ML RR-S2 CDA ammonium salt

• Molecular Formula: C20H30N12O10P2S2
• Molecular Weight: 724.6 g/mol
• SMILES: C1C2C(C(C(O2)N3C=NC4=C(N=CN=C43)N)OP(=O)(OCC5C(C(C(O5)N6C=NC7=C(N=CN=C76)N)O)OP(=S)(O1)[O-])[S-])O.[NH4+].[NH4+]
• InChI: InChI=1S/C20H24N10O10P2S2.2H3N/c21-15-9-17(25-3-23-15)29(5-27-9)19-12(32)13-8(38-19)2-36-42(34,44)40-14-11(31)7(1-35-41(33,43)39-13)37-20(14)30-6-28-10-16(22)24-4-26-18(10)30;;/h3-8,11-14,19-20,31-32H,1-2H2,(H,33,43)(H,34,44)(H2,21,23,25)(H2,22,24,26);2*1H3/t7-,8-,11?,12-,13-,14-,19-,20-,41?,42?;;/m1../s1
• InChIKey: VZYXAGGQHVUTHM-LJFXOJISSA-N

ML RR-S2 CDA Ammonium Salt Overview

ML RR-S2 CDA ammonium salt (CAS: 1638750-96-5; also known as ADU-S100 ammonium salt or MIW815 ammonium salt) is a synthetic cyclic dinucleotide (CDN) derivative that functions as a potent agonist of the stimulator of interferon genes (STING) pathway [1]. It is characterized by a mixed phosphodiester linkage (ML) architecture containing both 2′,5′ and 3′,5′ bonds in the [Rp,Rp] configuration, combined with double thiophosphate (S2) substitution . This compound represents the ammonium salt form of the first STING agonist to enter clinical development for immuno-oncology applications and is distinguished from natural CDNs and earlier-generation synthetic analogs by its engineered stability and potency profile [1].

STING pathway activation study fit

Mixed-linkage (ML) phosphodiester architecture

Double thiophosphate (S2) modification for stability research

Clinical-stage STING pathway research tool

• [1] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015 May 19;11(7):1018-30. doi: 10.1016/j.celrep.2015.04.031. View Source

Differentiation of ML RR-S2 CDA from Generic STING Agonists

Generic substitution among STING agonists is not scientifically valid due to fundamental differences in chemical architecture that dictate pharmacologic performance. Natural CDNs such as c-di-AMP (CDA) and cGAMP contain exclusively canonical 3′,5′ linkages and lack thiophosphate modifications, rendering them susceptible to rapid phosphodiesterase degradation and limiting their binding to human STING alleles—particularly the refractory hSTING^REF and hSTING^Q variants present in approximately 20% of the human population [1]. Even among synthetic CDN derivatives, variation in linkage stereochemistry (Rp vs Sp), the number and position of thiophosphate substitutions, and counterion selection (ammonium vs sodium vs free acid) produce substantial differences in solubility, thermal stability of the ligand-STING complex, allele activation breadth, and in vivo anti-tumor efficacy [2]. The quantitative evidence below establishes precisely why ML RR-S2 CDA ammonium salt cannot be arbitrarily replaced by its closest analogs without compromising experimental reproducibility and translational relevance.

Natural CDN analogs

Unmodified c-di-AMP or cGAMP lack thiophosphate groups and contain only canonical 3',5' linkages. Reported rapid degradation and limited activation of refractory hSTING alleles may not reproduce the pathway-engagement profile of this compound.

Stereochemical or counterion variants

Differences in Rp/Sp configuration, thiophosphate count, or counterion (sodium vs free acid) can shift solubility, thermal stability, and allele activation breadth. Model-response context may not transfer directly.

Alternative STING agonists (e.g., DMXAA)

DMXAA and other non-thio agonists show divergent anti-metastatic model response at higher doses; systemic model-response endpoints may differ significantly from the ML RR-S2 CDA profile.

• [1] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015 May 19;11(7):1018-30. doi: 10.1016/j.celrep.2015.04.031. View Source
• [2] Gajewski TF, Higgs EF, Leung MY, Zha Y, Karrison T, Luke JJ. Phase II Study of the STING Agonist MIW815 (ADU-S100) in Combination with Spartalizumab in Patients with Advanced Solid Tumors. J Clin Oncol. 2023;41(4_suppl):TPS273. doi: 10.1200/JCO.2023.41.4_suppl.TPS273. View Source

ML RR-S2 CDA: Quantitative Head-to-Head Evidence

Enhanced Type I IFN vs Unmodified CDA in THP-1 Cells

ML RR-S2 CDA ammonium salt demonstrates enhanced type I interferon (IFN) production compared to unmodified cyclic di-AMP (CDA; c-di-AMP) in THP-1 human monocytes. This differential activity is attributed to the mixed phosphodiester linkage (ML) architecture, which enhances binding to human STING relative to natural CDNs with exclusively 3′,5′ linkages .

