Tran et al. 2025 — Magnesium chelation of spent-brewer’s-yeast and soybean-meal peptides
Tran, Ha, Bui, Dam, To, Pham, and Le (School of Chemistry and Life Sciences, Hanoi University of Science and Technology) report a primary process-development study that compares spent brewer’s yeast (SBY) and soybean meal (SBM) — two protein-rich byproducts of Vietnamese food-industry waste streams — as substrates for Alcalase-hydrolysed peptide fractions that chelate magnesium(II), with the chelated peptide intended for animal-feed and human-nutrition magnesium supplementation rather than for heavy-metal sequestration. After autolysis and Alcalase hydrolysis followed by cross-flow filtration into 0.2 µm, 100 kDa, 10 kDa, and 3 kDa permeate fractions, the ≤3 kDa SBY peptide fraction reached >98 % magnesium chelating yield at pH 2 to 4 and Mg²⁺ loading 5 to 25 mM and outperformed all SBM fractions; FTIR amide-II (1579→1612 cm⁻¹), N–H stretch (3127→3271 cm⁻¹), and carboxyl (1398→1412 cm⁻¹) band shifts implicate amino and carboxyl groups as the Mg²⁺ binding sites; the spray-dried magnesium-chelated peptide product showed 1000 nm-scale particle size, in-vitro gastric stability, and HEK293 cytotoxicity below MgCl2 reference at 256 µg/mL. Magnesium is not among the ten HMI HMTc analytes (Pb, tAs, Cd, MeHg, tHg, iAs, Ni, Al, Cr-VI, Sn) and no contamination measurement is reported for any heavy metal in any matrix; the source enters the HMI corpus as upstream peptide-divalent-metal-chelation mechanism context alongside the peptide-Fe, peptide-Ca, peptide-Zn, peptide-Cu chelation literature already represented in the same Kimi peptide folder (han2025-peptide-zinc-complexation-aquatic-review, lu2022-metal-chelating-peptides-review, ding2015-soy-glycinin-thiol-peptides-heavy-metal-complexation, sosnowska2025-phage-display-peptide-copper-chemosensor).
Key numbers
| Step | Variable | Value |
|---|---|---|
| SBY raw substrate | Total protein, water content (page 2, §2.1) | 57.83 ± 2.00 % protein (Kjeldahl), 79.02 ± 0.32 % water |
| SBM raw substrate | Total protein, water content (page 2, §2.1) | 48.62 ± 0.44 % protein (Kjeldahl), 11.16 ± 0.57 % water |
| SBY autolysis end-state (12 h, pH 6, 50 °C) | Soluble protein increase from t=0 | 6.69-fold rise (Fig. 1A, page 3) |
| SBM hydrolysis end-state (8 h, pH 8, 50 °C) | Soluble protein increase from t=0 | 2.67-fold rise (Fig. 1B, page 3) |
| SBY hydrolysate Mg²⁺ chelating yield (20 h) vs SBM (8 h) | Endpoint chelating-yield comparison (Fig. 2, page 4) | SBY chelating yield was 1.34× that of SBM at endpoints (paper wording, page 4 §3.2: “the chelating yield of SBY was 1.34-time higher than that of SBM”) |
| SBM hydrolysate Mg²⁺ chelating yield, 5 h plateau | Chelating yield (Fig. 2B, page 4) | ≈40 % from h 5 through h 8 (Fig. 2B, chart read; no tabulated value in text) |
| SBY 3 kDa permeate (Pe 3 kDa) specific chelating yield (Fig. 3A; abstract page 1) | Mg²⁺ chelated per g protein | 94.44 mg/g protein (highest specific yield reported in the study) |
| SBY 3 kDa permeate optimal conditions | pH for peak yield (Fig. 4A, page 5; text page 6 §3.3) | pH 2 to 4 window; peak at pH 3 with 67.05 % chelating yield at 150 mM Mg²⁺ loading |
| SBM 3 kDa permeate optimal pH (Fig. 4D, page 5; text page 6 §3.3) | pH for peak yield | Peak at pH 3 with 51.52 % chelating yield at 150 mM Mg²⁺ loading |
| SBY 3 kDa permeate Mg²⁺ loading optimum (Fig. 4C, page 5; text page 6 §3.3) | Loading mM → yield | 5 mM and 25 mM Mg²⁺ → 99.83 % chelating yield peak; declining to 74.3 % at 150 mM |
