Hu 2024 — Engineered E. coli adsorption of Hg²⁺ and Cr⁶⁺ via MerR/ChrB

Shuting Hu and colleagues at Tianjin University constructed 24 engineered Escherichia coli BL21(DE3) strains that heterologously express the heavy-metal-binding proteins (HMBPs) MerR and ChrB from the chromium-resistant model bacterium Cupriavidus metallidurans CH34, then screened them for Hg²⁺ and Cr⁶⁺ adsorption from aqueous solution. Display mode (intracellular overexpression versus surface display via the Pseudomonas syringae ice-nucleation N-terminal anchor INPN), inter-domain linker chemistry (no linker, flexible Gly₄Ser, rigid proline-rich PAPAP, rigid helical AEAAAKEAAAKA, and a 96-bp intermediate repeat), and codon optimization (native versus codon-optimized merR_opt and chrB_opt) were varied combinatorially. The best mercury strain (M’-002: codon-optimized merR_opt surface-displayed via flexible linker on INPN) reached 658.66 µmol/g cell dry weight (CDW) at 300 µM Hg²⁺; the best chromium strain (B-008: intracellular chrB overexpression) reached 46.84 µmol/g CDW at 500 µM Cr⁶⁺. A mixed-bacterial system co-culturing M’-002 and B-008 adsorbed up to 1.14× and 4.09× more total ions than the single-strain controls in a four-metal Hg²⁺/Cr⁶⁺/Cd²⁺/As³⁺ cocktail. The paper enters the HMI corpus as upstream bioremediation-technology context for the metal-binding-peptide and HMBP literature already represented in the same Kimi peptide folder (luo2024-peptides-heavy-metal-remediation, urbina2018-biomining-peptide-metal-recovery, seregin2023-phytochelatins-sulfur-metal-chelating, ruttkay-nedecky2013-metallothionein-oxidative-stress, yang2024-metallothionein-comprehensive-review, thirumoorthy2007-metallothionein-overview). It does not measure heavy metals in any food, personal-care, or consumer matrix.

Key numbers

The paper reports adsorption capacities of engineered strains and the wild-type chassis under single-ion and mixed-ion conditions. No occurrence or exposure values for HMI heavy metals in any consumer matrix are reported; values below are the bioengineering parameters and AAS-quantified adsorption capacities the paper publishes.

Methods (brief)

Plasmid construction used pET28a as the expression vector. Native and codon-optimized merR (genome CP050332.1) and chrB (genome CP000355.2) were PCR-amplified from C. metallidurans CH34 plasmid pMOL28; codon-optimized merR_opt and chrB_opt were synthesized by GENEWIZ. EcoRI and HindIII restriction sites were appended to TF primers; HindIII and NheI flanked the linker; NheI and XhoI flanked TF primers. Linkers (FL Gly₄Ser, RL PAPAP, HL AEAAAKEAAAKA, 96-bp intermediate repeat) and the INPN anchor were inserted into five conservation plasmids (pE-NL, pE-FL, pE-RL, pE-HL, pKE-FL), ligated with T4 DNA ligase, and transformed into BL21(DE3). The resulting 24 strains include 10 surface-display constructs each for MerR and ChrB (5 linkers × 2 codon variants) plus 4 intracellular overexpression strains (M-006, M’-006, B-008, B’-008). Adsorption assays grew strains in LB broth (5 g/L yeast extract, 10 g/L NaCl, 10 g/L peptone, pH 7.2 to 7.4) to OD₆₀₀ 0.6 to 0.8 at 37 °C, cooled to 22 °C, induced with 0.2 mM IPTG (0.2 to 1.4 mM range in optimization), supplemented with target metal ions to 200 µM (50, 100, 300, 500, 1000, and 1500 µM range for concentration sweeps), and incubated for 12 h at 22 °C / 220 rpm. Cells were collected, washed three times with ddH₂O, oven-dried at 60 °C for 24 h, weighed, and microwave-digested for atomic absorption spectroscopy (AAS) quantitation of Hg²⁺ and Cr⁶⁺. Adsorption capacity (µmol/g CDW) was computed as total metal content divided by dry biomass. All experiments used three biological replicates. Pseudo-first-order and pseudo-second-order kinetic models were fitted; pseudo-first-order fit the data better but with substantial q_e,cal versus q_e,exp gaps that the authors attribute to model weakness.