IFN Production vs CDA

Reported

Directional enhancement of type I interferon (statistically significant) relative to unmodified cyclic di-AMP in THP-1 cells.

Supports human STING activation context

Quantitative fold-change not specified; source data to verify

Evidence Dimension	Type I interferon production in human monocytes
Target Compound Data	Enhanced IFN production (fold increase not numerically specified in primary source; directional enhancement reported)
Comparator Or Baseline	Unmodified cyclic di-AMP (CDA; Item No. 17753)
Quantified Difference	Enhanced type I IFN production relative to CDA (qualitative enhancement with statistical significance)
Conditions	THP-1 human monocyte cell line; in vitro stimulation assay

Why This Matters

This direct head-to-head comparison establishes that the ML modification is essential for achieving the level of human STING activation required for downstream immunomodulatory effects, rendering unmodified CDA an inadequate substitute.

Superior IFN-α Production vs ML cGAMP in Human PBMCs

In human peripheral blood mononuclear cells (PBMCs), ML RR-S2 CDA ammonium salt induces significantly higher levels of IFN-α compared to endogenous mixed-linkage cGAMP (ML cGAMP) when evaluated on a molar equivalent basis. At a 10 μM stimulation concentration for 6 hours, the compound produced measurably elevated IFN-α protein levels relative to ML cGAMP, as quantified by cytometric bead array [1].

IFN-α Induction vs ML cGAMP

Head-to-head

Significantly higher IFN-α protein levels (p < 0.05) at 10 µM in human PBMCs compared to ML cGAMP.

Supports non-redundant STING activation context

6h stimulation; cytometric bead array readout

Evidence Dimension	IFN-α protein production in human PBMCs
Target Compound Data	Elevated IFN-α levels (significantly higher than ML cGAMP)
Comparator Or Baseline	Endogenous ML cGAMP (mixed-linkage cyclic GMP-AMP)
Quantified Difference	Significantly higher levels (p < 0.05; statistical significance reported in primary literature)
Conditions	Cryopreserved hPBMCs; 10 μM stimulation; 6 hours; IFN-α quantification via Cytometric Bead Array (CBA) Human Flex Set with FACSVerse cytometer analysis

Why This Matters

This comparison directly addresses the question of whether an endogenous STING ligand can substitute for ML RR-S2 CDA; the data demonstrate that even the natural mixed-linkage ligand ML cGAMP produces inferior IFN-α induction, confirming that the thiophosphate modifications in ML RR-S2 CDA confer non-redundant potency advantages.

• [1] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015 May 19;11(7):1018-30. doi: 10.1016/j.celrep.2015.04.031. View Source

Thermal Stability of STING: Mixed-Linkage vs Canonical CDNs

The mixed phosphodiester linkage architecture of ML RR-S2 CDA—containing both 2′,5′ and 3′,5′ bonds—produces increased thermal stabilization of the human STING protein-ligand complex compared to CDNs with exclusively canonical 3′,5′ linkages, as measured by differential scanning fluorimetry (DSF) . This enhanced thermal stability reflects a more favorable binding interaction that translates to sustained pathway activation.

Thermal Stability vs Canonical CDNs

Class-level

Increased thermal stabilization of human STING-ligand complex by DSF relative to canonical 3',5' CDNs.

Supports sustained binding context

Exact ΔTm values not reported

Evidence Dimension	Thermal stability of human STING protein-ligand complex
Target Compound Data	Increased thermal stability (ΔTm shift) relative to canonical-linkage CDNs
Comparator Or Baseline	CDNs with exclusively canonical 3′,5′ phosphodiester linkages
Quantified Difference	Increased thermal stabilization observed in DSF assay (exact ΔTm values not reported in available sources; directional enhancement established)
Conditions	Differential scanning fluorimetry (DSF) assay with purified human STING protein

Why This Matters

Thermal stability of the STING-ligand complex is a direct biophysical correlate of binding affinity and residence time; this metric enables objective comparison across CDN analogs and predicts sustained pharmacodynamic effect duration, which is critical for experimental design and translational development.

