Fosgonimeton attenuates amyloid-beta toxicity in preclinical models of Alzheimer's disease

Abstract

Positive modulation of hepatocyte growth factor (HGF) signaling may represent a promising therapeutic strategy for Alzheimer's disease (AD) based on its multimodal neurotrophic, neuroprotective, and anti-inflammatory effects addressing the complex pathophysiology of neurodegeneration. Fosgonimeton is a small-molecule positive modulator of the HGF system that has demonstrated neurotrophic and pro-cognitive effects in preclinical models of dementia. Herein, we evaluate the neuroprotective potential of fosgonimeton, or its active metabolite, fosgo-AM, in amyloid-beta (Aβ)-driven preclinical models of AD, providing mechanistic insight into its mode of action. In primary rat cortical neurons challenged with Aβ (Aβ_1-42), fosgo-AM treatment significantly improved neuronal survival, protected neurite networks, and reduced tau hyperphosphorylation. Interrogation of intracellular events indicated that cortical neurons treated with fosgo-AM exhibited a significant decrease in mitochondrial oxidative stress and cytochrome c release. Following Aβ injury, fosgo-AM significantly enhanced activation of pro-survival effectors ERK and AKT, and reduced activity of GSK3β, one of the main kinases involved in tau hyperphosphorylation. Fosgo-AM also mitigated Aβ-induced deficits in Unc-like kinase 1 (ULK1) and Beclin-1, suggesting a potential effect on autophagy. Treatment with fosgo-AM protected cortical neurons from glutamate excitotoxicity, and such effects were abolished in the presence of an AKT or MEK/ERK inhibitor. In vivo, fosgonimeton administration led to functional improvement in an intracerebroventricular Aβ_25-35 rat model of AD, as it significantly rescued cognitive function in the passive avoidance test. Together, our data demonstrate the ability of fosgonimeton to counteract mechanisms of Aβ-induced toxicity. Fosgonimeton is currently in clinical trials for mild-to-moderate AD (NCT04488419; NCT04886063).

Keywords: Alzheimer's disease; Amyloid beta; Fosgonimeton; Hepatocyte growth factor (HGF); Neuroprotection; Neurotrophic factor.

Fig. 1

Fosgo-AM promotes neuronal survival, preserves neurite networks, and reduces tau hyperphosphorylation following Aβ_1-42 injury. Primary rat cortical neurons were treated with the active metabolite of fosgonimeton, fosgo-AM (100 ​nM), and Aβ_1-42 (15 ​μM; 2 ​μM oligomers) for 24 ​h and labeled with microtubule-associated protein-2 (MAP2; neuronal marker) and AT100 (marker for pTau-Thr212/Ser214). Scale bar ​= ​100 ​μm. (a) Representative images highlighting the effect of Aβ_1-42 oligomers on cortical neurons in the presence and absence of fosgo-AM. (b–d) Quantification of neuronal survival (i.e., number of MAP2+ neurons), neurite network (i.e., total length of MAP2+ neurites in μm), and pTau (i.e., overlap of AT100 and MAP2 area in μm²), expressed as percentage of normal control (100%). Data presented as mean ​± ​SEM; N ​= ​3 biological replicates (independent preparations of cortical neurons), n ​= ​4–6 technical replicates. Statistical differences were determined by one-way ANOVA followed by Fisher's LSD test. ∗∗∗p < 0.001, ∗∗∗∗p ​< ​0.0001 versus Aβ_1-42 control.

Fig. 2

Fosgo-AM attenuates Aβ-induced mitochondrial oxidative stress and cytochrome c release. (a) Primary rat cortical neurons were treated with the active metabolite of fosgonimeton, fosgo-AM, and Aβ_1-42 (15 ​μM; 2 ​μM oligomers) for 4 ​h and labeled with microtubule-associated protein-2 (MAP2; neuronal marker) and MitoSox (marker of mitochondrial ROS). Scale bar ​= ​100 ​μm. (b) Quantification of mitochondrial ROS (i.e., overlap of MitoSox and MAP2 area), expressed as percentage of normal control (100%). (c) Primary rat cortical neurons were treated with the fosgo-AM and Aβ_1-42 (15 ​μM; 2 ​μM oligomers) for 4 ​h and labeled with MAP2 and anti-CytC (marker of cytochrome c). Scale bar ​= ​100 ​μm. (d) Quantification of cytochrome c release (i.e., overlap of CytC and MAP2 area), expressed as percentage of normal control (100%). Data presented as mean ​± ​SEM; N ​= ​3 biological replicates (independent preparations of cortical neurons), n ​= ​4–6 technical replicates. Statistical differences were determined by one-way ANOVA followed by Fisher's LSD test. ∗∗p < 0.01, ∗∗∗p ​< ​0.001, ∗∗∗∗p ​< ​0.0001 versus Aβ_1-42 control.

