD versus healthy control subjects, illustrating another aspect of insulin resistance in AD. After oral simufilam, lymphocytes showed normalized basal mTOR activity and improved insulin-evoked mTOR activation in mTOR complex 1, complex 2, and upstream and downstream signaling components (Akt, p70S6K and phosphorylated Rictor). Suggesting mechanism, we showed that FLNA interacts with the insulin receptor until dissociation by insulin, but this linkage was elevated and its dissociation impaired in AD lymphocytes. Simufilam improved the insulin-mediated dissociation. Additionally, FLNA's interaction with Phosphatase and Tensin Homolog deleted on Chromosome 10 (PTEN), a negative regulator of mTOR, was reduced in AD lymphocytes and improved by simufilam. Discussion: Reducing mTOR's basal overactivity and its resistance to insulin represents another mechanism of simufilam to counteract aging and AD pathology. Simufilam is currently in Phase 3 clinical trials for AD dementia.

Keywords: Alzheimer’s disease; PTEN; aging; filamin A; insulin resistance; mTOR.

FIGURE 1

Simufilam treatment attenuated basal mTORC1 overactivation and restored its response to insulin in lymphocytes from AD subjects. Representative blot images from two AD and five healthy control subjects (A,E) show increased basal activation of mTORC1 by elevated levels of mTOR phosphorylation at S²⁴⁴⁸ (A) and by increased basal activation of mTORC1’s downstream kinase p70S6K, indicated by phosphorylation at T¹³⁸⁹ (E), in AD subjects versus healthy controls. Blots were analyzed by densitometric quantitation (B,D,F). With heightened basal activity of the mTORC1 signaling pathway in AD subject lymphocytes, its response to in vitro insulin was blunted. The elevated basal p70S6K, still lower than insulin-stimulated control levels, also showed a weak response to insulin. Oral simufilam treatment decreased the elevated basal mTORC1 activation and restored its response to insulin across both measures, with insulin-stimulated p70S6K levels slightly higher than untreated healthy controls on Day 28. The simufilam treatment effect is also seen in the dramatic improvement in percent stimulation by insulin (C,G). mTOR association with Raptor was lower in healthy controls but unaffected by insulin or simufilam (D). Data are means ± SEM. N = 13 for AD, 18 for control. *p < 0.05, ***p < 0.001, vs. respective value in healthy control by two-sided two-sample t-test; ^### p < 0.001 vs. respective AD value at Day 0 by two-sided paired t-test.

FIGURE 2

Simufilam treatment attenuated basal mTORC2 overactivation and restored its response to insulin in lymphocytes from AD subjects. Representative blot images from two AD and five healthy control subjects (A,E) show increased basal mTORC2 signaling by elevated levels of mTOR phosphorylation at S²⁴⁸¹ (A) and by increased phosphorylation of the mTORC2 component Rictor at T¹¹³⁵ (E) in AD subjects versus healthy controls. Blots were analyzed by densitometric quantitation (B,D,F). As with mTORC1 signaling, the heightened basal mTORC2 signaling in AD subjects left little room for insulin stimulation. The elevated levels of pT¹¹³⁵Rictor, still lower than insulin-stimulated control levels, again showed a weak response to insulin. Oral simufilam treatment decreased the elevated basal mTORC2 signaling and restored its response to insulin on both measures. Insulin-stimulated p S²⁴⁸¹mTOR was slightly higher on Day 28 than in untreated healthy controls. Percent stimulation of these mTORC2 signaling components by insulin improved dramatically after simufilam treatment (C,G). mTOR association with Rictor was unchanged in AD and unaffected by insulin or simufilam (D). Data are means ± SEM. N = 13 for AD, 18 for control. *p < 0.05, **p < 0.01, ***p < 0.001 vs. respective value in healthy control by two-sided two-sample t-test; ^## p < 0.01, ^### p < 0.001 vs. respective AD value at Day 0 by two-sided paired t-test.

