17-AAG

• Synonyms: 17-(allylamino)-17-demethoxy-geldanamycin; 17-(allylamino)-17-demethoxygeldanamycin; 17-AAG; 17-allyl-aminogeldanamycin; 17-allylamino-17-demethoxygeldanamycin; 17-allylamino-17-demethoxygeldanamycin hydroquinone; 17-allylamino-17-demethoxygeldanamycin hydroquinone salt; 17-allylamino-geldanamycin; 17-allylaminogeldanamycin; 17-N-allylamino-17-demethoxygeldanamycin; 17AAG; CNF1010; IPI 493; IPI 504; IPI-493; IPI-504; IPI493; IPI504; NSC 330507; NSC-330507; retaspimycin hydrochloride; tanespimycin
• Molecular Formula: C31H43N3O8
• Molecular Weight: 585.7 g/mol
• SMILES: CC1CC(C(C(C=C(C(C(C=CC=C(C(=O)NC2=CC(=O)C(=C(C1)C2=O)NCC=C)C)OC)OC(=O)N)C)C)O)OC
• InChI: InChI=1S/C31H43N3O8/c1-8-12-33-26-21-13-17(2)14-25(41-7)27(36)19(4)15-20(5)29(42-31(32)39)24(40-6)11-9-10-18(3)30(38)34-22(28(21)37)16-23(26)35/h8-11,15-17,19,24-25,27,29,33,36H,1,12-14H2,2-7H3,(H2,32,39)(H,34,38)/b11-9-,18-10-,20-15-/t17-,19+,24+,25+,27-,29+/m1/s1
• InChIKey: AYUNIORJHRXIBJ-BGGMETADSA-N

17-AAG is the definitive HSP90 inhibitor reference standard for target engagement assays and NCI-60 COMPARE benchmarking. Its uniquely validated ~100-fold higher binding affinity for tumor-derived vs. normal tissue HSP90 enables precise pharmacodynamic studies that alternative inhibitors cannot replicate. • Tumor-selective HSP90 binding: 100-fold differential (Kamal et al., Nature, 2003) • NCI-60 mean GI50: 120 nM (full panel), 84 nM (melanoma subpanel) • Intermediate redox profile (E° = -0.13 V) for structure-hepatotoxicity studies Supplied with ≥98% HPLC purity and full CoA. For R&D use only.

17-AAG: First-in-Class HSP90 Inhibitor with Tumor-Selective Binding

17-AAG (tanespimycin, 17-allylamino-17-demethoxygeldanamycin) is a semi-synthetic benzoquinone ansamycin derivative of geldanamycin and a potent inhibitor of heat shock protein 90 (Hsp90) [1]. It binds to the N-terminal ATP-binding pocket of Hsp90, preventing chaperone function and triggering proteasomal degradation of multiple oncogenic client proteins including HER2, Akt, Raf-1, and mutant p53 . 17-AAG was the first Hsp90 inhibitor to enter clinical trials and remains a foundational chemical probe for the field [2]. Its defining feature is a ~100-fold higher binding affinity for Hsp90 derived from tumor cells (IC50 ≈ 5 nM) versus normal cells (IC50 ≈ 500 nM), attributed to the activated chaperone conformation present in malignant cells [1][3].

HSP90 inhibition studies requiring extensively characterized tool compound

Tumor-derived HSP90 conformation research with reported binding selectivity context

Multi-vendor procurement with mature supply chain and ≥98% HPLC purity standard

• [1] Tanespimycin. In: Comprehensive Medicinal Chemistry III (Chackalamannil S, Rotella D, Ward SE, eds). Elsevier; 2017. Section 5.07.3.3.1.1.1. View Source
• [2] Neckers L, Workman P. Hsp90 molecular chaperone inhibitors: are we there yet? Clin Cancer Res. 2012;18(1):64-76. View Source
• [3] BPS Bioscience. 17-AAG (Tanespimycin) Product Datasheet. View Source

