gamma-Secretase modulator 4

• Molecular Formula: C23H19FN4O2
• Molecular Weight: 402.4 g/mol
• SMILES: CC1=C(C=CC=C1F)C2C3=NC(=CN3CCO2)C4=CC5=C(N4)C=CC(=C5OC)C#N
• InChI: InChI=1S/C23H19FN4O2/c1-13-15(4-3-5-17(13)24)22-23-27-20(12-28(23)8-9-30-22)19-10-16-18(26-19)7-6-14(11-25)21(16)29-2/h3-7,10,12,22,26H,8-9H2,1-2H3/t22-/m0/s1
• InChIKey: UBJJQXXWGRHIDV-QFIPXVFZSA-N

gamma-Secretase modulator 4 (GSM-4; CAS 1420200-82-3; CHEMBL4451760; C₂₃H₁₉FN₄O₂; MW 402.42) is a second-generation, non-NSAID, aryl imidazole–based small-molecule γ-secretase modulator. Unlike γ-secretase inhibitors (GSIs) that block all proteolytic activity, GSM-4 allosterically shifts the cleavage specificity of the γ-secretase complex on amyloid precursor protein (APP), preferentially reducing the production of the highly amyloidogenic Aβ42 peptide while leaving total Aβ output unchanged.

gamma-Secretase Modulator 4 Procurement Guide: Potency, Selectivity, and Comparator Data for Alzheimer's Disease Research

gamma-Secretase modulator 4 (GSM-4; CAS 1420200-82-3; CHEMBL4451760; C₂₃H₁₉FN₄O₂; MW 402.42) is a second-generation, non-NSAID, aryl imidazole–based small-molecule γ-secretase modulator . Unlike γ-secretase inhibitors (GSIs) that block all proteolytic activity, GSM-4 allosterically shifts the cleavage specificity of the γ-secretase complex on amyloid precursor protein (APP), preferentially reducing the production of the highly amyloidogenic Aβ42 peptide while leaving total Aβ output unchanged [1]. In cell-based assays, GSM-4 reduces Aβ42 levels with IC₅₀ values of 0.014 μM (human) and 0.017 μM (mouse) . The compound has been extensively characterized in human iPSC-derived neurons, transgenic mouse models, and multiple presenilin-1 (PS1) mutant backgrounds [1].

• [1] Liu Q, Waltz S, Woodruff G, et al. Effect of Potent γ-Secretase Modulator in Human Neurons Derived From Multiple Presenilin 1–Induced Pluripotent Stem Cell Mutant Carriers. JAMA Neurology. 2014;71(12):1481–1489. doi:10.1001/jamaneurol.2014.2482 View Source

Why Not All γ-Secretase Modulators Are Interchangeable: Critical Considerations for GSM-4 Procurement

γ-Secretase modulators (GSMs) and inhibitors (GSIs) differ profoundly in mechanism, Aβ species selectivity, and Notch-sparing properties, making class-level interchange scientifically unjustified. GSM-4 belongs to the aryl imidazole (non-carboxylic acid) second-generation GSM class, which binds directly to presenilin-1 (PS1) – the catalytic subunit of γ-secretase – and allosterically modulates APP cleavage [1]. In contrast, first-generation NSAID-based GSMs (e.g., tarenflurbil) target the APP substrate and exhibit sub-millimolar potency [1]. Carboxylic acid second-generation GSMs (e.g., GSM-1, GSM-2) and heterocyclic GSMs (e.g., E2012, BMS-932481) display distinct binding sites on the γ-secretase complex, producing different Aβ product signatures [1][2]. Furthermore, GSM-4 produces a unique biomarker profile in endogenous-APP human neurons – reducing Aβ42 and Aβ40 without concomitant elevation of Aβ38 – that differs from the canonical GSM signature observed in APP-overexpressing systems [3]. These mechanistic and pharmacological differences mean that procuring a generic 'GSM' without verifying the specific compound's quantitative profile against the intended assay context risks producing non-interpretable or contradictory results.

