Mirmahdi et al. 2021 — Biodecontamination of milk and dairy by probiotics: review (Italian J. Food Science)
This narrative review from an Iran–Pakistan author group, published in the Italian Journal of Food Science (Codon Publications, Special Issue 2, 2021), surveys (i) heavy metal and mycotoxin contamination patterns in milk and dairy products worldwide and (ii) the use of probiotic lactic acid bacteria (LAB), kefir grains, and yeast — especially Saccharomyces cerevisiae — as biosorbents for in-product decontamination. Heavy metals are the secondary focus; the dominant analytical focus is aflatoxin M1 (AFM1) and other mycotoxins. For the Heavy Metal Index the load-bearing content is Table 1 (country-level Pb / Cd / As concentrations in milk from ten primary studies, 2014–2021), Table 5 (heavy-metal removal efficiencies by LAB and yeasts in milk and kefir matrices), the cited International Dairy Federation (IDF) maximum permissible limits, and the FDA / EU AFM1 ceilings. No primary occurrence measurements are generated; the review is a directional gateway to the cited primary literature.
Key numbers
The review does not generate primary contamination data. Units in the original tables are heterogeneous (mg/L, mg/kg, µg/L, µg/kg, ng/L, ng/g) and are not harmonised in the review; values below are reproduced as printed.
International Dairy Federation maximum permissible limits in milk, as cited by the review from Malhat et al. 2012 (units as printed in the source). The source prints the assignments as: 2.6 µg/kg for copper, 10 µg/kg for cadmium, 20 µg/kg for lead, and 328 µg/kg for zinc. The Cu and Cd magnitudes as printed are inverted relative to the usual IDF orientation (Cd is typically the lower limit and Cu the higher) and should be checked against Malhat et al. 2012 before being treated as a regulatory anchor; the page reports the wording as the review prints it. See Verification notes.
FDA and European Commission maximum permissible limits for AFM1 in milk (review p. 79): FDA 0.5 µg/kg; European Commission 0.05 µg/kg.
Table 1 — milk heavy-metal contamination across 10 primary studies, 2014–2021 (units as printed; “lg” in the original is a typesetting artefact for µ).
| Country | Metal | Concentration | Primary reference cited |
|---|---|---|---|
| Egypt | Pb | 0.044–0.751 mg/L | Meshref et al. 2014 |
| Egypt | Cd | 0.008–0.179 mg/L | Meshref et al. 2014 |
| Serbia | Pb | 54.3–95.2 µg/kg | Suturović et al. 2014 |
| Serbia | Cd | 2.13–4.82 µg/kg | Suturović et al. 2014 |
| Iraq | Pb | 32 µg/L | Alani & Al-Azzawi 2015 |
| Pakistan | Pb | 0.014 mg/kg | Ismail et al. 2015 |
| Pakistan | Cd | 0.001 mg/kg | Ismail et al. 2015 |
| Bangladesh | Pb | 0.2 mg/L | Muhib et al. 2016 |
| Bangladesh | Cd | 0.073 mg/L | Muhib et al. 2016 |
| Iran | Pb | 14.0 µg/kg | Shahbazi et al. 2016 |
| Iran | Cd | 1.11 µg/kg | Shahbazi et al. 2016 |
| Brazil | Pb | 2.12–37.36 µg/L | de Oliveira et al. 2017 |
| Mexico | Pb | 0.03 mg/kg | Castro-González et al. 2018 |
| Mexico | As | 0.12 mg/kg | Castro-González et al. 2018 |
| Poland | Pb | 5.24 µg/L | Halagarda et al. 2018 |
| Poland | Cd | 0.15 µg/L | Halagarda et al. 2018 |
| Turkey | Pb | 0.0055 mg/L | Seğmenoğlu & Baydan 2021 |
| Turkey | Cd | 0.088 mg/L | Seğmenoğlu & Baydan 2021 |
| Turkey | As | 0.002 mg/L | Seğmenoğlu & Baydan 2021 |
Authors’ framing (p. 79–80): Pb in Iraq, Brazil, China, Spain, and Italy was reported above the IDF maximum permissible limit in the cited primary studies; Cd in Poland and Spain was above limit. AFM1 in China and India was above limit per the Table 2 mycotoxin synthesis (not reproduced here in full; covers 21 countries, 2014–2021, AFM1 / AFB1 / OA / ZEA / α-ZEA in milk).