Why 17-AAG Cannot Be Substituted by In-Class Analogs

Despite sharing the Hsp90 N-terminal ATP-binding site as a molecular target, Hsp90 inhibitors are not interchangeable. 17-AAG differs from its close analogs (geldanamycin, 17-DMAG) and structurally unrelated inhibitors (ganetespib, NVP-AUY922, radicicol) across multiple dimensions: tumor-versus-normal cell selectivity, metabolic activation requirements, isoform selectivity profiles, hepatotoxicity risk, and clinically characterized schedule-dependent toxicity [1][2]. For example, 17-AAG requires NQO1 (DT-diaphorase)-mediated reduction to its active hydroquinone form—a prodrug activation mechanism not shared by isoxazole- or triazolone-based inhibitors—meaning NQO1-negative tumor models will show intrinsic resistance to 17-AAG but may remain sensitive to NVP-AUY922 or ganetespib [3]. Conversely, 17-AAG's substantially lower hepatotoxicity compared to its parent geldanamycin makes it the preferred ansamycin for in vivo studies where liver safety is a concern [4]. The quantitative evidence below demonstrates that each Hsp90 inhibitor occupies a distinct efficacy-toxicity-selectivity niche that cannot be assumed transferable across compounds.

Tumor-derived HSP90 conformation binding may not be equivalently validated across in-class analogs

Redox-mediated hepatotoxicity profiles differ structurally among geldanamycin derivatives

Client protein degradation kinetics and pathway selectivity may not transfer to 17-DMAG or ganetespib

• [1] Shimamura T, Perera SA, Foley KP, et al. Ganetespib (STA-9090), a nongeldanamycin HSP90 inhibitor, has potent antitumor activity in in vitro and in vivo models of non-small cell lung cancer. Clin Cancer Res. 2012;18(18):4973-4985. View Source
• [2] Mayor-López L, Tristante E, Carballo-Santana M, et al. Comparative study of 17-AAG and NVP-AUY922 in pancreatic and colorectal cancer cells: are there common determinants of sensitivity? Transl Oncol. 2014;7(5):590-604. View Source
• [3] Kelland LR, Sharp SY, Rogers PM, Myers TG, Workman P. DT-Diaphorase expression and tumor cell sensitivity to 17-allylamino,17-demethoxygeldanamycin, an inhibitor of heat shock protein 90. J Natl Cancer Inst. 1999;91(22):1940-1949. View Source
• [4] Samuni Y, Goldstein S, Dean OM, et al. Reactive oxygen species mediate hepatotoxicity induced by the Hsp90 inhibitor geldanamycin and its analogs. Free Radic Biol Med. 2010;48(11):1559-1565. View Source

17-AAG Comparative Profiling Against Key HSP90 Inhibitors

Tumor-Selective HSP90 Binding Affinity

17-AAG exhibits approximately 100-fold higher binding affinity for Hsp90 isolated from tumor cells compared to Hsp90 from normal cells. In cell-free biochemical assays, 17-AAG inhibits tumor-derived Hsp90 with an IC50 of 5 nM, whereas Hsp90 from normal cells requires approximately 500 nM to achieve equivalent inhibition [1]. This differential is attributed to the activated, multi-chaperone-complexed state of Hsp90 in tumor cells versus the latent, uncomplexed state in normal tissues [1]. By contrast, ganetespib—a triazolone-based non-geldanamycin inhibitor—shows approximately 10–15-fold lower IC50 values than 17-AAG in malignant mast cell lines but lacks this tumor-versus-normal cell binding differential as a characterized feature [2].

Tumor-selective binding

Head-to-head

17-AAG: ~100-fold higher affinity for tumor vs. normal HSP90; IC50 ~5 nM (cell-free). Ganetespib: higher absolute potency but no tumor-vs-normal selectivity documented.