• [1] Crump CJ, Johnson DS, Li YM. Development and Mechanism of γ-Secretase Modulators for Alzheimer Disease. Biochemistry. 2013;52(19):3197–3216. doi:10.1021/bi400377p View Source
• [2] Borgegard T, Juréus A, Olsson F, et al. First and Second Generation γ-Secretase Modulators (GSMs) Modulate Amyloid-β (Aβ) Peptide Production through Different Mechanisms. Journal of Biological Chemistry. 2012;287(15):11810–11819. doi:10.1074/jbc.M111.305227 View Source
• [3] Liu Q, Waltz S, Woodruff G, et al. Effect of Potent γ-Secretase Modulator in Human Neurons Derived From Multiple Presenilin 1–Induced Pluripotent Stem Cell Mutant Carriers. JAMA Neurology. 2014;71(12):1481–1489. doi:10.1001/jamaneurol.2014.2482 View Source

gamma-Secretase Modulator 4: Quantitative Evidence for Differentiated Selection Against Closest Comparators

Aβ42 Reduction Potency: GSM-4 vs. Carboxylic Acid GSMs (GSM-1, GSM-2)

GSM-4 reduces Aβ42 with an IC₅₀ of 0.014 μM (14 nM) in human cell-based assays , representing a 25-fold potency advantage over the prototypical carboxylic acid GSM-1 (Aβ42 IC₅₀ = 348 nM [1]) and a 3.2-fold improvement over the more optimized carboxylic acid analog GSM-2 (Aβ42 IC₅₀ = 45 nM [1]). This quantitative potency differential is critical because GSM-1 and GSM-2 remain the most widely used reference compounds for the carboxylic acid GSM class in academic laboratories, yet their substantially higher IC₅₀ values necessitate the use of micromolar concentrations that may introduce off-target effects or solubility limitations not encountered with GSM-4 at its effective nanomolar range.

Evidence Dimension	Aβ42 reduction potency (IC₅₀)
Target Compound Data	IC₅₀ = 0.014 μM (14 nM) [human cell-based assay]
Comparator Or Baseline	GSM-1: IC₅₀ = 348 nM; GSM-2: IC₅₀ = 45 nM [human cell-based assays]
Quantified Difference	GSM-4 is 24.9-fold more potent than GSM-1; 3.2-fold more potent than GSM-2
Conditions	Human cell-based Aβ42 assays; GSM-1 and GSM-2 data from ACS Chemical Neuroscience comparator study (Nie et al., 2020; Peng et al., 2011)

Why This Matters

Higher potency at lower concentrations minimizes solvent (DMSO) interference, reduces the risk of compound precipitation in cell culture media, and enables robust pharmacology at concentrations far below those that may engage secondary targets.

• [1] Nie P, Vartak A, Li YM. γ-Secretase inhibitors and modulators: Mechanistic insights into drug discovery for Alzheimer's disease. ACS Chemical Neuroscience. 2020;11(19):2929–2947. GSM-1 Aβ42 IC₅₀ = 348 nM; GSM-2 Aβ42 IC₅₀ = 45 nM. View Source

Total Aβ Sparing: Direct Head-to-Head Comparison of GSM-4 vs. Semagacestat (γ-Secretase Inhibitor) in Human iPSC-Derived Neurons

In a direct head-to-head experiment using human iPSC-derived neurons expressing endogenous APP levels, GSM-4 reduced Aβ42 and Aβ40 in a dose-dependent manner while total Aβ levels remained unchanged ('spares the ε site') [1]. In stark contrast, the γ-secretase inhibitor semagacestat non-selectively suppressed Aβ42, Aβ40, Aβ38, and total Aβ levels in the same neuronal system [1]. This differential profile is mechanistically rooted in the fact that GSM-4 allosterically modulates γ-secretase cleavage specificity without inhibiting its proteolytic activity, whereas semagacestat competitively blocks the catalytic site and abolishes all γ-secretase-mediated cleavages, including Notch processing [1][2]. The preservation of total APP processing and Notch signaling by GSM-4 addresses the key toxicity mechanism (Notch-related gastrointestinal, immunological, and skin adverse events) that caused the Phase III failure of semagacestat [2].

