|Targeting Truncation-activated GSK-3β to Beat Alzheimer's Disease||Papers|
Glycogen synthase kinase-3 (GSK-3) is a protein kinase for over 100 different proteins in various signaling pathways. Virtually all age-related diseases involve hyperactive GSK-3, including Alzheimer's disease (AD) (Sayas and Ávila, 2021), heart disease (Lal et al., 2015), kidney disease (Ardalan et al., 2020), diabetes (MacAulay and Woodgett, 2008), cancer (He et al., 2020) and major depressive disorder (Duda et al., 2020). The natural substance, curcumin, is a potent GSK-3 inhibitor (Bustanji et al., 2009). Clinical trials have demonstrated its beneficial effects in cancer (Kong et al., 2021), diabetes (Pivari et al., 2019), cardiovascular disease (Singh et al., 2021) chronic kidney disease (Bagherniya et al., 2021) and major depressive disorder (Fusar-Poli et al., 2020). Surprisingly, the efficacy of curcumin for AD is limited and inconclusive (Sarker and Franks, 2018; Kuszewski et al., 2018). There was even a trend for curcumin-treated subjects to do worse than placebo-treated subjects on the Mini-Mental State Examination (Zhu et al., 2019).
In AD, hyperphosphorylation of the Tau protein by GSK-3 plays a central role. One would expect GSK-3 inhibitors to produce robust benefits. Unfortunately, therapeutic development along this line was disappointing. A series of highly selective and potent GSK-3 inhibitors have failed in pre-clinical assessment. (Cormier and Woodgett, 2017). Tideglusib is the only GSK-3 inhibitor that reached phase II clinical trials, but produced no clinical benefit for patients with mild to moderate AD (Lovestone et al., 2015).
This article will explain why both curcumin and Tideglusib are not effective for AD. GSK-3 remains a promising therapeutic target. However, GSK-3 can be activated through two entirely different mechanisms: dephosphorylation and truncation. For the treatment of AD, the inhibitors should target the truncation-activated GSK-3β.
Dephosphorylation- vs. Truncation-activated GSK-3β
GSK-3 has two isoforms: GSK-3α and GSK-3β. The main kinase that phosphorylates Tau protein is GSK-3β which consists of 420 amino acids. GSK-3β is constitutively active. It becomes inactivated upon phosphorylation at serine 9 (Ser9). Thus, the GSK-3β activity depends on the phosphorylation status at Ser9 which is regulated by a number of signaling pathways such as BDNF/TrkB/Akt. Activation of Akt inactivates GSK-3β via Ser9 phosphorylation.
Alternatively, high level of Ca2+ ions may activate calpain 1 to cleave GSK-3β between Ser381 and Ser382. The truncated GSK-3β becomes persistently active, independent of the phosphorylation status at Ser9 (Jin et al., 2015; Feng et al., 2013).
The binding site of curcumin to GSK-3β is not known. Tideglusib acts as an allosteric inhibitor, meaning that it does not bind to the substrate binding site (Balasubramaniam et al., 2020; Rippin and Eldar-Finkelman, 2021). In GSK-3β, the C-terminal serves as a regulatory domain. Removal of this domain by calpain 1 increases GSK-3β activity. It seems that both curcumin and Tideglusib cannot inhibit this activation process.
The BDNF deficiency could promote Ser9 dephosphorylation via the BDNF/TrkB/Akt pathway, thereby increasing GSK-3β activity. However, this does not seem to be the major mechanism in the pathogenic cascade of AD, for otherwise both curcumin and Tideglusib would be able to produce significant beneficial effects. In addition, calpain plays an important role in the pathogenesis of AD. Upregulation of calpain activity has been demonstrated to precede Tau hyperphosphorylation during the progression of AD (Kurbatskaya et al., 2016). Truncated GSK-3β has also been observed in the AD brain (Jin et al., 2015). Therefore, the truncation-activated GSK-3β is likely to be the major cause of AD. Further evidence comes from the efficacy of lithium in AD.
The Efficacy of Lithium in AD
Lithium is widely prescribed for the treatment of bipolar disorder. Its mechanisms of action are not fully understood. Lithium can inhibit Protein Kinase C, which could account for its beneficial effects on bipolar disorder (Saxena et al., 2017). Lithium is also a GSK-3β inhibitor. In animal models, lithium can block the accumulation of Aβ in the brains of mice (Phiel et al., 2003). Microdose lithium is sufficient to reduce Aβ plaques in rats (Wilson et al., 2020). In clinical studies, a recent review obtained two excellent clinical studies, all of which supported the effects of lithium for dementia prevention (Ishii et al., 2021).
The inhibition strength of lithium for GSK-3β is very weak compared to curcumin and Tideglusib. Quantitatively, the potency of an inhibitor can be measured by IC50, that is, the concentration (C) of an inhibitor required to inhibit (I) its target by 50%. Thus, smaller IC50 indicates stronger potency. The IC50 of lithium toward full-length GSK-3β is about 3 mM (Goñi-Oliver et al., 2007). Both curcumin and Tideglusib have IC50 of ~ 60 nM for full-length GSK-3β (Bustanji et al., 2009; selleckchem.com). Note that 1 mM = 106 nM. Hence, the potency of curcumin and Tideglusib is far stronger than lithium. Then, how can lithium produce more beneficial effects than curcumin and Tideglusib? Strikingly, lithium has been demonstrated to inhibit both full-length and truncated GSK-3β with approximately the same IC50 (Goñi-Oliver et al., 2007).