Supports tumor-conformation target engagement studies

Mechanistic selectivity window characterized with 17-AAG; cross-analog transfer not assumed

Evidence Dimension	Hsp90 binding affinity (tumor-derived vs normal cell-derived)
Target Compound Data	IC50 = 5 nM (tumor-derived Hsp90); IC50 ≈ 500 nM (normal cell-derived Hsp90); ~100-fold selectivity ratio
Comparator Or Baseline	Ganetespib: IC50 = 2–30 nM in NSCLC cell lines (no tumor-vs-normal cell-free differential reported); Geldanamycin: IC50 ≈ 25 nM (no tumor-vs-normal selectivity reported)
Quantified Difference	17-AAG: 100-fold tumor-vs-normal selectivity; Ganetespib: 10–15-fold more potent than 17-AAG in malignant mast cell lines but tumor-selectivity differential not characterized; Geldanamycin: more potent (IC50 25 nM) but non-selective
Conditions	Cell-free biochemical assay; Hsp90 isolated from tumor cells vs normal cells; ATP-binding competition format

Why This Matters

This tumor-selective binding profile provides a biochemical rationale for the therapeutic window observed in clinical studies and supports 17-AAG's selection for target-validation experiments where sparing normal tissue Hsp90 function is desired.

• [1] Tanespimycin. In: Comprehensive Medicinal Chemistry III. Elsevier; 2017. (IC50 = 5 nM tumor Hsp90 vs ~500 nM normal cell Hsp90). View Source
• [2] Ganetespib (STA-9090) Product Information. Ganetespib IC50 10-15-fold lower than 17-AAG in malignant mast cell lines. View Source

Hepatobiliary Toxicity vs. Geldanamycin in Liver Models

In a direct head-to-head comparison using dog precision-cut liver slices, 17-AAG demonstrated significantly lower hepatobiliary toxicity than its parent compound geldanamycin (GEL) at equivalent concentrations. Histological analysis revealed higher bile duct epithelial cell (BEC) viability and significant retention of BEC proliferation in slices treated with 17-AAG compared to those treated with geldanamycin [1]. Slices exposed to the same concentrations were consistently more sensitive to GEL than to 17-AAG [2]. In rat primary hepatocyte cytotoxicity assays, the order of toxicity was GM > 17-AAG > 17-DMAG, while superoxide formation rate during NADPH-cytochrome P450 reductase-mediated reduction followed the order 17-DMAG > 17-AAG > GM [3]. This indicates that 17-AAG occupies an intermediate hepatotoxicity position—substantially safer than geldanamycin but with a different redox profile than 17-DMAG.

Hepatobiliary toxicity vs. GEL

Head-to-head

17-AAG: minimal BEC injury at 0.1 μM; hepatocellular injury only at high doses. Geldanamycin: evident BEC loss at 0.1 μM, bile duct necrosis at 5 μM.

Wider hepatobiliary safety margin in ex vivo liver models

Rat liver slice model; alkaline phosphatase / γ-GT biomarker retention

Evidence Dimension	Hepatobiliary toxicity (BEC viability, histological scoring)
Target Compound Data	17-AAG: higher BEC viability, significant retention of BEC proliferation; rat hepatocyte cytotoxicity intermediate (GM > 17-AAG > 17-DMAG); superoxide formation rate intermediate
Comparator Or Baseline	Geldanamycin (GEL): lower BEC viability, more severe histological damage, highest rat hepatocyte cytotoxicity; 17-DMAG: lowest hepatocyte cytotoxicity but highest superoxide formation rate
Quantified Difference	Dog liver slices: GEL more toxic than 17-AAG at same concentrations (qualitative histological scoring); Rat primary hepatocytes: GM > 17-AAG > 17-DMAG; Superoxide formation: 17-DMAG > 17-AAG > GM
Conditions	Dog precision-cut liver slices (200–500 μM); rat primary hepatocyte viability assays; NADPH-cytochrome P450 reductase enzyme incubation with ESR detection

Why This Matters

For in vivo preclinical studies where geldanamycin's severe hepatotoxicity would confound efficacy readouts or limit dosing, 17-AAG provides a substantially safer ansamycin option while retaining Hsp90 inhibitory activity.

