TMI-1

• Molecular Formula: C17H22N2O5S2
• Molecular Weight: 398.5 g/mol
• InChI: InChI=1S/C17H22N2O5S2/c1-4-5-11-24-13-6-8-14(9-7-13)26(22,23)19-10-12-25-17(2,3)15(19)16(20)18-21/h6-9,15,21H,10-12H2,1-3H3,(H,18,20)/t15-/m0/s1
• InChIKey: CVZIHNYAZLXRRS-HNNXBMFYSA-N

TMI-1 Procurement Guide: Dual TACE/MMP Inhibitor

TMI-1 (WAY-171318; CAS 287403-39-8) is a thiomorpholine hydroxamate small molecule that acts as a dual inhibitor of TNF-α converting enzyme (TACE/ADAM17) and matrix metalloproteinases (MMPs) [1]. It was discovered using structure-based design by Wyeth (Pfizer) and exhibits nanomolar potency against TACE (IC50 8.4 nM) and multiple MMPs (IC50 3–26 nM) in cell-free enzyme assays [1]. TMI-1 is orally bioavailable and has demonstrated efficacy in suppressing TNF-α secretion in vitro in human primary monocytes, human whole blood, and RA patient synovial explants, as well as reducing clinical severity in collagen-induced arthritis models in vivo [1].

Dual TACE/MMP pathway inhibition study fit for inflammatory and oncologic signaling research

Compatible with cell-free enzyme assays, cell-based models, and human tissue explant studies

Orally administered research tool; reported in vivo model-response endpoints in rodent arthritis and tumor models

• [1] Zhang, Y., Xu, J., Levin, J., Hegen, M., Li, G., Robertshaw, H., ... & Lin, L. L. (2004). Identification and characterization of 4-[[4-(2-butynyloxy)phenyl]sulfonyl]-N-hydroxy-2,2-dimethyl-(3S)thiomorpholinecarboxamide (TMI-1), a novel dual tumor necrosis factor-alpha-converting enzyme/matrix metalloprotease inhibitor for the treatment of rheumatoid arthritis. Journal of Pharmacology and Experimental Therapeutics, 309(1), 348-355. View Source

TMI-1 Differentiation: Class-Level Substitution Risks

The TACE/ADAM17 inhibitor class spans compounds with dramatically different selectivity profiles, oral bioavailability characteristics, and functional outcomes. Broad-spectrum MMP inhibitors such as GM6001 and TAPI-2 carry pan-MMP inhibition liabilities that are clinically associated with musculoskeletal toxicity, whereas highly selective ADAM17 inhibitors like KP-457 lack the MMP co-inhibition that may limit tissue destruction in inflammatory arthritis [1]. Within the Wyeth TMI series, TMI-2 and TMI-005 (apratastat) are more TACE-selective and fail to induce caspase-dependent apoptosis in triple-negative breast cancer cells, a functional distinction that TMI-1 uniquely delivers [2]. The quantitative evidence below demonstrates that TMI-1 occupies a specific dual TACE/MMP inhibition window with slow-binding kinetics at ADAM17 and oral in vivo efficacy, a combination that neither more selective nor more broadly acting analogs replicate.

Target

Substitute

Mismatch Context

TMI-1

TMI-2 / TMI-005

TACE-selective analogs; apoptosis and tumor-cell endpoint profiles may differ substantially

TMI-1

KP-457

ADAM17-selective; MMP co-inhibition context may not transfer to dual-pathway research models

TMI-1

GM6001 / TAPI-2

Pan-MMP inhibitors with differing binding kinetics; sustained target engagement profile may not replicate

• [1] Coussens, L. M., Fingleton, B., & Matrisian, L. M. (2002). Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science, 295(5564), 2387-2392. View Source
• [2] Mezil, L., Berruyer-Pouyet, C., Cabaud, O., Josselin, E., Combes, S., Brunel, J. M., ... & Lopez, M. (2012). Tumor selective cytotoxic action of a thiomorpholin hydroxamate inhibitor (TMI-1) in breast cancer. PLoS One, 7(9), e43409. View Source

TMI-1 Comparator Evidence: Five Key Dimensions

ADAM17 Slow-Binding Kinetics vs. TAPI-2 and GM6001

TMI-1 inhibits ADAM17 with a Ki of 0.079 nM and exhibits slow-binding kinetics, whereas the commonly used comparator TAPI-2 has a reported Ki of 120 nM (0.12 μM), representing a >1500-fold weaker affinity [1][2]. GM6001, a broad-spectrum MMP inhibitor, has a Kd <100 nM for ADAM17 but lacks slow-binding behavior [3]. TMI-1's slow-binding kinetics, verified by progress-curve analysis, is specific to ADAM17 and not observed for ADAM8, 10, or 12, giving it a kinetic selectivity advantage in shedding assays [1].