In Vivo Anti-Tumor Efficacy vs DMXAA in Metastasis Models

In a B16.F10 melanoma experimental metastasis model, ML RR-S2 CDA ammonium salt (50 μg intratumoral dosing) produced superior reduction in lung metastatic nodules compared to the murine STING agonist DMXAA (150 μg intratumoral dosing). The histogram analysis of total lung nodule counts demonstrated that ML RR-S2 CDA treatment reduced metastatic burden to near-background levels, whereas DMXAA treatment produced only modest reduction compared to HBSS vehicle control [1].

Anti-Metastatic Effect vs DMXAA

Head-to-head

Near-complete suppression of B16.F10 lung nodules (50 µg) vs modest reduction with DMXAA (150 µg); p < 0.01.

Supports in vivo model-response context

C57BL/6 metastasis model; intratumoral dosing

Evidence Dimension	Reduction of B16.F10 lung metastatic nodules
Target Compound Data	Near-complete suppression of lung metastatic nodules (50 μg dose; 3 doses on days indicated)
Comparator Or Baseline	DMXAA (vadimezan; ASA-404) at 150 μg dose
Quantified Difference	ML RR-S2 CDA (50 μg) produced markedly greater nodule reduction than DMXAA (150 μg) at one-third the dose; statistical significance **p < 0.01, ***p < 0.001 relative to control
Conditions	C57BL/6 mice implanted with 5×10⁴ B16.F10 melanoma cells on right flank at day 0, IV challenge with 1×10⁵ cells at day 7; intratumoral treatment of flank tumor with ML RR-S2 CDA (50 μg), DMXAA (150 μg), or HBSS control; lung nodules counted at day 28

Why This Matters

This direct head-to-head comparison in a clinically relevant metastasis model demonstrates that DMXAA—a frequently cited alternative STING agonist—cannot recapitulate the anti-metastatic efficacy of ML RR-S2 CDA, even when used at triple the dose. Procurement of DMXAA for experiments requiring robust systemic anti-tumor immunity would therefore represent a scientifically suboptimal choice.

• [1] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015 May 19;11(7):1018-30. doi: 10.1016/j.celrep.2015.04.031. Figure 5E. View Source

Refractory STING Allele Activation vs Canonical CDNs

ML RR-S2 CDA ammonium salt potently activates all five known human STING alleles, including the refractory hSTING^REF and hSTING^Q alleles that show limited or no response to canonical cyclic dinucleotides. In contrast, natural CDNs with exclusively 3′,5′ linkages demonstrate markedly reduced activation of these refractory alleles, which are present in approximately 20% of the human population [1]. This broad allele coverage is a direct consequence of the dithio, mixed-linkage chemical architecture.

Allele Coverage vs Canonical CDNs

Class-level

Potent activation of all five hSTING alleles, including refractory hSTINGREF and hSTINGQ, reported across multiple studies.

Supports broad human STING research context

~20% population carry refractory alleles

Evidence Dimension	Activation breadth across human STING allelic variants
Target Compound Data	Potent activation of all five hSTING alleles including hSTINGREF and hSTINGQ refractory alleles
Comparator Or Baseline	Natural CDNs with exclusively canonical 3′,5′ linkages
Quantified Difference	Full activation of refractory alleles vs minimal/non-activation by canonical CDNs (reported across multiple publications)
Conditions	In vitro STING activation assays across human STING allele variants (WT/WT, WT/REF, WT/HAQ, REF/REF, Q/Q genotypes)

Why This Matters

For studies involving human-derived cells or translational research aimed at clinical applicability, the ability to activate refractory STING alleles is non-negotiable. Use of natural CDNs or other analogs lacking this breadth of allele coverage introduces significant risk of false-negative results when working with cells from donors carrying refractory genotypes—a variable that cannot be controlled for without pre-screening donor STING genotypes.

• [1] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015 May 19;11(7):1018-30. doi: 10.1016/j.celrep.2015.04.031. View Source

Phosphodiesterase Resistance: vs Non-Thio Analogs

The double thiophosphate (S2) substitution in ML RR-S2 CDA confers optimal resistance to phosphodiesterase-mediated degradation compared to non-thiophosphate-containing CDN analogs. This chemical modification directly addresses the primary metabolic liability of natural cyclic dinucleotides, which are rapidly hydrolyzed by phosphodiesterases in biological matrices . The enhanced cellular potency of ML RR-S2 CDA is directly attributed to this phosphodiesterase-resistant architecture combined with the mixed phosphodiester linkage in the [Rp,Rp] configuration.

PDE Resistance vs Non-Thio Analogs

Class-level

Enhanced cellular potency and 100% tumor growth inhibition in B16.F10, 4T1, CT26 models attributed to S2 thiophosphate modification.

