Bacteria-initiated reversible addition-fragmentation chain-transfer (RAFT) polymerization, first demonstrated with the iron-reducing bacterium Cupriavidus metallidurans CH34 in a modified Fenton system (Bennett et al., 2022), converts cellular metabolism into controlled radical polymerization under ambient, oxygen-tolerant conditions, but the original system is bounded by a ~40-50% conversion plateau, monomer toxicity toward the cells, poor blocking efficiency in chain extension from macro-chain-transfer agents, and unexplained cycle-to-cycle variability when the cells are recycled. A 2025 redesign (Li et al., Nat. Commun. 16, 10257) using genetically engineered Shewanella oneidensis, whose secreted flavins directly reduce chain-transfer agents under light, resolved several platform-level strands: conversion above 90% with dispersity below 1.20, controlled polymerization up to degree of polymerization 2000, and clean block-copolymer synthesis. The surviving open core concerns the mechanistic understanding of cell-mediated polymerization that the 2022 work asked for and the redesign bypassed rather than answered: the metabolic or growth-phase basis of recycling variability, extension to monomers toxic to the producer cells, adjudication among the competing hypotheses for the poor blocking efficiency of bacterial Fenton-RAFT, and cell-autonomous spatial programming of polymer synthesis in engineered living materials.

Background

Reversible-deactivation radical polymerization (RDRP) is a family of chain-growth methods in which a dormant equilibrium keeps only a small fraction of chains propagating, so that average molar mass tracks conversion and molar-mass distributions stay narrow. RAFT (reversible addition-fragmentation chain-transfer) polymerization is the most implemented variant: a thiocarbonylthio chain-transfer agent (CTA), typically a trithiocarbonate, is shuttled between dormant and active chains, giving predictable molar masses, low dispersity (Đ, the ratio of weight-average to number-average molar mass), and retention of a functional chain end. That chain-end fidelity is what allows sequential addition of a second monomer to a purified macro-CTA (a polymer chain that still carries the thiocarbonylthio chain-transfer end group) to form block copolymers, and under suitable conditions polymerization-induced self-assembly (PISA), in which the growing block self-assembles in situ into polymer nanoparticles.

A separate line of work has sought to let living cells, rather than purified reagents, generate the radicals that start and sustain these polymerizations — a route toward engineered living materials (ELMs), hybrid materials in which living cells synthesize or remodel abiotic components. Prior demonstrations include Shewanella oneidensis biofilms acting as living electrodes for electro-mediated polymerization (Fan et al., 2018), aerobic radical polymerization driven by microbial metabolism (Fan et al., 2020), bacteria-guided glycopolymer synthesis (Luo et al., 2019), uncontrolled iron-catalysed free-radical polymerization by living C. metallidurans (Bennett et al., 2020), and E. coli terminal electron flux reducing aryl diazonium salts to initiating aryl radicals for RAFT of a methacrylate (Nothling et al., 2021). Bennett et al. (ACS Macro Lett. 2022) combined the iron-reducing bacterium C. metallidurans CH34 with ferric iron and glucose oxidase (GOx, which consumes glucose and dissolved oxygen to produce hydrogen peroxide): the cells continuously reduce Fe3+\mathrm{Fe}^{3+} to Fe2+\mathrm{Fe}^{2+}, and the resulting Fenton reaction (Fe2++H2O2→Fe3++OH−+OH⋅\mathrm{Fe}^{2+}+\mathrm{H_2O_2}\rightarrow\mathrm{Fe}^{3+}+\mathrm{OH^-}+\mathrm{OH}\cdot) generates hydroxyl radicals that initiate RAFT in air at 30 °C. This gave water-soluble polymers with dispersity as low as 1.12, trithiocarbonate retention, PISA nanoparticles, and reusable initiating cells.