Table 5 — heavy metal bioremoval in milk and dairy products (review p. 84).
| Matrix | Microorganism | Metal | Removal % w/w | Conditions | Primary reference cited |
|---|---|---|---|---|---|
| Milk | Saccharomyces cerevisiae | Pb | 70% | Yeast inoculation 22×10⁸ CFU; lead spike 70 µg/L | Massoud et al. 2019 |
| Kefir | Lactococcus lactis, Kluyveromyces marxianus co-culture | Ni | 81.53% | 10 d contact | Cherni et al. 2020 |
| Kefir | same co-culture | Cu | 73.45% | 10 d contact | Cherni et al. 2020 |
| Kefir | same co-culture | Cd | 79.48% | 10 d contact | Cherni et al. 2020 |
| Kefir | same co-culture | Pb | 68.53% | 10 d contact | Cherni et al. 2020 |
| Kefir | same co-culture | Fe | 58.17% | 10 d contact | Cherni et al. 2020 |
| Milk | S. cerevisiae | Cd | 70% | Cd spike 80 µg/L; 30×10⁸ CFU; 4 d storage | Masoud et al. 2020 |
| Milk | Lactobacillus acidophilus | Pb | 80% | 1×10¹² CFU; Pb spike 100 µg/L; 4 d | Massoud et al. 2020b |
| Milk | L. acidophilus | Cd | 75% | 1×10¹² CFU; Cd spike 100 µg/L; 4 d | Massoud et al. 2020b |
| Milk | S. cerevisiae | Hg | 70% | Hg spike 80 µg/L; 22×10⁸ CFU biomass; 30 min contact | Massoud et al. 2021 |
The review’s qualitative summary (p. 86 Conclusions) is that kefir grains had the best ability across metals among the methods reviewed and that S. cerevisiae and L. acidophilus are the most promising single-organism biosorbents for Pb / Cd / Hg in milk under the spiked-laboratory conditions of the cited primary studies. The cited studies are all laboratory spike–and–removal experiments, not field decontamination of naturally contaminated milk.
Mechanism statements made by the review (p. 85, not pooled measurements):
- Bioremoval is dominated by adhesion of toxins and metal ions to cell-wall components of the microorganism (mannans and β-1,3-glucan / β-1,6-glucan chains in S. cerevisiae; peptidoglycan and teichoic / lipoteichoic acids in LAB; a proteinaceous S-layer and neutral polysaccharides as minor components).
- Cell viability is not required for binding; nonviable, heat-, acid-, and ultrasound-treated cells often show higher binding capacity than viable cells because treatment denatures proteins, exposes binding sites, and increases cell-wall permeability (Karazhiyan et al. 2016; Haskard et al. 2001).
- AFM1 binding is reported as a weak noncovalent interaction with cell-wall components; binding capacity varies by strain (El Khoury et al. 2011; Turbic et al. 2002).
- Adsorption of patulin, AFM1, and other mycotoxins by LAB and by kefir grains in simulated gastrointestinal conditions is reversible (Zoghi et al. 2020; Taheur et al. 2017); at pH 3, further amounts of toxins are released.
Table 3 — fermented dairy products and their starter cultures (review p. 81; not contamination data, supply-chain context only): Acidophilus milk; buttermilk (Egypt and Ethiopia); cheddar cheese; matzoon (Armenia); Leben (Arab world, “from camel milk”, Lactococcus lactis, Lactobacillus pentosus, L. plantarum, L. brevis, Pediococcus pentosaceus); kishk (Arab world); kumis (Central Asia, Turkic countries; Lactococcus lactis, Streptococcus thermophilus, Lactobacillus delbrueckii subsp. bulgaricus, L. helveticus, L. casei subsp. Pseudoplantarum, L. brevis, Kluyveromyces marxianus var. lactis, Saccharomyces cerevisiae, Candida inconspicua, C. maris); ymer (Denmark); kefir (Balkans and Baltics); dahi and mishti doi (India); matsoni (Georgia); wara, biruni, mish, rob (Africa); doogh (Iran, Lactobacillus acidophilus, L. rhamnosus, L. casei, Bifidobacterium lactis); yogurt (Serbia); clabber (United States).
Note that the Leben row is the only mention of camel milk in the review and is a single-row supply-chain note; the filename of the source PDF (mirmahdi2021-camel-milk-heavy-metals.pdf) is misleading on this point. See Verification notes.
Methods (brief)
Narrative review (declared “review article” by the journal; received 17 April 2021, accepted 31 May 2021, published 1 July 2021). No systematic-review protocol, PRISMA flow, or quantitative pooling. Five tables aggregate primary-study findings on (1) milk heavy metals 2014–2021, (2) milk mycotoxins 2014–2021, (3) fermented dairy products and starters, (4) aflatoxin decontamination by LAB / yeast / kefir, and (5) heavy-metal decontamination by LAB and yeast. Evidence tier B: useful directional gateway to the cited primary literature on probiotic / yeast bioremoval of heavy metals in milk and to country-level milk Pb / Cd / As datapoints, but the review does not normalise units across cited studies, does not assess quality of the primary studies, and does not derive pooled occurrence or pooled removal estimates. Funding: none declared. Competing interests: none declared.
