Minimalist d‑Tripeptide Prodrug Mimetics of the Myostatin Prodomain as Selective Inhibitors of Skeletal Muscle Atrophy Signaling

Abstract

Skeletal muscle atrophy is driven in part by unrestrained myostatin (MSTN) signaling. We designed all-d tripeptide prodrug mimetics of the MSTN prodomain minimum active 23-mer fragment. Compounds 1 (H-d-Arg-d-Leu-d-Ala-OiPr) and 2 (H-d-Arg-d-Abu-d-Ala-OiPr) inhibited MSTN-induced myotube atrophy with sub-micromolar potency, while free-acid analogues (3, 4) and scrambled control (scr) were less active. All-d configuration conferred proteolytic stability exceeding 24 h. In vivo, compound 1 increased soleus mass and grip strength in mice. This work establishes compound 1 as a lead prodrug for MSTN antagonism in muscle atrophy.

Myostatin (MSTN, GDF-8) is a member of the TGF-β superfamily that negatively regulates skeletal muscle mass, and its overactivation contributes to disuse, cachectic, and sarcopenic muscle wasting. , The latent MSTN prodomain noncovalently associates with the mature growth factor and can itself antagonize receptor engagement, providing a template for peptide-based inhibitor design. ,, We previously mapped the minimum active 23-mer fragment (H-WRQ­NTR­YSR­IEA­IKI­QIL­SKL­RL-NH2, positions 21–43 of the mouse MSTN prodomain) that retains the atrophy-suppressing activity of the full-length prodomain.

A critical liability of l-peptide leads is their rapid proteolytic degradation in biological matrixes, which limits pharmacological utility. ,, All-d peptide isomers resist recognition by proteolytic enzymes and can retain the pharmacophoric display of the parent sequence when built on a retro-inverso or minimalist topology. ,,, Building on structure–activity relationships within the 23-mer, we synthesized a series of minimalist all-d tripeptide isopropyl ester prodrugs designed to mimic the essential pharmacophore of the parent peptide within a dramatically reduced scaffold.

Here, we report the design, solid-phase synthesis, and biochemical and cell-based characterization of these d-tripeptide mimetics. We show that isopropyl ester prodrugs 1 and 2 inhibit MSTN-induced myofiber atrophy with submicromolar potency, while the corresponding free acids (3, 4) and a scrambled control (scr) are markedly less potent, consistent with an ester-prodrug mechanism of cellular delivery.

Guided by the 23-mer pharmacophore, we designed a minimalist all-d tripeptide scaffold pairing a cationic d-Arg residue with a hydrophobic d-Leu or d-Abu residue and a C-terminal d-Ala isopropyl ester or free acid (Table and Figure ). Peptides 1, 2, 3, and 4 and the scrambled control scr were assembled on Wang resin (100–200 mesh, loading 0.65 mmol/g; Novabiochem) using HBTU/HOBt/DIPEA activation (Figure ). Wang resin was selected because TFA-mediated cleavage releases a C-terminal free carboxylic acid, which was subsequently esterified with isopropanol under Mitsunobu or DCC/DMAP conditions to install the prodrug ester for compounds 1 and 2.

1. Designed d-Tripeptide Sequences .

Compound	Sequence	C-Terminus	Role
1	H-d-Arg-d-Leu-d-Ala-OiPr	Isopropyl ester	Lead prodrug
2	H-d-Arg-d-Abu-d-Ala-OiPr	Isopropyl ester	Lead prodrug
3	H-d-Arg-d-Leu-d-Ala-OH	Free acid	Active-metabolite control
4	H-d-Arg-d-Abu-d-Ala-OH	Free acid	Active-metabolite control
scr	H-d-Ala-d-Arg-d-Leu-OiPr	Isopropyl ester	Scrambled control

^aAll residues are d configured. OiPr = isopropyl ester; OH = free carboxylic acid. scr = scrambled sequence isopropyl ester control.

1.