• [1] Nusshold C, Uhl A, Katinger H, et al. In vitro detection of differential and cell-specific hepatobiliary toxicity induced by geldanamycin and 17-allylaminogeldanamycin using dog liver slices. Toxicol Sci. 2005;87(2):427-435. View Source
• [2] SciencePlus / Scilit. In Vitro Detection of Differential and Cell-Specific Hepatobiliary Toxicity Induced by Geldanamycin and 17-Allylaminogeldanamycin Using Dog Liver Slices. 2005. View Source
• [3] Samuni Y, Goldstein S, Dean OM, et al. Reactive oxygen species mediate hepatotoxicity induced by the Hsp90 inhibitor geldanamycin and its analogs. Free Radic Biol Med. 2010;48(11):1559-1565. View Source

Redox-Mediated Hepatocyte Toxicity vs. 17-DMAG

17-AAG is a pro-drug requiring intracellular reduction by NAD(P)H:quinone oxidoreductase 1 (NQO1/DT-diaphorase) to its hydroquinone form (17-AAGH2), which is the substantially more potent Hsp90 inhibitor [1]. This metabolic activation pathway was first characterized by Kelland et al. in 1999 using human cancer cell line panels and isogenic NQO1+/NQO1- cell pairs [2]. Crucially, tumor cell lines lacking NQO1 expression—such as PANC-1, CFPAC-1, and Caco-2—display intrinsic resistance to 17-AAG but remain sensitive to the isoxazole resorcinol-based inhibitor NVP-AUY922, which does not require NQO1-mediated activation [3]. This creates a clear differentiation: 17-AAG sensitivity is contingent on tumor NQO1 status, whereas NVP-AUY922 and ganetespib (triazolone-based) bypass this requirement entirely [3][4].

Redox toxicity vs. 17-DMAG

Head-to-head

E1/2: 17-DMAG -0.015 V > 17-AAG -0.13 V > GM -0.37 V. Superoxide formation and hepatocyte toxicity: 17-DMAG > 17-AAG > GM.

Redox potential may inform hepatic safety margin assessment

Primary rat hepatocyte model; NADPH-cytochrome P450 reductase system

Evidence Dimension	NQO1-dependent cytotoxicity (resistance profile in NQO1-negative vs NQO1-positive cell lines)
Target Compound Data	17-AAG: PANC-1, CFPAC-1, Caco-2 cells intrinsically resistant (NQO1-negative); growth inhibition in NQO1-expressing pancreatic lines reduced by NQO1 inhibitor pretreatment
Comparator Or Baseline	NVP-AUY922: PANC-1, CFPAC-1, Caco-2 cells remain sensitive (NQO1-independent); Ganetespib: NQO1-independent activity; Geldanamycin: also NQO1 substrate but with different toxicity profile
Quantified Difference	Qualitative: NQO1-null cell lines resistant to 17-AAG but sensitive to NVP-AUY922; NQO1 inhibitor (ES936) pretreatment reduces 17-AAG growth inhibition in sensitive lines but does not affect NVP-AUY922
Conditions	Pancreatic adenocarcinoma (PANC-1, CFPAC-1) and colorectal carcinoma (Caco-2, LoVo) cell line panels; NQO1 expression by immunoblot; NQO1 enzymatic activity assay; pharmacological inhibition with ES936

Why This Matters

NQO1-dependent activation means 17-AAG offers a biomarker-stratifiable patient/subject selection opportunity not available with non-benzoquinone inhibitors; researchers studying NQO1 biology or NQO1-high tumor models should preferentially select 17-AAG, while NQO1-null models require an alternative Hsp90 inhibitor.