| SBM 3 kDa permeate Mg²⁺ loading optimum (Fig. 4F, page 5; text page 6 §3.3) | Loading mM → yield | 25 mM Mg²⁺ → 87.39 % peak; declining to 42.05 % at 150 mM |
| Reaction time × temperature (Fig. 4B SBY / 4E SBM) | Effect | SBY peaks 74.8 % at 30 min / 40 °C; SBM peaks 47.81 % at 30 °C; temperature and time effects “minor” per discussion (page 6) |
| FTIR amide II band shift (Fig. 5, page 6) | SBYP → Mg-SBYP | 1579.25 cm⁻¹ → 1611.71 cm⁻¹ (peptide–Mg interaction, paper’s framing) |
| FTIR N–H stretch shift | SBYP → Mg-SBYP | 3127.17 cm⁻¹ → 3271.24 cm⁻¹ (N–H vibration confirms amino-group involvement) |
| FTIR carboxyl C–O shift | SBYP → Mg-SBYP | 1398.28 cm⁻¹ → 1411.58 cm⁻¹ (–COOH interaction with Mg²⁺); additional shifts 1053.41 → 1023.34 cm⁻¹ attributed to C–O stretching of hydroxyl/carboxyl |
| Particle size distribution (PSD, Fig. 6, page 7) | Spray-dried Mg-SBYH (chelated hydrolysate) range | ≈1400 nm to ≈3500 nm (multimodal) |
| Particle size distribution | Spray-dried Mg-SBYP (chelated 3 kDa permeate) range | ≈770 nm to ≈1200 nm |
| In-vitro gastrointestinal stability (Fig. 7, page 7) | Free Mg²⁺ retention at end of 6 h simulation, MgCl2 control | 89.15 % |
| In-vitro gastrointestinal stability | Free Mg²⁺ retention at end of 6 h simulation, Mg-SBYP | 56.85 % after first 30 min gastric, hovering ≈50 % through end (lowest of the three samples) |
| Cytotoxicity (HEK293, MTT, Table 2, page 7) | Negative control, ellipticine positive control | 0 %, 93.04 ± 2.12 % cell death |
| Cytotoxicity at 256 µg/mL | MgCl2 / SBY peptide / Mg-SBYP cell death | 1.09 ± 0.02 % / 0.33 ± 0.03 % / 0.11 ± 0.02 % |
| Cytotoxicity IC50 (256 µg/mL exposure) | MgCl2, SBY peptide, Mg-SBYP | All >256 µg/mL |
| SBY 3 kDa permeate amino-acid profile, key residues (Table 1, page 4) | Glu, Asp, Lys, Arg, His | Glu 14.88 %, Asp 10.94 %, Lys 6.49 %, Arg 7.07 %, His 3.19 % |
| SBM 3 kDa permeate amino-acid profile, key residues (Table 1, page 4) | Glu, Asp, Lys, Arg, His | Glu 15.82 %, Asp 10.55 %, Lys 6.01 %, Arg 8.07 %, His 3.04 % |
Methods (brief)
Spent brewer’s yeast was Saccharomyces pastorianus collected from Habeco Co. Ltd (Hanoi) at the last day of main fermentation, transported cold, washed 1:4 v/v in distilled water, centrifuged 6,000 rpm to remove supernatant, and stored at −20 °C; soybean meal was supplied by Dabaco Co. Ltd (Vietnam), milled and sifted through mesh 80 sieve. Total protein on both substrates by Kjeldahl. Autolysis of SBY ran in a Biostat B plus bioreactor at 7 % dry-matter, pH 6.0, 50 °C, 600 rpm, 1000 mL working volume, with hourly aliquots boiled 10 min to inactivate enzymes; after 12 h the autolysate was centrifuged 10,000 rpm × 10 min and the supernatant transferred to Alcalase (Novozymes; 30 mg/g protein dosage) hydrolysis at pH 8.0, 50 °C, 600 rpm, 8 h. SBM was autoclaved at 121 °C × 15 min before enzymatic hydrolysis under the same Alcalase conditions. Hydrolysates were centrifuged 10,000 rpm × 15 min and cross-flow filtered (Quixstand bench-top CFF) through 0.2 µm (CFP-2-E-3MA), 100 kDa (UFP-100-E-3MA), 10 kDa (UFP-10-E-3MA), and 3 kDa (UFP-3-C-3MA) cartridges.
Chelation reactions ran at micro-scale (1 mL total: 50 µL of 3 M MgCl2 + 950 µL peptide solution, vortex 3 min) and macro-scale (1000 mL in Biostat B plus, 300 rpm, 10 min); chelation mixtures were centrifuged 10,000 rpm × 5 min to remove precipitate. Free Mg²⁺ was determined by colorimetric assay with 4-(2-pyridylazo) resorcinol, absorbance 506 nm on a GeneQuant 1300. Chelating yield is the ratio chelated/loaded Mg, in %; specific chelating yield is mg Mg(II) chelated per g peptide.