Fig. 3

Fosgo-AM increases AKT and ERK phosphorylation and reduces GSK3β activity and tau phosphorylation following Aβ_1-42 injury. Primary rat cortical neurons were treated with the active metabolite of fosgonimeton, fosgo-AM, and Aβ_1-42 (15 ​μM; 2 ​μM oligomers) for 24 ​h and protein lysates were analyzed for total AKT, phospho-AKT (pAKT^Ser473), total ERK, phospho-ERK (pERK^{Thr202/Tyr204}), and phospho-GSK3β (pGSK3β^Tyr216 or pGSK3β^Ser389), or phospho-Tau (pTau^{Thr212/Ser214}) via Simple Western. (a, b) Representative images of Simple Westerns and corresponding quantification showing AKT phosphorylation as pAKT/AKT. (c, d) Representative images of Simple Westerns and corresponding quantification showing ERK phosphorylation as pERK/ERK. (e, f) Representative images of Simple Westerns and corresponding quantification showing GSK3β activity as pGSK3β^Tyr216(active)/pGSK3β^Ser389(inactive). (g, h) Representative images of Simple Westerns and corresponding quantification showing levels of pTau, normalized to GAPDH level. Borders for representative images highlight that each protein was evaluated independently in a separate capillary system. Lanes: 1, normal control; 2, Aβ_1-42 control; 3, Aβ_1-42 ​+ ​fosgo-AM (10 ​nM); 4, Aβ_1-42 ​+ ​fosgo-AM (100 ​nM), Aβ_1-42 ​+ ​fosgo-AM (1 ​μM). Data presented as mean ​± ​SEM; N ​= ​3–4 biological replicates (independent preparations of cortical neurons). Statistical differences were determined by one-way ANOVA followed by Fisher's LSD test. ∗p ​< ​0.05, ∗∗p ​< ​0.01, ∗∗∗p ​< ​0.001, ∗∗∗∗p ​< ​0.0001 versus Aβ_1-42 control.

Fig. 4

Fosgo-AM increases levels of ULK1 and Beclin-1 following Aβ_1-42 injury. Primary rat cortical neurons were treated with the active metabolite of fosgonimeton, fosgo-AM, and Aβ_1-42 (15 ​μM; 2 ​μM oligomers) for 24 ​h and protein lysates were analyzed for ULK1 and Beclin-1 via Simple Western. (a, b) Representative image of Simple Westerns and corresponding quantification showing levels of ULK1, normalized to GAPDH levels. (c, d) Representative images of Simple Westerns and corresponding quantification showing levels of Beclin-1, normalized to GAPDH levels. Borders for representative images highlight that each protein was evaluated independently in a separate capillary system. Lanes: 1, normal control; 2, Aβ_1-42 control; 3, Aβ_1-42 ​+ ​fosgo-AM (10 ​nM); 4, Aβ_1-42 ​+ ​fosgo-AM (100 ​nM); 5, Aβ_1-42 ​+ ​fosgo-AM (1 ​μM). Data presented as mean ​± ​SEM; N ​= ​3–4 biological replicates (independent preparations of cortical neurons). Statistical differences were determined by one-way ANOVA followed by Fisher's LSD test. ∗p ​< ​0.05, ∗∗p ​< ​0.01, ∗∗∗p ​< ​0.001 versus Aβ_1-42 control.

Fig. 5

Fosgo-AM promotes neuronal survival, preserves neurite networks, and reduces tau hyperphosphorylation following glutamate injury. Primary rat cortical neurons were treated with the active metabolite of fosgonimeton, fosgo-AM (100 ​nM), and glutamate (20 ​μM) for 24 ​h and labeled with microtubule-associated protein-2 (MAP2; neuronal marker) and AT100 (marker for pTau). Scale bar ​= ​100 ​μm. (a) Representative images highlighting the effect of glutamate on cortical neurons in the presence and absence of fosgo-AM. (b–d) Quantification of neuronal survival (i.e., number of MAP2+ neurons), neurite network (i.e., total length of MAP2+ neurites in μm), and pTau (i.e., overlapping area of AT100 and MAP2 in μm²), expressed as percentage of normal control (100%). Data presented as mean ​± ​SEM; N ​= ​3 biological replicates (independent preparations of cortical neurons), n ​= ​4–6 technical replicates. Statistical differences were determined by one-way ANOVA followed by Fisher's LSD test. ∗∗∗p < 0.001, ∗∗∗∗p ​< ​0.0001 versus glutamate control.

Fig. 6

Neuroprotective actions of fosgo-AM versus glutamate excitotoxicity are mediated by AKT and ERK signaling. Primary rat cortical neurons were treated with AKT inhibitor (GSK-690963) or MEK/ERK inhibitor (PD98059), followed by fosgo-AM (100 ​nM), and glutamate (20 ​μM) for 24 ​h. Cells were labeled with microtubule-associated protein-2 (MAP2; neuronal marker) to assess neuronal survival (i.e., number of MAP2+ neurons) and neurite network (i.e., total length of MAP2+ neurites in μm). (a–b) Effect of fosgo-AM on (a) neuronal survival and (b) neurite networks following glutamate injury, in the presence of the AKT inhibitor or the MEK/ERK inhibitor. Data presented as mean ​± ​SEM; N ​= ​3 biological replicates (independent preparations of cortical neurons), n ​= ​4–6 technical replicates. Statistical differences were determined by one-way ANOVA followed by Dunnett's multiple comparisons test. ∗∗∗∗p ​< ​0.0001 versus glutamate control.

Fig. 7

Fosgonimeton improves cognitive performance in passive avoidance retention in ICV-Aβ_25-35-exposed rats. Cognitive performance was assessed in rats administered with ICV-Aβ_25-35 or sham and treated with fosgonimeton or vehicle for 14 days. (a) Step-through latency in passive avoidance retention trial. Fosgonimeton at all tested doses resulted in significant rescue of cognitive impairment, as indicated by longer step-through latencies. (b) Data presented as percent recovery, normalized to Sham ​+ ​Vehicle (100%) and ICV-Aβ ​+ ​Vehicle (0%) groups. Data presented as mean ​± ​SEM; n ​= ​12 rats per group. Statistical differences were determined by one-way ANOVA followed by Dunnett's multiple comparisons test for step-through latency, and by Kruskal-Wallis followed by Dunn's multiple comparisons for percent recovery. ∗p ​< ​0.05, ∗∗p ​< ​0.01, ∗∗∗p ​< ​0.001, ∗∗∗∗p ​< ​0.0001 versus ICV-Aβ ​+ ​Vehicle.