ADAM17 Binding Kinetics

Head-to-head

Ki 0.079 nM, slow-binding kinetics reported

Supports sustained target engagement assay context

TAPI-2 Ki 120 nM for comparison; slow-binding specific to ADAM17

Evidence Dimension	ADAM17 inhibition constant (Ki) and binding kinetics
Target Compound Data	Ki = 0.079 nM; slow-binding kinetics observed
Comparator Or Baseline	TAPI-2: Ki = 120 nM (0.12 μM), no slow-binding reported; GM6001: Kd <100 nM for ADAM17, no slow-binding
Quantified Difference	~1500-fold lower Ki vs. TAPI-2
Conditions	Cell-free enzyme assay using fluorescent substrates; progress-curve analysis (Moss & Rasmussen, 2007)

Why This Matters

For shedding assays where sustained ADAM17 blockade is required, TMI-1's sub-nanomolar Ki and slow off-rate translate into prolonged target residence time, reducing the need for frequent re-dosing and minimizing off-target rebound.

• [1] Moss, M. L., & Rasmussen, F. H. (2007). Fluorescent substrates for the proteinases ADAM17, ADAM10, ADAM8, and ADAM12 useful for high-throughput inhibitor screening. Analytical Biochemistry, 366(2), 144-148. View Source
• [2] Anjiechem TMI-1 Product Page, citing Ki values from Moss & Rasmussen (2007) for TMI-1: ADAM8 (21 nM), ADAM10 (16 nM), ADAM12 (1.8 nM), ADAM17 (0.079 nM), and noting slow-binding inhibition of ADAM17/TACE but not other ADAM enzymes. View Source
• [3] HMS LINCS Database. GM6001 Target Affinity Spectrum. Harvard Medical School. View Source

Whole-Blood TNF-α Suppression vs. Apratastat

TMI-1 inhibits LPS-induced TNF-α secretion in human whole blood with an IC50 of 300 nM [1]. In contrast, apratastat (TMI-005) inhibits TNF-α release in human whole blood with an IC50 of 144 ng/mL, equivalent to ~330 nM at its molecular weight of 436.5 Da . While whole-blood potency appears broadly similar, the pivotal differentiation is that TMI-1 also potently inhibits TNF-α secretion in human synovium tissue explants from RA patients at IC50 <100 nM [1], whereas TMI-005's ex vivo IC50 of 81.7 ng/mL (~187 nM) reflects a less favorable translation to diseased human tissue [2].

TNF-α Suppression Context

Cross-study

Whole blood IC50 300 nM; RA synovium explant suppression reported

Reported tissue-level target engagement context

Apratastat whole blood IC50 ~330 nM; synovium explant context differs

Evidence Dimension	TNF-α suppression in translational human tissue models
Target Compound Data	Human whole blood IC50 = 300 nM; human RA synovium explant IC50 <100 nM
Comparator Or Baseline	Apratastat (TMI-005): human whole blood IC50 = 144 ng/mL (~330 nM); ex vivo IC50 = 81.7 ng/mL (~187 nM)
Quantified Difference	~2-fold superior tissue-level potency (RA synovium IC50 <100 nM vs. TMI-005 ex vivo ~187 nM)
Conditions	LPS-stimulated TNF-α secretion in human whole blood and RA patient synovium tissue explants

Why This Matters

For translational inflammation research, TMI-1's demonstrated potency in human disease-relevant synovium tissue provides stronger evidence of target engagement in the pathological microenvironment than whole-blood data alone.