For mixed-ion adsorption, the four-metal cocktail (Hg²⁺/Cr⁶⁺/Cd²⁺/As³⁺) was prepared at 40, 100, and 200 µM per ion (160, 400, and 800 µM total). For the mixed-bacterial system, the two optimal strains M’-002 and B-008 were separately activated, then co-cultured in fresh LB at 1:1 v/v at 1% inoculation ratio.

Implications

The paper is a primary bioremediation-engineering study: it constructs, optimizes, and benchmarks engineered E. coli strains for aqueous Hg²⁺ and Cr⁶⁺ removal. It does not measure Hg, Cr, Pb, Cd, As, or any other HMI-tracked analyte in any food, supplement, personal-care, or consumer matrix. Its value to the HMI corpus is upstream context for the metal-binding-peptide and HMBP literature that the same Kimi peptide folder is populating: it shows what a state-of-the-art recombinant bacterial sorbent achieves under controlled laboratory conditions, against which broader review claims about peptide- and protein-based heavy-metal remediation can be calibrated. The paper itself notes that even the best-performing strain (M’-006) fails the WHO drinking-water removal-rate target at higher influent concentrations, which bounds practitioner expectations for engineered bacterial sorbents under real wastewater conditions.

Within the wiki/HMTc firewall, this source bears on no per-row P97/P45 percentile calculation, no regulatory crosswalk, and no occurrence summary for any product category. It belongs on metals/lead, metals/mercury, metals/chromium, or any ingredient or product page only as a methods reference if a future synthesis pass on engineered-bacterial bioremediation needs to cite the underlying construction strategy. The routing layer should leave this source unattached to consumer products and ingredients.