Evidence Dimension	Effect on total Aβ levels (Aβ processing preservation)
Target Compound Data	GSM-4: Total Aβ levels unaffected at concentrations producing near-maximal Aβ42 reduction [human iPSC-derived neurons]
Comparator Or Baseline	Semagacestat: Total Aβ levels dose-dependently decreased; Aβ42, Aβ40, and Aβ38 all reduced [human iPSC-derived neurons]
Quantified Difference	Qualitative categorical difference: GSM-4 spares total Aβ and ε-site cleavage; semagacestat inhibits all cleavages
Conditions	Human iPSC-derived neurons from NDC (non-demented control) and PS1 A246E mutant backgrounds; 10-point concentration–response curves; 3-day compound treatment

Why This Matters

For researchers investigating APP processing biology or screening for Alzheimer's therapeutics, GSM-4 enables selective reduction of amyloidogenic Aβ species without the confounding pleiotropic effects of global γ-secretase inhibition on Notch, EphB2, EphA4, and other substrates.

• [1] Liu Q, Waltz S, Woodruff G, et al. Effect of Potent γ-Secretase Modulator in Human Neurons Derived From Multiple Presenilin 1–Induced Pluripotent Stem Cell Mutant Carriers. JAMA Neurology. 2014;71(12):1481–1489. doi:10.1001/jamaneurol.2014.2482 View Source
• [2] Doody RS, Raman R, Farlow M, et al. A Phase 3 Trial of Semagacestat for Treatment of Alzheimer's Disease. New England Journal of Medicine. 2013;369:341–350. doi:10.1056/NEJMoa1210951 View Source

Notch Signaling Sparing: Quantitative Selectivity of GSM-4 vs. Semagacestat

GSM-4 does not inhibit Notch cleavage at concentrations up to 25 μM, as established for this class of Wagner-Tanzi second-generation GSMs [1]. Semagacestat, by comparison, inhibits Notch signaling with an IC₅₀ of 14.1 nM in H4 human glioma cells , yielding a Notch/Aβ42 selectivity ratio of merely 1.3 . This near-equipotent inhibition of Notch and APP processing by semagacestat mechanistically underpins the dose-limiting toxicities observed in clinical trials, including increased risk of skin cancers, gastrointestinal bleeding, and cognitive worsening [2]. The >1,700-fold selectivity window of GSM-4 for Aβ42 reduction over Notch inhibition (calculated using 25 μM no-effect concentration for Notch vs. 14 nM IC₅₀ for Aβ42) contrasts sharply with semagacestat's near-equipotency.

Evidence Dimension	Notch processing selectivity index (Notch IC₅₀ / Aβ42 IC₅₀)
Target Compound Data	Notch IC₅₀ >25 μM; Aβ42 IC₅₀ = 0.014 μM → selectivity index >1,786 [GSM-4]
Comparator Or Baseline	Semagacestat: Notch IC₅₀ = 14.1 nM; Aβ42 IC₅₀ = 10.9 nM → selectivity index = 1.3 [H4 human glioma cells]
Quantified Difference	GSM-4 exhibits >1,700-fold selectivity for Aβ42 over Notch; semagacestat exhibits only 1.3-fold selectivity
Conditions	GSM-4: Notch cleavage assay (NICD generation), Wagner et al. class-level data; Semagacestat: H4 human glioma cells stably overexpressing wild-type human APP

Why This Matters

For preclinical drug discovery programs transitioning from GSIs to safer GSM-based strategies, GSM-4 provides a validated tool compound with a documented >1,000-fold therapeutic window between desired Aβ modulation and undesired Notch inhibition, enabling meaningful safety pharmacology benchmarking.