The same paper documented the system's limits, which constitute the open question at issue. Monomer scope is capped by toxicity toward the cells (for example, NN-acryloylmorpholine had a half-maximal inhibitory concentration (MIC50, the monomer concentration at which bacterial growth is halved) of 42.5 mM, versus >100 mM for N,NN,N-dimethylacrylamide). Conversion plateaus near 40-50%, attributed to depletion of the glucose feedstock consumed by GOx and poor cell tolerance of high monomer concentrations, which in turn caps attainable chain length (control was already degraded at target degree of polymerization (DP) 800, Đ = 1.7). Chain extension from a pDMA (poly(N,NN,N-dimethylacrylamide)) macro-CTA gave a bimodal molar-mass distribution in its size-exclusion-chromatography (SEC, also called gel-permeation chromatography, GPC) trace — poor blocking efficiency — whose origin the authors could not assign among radical oxidation of the trithiocarbonate chain end, chain-length-dependent retardation of chain transfer, and adhesion of bacterially synthesized polymer to the cell surface, with termination-driven loss of the RAFT agent raised as a further possibility. Recycling the cells across three polymerization cycles gave conversions of 40, 80, and 50% with varying molar masses, attributed only speculatively to proliferation or growth-phase changes. In 2025, a redesigned system (Li et al.) showed that several of these platform-level limits are not fundamental: engineered S. oneidensis secreting flavin electron shuttles, photoexcited to directly reduce CTAs, achieved >90% conversion with Đ < 1.20, controlled polymerization from DP 100 to DP 2000, eight-plus monomers including methyl methacrylate by emulsion, in-situ diblock and triblock synthesis with clean chromatographic shifts, light-gated temporal control, and mapped the electron-transfer route (D-lactate metabolism via the CymA and Mtr membrane-protein conduits of the S. oneidensis electron-transfer chain to secreted flavins). What that redesign did not address — the metabolic basis of recycling variability, monomers toxic to the cells, and the mechanistic adjudication of the Fenton system's blocking failure — together with cell-directed spatial programming of polymer synthesis, defines the surviving open core.

Problem Statement

Consider RAFT polymerization in which living bacteria generate the initiating radicals — as in the oxygen-tolerant C. metallidurans/glucose-oxidase Fenton system of Bennett et al. (ACS Macro Lett. 2022) — and establish the mechanistic understanding needed to overcome the system's biological and chemical limits. Following the partial resolutions delivered by the engineered S. oneidensis flavin/electron-transfer system of Li et al. (2025), which raised conversion above 90%, restored chain-end fidelity and block-copolymer synthesis, and demonstrated genetic engineering of the electron-mediator pathway, the surviving open core is: (1) identify the metabolic and growth-phase parameters that cause the large cycle-to-cycle variation in monomer conversion and molar mass when living cells are recycled across successive polymerizations (40, 80, and 50% conversion over three cycles in the original system), and determine how polymerization performance depends on cell state; (2) extend metabolically initiated controlled radical polymerization to monomers whose toxicity toward the producer cells currently excludes them while retaining control over molar mass and dispersity; (3) determine which of the candidate mechanisms — radical oxidation of the trithiocarbonate chain end, chain-length-dependent retardation of chain transfer, adhesion of bacterially synthesized polymer to the cell surface, or another cause such as termination-driven loss of the RAFT agent — actually causes the bimodal, poorly blocked chain extension in bacterial Fenton-RAFT, so that initiator-based cell systems can also produce well-defined block copolymers; and (4) achieve cell-autonomous spatial (not merely light-gated temporal) control over where and when synthetic polymer synthesis occurs, the step required for truly programmable engineered living materials in which living cells direct polymerization. These four strands are independent open sub-questions: resolving at least one of them leaves the others open, and any resolution must keep polymerization dependent on living, metabolizing cells.