Implications
Certification: This review is not a primary occurrence dataset. It is useful for HMT&C as a directional pointer to ten country-level milk Pb / Cd / As primary studies (Table 1) and to laboratory-scale bioremoval evidence (Table 5). It does not, on its own, support a per-row standard. The IDF maximum permissible limits as printed in the review should be checked against the original Malhat et al. 2012 source before being used as a regulatory anchor for any HMT&C dairy row (see Verification notes on the apparent Cu / Cd magnitude inversion in the printed text).
Courses: Useful for a supply-chain and biotechnology module showing (a) that bioremoval of metals and aflatoxins in fermented dairy is an active research area, (b) that kefir-grain and S. cerevisiae systems have laboratory-spike removal efficiencies of 58–82 % across Pb / Cd / Hg / Ni / Cu / Fe, and (c) that cell-wall adhesion (mannans, β-glucans, peptidoglycan, teichoic acids) is the proposed mechanism, with viability not required.
App: Not a source of usable ppb occurrence values for contamination_profile. Useful as a citation hub for the milk-and-dairy contamination profile when the underlying primary references (Meshref 2014, Suturović 2014, Alani & Al-Azzawi 2015, Ismail 2015, Muhib 2016, Shahbazi 2016, de Oliveira 2017, Castro-González 2018, Halagarda 2018, Seğmenoğlu & Baydan 2021) are later ingested in their own right.
Wiki pages this source may touch
Verification notes
- Filename mismatch. The source PDF was filed in
raw/Manual Fetch Discovery/under the filenamemirmahdi2021-camel-milk-heavy-metals.pdf, which suggests a camel-milk study. The paper is in fact a general dairy biodecontamination review; camel milk is mentioned only as the substrate for Leben in Table 3’s supply-chain inventory. The cite-key was set tomirmahdi2021-dairy-decontamination-probiotics-reviewto reflect actual scope; the raw file path is preserved as-fetched. - IDF Cu / Cd magnitude orientation. The review prints (p. 79): “Maximum permissible limits of heavy metal contents in milk (considered by International Dairy Federation) are 2.6 µg/kg for Copper, 10 µg/kg for Cadmium, 20 µg/kg for lead, and 328 µg/kg for zinc.” The 2.6 µg/kg Cu / 10 µg/kg Cd assignment is unusual — the IDF / Codex Cd ML for milk is commonly cited at lower magnitude than Cu. The wording is reproduced as printed; the IDF reference should be cross-checked against Malhat et al. 2012 before this is used as a regulatory anchor.
- Table 1 unit and country attribution preserved verbatim. “lg/kg” in the Serbia row of the printed table is a typesetting artefact for µg/kg. The Iraq / Brazil rows use µg/L rather than mg/L; Pakistan / Mexico use mg/kg; Egypt / Bangladesh use mg/L; Iran / Serbia use µg/kg. Reading-across countries without unit harmonisation will give wrong magnitudes — this is the review’s own framing, not a transcription change.
- Brand firewall (Part 12). No commercial dairy brand names appear in the source or this page. Scientific vendor / strain names retained per the 2026-05-17 carve-out (e.g., L. acidophilus, S. cerevisiae, Lactococcus lactis, Kluyveromyces marxianus).
- Wiki/HMT&C firewall (Part 2). This page reports what the review reports; no synthesis across other wiki sources, no HMT&C threshold proposals.
- Products array intentionally empty. The review covers milk and several fermented dairy products at the supply-chain level but does not present primary contamination data tied to a specific HMTc-locked product row. Routing to ingredient pages and metal pages is the appropriate scope.
- Mycotoxin content not extracted in detail. Table 2 (AFM1 / AFB1 / OA / ZEA / α-ZEA across 21 countries) and Table 4 (aflatoxin decontamination by LAB / yeast / kefir) are summarised in the body text but not transcribed in full; HMI’s metals-and-supply-chain remit deprioritises mycotoxin primary numbers. The Table 2 country set is referenced (CN, HR, RS, IR, MK, PK, AR, BA, IT, TZ, MY, XK, SV, TR, ET, KE, BR, EC, ES, IN, MA, MW) but kept off the
jurisdictions:frontmatter to avoid implying heavy-metal coverage for those countries.
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.