Spray-dried magnesium chelating peptide (MgCP) was prepared on a Buchi C-290 with 0.1 % magnesium stearate, inlet 120 °C, outlet 60 °C; total Mg in spray-dried MgCP by AOAC 968.08. Soluble protein by Lowry; α-amino by ninhydrin with glycine standard; amino-acid profile by ion-exchange chromatography (Hitachi L-8800) per ISO 13903:2005 (17 amino acids including Cys/cystine sum) and AOAC 994.12 (Cys, Met). FTIR spectra on NICOLET IS20 (Thermo Fisher) in ATR mode, 500–4000 cm⁻¹. Particle size distribution on a Zetasizer Blue (Malvern, MAL1257886) at 25 °C, refractive indices 1.45 (peptide) / 1.33 (water), equilibration 30 s.
In-vitro gastrointestinal stability followed Brodkorb et al. INFOGEST static protocol: MgCP dissolved in 0.06 N HCl pH 2 to final 10 mg/mL, pre-warmed 37 °C × 30 min, pepsin (P7000-Sigma Aldrich, 599 U/mg, final 0.05 mg/mL) added at pH 2 for 120 min at 37 °C, 120 rpm; bile salts (Merck 1.03756.0500) added to pH 7.4 with NaOH, then pancreatin (Merck 1.071300.1000) at final 0.05 mg/mL for intestinal stage 4 h at 37 °C. Hourly aliquots boiled to stop enzymatic activity before free Mg²⁺ measurement; MgCl2 3 M control. Cytotoxicity on HEK293 in DMEM + 2 mM L-glutamine + penicillin/streptomycin + 10 % fetal bovine serum, seeded 1.5 × 10⁵ cells/well in 96-well plates; MgCP exposure at 256 to 8 µg/mL; ellipticine positive control; DMSO ≤1 % negative control; MTT assay read at 540/720 nm on Tecan Spark; n = 3 independent replicates. Cell-inhibition >50 % triggered IC50 calculation by TableCurve AISN.
Statistics: Tukey test for between-treatment significance (p < 0.05) in SPSS; descriptive statistics in Megastat (Excel add-in). All experiments in triplicate.
Implications
The paper is a single-laboratory process-development study whose central claim — that ≤3 kDa SBY peptide permeate chelates Mg(II) at >98 % yield under acidic loading and that the spray-dried product is non-cytotoxic to HEK293 at 256 µg/mL — is a feasibility demonstration for valorising brewery and soy-processing waste streams into a magnesium supplement intended for animal-feed and human-nutrition use, not a contamination measurement on any HMI ingredient or product. Magnesium is not among the ten HMTc analytes (Pb, tAs, Cd, MeHg, tHg, iAs, Ni, Al, Cr-VI, Sn) and the source contributes no occurrence data to any HMI metal, ingredient, or product page, does not constrain any HMTc per-row standard, and does not warrant a Part 9 synthesis trigger on any cell of the contamination index. The mechanism finding (carboxyl groups of Glu/Asp side chains and amino groups of Lys/Arg/His N-terminal and side chains as the dominant Mg²⁺ binding sites, confirmed by FTIR amide-II and N–H stretch shifts) is the same coordination chemistry framework already documented for peptide-Fe, peptide-Ca, peptide-Zn, and peptide-Cu chelates in the adjacent Kimi-folder corpus and applies in principle to divalent heavy metals (Pb²⁺, Cd²⁺, Ni²⁺), but the paper itself does not test the specificity of SBY/SBM peptides against any HMI analyte. As a corpus addition this is upstream peptide-divalent-metal coordination context, not occurrence data; downstream wiki implications are confined to the metal-binding-peptide review pages and would only enter the contamination synthesis if a follow-up paper tested the same SBY/SBM peptide fractions against Pb²⁺ or Cd²⁺.
Verification notes
- PDF read in full (
9 pages) via the Read tool: title and abstract (page 1), introduction and Materials-and-Methods §2.1–§2.11 (pages 1–3), Results §3.1 peptide recovery with Fig. 1 soluble protein/α-amino curves and Table 1 amino-acid profile (pages 3–4), §3.2 chelating ability with Fig. 2 chelating-yield curves and Fig. 3 specific chelating yield by fraction (page 4), §3.3 chelating conditions with Fig. 4 pH/T/loading panels for SBY and SBM (page 5), §3.4 FTIR spectra with Fig. 5 SBYP vs Mg-SBYP overlay (page 6), §3.5 particle size with Fig. 6 PSD before/after spray drying (page 7), §3.6 in-vitro digestion with Fig. 7 free-Mg²⁺ evolution (page 7), §3.7 cytotoxicity with Table 2 HEK293 viability (page 7), conclusion (page 8), and the full 38-entry reference list (pages 8–9) were all reviewed. Tables 1 and 2 read for exact figures. - DOI
10.1016/j.jgeb.2025.100528, raw handleMFK_57-magnesium-chelation-of-low-molecular-weight-pep, and cite-key stemtran2025-magnesium-chelationchecked againstwiki/sources/(grep -l for DOI, raw_handle, and author/year cite-key stem); no existing page on any of the three identity dimensions. - Open access under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 (CC-BY-NC-ND 4.0) per the first-page rights notice (“This is an open access article under the CC BY-NC-ND license”); recorded as
license: cc-by-nc-nd-4.0. Same constraint as sosnowska2025-phage-display-peptide-copper-chemosensor in the same Kimi peptide folder. - Metals discussed in the paper are magnesium (the central analyte) with passing references to iron, calcium, and zinc peptide chelates as the literature comparators cited from references 2, 3, 9, 10, and 11. Magnesium is not among the ten HMI HMTc analytes (Pb, tAs, Cd, MeHg, tHg, iAs, Ni, Al, Cr-VI, Sn); the other divalent metals appear only as analogue-chemistry context, not as measurements in any matrix.