• [1] Zhang, Y., et al. (2004). JPET 309(1), 348-355. Reports TMI-1 whole blood IC50 = 300 nM and RA synovium explant IC50 <100 nM. View Source
• [2] PubMed Summary for Apratastat: in vitro human whole blood IC50 = 144 ng/mL; ex vivo IC50 = 81.7 ng/mL. View Source

Tumor-Selective Cytotoxicity vs. TMI-2 and TMI-005

In a direct head-to-head experiment, TMI-1 induced dose-dependent apoptosis and caspase-3/7 activation in triple-negative SUM149 breast cancer cells, whereas TMI-2 and TMI-005 failed to induce either apoptosis or caspase-3/7 activity at equivalent concentrations (2.5–20 μM) [1][2]. Across a panel of 34/40 tumor cell lines of diverse origins, TMI-1 exhibited selective cytotoxicity with ED50 values ranging from 0.6 μM to 12.5 μM, while non-malignant cells remained resistant even at high concentrations [1]. In contrast, conventional cytotoxic drugs and the TACE-selective analogs TMI-2/TMI-005 lack this tumor-selective apoptosis phenotype [1].

Apoptosis Endpoint Context

Head-to-head

Caspase-3/7 activation reported in SUM149 cells; TMI-2 and TMI-005 did not activate

Supports apoptosis pathway-response interpretation

SUM149 triple-negative breast cancer cell model; 24–48 h exposure

Evidence Dimension	Cancer cell apoptosis induction (caspase-3/7 activity) and tumor-selective cytotoxicity
Target Compound Data	TMI-1: dose-dependent caspase-3/7 activation in SUM149 cells; ED50 0.6–12.5 μM across 34/40 tumor lines; non-malignant cells resistant
Comparator Or Baseline	TMI-2: no caspase-3/7 activation, no apoptosis; TMI-005 (apratastat): no caspase-3/7 activation, no apoptosis
Quantified Difference	Qualitative functional switch: TMI-1 induces caspase-dependent apoptosis; TMI-2 and TMI-005 do not
Conditions	SUM149 triple-negative breast cancer cells; 24–48 h treatment; annexin V/7-AAD staining; caspase-3/7 activity assay

Why This Matters

This is a go/no-go functional differentiation. Researchers selecting a TACE/MMP inhibitor for oncology applications must use TMI-1 specifically, as TMI-2 and TMI-005 lack the pro-apoptotic phenotype required for anti-cancer activity.

• [1] Mezil, L., Berruyer-Pouyet, C., Cabaud, O., Josselin, E., Combes, S., Brunel, J. M., ... & Lopez, M. (2012). Tumor selective cytotoxic action of a thiomorpholin hydroxamate inhibitor (TMI-1) in breast cancer. PLoS One, 7(9), e43409. View Source
• [2] PLoS One Figure 2: TMI-1, not TMI-2 and TMI-005, induces dose-dependent caspase-3/7 activity after 24 h in SUM149 cells. View Source

In Vivo Arthritis Efficacy vs. TMI-2

TMI-1 reduced clinical severity scores in a prophylactic collagen-induced arthritis (CIA) mouse model at oral doses of 5, 10, and 20 mg/kg b.i.d., and demonstrated efficacy in a therapeutic CIA model at 100 mg/kg p.o. b.i.d. [1]. TMI-1's in vivo LPS-induced TNF-α suppression ED50 is 5 mg/kg (p.o.) [1]. For comparison, TMI-2 exhibits an ED50 of 3 mg/kg for LPS-induced TNF-α inhibition in mice, but requires 100 mg/kg b.i.d. in the therapeutic CIA model, and critically fails to induce any tumor apoptosis [2][3]. Thus, TMI-1 achieves comparable or superior anti-arthritic efficacy at lower doses (prophylactic 5 mg/kg effective dose vs. TMI-2's 30 mg/kg minimal effective dose in rat adjuvant arthritis at 30 mg/kg b.i.d.) while uniquely providing tumor-selective cytotoxicity [1][2].

Arthritis Model Response

Cross-study

Reported severity-score reduction in prophylactic CIA model; oral administration

Reported arthritis model-response context

Prophylactic and therapeutic CIA models; lower-dose response vs comparator context

Evidence Dimension	In vivo oral anti-arthritic efficacy (CIA model severity score reduction)
Target Compound Data	Prophylactic CIA: effective at 5, 10, 20 mg/kg p.o. b.i.d.; Therapeutic CIA: effective at 100 mg/kg p.o. b.i.d.; LPS-TNF ED50 = 5 mg/kg
Comparator Or Baseline	TMI-2: Rat adjuvant arthritis effective at 30 and 100 mg/kg p.o. b.i.d.; Therapeutic CIA effective at 100 mg/kg p.o. b.i.d.; LPS-TNF ED50 = 3 mg/kg
Quantified Difference	6-fold lower minimum effective dose in arthritis (TMI-1 5 mg/kg vs. TMI-2 30 mg/kg)
Conditions	Mouse prophylactic and therapeutic collagen-induced arthritis (CIA) models; oral gavage administration

Why This Matters

Lower effective dosing in vivo translates to reduced compound consumption and potentially lower systemic MMP-related toxicity, making TMI-1 more cost-effective for long-term animal studies.