Verification notes

• PDF read in full (13 pages: abstract, introduction, materials and methods, plasmid construction, cell growth and protein expression, metal adsorption, results across construction-and-pre-selection, growth, pre-screening, induction factors, concentration sweeps, mixed-ion adsorption, removal-rate comparison, mixed-bacterial performance, discussion, conclusions, references, supplementary information cover). Tables 1 (biosorption-by-microbes context), 2 (plasmids used), and 3 (kinetics parameters) and Figures 1 through 5 (construction profile and pre-selection bar charts; IPTG/temperature/time optimization curves; concentration sweeps; mixed-ion adsorption; adsorption-rate comparisons) reviewed.
• DOI 10.1186/s12896-024-00842-9, raw handle MFK_51-adsorption-of-hg2cr6-by-metal-binding-proteins-, and cite-key hu2024-engineered-ecoli-hg-cr-adsorption checked against wiki/sources/; no existing page (DOI, handle, and cite-key greps all empty at ingest time). The distinct hu2024-ecl-pb-lycium-glycyrrhiza page is a different first-author Hu on a Lycium/Glycyrrhiza Pb paper; cite-key collision avoided by the specific suffix here.
• Publication is BMC Biotechnology 2024;24:15, Open Access under CC BY 4.0 per the BMC license footer; license recorded as CC BY 4.0.
• HMI’s 10 HMTc analytes (Pb, tAs, Cd, MeHg, tHg, iAs, Ni, Al, Cr-VI, Sn) are referenced in the paper as bioremediation targets (Hg²⁺ and Cr⁶⁺ are the operational targets; Cd²⁺ and As³⁺ enter the mixed-ion experiments at 40 to 200 µM per ion as bystanders for selectivity assessment). None are measured in any consumer matrix. metals: is therefore [].
• ingredients: and products: are []: no food, supplement, personal-care, or consumer matrix is present. The matrix is engineered bacterial cells in LB broth spiked with heavy-metal salts.
• matrices: [engineered-ecoli, surface-display, aqueous-sorption-test, wastewater-remediation] follows the existing pattern for biosorption / bioremediation matrices in the corpus (e.g., iris-sibirica-biomass, biochar, aqueous-sorption-test, wastewater-remediation; silk-fibroin, protein-inorganic-hybrid-nanoflower, aqueous-sorption-test, wastewater-remediation).
• jurisdictions: [CN, GLOBAL] because the lab is at Tianjin University (Project No. 2018YFA0902100, National Key R&D Program of China; NSFC 22178262) and the engineered-strain technology has no jurisdiction-specific implications.
• Evidence tier C: primary research paper, but it generates no primary occurrence data for any HMI analyte in any HMI matrix; its value is upstream methods reference rather than direct evidence for any contamination claim.
• Brand firewall: no consumer brand names appear in the paper. Reagent and instrument vendors named in Materials and Methods (TransGen Biotech Co., Ltd; GENEWIZ Co., Ltd; pET28a vector; ATCC 43123D-5 for C. metallidurans plasmid pMOL28) are scientific suppliers, not consumer brands. Per the scientific-method-vendor exception in docs/gpt-collaboration/verification-checklist.md (Exception 2, locked 2026-05-17), these vendor names are appropriate to preserve as methods detail; they are not attributable to any contamination value.
• Wiki/HMTc firewall: no synthesis claims about contamination thresholds, occurrence values, or HMTc standards imported. The source does not bear on any HMI per-row P97/P45 calculation.
• Source-type: peer-reviewed (primary research article, not a review).
• Paper-internal data integrity flags (recorded, not stop-condition):
  • Abstract reports M’-002 maximal Hg²⁺ adsorption of 658.40 µmol/g; results section (p. 8) reports 658.66 µmol/g. Rounding difference, same data; both recorded.
  • Abstract and Conclusions report B-008 maximal Cr⁶⁺ adsorption of 46.84 µmol/g at 500 µM; results section (p. 8) reports 46.73 µmol/g at 500 µM. Sub-1% discrepancy, same data point; both recorded.
  • Discussion p. 11 quotes “WHO GDWQ” Hg²⁺ at 50 µg/L (0.249 µM) and Cr⁶⁺ at 6 µg/L (0.115 µM). WHO GDWQ 4th ed actually gives Hg at 6 µg/L and provisional total Cr at 50 µg/L (i.e., the paper appears to have transposed the values). The molar conversions stated in the paper are internally consistent with the numbers as the paper writes them. Recorded as the paper states without correction since this is a quoted regulatory reference inside a bioengineering paper, not a primary regulatory citation; downstream pages that need WHO drinking-water values should cite the WHO Guidelines for Drinking-water Quality directly, not this paper.
• Funding declared: National Key R&D Program of China Project 2018YFA0902100; National Natural Science Foundation of China 22178262.
• Audit subagent (2026-06-08) flagged two ⚠️ Check 1 items with REVISE verdict:
  • Mixed-bacterial 449.04/50.71 µmol/g labeled “at 500 µM total mixed ions”: verified against PDF p. 10 (“The adsorption capacity of the mixed bacterial system for a single ion was firstly examined. As shown in Fig. 3c and d… when the ion concentration increased to 500 µM, the adsorption capacities of the mixed bacteria for Hg²⁺ and Cr⁶⁺ reached a maximum of 449.04 and 50.71 µmol/g CDW”) — the 500 µM is single-ion (Hg²⁺ or Cr²⁺ alone) with the mixed-bacterial co-culture, not 500 µM total across the four-metal cocktail (the four-metal cocktail experiments are at 160/400/800 µM total per Fig. 4). Subagent finding verified correct; corrected the two Key numbers rows and added a dedicated four-metal cocktail concentration row.
  • BL21 Hg²⁺ baseline 273.2 µmol/g attributed to Fig. 3a: re-verification against Table 1 shows 273.2 µmol/g is the value for Penicillium canescens (ref [20], Say/Yilmaz/Denizli 2003), not for E. coli BL21. The Table 1 BL21 row from ref [22] reports only a 43.7% removal rate without a µmol/g capacity. The paper’s own BL21 measurements appear only as the wild-type control curve in Fig. 3a, where BL21 visually plateaus around 270–280 µmol/g without a precise text value stated. The subagent’s confirmation of “273.2 from Table 1 ref [22]” was itself based on a misread of Table 1 (the audit attributed the Penicillium canescens row’s value to the BL21 ref [22] row), but the underlying issue — the 273.2 µmol/g number cannot be attributed to BL21 — is real. Corrected the Key numbers row to “~270 to 280 µmol/g CDW at 500 µM Hg²⁺ (Fig. 3a curve only; paper text does not state a precise numerical baseline for BL21).”
• Subagent verdict: REVISE; both findings applied with verification against the source. No QUARANTINE triggers (no ❌ Check 1 cluster, no Check 4 or Check 5 ❌).

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.