metals: []is therefore correct under the same convention used for sosnowska2025-phage-display-peptide-copper-chemosensor (Cu, also non-HMTc) and han2025-peptide-zinc-complexation-aquatic-review (Zn). - No ingredient or product is measured for heavy-metal contamination;
ingredients: []andproducts: []. Spent brewer’s yeast and soybean meal are food-industry byproducts that could appear as substrate inputs to dietary supplements or animal feed, but the paper does not measure them for contamination and does not place a finished product on the market. The matrices field uses descriptive placeholdersmethodology-contextandindustrial-protein-byproductbecause neither of the two substrates has a current taxonomy slug; this follows the precedent in sosnowska2025-phage-display-peptide-copper-chemosensor which usedsynthetic-peptide-library, methodology-contextfor a similarly out-of-vocabulary methodology paper. No new ingredient or product page is proposed; per Part 10 of CLAUDE.md that decision is Karen’s. jurisdictions: [GLOBAL]because the work is methodology with no jurisdiction-specific occurrence claims; all seven authors are at Hanoi University of Science and Technology and the SBY and SBM substrates come from Hanoi-area suppliers (Habeco brewery and Dabaco Co. Ltd), but the conclusions are presented as a general process-development result not tied to a Vietnamese regulatory framework. Funding via the Vietnamese Ministry of Education and Training project B2023-BKA-17.- Brand firewall: the named commercial substrates (Habeco brewery yeast, Dabaco soybean meal) are substrate-supplier identifiers in a process-development paper, not branded finished products being benchmarked for contamination. Per Part 12 they are preserved in Methods because removing them would impair reproducibility of the substrate sourcing, and there is no contamination measurement being attributed to either supplier. Instrument and reagent vendors (Novozymes Alcalase, Buchi C-290, NICOLET IS20, Malvern Zetasizer Blue, Quixstand CFF, Hitachi L-8800, Tecan Spark, Sigma Aldrich pepsin, Merck bile salts and pancreatin, GeneQuant 1300, AOAC 968.08, ISO 13903:2005, AOAC 994.12) are scientific-method vendor names and reference materials per the Part 12 Exception 2 carve-out for scientific reproducibility; preserved in Methods.
- Wiki/HMTc firewall: no HMTc threshold proposals, no consumer translations, no synthesis claims about magnesium toxicology or about whether peptide-magnesium chelates could sequester heavy-metal contaminants. The paper’s claim that the spray-dried Mg-SBYP product is “suitable for animal and human consumption” is preserved verbatim as the authors’ own conclusion, not adopted as an HMI position.
- Speciation: the paper measures total Mg by colorimetric 4-(2-pyridylazo) resorcinol assay for free Mg²⁺ in solution and by AOAC 968.08 for total Mg in the spray-dried product; the chelating-yield calculation depends on the difference between loaded and free Mg²⁺ after centrifugation. No HMI speciation distinction (iAs/tAs, MeHg/tHg/iHg, Cr-VI/Cr) applies because the analyte is magnesium.
- Evidence tier: C (single-laboratory primary process-development study with triplicate replicates per condition, no inter-laboratory replication, no environmental or food-matrix validation against heavy metals, no orthogonal analytical confirmation against ICP-MS or AAS for the chelated product’s metal content beyond AOAC 968.08).
- Adjacent context: see han2025-peptide-zinc-complexation-aquatic-review and lu2022-metal-chelating-peptides-review for peptide-divalent-metal coordination-chemistry reviews, ding2015-soy-glycinin-thiol-peptides-heavy-metal-complexation for the upstream soy-protein peptide-metal binding mechanism, sosnowska2025-phage-display-peptide-copper-chemosensor for a parallel single-metal-target peptide methodology study in the same Kimi peptide folder, and luo2024-peptides-heavy-metal-remediation for the application-side review on peptide-mediated heavy-metal remediation.
Page history
The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.
| Commit | Date | Description |
|---|---|---|
| 1476f44 | 2026-06-09 | ingest: cacic2019-hemp-heavy-metals fresh from MFK/June 9 |