Supports metabolic stability context

Specific PDE half-life not reported

Evidence Dimension	Phosphodiesterase resistance and resultant cellular potency
Target Compound Data	Optimal phosphodiesterase resistance conferred by double thiophosphate substitution (S2) in [Rp,Rp] configuration
Comparator Or Baseline	Non-thiophosphate CDN analogs with native phosphodiester bonds
Quantified Difference	Enhanced cellular potency (100% tumor growth inhibition observed in B16.F10, 4T1, and CT26 models at 3×50 μg intratumoral dosing); specific phosphodiesterase half-life values not reported in available sources
Conditions	In vitro cellular assays and in vivo murine tumor models; CDN degradation susceptibility assessment

Why This Matters

Phosphodiesterase susceptibility is the single greatest determinant of CDN in vivo half-life and bioavailability. Non-thio analogs—including natural CDNs—undergo rapid degradation in serum and tissues, producing variable and often non-reproducible experimental outcomes. Procurement of ML RR-S2 CDA ammonium salt with the validated S2 modification ensures the metabolic stability required for consistent in vivo pharmacology studies.

ML RR-S2 CDA: Research & Industrial Applications

In Vivo Tumor Immunotherapy with Reproducible Efficacy

Based on direct head-to-head evidence demonstrating superior reduction of B16.F10 lung metastatic nodules relative to DMXAA at one-third the dose (50 μg vs 150 μg) [1], ML RR-S2 CDA ammonium salt is the evidence-supported choice for murine tumor models—including B16 melanoma, 4T1 mammary carcinoma, and CT26 colon carcinoma—where consistent tumor regression and immune-mediated rejection are required. The compound's documented 100% growth inhibition in established B16.F10, 4T1, and CT26 tumors following 3×50 μg intratumoral dosing provides a validated dosing regimen that ensures reproducible results across independent laboratories.

STING Activation in Genetically Diverse Human Primary Cells

For experiments involving human peripheral blood mononuclear cells (PBMCs), THP-1 monocytes, or dendritic cells from genetically diverse donor populations, ML RR-S2 CDA ammonium salt is the only evidence-supported selection among STING agonists due to its demonstrated ability to potently activate all five hSTING alleles, including the refractory hSTING^REF and hSTING^Q variants present in ~20% of humans [1]. Studies using canonical CDNs or non-dithio analogs risk false-negative results from donors carrying refractory alleles—a confounding variable that ML RR-S2 CDA eliminates. The documented significantly higher IFN-α induction relative to ML cGAMP in human PBMCs at 10 μM further supports this selection for human immunology studies.

PDE-Resistant STING Agonist for Extended-Duration Studies

For experimental protocols requiring sustained STING pathway activation over extended timeframes—such as multi-day in vitro stimulation assays, ex vivo tissue culture systems, or in vivo studies with infrequent dosing intervals—ML RR-S2 CDA ammonium salt offers the validated phosphodiesterase resistance conferred by its double thiophosphate (S2) substitution in the [Rp,Rp] configuration [1]. Non-thiophosphate analogs undergo rapid enzymatic degradation, producing variable and often non-reproducible pharmacodynamic profiles. The documented thermal stabilization of the human STING-ligand complex in DSF assays provides additional biophysical evidence of sustained target engagement that predicts prolonged pathway activation.

Comparative Benchmarking and SAR Studies of STING Agonists

ML RR-S2 CDA ammonium salt serves as the validated reference standard for comparative STING agonist studies due to its well-characterized chemical architecture and extensive documentation as the first STING agonist to enter clinical development [1]. Its defined mixed-linkage (ML) and dithio (S2) features provide a clear structural baseline against which novel analogs can be benchmarked for thermal stability (via DSF), allele activation breadth (via hSTING variant panels), and in vivo anti-tumor efficacy (via established murine models). The compound's quantitative performance data across multiple comparators—including unmodified CDA, ML cGAMP, and DMXAA —establishes a robust evidentiary foundation for SAR studies.

Application

Selection Property

Validation Focus

Murine tumor model research

Reproducible model-response evidence

Tumor regression and immune-mediated rejection endpoints

Human primary cell STING pathway studies

Broad hSTING allele activation profile

Interferon response across diverse donor genotypes

Extended-duration STING activation assays

Phosphodiesterase-resistant architecture

Sustained target engagement and pathway activation

STING agonist comparator and SAR studies

Well-characterized mixed-linkage dithio architecture

Thermal stability, allele breadth, and in vivo model-response benchmarking

• [1] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ, Metchette K, Dubensky TW Jr, Gajewski TF. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep. 2015 May 19;11(7):1018-30. doi: 10.1016/j.celrep.2015.04.031. View Source