Chemical structures of d-tripeptide compounds 1, 2, 3, 4, and scr, illustrating the isopropyl ester prodrug moiety (compounds 1, 2, scr) and the corresponding free-acid forms (compounds 3 and 4). All stereocenters are d-configured as drawn. Structures generated from SMILES with RDKit (2D depiction); molecular formulas match Table .

2.

Asymmetric solid-phase synthesis of compound 1 (H-d-Arg-d-Leu-d-Ala-OiPr) via Fmoc-SPPS on Wang resin, TFA cleavage, solution-phase esterification, and RP-HPLC purification. Structural variants (compounds 2, 3, 4, and scr) were prepared in parallel.

All final compounds were purified by preparative reversed-phase HPLC and lyophilized. Chiral purity was confirmed by chiral HPLC (ee >98% for all compounds), and identity was confirmed by high-resolution mass spectrometry (Table ).

2. Molecular Formulas, Masses, and Purity of d-Tripeptides .

Compound	Sequence	Formula	[M + H]+ Calcd	[M + H]+ Found	Δppm	ee (%)	Purity (%)
1	H-d-Arg-d-Leu-d-Ala-OiPr	C18H36N6O4	401.2871	401.2878	+1.9	>98	>99
2	H-d-Arg-d-Abu-d-Ala-OiPr	C16H32N6O4	373.2558	373.2561	+0.8	>98	>99
3	H-d-Arg-d-Leu-d-Ala-OH	C15H30N6O4	359.2401	359.2406	+1.3	>99	>99
4	H-d-Arg-d-Abu-d-Ala-OH	C13H26N6O4	331.2088	331.2092	+1.2	>99	>99
scr	H-d-Ala-d-Arg-d-Leu-OiPr	C18H36N6O4	401.2871	401.2879	+2.1	>99	>99

^aCompound and Sequence columns use d-configured residue notation with proper hyphenation (e.g., d-Ala). [M + H]⁺ values are m/z. Δppm = (found – calcd)/calcd × 10⁶. ee, enantiomeric excess by chiral HPLC; purity by analytical RP-HPLC (214 nm).

High-resolution ESI-MS confirmed the expected molecular formulas for all compounds, with observed [M + H]⁺ masses within 2.1 ppm of calculated values (Table ).

For the assay, cells were seeded in differentiation medium and treated with recombinant MSTN in the presence or absence of test compounds. Compounds 1 and 2 inhibited MSTN-induced myofiber atrophy with IC₅₀ values of 0.42 ± 0.08 and 0.61 ± 0.11 μM, respectively (Figure and Figure S10), and increased myotube cross-sectional area (CSA) at concentrations at or above the EC₈₀ for atrophy-gene suppression. In contrast, free-acid analogues 3 and 4 and scrambled control scr showed substantially reduced activity, consistent with an ester-prodrug-dependent mechanism of cellular entry and intracellular hydrolysis. Compound 1 also suppressed MSTN-induced Smad2/3 phosphorylation in a concentration-dependent manner (Figure ; see also Figure S17). , In selectivity profiling, compound 1 preferentially inhibited MSTN/GDF-8 signaling relative to GDF-11, Activin A, and TGF-β1 (Figure ).

3.

Dose-dependent inhibition of MSTN-induced myofiber atrophy by compounds 1, 2, 3, 4, and scr. Individual data points represent independent biological replicates.

4.

Concentration-dependent suppression of MSTN signaling by compound 1 in C2C12 myotubes. (A) Representative Western blot for p-Smad2/3, total Smad2/3, and β-actin across increasing concentrations. (B) Quantification of p-Smad2/3/total Smad2/3 ratio. Data are mean ± SD (n = 3).

7.

TGF-β superfamily selectivity profiling of compound 1. (A) Bar chart comparing IC50 values across MSTN/GDF-8, GDF-11, Activin A, and TGF-β1. (B) Dose–response curves for the four cytokines.

Individual data points shown in the figures represent independent biological replicates (distinct wells derived from separate myofiber differentiations) and not technical replicates of the same sample; this distinction is stated explicitly in each figure legend.