References

  1. Mechelle R. Bennett, Cara Moloney, Francesco Catrambone, Federico Turco, Benjamin Myers, Katalin Kovacs, Philip J. Hill, Cameron Alexander, Frankie J. Rawson, Pratik Gurnani, 'Oxygen-Tolerant RAFT Polymerization Initiated by Living Bacteria', ACS Macro Letters 11 (2022) 954–960, DOI 10.1021/acsmacrolett.2c00372, https://doi.org/10.1021/acsmacrolett.2c00372
  2. Mechelle R. Bennett, Pratik Gurnani, Phil J. Hill, Cameron Alexander, Frankie J. Rawson, 'Iron-Catalysed Radical Polymerisation by Living Bacteria', Angewandte Chemie International Edition 59 (2020) 4750–4755, DOI 10.1002/anie.201915084, https://doi.org/10.1002/anie.201915084
  3. Gang Fan, Christopher M. Dundas, Austin J. Graham, Nathaniel A. Lynd, Benjamin K. Keitz, 'Shewanella oneidensis as a living electrode for controlled radical polymerization', Proceedings of the National Academy of Sciences 115 (2018) 4559–4564, DOI 10.1073/pnas.1800869115, https://doi.org/10.1073/pnas.1800869115
  4. Gang Fan, Austin J. Graham, Jayaker Kolli, Nathaniel A. Lynd, Benjamin K. Keitz, 'Aerobic radical polymerization mediated by microbial metabolism', Nature Chemistry 12 (2020) 638–646, DOI 10.1038/s41557-020-0460-1, https://doi.org/10.1038/s41557-020-0460-1
  5. Yan Luo, Yan Gu, Ruyan Feng, John Brash, Ahmed M. Eissa, David M. Haddleton, Gaojian Chen, Hong Chen, 'Synthesis of glycopolymers with specificity for bacterial strains via bacteria-guided polymerization', Chemical Science 10 (2019) 5251–5257, DOI 10.1039/C8SC05561K, https://doi.org/10.1039/C8SC05561K
  6. Mitchell D. Nothling, Hanwei Cao, Thomas G. McKenzie, Dianna M. Hocking, Richard A. Strugnell, Greg G. Qiao, 'Bacterial Redox Potential Powers Controlled Radical Polymerization', Journal of the American Chemical Society 143 (2021) 286–293, DOI 10.1021/jacs.0c10673, https://doi.org/10.1021/jacs.0c10673
  7. Chao Li, Jing Liu, Wenchang Hu, Lin Xiao, Feng Li, Qijing Liu, Junqi Zhang, Huan Yu, Baocai Zhang, Dake Xu, Shaoan Cheng, Wen-Wei Li, Kenneth H. Nealson, Hao Song, 'Photo-excited extracellular electron transfer of electroactive microorganism triggers RAFT polymerization', Nature Communications 16 (2025) 10257, DOI 10.1038/s41467-025-65119-x, https://doi.org/10.1038/s41467-025-65119-x
  8. Austin J. Graham, Gina Partipilo, Christopher M. Dundas, Ismar E. Miniel Mahfoud, Kathleen N. Halwachs, Alexis J. Holwerda, Trevor R. Simmons, Thomas M. FitzSimons, Sarah M. Coleman, Rebecca Rinehart, Darian Chiu, Avery E. Tyndall, Kenneth C. Sajbel, Adrianne M. Rosales, Benjamin K. Keitz, 'Transcriptional regulation of living materials via extracellular electron transfer', Nature Chemical Biology 20 (2024) 1329–1340, DOI 10.1038/s41589-024-01628-y, https://doi.org/10.1038/s41589-024-01628-y
  9. Xianfeng Chu, Xiaowen Dou, Jiaying Yu, Junpeng Zhou, Di Ma, Miao Miao, Shaojun Hu, Kai Sun, Shihong Zhu, Qi Liu, Xiuming Zhang, Yong Jiang, Zhi-Fei Wang, 'Synthesis of Polymers via Cancer Cell Metabolism-Mediated Controlled Radical Polymerization and Application in Engineering of Cell Surface', Biomacromolecules 26 (2025) 238–247 (published online December 2024), DOI 10.1021/acs.biomac.4c01020, https://doi.org/10.1021/acs.biomac.4c01020
  10. Dan Wu, Jiaqi Lei, Zhankui Zhang, Feihe Huang, Marija Buljan, Guocan Yu, 'Polymerization in living organisms', Chemical Society Reviews 52 (2023) 2911–2945, DOI 10.1039/d2cs00759b, https://doi.org/10.1039/d2cs00759b
  11. Vianna F. Jafari, James L. Grace, Jiajia Li, Joji Tanaka, Glen R. Jones, Athina Anastasaki, Wei You, Jian Zhu, Masami Kamigaito, Cyrille Boyer, Greg G. Qiao, 'Beyond Traditional RAFT Polymerization: Emerging Strategies and Future Perspectives; A Third Update', Advanced Science 13 (2026) e20657, DOI 10.1002/advs.202520657, https://doi.org/10.1002/advs.202520657
  12. Yao Fu, Xinyan Dai, Ka Leung Lam, Soham Das, Tan Zhang, 'Initiating radical polymerization at room temperature: Why and How?', RSC Advances 16 (2026) 26988–27007, DOI 10.1039/d5ra08751a, https://doi.org/10.1039/d5ra08751a