• [1] Zhang, Y., et al. (2004). JPET 309(1), 348-355. CIA model data for TMI-1: prophylactic 5–20 mg/kg b.i.d.; therapeutic 100 mg/kg b.i.d.; LPS-TNF ED50 = 5 mg/kg. View Source
• [2] Levin, J. I., et al. (2004). Characterization of (2R,3S)-2-([[4-(2-butynyloxy)phenyl]sulfonyl]amino)-N,3-dihydroxybutanamide (TMI-2), a potent and selective inhibitor of TNF-alpha converting enzyme. International Immunopharmacology, 4(14), 1845-1854. View Source
• [3] Mezil, L., et al. (2012). PLoS One 7(9), e43409. Figure 2 shows TMI-2 fails to induce caspase-3/7 activity or apoptosis in SUM149 cells. View Source

Target Selectivity Profile vs. KP-457

KP-457 is a highly selective ADAM17 inhibitor (IC50 11.1 nM) with >67-fold selectivity over ADAM10 (IC50 748 nM) and >64-fold selectivity over MMP2 (IC50 717 nM), but negligible MMP inhibition at its effective concentrations [1]. TMI-1, in contrast, exhibits a broader dual inhibition profile: TACE IC50 8.4 nM, MMP-13 IC50 3 nM, MMP-2 IC50 4.7 nM, MMP-1 IC50 6.6 nM, MMP-9 IC50 12 nM, MMP-7 IC50 26 nM, and MMP-14 IC50 26 nM [2]. In the rheumatoid arthritis context, this broader MMP coverage may provide additional benefit by inhibiting cartilage-degrading MMP-13 and MMP-9 within the inflamed joint [3]. In the cancer context, the dual TACE/MMP profile of TMI-1, but not the selective profile of KP-457, is associated with tumor-selective cytotoxicity [4].

Kinase Selectivity Window

Class-level

TACE IC50 8.4 nM; broad MMP coverage vs KP-457 selective profile

Supports dual TACE/MMP pathway research context

Selectivity context determines experimental application fit

Evidence Dimension	Selectivity window across ADAM17 vs. off-target ADAM10 and MMP panel
Target Compound Data	TMI-1: TACE IC50 8.4 nM; ADAM10 Ki 16 nM (~200-fold weaker than ADAM17 Ki); MMP-13 3 nM; MMP-2 4.7 nM; MMP-1 6.6 nM; MMP-9 12 nM
Comparator Or Baseline	KP-457: ADAM17 IC50 11.1 nM; ADAM10 IC50 748 nM (~67-fold); MMP2 717 nM; MMP13 930 nM; MMP9 5410 nM
Quantified Difference	TMI-1 inhibits MMP-13 with 310-fold greater potency than KP-457 (3 nM vs. 930 nM); TMI-1 inhibits MMP-9 with 450-fold greater potency (12 nM vs. 5410 nM)
Conditions	Cell-free enzyme assays; reported IC50 values from primary literature and vendor datasheets

Why This Matters

The selectivity profile directly determines the experimental application: KP-457 is suitable for studies requiring isolated ADAM17 inhibition with minimal MMP interference, while TMI-1 is preferred for disease models where dual TACE/MMP inhibition and/or tumor-killing activity are required.

• [1] Adooq Bioscience. KP-457 product page: ADAM17 IC50 = 11.1 nM, ADAM10 IC50 = 748 nM, MMP2 IC50 = 717 nM, MMP13 IC50 = 930 nM, MMP9 IC50 = 5410 nM. View Source
• [2] Zhang, Y., et al. (2004). JPET 309(1), 348-355. TMI-1 IC50 values for TACE (8.4 nM) and multiple MMPs. View Source
• [3] Murphy, G., & Nagase, H. (2008). Reappraising metalloproteinases in rheumatoid arthritis and osteoarthritis: destruction or repair? Nature Clinical Practice Rheumatology, 4(3), 128-135. View Source
• [4] Mezil, L., et al. (2012). PLoS One 7(9), e43409. TMI-1's dual TACE/MMP inhibition is linked to its tumor-selective cytotoxicity phenotype. View Source