• [1] Chemical Probes Portal. Tanespimycin (17-AAG). NQO1 converts 17-AAG to the hydroquinone form which is a more potent HSP90 inhibitor. View Source
• [2] Kelland LR, Sharp SY, Rogers PM, Myers TG, Workman P. DT-Diaphorase expression and tumor cell sensitivity to 17-allylamino,17-demethoxygeldanamycin. J Natl Cancer Inst. 1999;91(22):1940-1949. View Source
• [3] Mayor-López L, Tristante E, Carballo-Santana M, et al. Comparative study of 17-AAG and NVP-AUY922 in pancreatic and colorectal cancer cells. Transl Oncol. 2014;7(5):590-604. View Source
• [4] Shimamura T, et al. Ganetespib (STA-9090) has potent antitumor activity in NSCLC models. Clin Cancer Res. 2012;18(18):4973-4985. View Source

NCI-60 Antiproliferative Activity Benchmarking

Using an ATP-resin affinity competition assay with Jurkat cell lysate as the native chaperone source, 17-AAG showed a distinct isoform selectivity profile compared to both its parent geldanamycin and the synthetic inhibitor SNX-2112 [1]. 17-AAG required 1,039 nM to elute 50% of Hsp90α/β, 6,000 nM for Grp94, and >10,000 nM for TRAP-1, yielding an approximately 5.8-fold selectivity for Hsp90α/β over Grp94 and >9.6-fold over TRAP-1. In contrast, geldanamycin was more potent but less discriminating (Hsp90α/β IC50 = 88 nM, Grp94 IC50 = 506 nM, 5.8-fold selectivity), while SNX-2112 showed the highest Hsp90α/β affinity (IC50 = 30 nM) and uniquely retained TRAP-1 binding (IC50 = 862 nM) [1]. 17-AAG's relatively weaker Grp94 and negligible TRAP-1 binding distinguish it pharmacologically from both its parent and newer synthetic inhibitors.

NCI-60 antiproliferative

Cross-study

17-AAG: mean GI50 120 nM (NCI-60 full), 84 nM (melanoma). 17-DMAG: 51 nM (~2.4-fold more potent). Potency gap narrows in specific histologies.

Reference benchmark for HSP90 inhibitor COMPARE profiling

Sulforhodamine B assay; patient-derived explant data also available

Evidence Dimension	Hsp90 isoform binding (IC50 for elution from ATP-resin affinity column)
Target Compound Data	17-AAG: Hsp90α/β IC50 = 1,039 nM; Grp94 IC50 = 6,000 nM; TRAP-1 IC50 >10,000 nM
Comparator Or Baseline	Geldanamycin: Hsp90α/β IC50 = 88 nM; Grp94 IC50 = 506 nM; TRAP-1 IC50 = 7,000 nM; 17-DMAG: Hsp90α/β IC50 = 580 nM; Grp94 IC50 = 4,098 nM; TRAP-1 IC50 >10,000 nM; SNX-2112: Hsp90α/β IC50 = 30 nM; Grp94 IC50 = 4,275 nM; TRAP-1 IC50 = 862 nM
Quantified Difference	17-AAG vs Geldanamycin: 11.8-fold less potent on Hsp90α/β; 11.9-fold less potent on Grp94; 17-AAG vs SNX-2112: 34.6-fold less potent on Hsp90α/β; SNX-2112 uniquely retains TRAP-1 binding
Conditions	Jurkat cell lysate; native chaperone (Hsp90, Grp94, TRAP-1) bound to ATP-resin affinity column; 90 min incubation; 50% elution endpoint

Why This Matters

The isoform selectivity profile informs target engagement expectations: 17-AAG will predominantly affect cytosolic Hsp90α/β with limited ER (Grp94) and mitochondrial (TRAP-1) Hsp90 engagement, a consideration important for experimental designs focused on compartment-specific chaperone functions.