• [1] Tanzi RE, Wagner SL, Kounnas MZ. gamma-secretase modulation: from bench to clinic. Alzheimer's & Dementia. 2025;20(Suppl 6):e090509. doi:10.1002/alz.090509. Data: GSMs do not significantly inhibit Notch cleavage at concentrations up to 25 μM. View Source
• [2] Doody RS, Raman R, Farlow M, et al. A Phase 3 Trial of Semagacestat for Treatment of Alzheimer's Disease. NEJM. 2013;369:341–350. View Source

Unique Aβ Product Signature: GSM-4 in Human iPSC Neurons vs. Canonical GSM Profile in Overexpression Models

In human iPSC-derived neurons expressing endogenous APP levels, GSM-4 reduced Aβ42 (most potently), followed by Aβ40, while Aβ38 levels were not significantly altered [1]. This differs qualitatively from the canonical GSM signature observed in APP-overexpressing cell lines and transgenic mice, where GSMs typically lower Aβ42 while reciprocally increasing Aβ38 [1]. Liu et al. (2014) explicitly demonstrated this discrepancy: 'treatment with GSM-4 lowered the levels of secreted Aβ42 and Aβ40, consistent with previously reported studies, but not Aβ38, which differed from previously reported studies using overexpressing human APP cell lines and transgenic mice' [1]. This result indicates that the Aβ product signature of GSM-4 is dependent on APP expression context and that data from overexpression systems cannot be directly extrapolated to physiological conditions.

Evidence Dimension	Aβ38 modulation (increase vs. no change) under endogenous vs. overexpressed APP conditions
Target Compound Data	GSM-4: Aβ42 decreased, Aβ40 decreased, Aβ38 NOT significantly changed; total Aβ unchanged [human iPSC-derived neurons with endogenous APP]
Comparator Or Baseline	GSM-4 in APP-overexpressing systems: Aβ42 decreased, Aβ40 decreased, Aβ38 increased [Tg2576 mouse mixed brain cultures and APP-overexpressing cell lines]
Quantified Difference	Qualitative difference: Aβ38 unchanged in endogenous-APP human neurons vs. Aβ38 increased in APP-overexpressing models
Conditions	Human iPSC-derived neurons from NDC and PS1 A246E, H163R, M146L mutant backgrounds; 3-day GSM-4 treatment; MesoScale Aβ 3-plex assay

Why This Matters

This context-dependent Aβ signature means GSM-4 is the appropriate compound for studies seeking to replicate physiologically relevant APP processing in human neuronal models, and that researchers must not assume the Aβ38-elevating profile reported for other GSMs in overexpression systems will apply.

• [1] Liu Q, Waltz S, Woodruff G, et al. Effect of Potent γ-Secretase Modulator in Human Neurons Derived From Multiple Presenilin 1–Induced Pluripotent Stem Cell Mutant Carriers. JAMA Neurology. 2014;71(12):1481–1489. doi:10.1001/jamaneurol.2014.2482 View Source

Enhanced Potency Against Pathogenic Presenilin-1 Mutations: GSM-4 IC₅₀ in PS1 A246E vs. Wild-Type Neurons

GSM-4 exhibits differential potency based on PS1 genetic background, with a lower IC₅₀ observed in PS1 A246E mutant neurons compared to wild-type (NDC1) control neurons, indicating enhanced effectiveness against this familial Alzheimer's disease (FAD)-linked mutation [1]. The study further demonstrated that GSM-4 effectively reduced the pathologically elevated Aβ42:Aβ40 ratio across three distinct PS1 mutations (A246E, H163R, and M146L) in human iPSC-derived neurons [1]. This is in direct contrast to earlier-generation GSMs (particularly NSAID-type first-generation compounds), which showed markedly reduced or absent efficacy against certain PS1 mutations such as L166P and ΔExon9 [2]. The retained potency of GSM-4 across multiple PS1 FAD mutations distinguishes it as a more robust tool compound for studying γ-secretase biology across diverse genetic backgrounds.