TMI-1 Application Scenarios

Rheumatoid Arthritis & Inflammation Modeling

TMI-1 is the preferred tool compound for oral TNF-α suppression studies in rodent arthritis models. With an in vivo LPS-TNF ED50 of 5 mg/kg (p.o.) and demonstrated efficacy at 5 mg/kg b.i.d. in the prophylactic CIA model [1], TMI-1 requires 6-fold lower dosing than TMI-2 to achieve anti-arthritic activity [2]. Its ability to suppress TNF-α secretion in human RA synovium explants (IC50 <100 nM) provides direct translational relevance [1]. Unlike TAPI-2 or GM6001, TMI-1 offers oral route compatibility, enabling chronic dosing studies without the complications of parenteral administration.

Triple-Negative & ERBB2+ Breast Cancer Research

For cancer research groups studying ADAM17/MMP-targeted apoptosis, TMI-1 is uniquely qualified among the TMI series. Only TMI-1—not TMI-2 or TMI-005—induces caspase-dependent apoptosis in triple-negative SUM149 and ERBB2-overexpressing breast tumor cell lines [3]. TMI-1 is active in 85% (34/40) of tested tumor cell lines with ED50 values of 0.6–12.5 μM, while sparing non-malignant cells [3]. It synergizes with docetaxel, doxorubicin, and lapatinib and demonstrates in vivo tumor growth inhibition in MMTV-ErbB2/neu transgenic mice at 100 mg/kg/day without adverse effects [3].

ADAM17 Shedding with Sustained Target Engagement

In biochemical shedding assays, TMI-1's sub-nanomolar Ki (0.079 nM) at ADAM17 and slow-binding kinetics differentiate it from commonly used comparators TAPI-2 (Ki 120 nM) and GM6001 (no slow-binding) [4][5]. The slow-binding property, specific to ADAM17 and absent for ADAM8/10/12, ensures that TMI-1 maintains target coverage during prolonged incubation, reducing ligand-rebinding artifacts in continuous shedding protocols [4]. TMI-1 inhibits all ADAMs tested (ADAM8, 10, 12, 17) with nanomolar potency, making it suitable as a pan-ADAM control in contrast to the ADAM10-selective GI 254023X.

Chemotherapy-Induced Neuroprotection

Emerging evidence demonstrates that TMI-1 reverses paclitaxel-induced TRPV1 upregulation and lowers inflammatory factors (TNF-α, IL-1β, IL-6) in DRG neuronal cells in vitro . This application leverages TMI-1's dual TACE/MMP inhibition profile to address neuroinflammation, a distinct use case not documented for the more TACE-selective analogs TMI-2, KP-457, or GW280264X.

Application

Selection Property

Validation Focus

Arthritis inflammation research models

Dual TACE/MMP inhibition profile

Synovial TNF-α and disease-score endpoints

Breast cancer cell-model studies

Apoptosis pathway-response context

Caspase activation and cell-viability endpoints

ADAM17 shedding biochemical studies

Sustained target engagement kinetics

Slow-binding ADAM17 occupancy verification

Chemotherapy-induced neuroinflammation research

Neuroinflammatory mediator profile

TRPV1 and inflammatory cytokine endpoints

• [1] Zhang, Y., et al. (2004). JPET 309(1), 348-355. CIA model, LPS-TNF ED50, RA synovium explant, and whole blood data for TMI-1. View Source
• [2] Levin, J. I., et al. (2004). Int Immunopharmacol 4(14), 1845-1854. TMI-2 rat adjuvant arthritis efficacy at 30 mg/kg b.i.d. View Source
• [3] Mezil, L., et al. (2012). PLoS One 7(9), e43409. Tumor-selective cytotoxicity data: 34/40 cell lines ED50 0.6–12.5 μM; caspase-3/7 activation by TMI-1 not TMI-2/TMI-005; MMTV-ErbB2/neu in vivo model. View Source
• [4] Moss, M. L., & Rasmussen, F. H. (2007). Anal Biochem 366(2), 144-148. Ki values and slow-binding kinetics for TMI-1 vs. TAPI-2 and GM6001. View Source
• [5] Anjiechem TMI-1 Product Page, summarizing Ki values from Moss & Rasmussen (2007): ADAM8 (21 nM), ADAM10 (16 nM), ADAM12 (1.8 nM), ADAM17 (0.079 nM); slow-binding specific to ADAM17. View Source