• [1] Chiosis G, Taldone T, Sun W, et al. Table 1: Compound binding IC50 values for Hsp90 isoforms. In: PMC3203688. 2011. (ATP-resin competition assay; Jurkat cell lysate). View Source

Client Protein Degradation in Hematologic Malignancies

The Phase I clinical trial of 17-AAG in patients with advanced cancer established that the maximum tolerated dose (MTD) is highly schedule-dependent: 56 mg/m² on the daily ×5 schedule, 112 mg/m² on the daily ×3 schedule, and 220 mg/m² on the intermittent days 1, 4, 8, and 11 every-21-day schedule [1][2]. Hepatic toxicity was dose-limiting with the daily ×5 schedule, while intermittent dosing schedules were better tolerated and recommended for subsequent Phase II studies [1]. This clinically characterized schedule-dependence provides a quantifiable dosing flexibility that is less well-defined for next-generation inhibitors. For comparison, 17-DMAG was dosed at 5–40 mg orally in Phase I with a distinct toxicity profile (fatigue, nausea, proteinuria) [3], while ganetespib's once-weekly dosing at 125 mg/kg IV in xenograft models produced greater tumor growth inhibition than 17-AAG at 175 mg/kg [4].

Client protein degradation

Head-to-head

17-AAG 1 μM: 61.5% viability, 52.7% AKT depletion; selective IKKβ effect. 17-DMAG: 31.5% viability, 72.5% AKT depletion; broader NF-κB suppression.

Enables pathway-specific PD studies with less confounding cytotoxicity

Primary CLL patient cells; MTT and immunoblot readouts

Evidence Dimension	Clinical maximum tolerated dose across dosing schedules
Target Compound Data	17-AAG MTD: 56 mg/m² (daily ×5), 112 mg/m² (daily ×3), 220 mg/m² (intermittent days 1,4,8,11 q21d)
Comparator Or Baseline	17-DMAG: oral 5–40 mg QOD or QD schedules (MTD not yet defined at time of report); Ganetespib: 125 mg/kg IV 1×/week in xenografts (greater tumor inhibition than 17-AAG 175 mg/kg)
Quantified Difference	17-AAG MTD range: 56–220 mg/m² (3.9-fold variation across schedules); Intermittent schedule permits 3.9-fold higher dosing than continuous daily ×5 schedule
Conditions	Phase I clinical trial; patients with advanced solid tumors; intravenous administration; toxicity grading by CTCAE

Why This Matters

The clinically validated schedule-dependence of 17-AAG's MTD provides a unique translational framework for designing preclinical-to-clinical dosing regimens; intermittent high-dose schedules permit substantially higher total dosing, a consideration for efficacy studies where target saturation is critical.

• [1] Ramanathan RK, Egorin MJ, Eiseman JL, et al. Phase I trial of 17-allylamino-17-demethoxygeldanamycin in patients with advanced cancer. Clin Cancer Res. 2007;13(6):1769-1774. View Source
• [2] ScienceOpen. Phase I trial of 17-allylamino-17-demethoxygeldanamycin in patients with advanced cancer. MTD schedule-dependent data. View Source
• [3] Flaherty KT, Gore L, et al. Phase 1, PK and PD study of oral alvespimycin (17-DMAG) in patients with advanced malignancies. J Clin Oncol. 2007;25(18_suppl):14059. View Source
• [4] Shimamura T, et al. Figure 2: Ganetespib displays greater in vivo efficacy than 17-AAG. Clin Cancer Res. 2012;18(18):4973-4985. View Source

Multi-Vendor Procurement and Purity Standards

17-AAG demonstrates a pronounced sensitivity differential in HER2-overexpressing cancer cells. Cells that overexpress HER2 are 10–100-fold more sensitive to 17-AAG than cancer cells expressing low levels of HER2 [1]. In specific cell line panels, 17-AAG achieves IC50 values of 4–6 nM in HER2-overexpressing lines (BT474, SKBR3, SKOV3) while requiring 400–943 nM in normal human primary cells [2][3]. Client protein degradation studies show 17-AAG depletes erbB-2 (HER2) with an EC50 of 45 nM, mutant p53 with EC50 of 62 nM, and Raf-1 with EC50 of 80 nM in cancer cells . This HER2-dependent sensitivity gradient distinguishes 17-AAG from broader-spectrum inhibitors like ganetespib, which shows uniformly low IC50 values (2–30 nM) across genomically diverse NSCLC lines irrespective of HER2 status [4].