Evidence Dimension	IC₅₀ shift between PS1 mutant and wild-type neurons
Target Compound Data	GSM-4: IC₅₀ lower in PS1 A246E mutant neurons vs. NDC1 wild-type neurons; effective across A246E, H163R, and M146L mutations [human iPSC-derived neurons]
Comparator Or Baseline	NSAID-type first-generation GSMs: Markedly reduced or no efficacy against PS1 mutations L166P, ΔExon9 [Kretner et al., class-level data]
Quantified Difference	GSM-4 retains efficacy across 3 tested PS1 FAD mutations; first-generation GSMs show loss of efficacy against resistant mutations
Conditions	Human iPSC-derived sorted neurons; 3 different PS1 mutations (A246E, H163R, M146L); 10-point concentration–response curves; Aβ42:Aβ40 ratio measurement

Why This Matters

For research programs investigating familial AD mechanisms or developing mutation-specific therapeutic strategies, GSM-4 provides a tool that reliably modulates γ-secretase activity across multiple clinically relevant PS1 mutations, avoiding the mutation-dependent efficacy limitations of earlier GSMs.

• [1] Liu Q, Waltz S, Woodruff G, et al. Effect of Potent γ-Secretase Modulator in Human Neurons Derived From Multiple Presenilin 1–Induced Pluripotent Stem Cell Mutant Carriers. JAMA Neurology. 2014;71(12):1481–1489. doi:10.1001/jamaneurol.2014.2482 View Source
• [2] Kretner B, Fukumori A, Gutsmiedl A, et al. Differential responses to γ-secretase modulators between presenilin-1 mutations: Implications for mechanism of action. EMBO Molecular Medicine. 2011. View Source

Structural Class Differentiation: Aryl Imidazole GSM-4 vs. Carboxylic Acid GSMs and Imidazole GSM E2012

GSM-4 belongs to the aryl imidazole structural class of γ-secretase modulators, characterized by a cyanophenyl-indole-imidazole scaffold (SMILES: N#CC1=C(OC)C2=C(C=C1)NC(C3=CN(CCO[C@H]4C5=CC=CC(F)=C5C)C4=N3)=C2) . This scaffold is structurally and mechanistically distinct from the carboxylic acid GSM class (e.g., GSM-1, GSM-2, BIIB042) and from the Eisai imidazole GSM E2012 [1]. Radioligand binding and cross-competition experiments have established that aryl imidazole GSMs and carboxylic acid GSMs bind to distinct, non-overlapping sites within the γ-secretase complex, with different conformational effects on the enzyme [2]. Furthermore, carboxylic acid GSMs such as GSM-1 show competitive binding interactions with E2012-class imidazole GSMs, but non-competitive interactions with first-generation NSAID GSMs, confirming distinct allosteric binding sites [2]. The aryl imidazole scaffold of GSM-4 therefore occupies a unique chemical space within the GSM pharmacopeia, with a binding mode and Aβ product profile that cannot be replicated by simply selecting any carboxylic acid GSM.

Evidence Dimension	Structural class and binding site distinctiveness
Target Compound Data	Aryl imidazole scaffold; cyanophenyl-indole core; MW 402.42; binding site distinct from carboxylic acid and E2012-class GSMs
Comparator Or Baseline	Carboxylic acid GSMs (GSM-1, GSM-2, BIIB042): acid-derived chemotype; E2012: Eisai imidazole class; NSAID GSMs: APP-targeting
Quantified Difference	Qualitative differentiation by chemical scaffold, binding site competition profile, and conformational effect
Conditions	Radioligand binding assays with [³H]AZ8349 GSM probe; cellular cross-competition experiments; structure-activity relationship analyses

Why This Matters

For structure-activity relationship (SAR) studies or chemical biology investigations of γ-secretase allostery, procuring GSM-4 provides access to a binding site and conformational pharmacology that is distinct from carboxylic acid GSMs, enabling orthogonal validation experiments within the same modulator class.