All-d configuration conferred marked resistance to trypsin- and chymotrypsin-mediated proteolysis, with >90% of compounds 1–4 remaining intact after 24 h incubation (Figures S11–S15 and S19), in contrast to the corresponding l-peptide, which was fully degraded within 30 min.

Because compounds 1 and 2 contain a hydrolyzable isopropyl ester, we additionally evaluated their stability in mouse serum. Both esters showed measurable but incomplete hydrolysis over 4 h (t _1/2 > 4 h), indicating that serum esterase activity alone does not fully account for their reduced cellular potency relative to compounds 3 and 4; the serum stability data are provided in the Supporting Information (Figure S16).

The markedly greater potency of the isopropyl ester prodrugs (1, 2) relative to that of the free acids (3, 4) is consistent with ester-mediated enhancement of membrane permeability and subsequent intracellular activation. Esterase rescue experiments with the esterase inhibitor BNPP and LC–MS/MS detection of intracellular free-acid formation further support this prodrug mechanism (Figures and ). Because inhibition by the free acids was not fully abolished, we cannot strongly argue against a minor contribution from direct extracellular activity of the free acid; rather, our data support a predominant but not exclusive prodrug-dependent mechanism. Attribution of the potency difference to non-esterase-mediated uptake differences (e.g., passive permeability) rather than solely to intracellular ester hydrolysis also remains possible and warrants direct mechanistic follow-up (e.g., LC–MS/MS quantification of intracellular free-acid accumulation).

5.

Intracellular prodrug activation of compound 1. (A) Extracted ion chromatograms from C2C12 myotube lysates showing detection of compound 3 (active metabolite) after treatment with compound 1. (B) Dose–response curves for MSTN signaling inhibition by compounds 1 and 3, with and without the esterase inhibitor BNPP, demonstrating esterase-dependent activation.

6.

Prodrug mechanism: BNPP esterase rescue and LC–MS/MS intracellular detection. (A) Esterase rescue assay in C2C12 myotubes showing dose–response plots for compounds 1–4 with and without BNPP. (B) LC–MS/MS chromatograms from C2C12 cell lysates treated with compound 1, showing parent compound and intracellularly formed metabolite.

The reduced activity of the scrambled control scr (identical composition, non-native sequence) argues against a purely nonspecific, cationic-amphiphile mechanism and supports sequence-dependent target engagement, although we cannot rule out a minor contribution from sequence-independent cationic uptake. In vivo, 28-day administration of compound 1 (0.01 mg/kg, i.p., every 48 h) increased soleus muscle weight and grip strength in BALB/c mice (Figure ; see also Figure S18).

8.

In vivo muscle hypertrophy in BALB/c mice following 28-day administration of compound 1. (A) Experimental timeline (0.01 mg/kg, i.p., q48h × 28 days; n = 8/group). (B) Body weight monitoring over 28 days. (C) Normalized soleus muscle weight. (D) Grip strength measurement. (E) Representative H&E muscle cross-sections and quantification of myofiber cross-sectional area (CSA).

This study has several limitations that should be considered in interpreting the results. First, direct biophysical confirmation of binding between the d-tripeptides and the MSTN prodomain-binding site (e.g., surface plasmon resonance, isothermal titration calorimetry, or NMR) was not performed. Activity was inferred from cell-based functional assays alone.

Second, in vivo pharmacokinetic/pharmacodynamic characterization was not performed. Plasma and skeletal muscle tissue exposure of compounds 1 and 2 were not measured, so the relationship between in vitro potency and achievable in vivo exposure remains unknown.

Third, membrane permeability was not directly assessed. PAMPA or Caco-2 permeability data would help clarify the extent to which ester prodrug design improves passive diffusion versus other uptake mechanisms.

Fourth, compound 3 (the nonhydrolyzable-ester-lacking free acid analogue of compound 1) was not evaluated in vivo as a negative control, which would strengthen causal attribution of any in vivo effect to the prodrug mechanism rather than to off-target or free-acid activity.