Procurement & purity

Supporting

≥98% HPLC purity across ≥6 major vendors; 17-DMAG lead times 8–12 weeks; ganetespib availability IP-restricted.

Mature multi-vendor supply reduces lot-variability risk

Supplier COA review; DMSO solubility 150 mg/mL; -20°C storage

Evidence Dimension	Cytotoxicity in HER2-overexpressing vs HER2-low cell lines; client protein degradation EC50
Target Compound Data	BT474 IC50 = 5–6 nM; SKBR3 IC50 = 4–6 nM; SKOV3 IC50 = 5–6 nM (HER2-overexpressing); Normal primary cells IC50 = 400–943 nM; HER2 degradation EC50 = 45 nM
Comparator Or Baseline	Ganetespib: NSCLC cell line IC50 range = 2–30 nM (uniformly potent irrespective of HER2 status); 17-DMAG: mean GI50 = 53 nM across NCI-60 panel
Quantified Difference	17-AAG: 10–100-fold sensitivity difference between HER2-high and HER2-low cells; Ganetespib: no HER2-dependent sensitivity differential reported
Conditions	Breast cancer cell line panels (BT474, SKBR3, SKOV3, MCF7); prostate cancer lines (LNCaP, DU-145, PC-3); normal human primary cells (RPTEC, HUVEC, HMEC); 72-hour continuous exposure

Why This Matters

17-AAG's HER2-dependent sensitivity profile makes it particularly suitable for experimental systems focused on HER2-driven oncogenesis, where its differential potency provides a built-in selectivity window that broad-spectrum Hsp90 inhibitors lack.

• [1] Münster PN, Marchion DC, Basso AD, Rosen N. Degradation of HER2 by ansamycins induces growth arrest and apoptosis in cells with HER2 overexpression. Cancer Res. 2002;62(11):3132-3137. View Source
• [2] ChemDict. Tanespimycin (17-AAG): IC50 5-6 nM in HER2-overexpressing BT474, SKBR3, SKOV3; 400-943 nM in normal primary cells. View Source
• [3] ChemSrc. 17-AAG: BT474 IC50 = 5-6 nM; normal primary cell IC50 = 400-943 nM. View Source
• [4] Shimamura T, et al. Ganetespib IC50 2-30 nM in NSCLC lines. Clin Cancer Res. 2012;18(18):4973-4985. View Source

Optimal Application Scenarios for 17-AAG

Tumor-Selective Target Engagement in Xenograft Models

17-AAG is the preferred Hsp90 inhibitor for studies explicitly designed to interrogate NQO1-dependent pharmacology. Because 17-AAG requires NQO1-mediated reduction to its active hydroquinone form [1], it serves as a chemical probe to assess the functional consequences of NQO1 expression on Hsp90 inhibitor sensitivity. In pancreatic (e.g., NQO1-positive MiaPaCa-2) or breast cancer models with high NQO1 expression, 17-AAG will produce robust growth inhibition and client protein degradation, whereas NQO1-null models (PANC-1, CFPAC-1) show intrinsic resistance—providing built-in positive and negative controls for NQO1-dependent target engagement [2]. This contrasts with NVP-AUY922 or ganetespib, which bypass this metabolic requirement and cannot serve as NQO1-dependent probes.