• [1] Peng H, Talreja T, Xin Z, et al. Discovery of BIIB042, a Potent, Selective, and Orally Bioavailable γ-Secretase Modulator. ACS Medicinal Chemistry Letters. 2011;2(10):786–791. doi:10.1021/ml200175q View Source
• [2] Borgegard T, Juréus A, Olsson F, et al. First and Second Generation γ-Secretase Modulators (GSMs) Modulate Amyloid-β (Aβ) Peptide Production through Different Mechanisms. Journal of Biological Chemistry. 2012;287(15):11810–11819. doi:10.1074/jbc.M111.305227 View Source

gamma-Secretase Modulator 4: Recommended Application Scenarios Based on Verified Quantitative Evidence

Human iPSC-Derived Neuronal Models of Familial Alzheimer's Disease for Mechanistic Studies of γ-Secretase Biology

GSM-4 is the validated compound of choice for studies using human iPSC-derived neurons expressing endogenous APP levels, particularly from carriers of PS1 mutations (A246E, H163R, M146L) [1]. Unlike APP-overexpressing cell lines that produce artefactual Aβ38 elevation upon GSM treatment, human iPSC neurons treated with GSM-4 reveal the physiologically relevant Aβ product signature: robust Aβ42 and Aβ40 reduction without Aβ38 alteration and with preserved total Aβ levels [1]. The enhanced potency of GSM-4 against PS1 A246E mutant neurons (lower IC₅₀ than wild-type) makes it especially suitable for FAD mechanistic studies [1].

Notch-Sparing Preclinical Drug Discovery: Benchmarking γ-Secretase Target Engagement Without Notch-Related Toxicity Confounds

For drug discovery programs requiring γ-secretase target engagement without the confounding inhibition of Notch signaling, GSM-4 provides a >1,700-fold selectivity window (Notch IC₅₀ >25 μM vs. Aβ42 IC₅₀ = 14 nM) [2][3]. This contrasts with the γ-secretase inhibitor semagacestat (Notch/Aβ42 selectivity ratio = 1.3) [3], enabling researchers to dissect APP-specific pharmacology from Notch-mediated adverse effects. GSM-4 can serve as a positive control for Notch-sparing modulation in assays that screen for safer γ-secretase-targeting compounds [1].

Comparator Studies Differentiating Allosteric Modulation from Catalytic Inhibition of γ-Secretase

GSM-4 is an optimal tool for head-to-head mechanistic comparisons against γ-secretase inhibitors (e.g., semagacestat, DAPT, L-685,458). In the same human iPSC neuronal system, GSM-4 reduces amyloidogenic Aβ species while sparing total Aβ production and ε-site cleavage, whereas semagacestat indiscriminately suppresses all Aβ species and total Aβ [1]. This binary experimental design (GSM-4 treatment vs. GSI treatment) enables clean dissection of allosteric modulation vs. catalytic inhibition in downstream readouts including gene expression, Notch target gene activation, and substrate processing kinetics [1][2].

Chemical Biology Studies of γ-Secretase Allosteric Binding Sites Using Orthogonal GSM Chemotypes

GSM-4 (aryl imidazole scaffold) occupies a distinct allosteric binding site on the γ-secretase complex compared to carboxylic acid GSMs (GSM-1, GSM-2, BIIB042) and first-generation NSAID GSMs [2][3]. Researchers investigating the structural pharmacology of γ-secretase allostery can employ GSM-4 as an orthogonal chemotype tool to probe binding site-specific conformational effects, using competitive binding assays with [³H]-labeled GSM probes and cross-competition experiments against carboxylic acid and Eisai E2012-class GSMs [3]. This orthogonal approach enables mapping of distinct allosteric pockets and their differential effects on Aβ product spectra [2].

• [1] Liu Q, Waltz S, Woodruff G, et al. Effect of Potent γ-Secretase Modulator in Human Neurons Derived From Multiple Presenilin 1–Induced Pluripotent Stem Cell Mutant Carriers. JAMA Neurology. 2014;71(12):1481–1489. doi:10.1001/jamaneurol.2014.2482 View Source
• [2] Borgegard T, Juréus A, Olsson F, et al. First and Second Generation γ-Secretase Modulators (GSMs) Modulate Amyloid-β (Aβ) Peptide Production through Different Mechanisms. Journal of Biological Chemistry. 2012;287(15):11810–11819. doi:10.1074/jbc.M111.305227 View Source
• [3] Tanzi RE, Wagner SL, Kounnas MZ. gamma-secretase modulation: from bench to clinic. Alzheimer's & Dementia. 2025;20(Suppl 6):e090509. doi:10.1002/alz.090509 View Source