FIGURE 3

Simufilam treatment attenuates mTORC2-mediated Akt1 activation. Akt1 is activated by mTORC2 signaling by phosphorylation on S⁴⁷³ to further activate mTORC1. Representative blot images from two AD and five healthy control subjects (A) show increased basal Akt1 activation, as indicated by its phosphorylation at S⁴⁷³, in AD subjects versus healthy controls. Blots were analyzed by densitometric quantitation (B). Showing a similar pattern to downstream effectors p70S6K and pT¹¹³⁵Rictor, basal Akt1 activity was elevated in AD subjects without nearing insulin-stimulated control levels, and its response to insulin was still blunted. Oral simufilam treatment decreased the elevated Akt1 activation and restored its response to insulin to a level not different from healthy control. Percent stimulation of Akt1 by insulin improved dramatically after simufilam treatment (C). Data are means ± SEM. N = 13 for AD, 18 for control. *p < 0.05, ***p < 0.001 vs. respective value in healthy control by two-sided two-sample t-test; ^# p < 0.05, ^### p < 0.001 vs. respective AD value at Day 0 by two-sided paired t-test.

FIGURE 4

Simufilam treatment restores the insulin-stimulated reduction in FLNA linkage to insulin receptor beta (IRβ). Representative blot images from two AD and four healthy control subjects (A) show higher levels of the FLNA—IRβ interaction and impaired dissociation of this linkage upon insulin stimulation in AD subjects versus healthy controls. Simufilam treatment improved the insulin-stimulated dissociation of FLNA—IRβ interaction. Blots were analyzed by densitometric quantitation (B). Percent inhibition by insulin in AD subjects improved dramatically with simufilam treatment (C). Data are means ± SEM. N = 13 for AD, 18 for control. ***p < 0.001 vs. respective value in healthy control by two-sided two-sample t-test; ^### p < 0.001 vs. respective AD value at Day 0 by two-sided paired t-test; +++p < 0.001 insulin vs. basal value at respective timepoint by two-sided paired t-test.

FIGURE 5

Elevated pS²¹⁵²FLNA and reduced FLNA—PTEN linkage in lymphocytes of AD subjects were improved by oral simufilam. Increased levels of FLNA phosphorylation at S²¹⁵² were coincident with a reduced FLNA—PTEN linkage in AD versus age-matched control lymphocytes. Exogenous Aβ₄₂ increased S²¹⁵²FLNA to AD-like levels but did not reduce the FLNA—PTEN linkage in control lymphocytes. Aβ₄₂ had no further effect on AD lymphocytes. [(A) representative blot images; (B) densitometric quantitation of blot images]. Oral simufilam to AD subjects reduced the high levels of S²¹⁵²FLNA and improved PTEN tethering to FLNA in their lymphocytes [(C) representative blot images; (D) densitometric quantitation of blot images]. Data are means ± SEM. N = 4 for AD and age-matched healthy control (A,B), and N = 13 for AD patients treated with simufilam (C,D). *p < 0.05 vs. vehicle by two-sided paired t-test (B); ^# p < 0.05, ^## p < 0.01 vs. control by two-sided two-sample t-test (B); ***p < 0.001, **p < 0.01 vs. Day 0 by two-sided paired t-test (D).

FIGURE 6

Simufilam reduced S²¹⁵² phosphorylation of FLNA in postmortem human AD and amnestic MCI brain tissue. FLNA is phosphorylated at S²¹⁵² in postmortem brain tissue of AD and amnestic MCI subjects compared to non-amnestic MCI and age-matched healthy control subjects [(A) representative immunoblot images; (B) densitometric quantitation of blot images]. Ex vivo incubation with simufilam (1 nM) reduced levels of pS²¹⁵²FLNA in AD and amnestic MCI with no effect on the already low levels in non-amnestic MCI and age-matched healthy control brain tissue. In contrast, incubation with Aβ₄₂ increased pS²¹⁵²FLNA levels in non-amnestic MCI and age-matched healthy control subjects but did not further increase the elevated levels in AD or amnestic MCI. Vertical lines indicate separation between 2 blots (9 lanes of one blot with Controls, Amnestic MCI and Non-amnestic MCI and 3 lanes of another blot with AD). Data are means ± SEM. N = 6. **p < 0.01 vs. Kreb’s-Ringer (K-R) in control group by two-sided two-sample t-test; ^## p < 0.01 vs. K-R in respective group by two-sided paired t-test.