Assessment in pathological atrophy models will be warranted to fully substantiate therapeutic claims, including disuse (hindlimb immobilization/denervation), cachectic, and aged sarcopenic models. Taken together, these findings should be regarded as qualitative, proof-of-concept validation of a minimalist d-tripeptide prodrug strategy for MSTN prodomain mimicry rather than a definitive demonstration of therapeutic efficacy.

In conclusion, we have designed, synthesized, and characterized minimalist all-d tripeptide prodrug mimetics of the MSTN prodomain-derived minimum active 23-mer fragment (H-WRQ­NTR­YSR­IEA­IKI­QIL­SKL­RL-NH2, positions 21–43 of the mouse MSTN prodomain). Isopropyl ester prodrugs 1 and 2 selectively inhibited MSTN-induced atrophy signaling with submicromolar potency and superior proteolytic stability relative to the parent l-peptide. Compound 1 is nominated as a lead for further optimization. Given the limitations discussed aboveabsence of direct biophysical binding data and of in vivo pharmacokinetic and compound 3 control studiesthis work should be viewed and interpreted as a qualitative validation of the minimalist d-tripeptide prodrug concept, providing a chemical foundation for future mechanistic and in vivo studies rather than a stand-alone therapeutic claim.

Experimental Methods

Peptide Synthesis. Wang resin (0.2 mmol scale) was swollen in DMF (3 × 5 min) prior to Fmoc solid-phase peptide synthesis using standard HBTU/HOBt/DIPEA coupling and 20% piperidine/DMF deprotection. Fmoc-d-amino acids (d-Arg­(Pbf), d-Leu, d-Abu, d-Ala) were coupled sequentially. Cleavage from resin and side-chain deprotection were performed with TFA/TIS/H2O (95:2.5:2.5, v/v/v, 2 h), affording the C-terminal free acid directly from Wang resin. Crude peptides were precipitated in cold diethyl ether, purified by preparative RP-HPLC, and lyophilized. Isopropyl esters (1, 2, scr) were prepared from the purified free acids (3, 4-type precursors) by Fischer esterification with isopropanol/thionyl chloride, followed by a second round of RP-HPLC purification.

Cell Culture and Atrophy Assay. For the assay, cells were seeded as C2C12 myoblasts and differentiated into myofibers over 4–5 days in low-serum differentiation medium. Differentiated myofibers were treated with recombinant mouse MSTN (10 ng/mL) in the presence or absence of test compounds for 48 h, after which myofiber cross-sectional area (CSA) and atrophy-associated gene expression (MuRF1, Atrogin-1) were quantified.

Proteolytic Stability. Compounds were incubated with trypsin or α-chymotrypsin (1:50 enzyme:substrate, w/w) in PBS (pH 7.4, 37 °C) and sampled at 0, 0.5, 1, 2, 4, 8, and 24 h; remaining intact peptide was quantified by analytical RP-HPLC.

Serum Ester Stability. Compounds 1 and 2 were incubated in 90% mouse serum at 37 °C for up to 4 h; ester hydrolysis to the corresponding free acid was monitored by LC–MS, confirming measurable but incomplete esterase-mediated conversion over the assay window (see Supporting Information, Figure S16).

Statistical Analysis. Data are presented as mean ± SEM from independent biological replicates as indicated in each figure legend. Statistical comparisons were performed by one-way ANOVA with Tukey’s post hoc test; p < 0.05 was considered significant. IC₅₀ and EC₈₀ values were derived from four-parameter logistic fits, normalized and minimized against vehicle and untreated controls, respectively.

Safety. No unexpected or unusually high safety hazards were encountered.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.6c00358.

• Materials and methods, solid-phase peptide synthesis details, HPLC, HRMS, cell culture and atrophy assay protocols, proteolytic stability assays, serum ester stability assay; figures showing concentration-dependent inhibition of MSTN-induced Smad2/3 phosphorylation, soleus muscle wet weight during unloading, proteolytic stability; paragraph describing NMR spectra purity data (DOCX)

The authors declare no competing financial interest.