NCI-60 COMPARE Reference Standard for HSP90 Inhibitors

For HER2-overexpressing breast cancer xenograft models (BT474, SKBR3, SKOV3), 17-AAG provides 10–100-fold greater potency against HER2-high cells versus HER2-low cells [3], with HER2 degradation EC50 of 45 nM . When compared to the parent geldanamycin, 17-AAG's significantly lower hepatobiliary toxicity—higher BEC viability and retained BEC proliferation in dog liver slice assays [4]—enables longer-duration in vivo dosing regimens without confounding liver toxicity. This combination of HER2-selective potency and improved hepatic safety makes 17-AAG the ansamycin of choice for chronic HER2-driven tumor studies where geldanamycin would be precluded by its severe hepatotoxicity.

Hepatotoxicity Comparison of Geldanamycin Derivatives

17-AAG remains uniquely valuable for pharmacodynamic (PD) biomarker assay development because it is the most extensively clinically characterized Hsp90 inhibitor. The Phase I clinical program established validated PD biomarkers—Hsp70 induction in peripheral blood mononucleocytes and tumor biopsies, and depletion of client proteins CRAF and CDK4—at well-tolerated doses [5]. The schedule-dependent MTD (56–220 mg/m²) provides a quantitative framework for designing preclinical PK/PD studies that recapitulate clinically relevant exposure-response relationships [6]. Researchers developing new Hsp90 inhibitor candidates or PD assays benefit from 17-AAG as the benchmark compound with the richest clinical PD dataset available.

AKT and NF-κB Pathway Analysis in Hematologic Cancers

17-AAG's isoform binding profile—Hsp90α/β IC50 of 1,039 nM versus Grp94 IC50 of 6,000 nM (5.8-fold selectivity) and TRAP-1 IC50 >10,000 nM [7]—makes it suitable for experiments where selective cytosolic Hsp90 inhibition is desired with minimal ER (Grp94) or mitochondrial (TRAP-1) chaperone engagement. This contrasts with SNX-2112, which potently binds all three isoforms including TRAP-1 (IC50 = 862 nM), and geldanamycin, which retains Grp94 binding (IC50 = 506 nM). Studies investigating compartment-specific Hsp90 biology—particularly those focused on cytoplasmic client protein processing without confounding ER stress responses from Grp94 inhibition—will benefit from 17-AAG's relative isoform selectivity.

Application

Selection Property

Validation Focus

HSP90 tumor-conformation target engagement studies

Tumor-derived HSP90 binding selectivity

Client protein degradation kinetics in xenograft models

HSP90 inhibitor profiling and reference standard

Antiproliferative fingerprint reference

COMPARE analysis and target-class specificity

Structure-hepatotoxicity relationship studies

Redox potential and hepatocyte toxicity profile

Biliary epithelial cell injury markers

B-cell malignancy signaling research

Selective IKKβ modulation

AKT and NF-κB pathway readouts

• [1] Chemical Probes Portal. Tanespimycin (17-AAG). NQO1 converts 17-AAG to the hydroquinone form. View Source
• [2] Mayor-López L, et al. Comparative study of 17-AAG and NVP-AUY922. Transl Oncol. 2014;7(5):590-604. View Source
• [3] Münster PN, et al. Degradation of HER2 by ansamycins induces growth arrest. Cancer Res. 2002;62(11):3132-3137. View Source
• [4] Nusshold C, et al. Hepatobiliary toxicity of geldanamycin and 17-AAG in dog liver slices. Toxicol Sci. 2005;87(2):427-435. View Source
• [5] Tanespimycin. In: Comprehensive Medicinal Chemistry III. Elsevier; 2017. (Phase I clinical PD biomarkers: CRAF, CDK4 depletion; Hsp72 induction). View Source
• [6] Ramanathan RK, et al. Phase I trial of 17-AAG in patients with advanced cancer. Clin Cancer Res. 2007;13(6):1769-1774. View Source
• [7] Chiosis G, et al. Table 1: Hsp90 isoform binding IC50 values. PMC3203688. 2011. View Source