Is Amyloid Beta an Attractive Target?
Alzheimer's disease is characterized by amyloid plaques and neurofibrillary tangles (NFTs), comprising amyloid beta (Aβ) and Tau protein, respectively. Their causal relationship is still being debated. The "Amyloid Cascade Hypothesis" postulated that AD was initiated by aggregates of Aβ and thus Aβ should be an attractive therapeutic target (Hardy and Higgins, 1992). In subsequent three decades, many drugs were developed on the basis of this hypothesis. Unfortunately, to date, the only Aβ-directed drug ever approved by the U.S. Food and Drug Administration (FDA) is Aduhelm under the "accelerated approval pathway," which requires the company, Biogen, to conduct a new randomized, controlled clinical trial to verify the drug’s clinical benefit. If the trial fails to verify clinical benefit, the FDA may initiate proceedings to withdraw approval of the drug (press release, June 07, 2021).
The outlook of Aduhelm is not promising, as compelling evidence indicates that amyloid plaques are not the cause of AD, but simply a by-product of its pathogenic cascade. It has been shown that Aβ toxicity requires the presence of Tau protein (Rapoport et al., 2002; Ittner et al., 2010; Roberson et al., 2011). Furthermore, Aβ should be able to drive Tau pathology (NFTs) if AD originates from Aβ aggregates. In the past three decades, Aβ proponents have been trying to demonstrate this prediction experimentally, but failed by using traditional modeling systems. In 2011, a novel technology, called induced pluripotent stem cell (iPSC), was found to have great potential in modeling AD (Yagi et al., 2011). However, the original iPSC system still failed to recapitulate Tau pathology. In the following years, researchers continued to improve the modeling systems. A 3D cell culture system finally succeeded in producing robust NFTs. Why is 3D cell culture a better system for modeling AD? The answer lies in the 4-repeat Tau (D'Avanzo et al., 2015):
These findings support the BDNF Cascade Hypothesis (next section) wherein the Tau-dependent neuronal hyperexcitation (see Paper 2) plays a key role in AD.
The Achilles' Heel of Alzheimer's Disease
On the basis of the BDNF Cascade Hypothesis (Figure 2), the Achilles' heel of AD is likely to be the truncation-activated GSK-3β, as it can cause both Tau pathology and Aβ pathology. Hyperphosphorylation of Tau by GSK-3β may form Tau oligomers and aggregate into NFTs. Tau oligomers are more toxic than NFTs (Shafiei et al., 2017). On the other hand, the Tau-dependent hyperexcitation in the upstream explains why Aβ toxicity requires the presence of Tau protein (Rapoport et al., 2002; Ittner et al., 2010; Roberson et al., 2011). Substantial evidence indicates that oxidative stress promotes the production of Aβ (Arimon et al., 2015). Mitochondria are a major source of reactive oxygen species, driven primarily by excessive mitochondrial fission (Ježek et al., 2018). The protein, dynamin-related protein 1 (Drp1), plays a critical role in mitochondrial fission. Activation of Drp1 promotes mitochondrial fission, and contributes to Aβ pathology in AD as inhibition of Drp1 has been shown to ameliorate Aβ deposition, synaptic depression and cognitive Impairment in an AD model (Baek et al., 2017). Remarkably, GSK-3β can activate Drp1 by phosphorylating it at a well-established site, Ser616 (Huang et al., 2015; Baum and Gama, 2021). GSK-3β could also phosphorylate Ser40 and Ser44 to enhance Drp1 activity, thereby exacerbating Aβ toxicity (Yan et al., 2015).
Why not Target Calpain?
Calpain activation is a critical step in the pathogenic cascade of AD (Figure 2). It appears that calpain could also be a promising target. However, calpain has numerous substrates involved in a wide variety of normal physiological processes. High dose of calpain will definitely produce serious side effects. To date, Alicapistat (ABT-957) is the only calpain inhibitor that has reached clinical phase I. At a tolerable dose, the drug did not produce expected effects (Lon et al., 2019).
In contrast, truncation-activated GSK-3β is a pathological condition. Selective drugs to inhibit GSK-3β truncation may not cause severe side effects. Currently, researchers are developing inhibitors that target the substrate binding site (Rippin et al., 2020). Direct binding to the substrate binding site will inhibit GSK-3β, regardless of its activation mechanisms. Another approach is to block the interaction between calpain and GSK-3β. Moreover, knowing exactly how the C-terminal domain regulates GSK-3β activity could also inspire novel drug discovery.
In conclusion, excessive activation of GSK-3β can lead to both Tau pathology and Aβ pathology in Alzheimer's disease. Thus, inhibition of GSK-3β should be on the right track. In AD brain, truncated GSK-3β has been observed, suggesting that excessive GSK-3β activity in AD is caused primarily by calpain cleavage, rather than dephosphorylation. Curcumin and Tideglusib, while potent for the inhibition of full-length GSK-3β, may not be able to inhibit the truncation-activated GSK-3β. Lithium has been shown to inhibit truncation-activated GSK-3β. However, its potency and selectivity are insufficient for the treatment of AD. Potent and selective inhibitors of truncation-activated GSK-3β hold promise to beat Alzheimer's disease.
Author: Frank Lee