Living Bacteria-Mediated Aerobic Photoinduced Radical Polymerization for in Situ Bacterial Encapsulation and Differentiation

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Jul 2021Living Bacteria-Mediated Aerobic Photoinduced Radical Polymerization for in Situ Bacterial Encapsulation and Differentiation Huan Lu†, Yiming Huang†, Fengting Lv, Libing Liu, Yuguo Ma and Shu Wang Huan Lu† Beijing National Laboratory for Molecular Sciences, Beijing 100190 Centre for the Soft Matter Science and Engineering, The Key Lab of Polymer Chemistry & Physics of the Ministry of Education, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871 , Yiming Huang† *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Beijing National Laboratory for Molecular Sciences, Beijing 100190 Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 , Fengting Lv Beijing National Laboratory for Molecular Sciences, Beijing 100190 Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 , Libing Liu Beijing National Laboratory for Molecular Sciences, Beijing 100190 Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 , Yuguo Ma *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Beijing National Laboratory for Molecular Sciences, Beijing 100190 Centre for the Soft Matter Science and Engineering, The Key Lab of Polymer Chemistry & Physics of the Ministry of Education, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871 and Shu Wang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Beijing National Laboratory for Molecular Sciences, Beijing 100190 Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 https://doi.org/10.31635/ccschem.021.202100957 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Conventional polymerizations mediated by living cells typically require synthetic transition-metal complexes or photoredox catalysts. Herein, we report an alternative photoinduced polymerization strategy for preparing functional polymer hydrogels through bacteria-initiated radical polymerization of acrylamides in ordinary culture media. Upon light irradiation under ambient conditions, polyacrylamides were obtained with molecular weights of over 150 kDa using various bacteria. Electron spin resonance spectra data suggested that photosensitization in culture media converted the dissolved oxygen to hydroxyl radicals and initiated polymerization. Although oxygen inhibited the polymerization, living facultative anaerobic bacteria could deplete oxygen and allow the radical propagation to proceed. With a bisacrylamide cross-linker, hydrogels could be achieved, with the bacteria trapping themselves inside simultaneously, and maintaining high viability (>94%). Various bacteria species were differentiated based on their gelation behavior properties, relative to their oxygen tolerance under appropriate conditions. This process resembles silkworms' behavior for spinning a cocoon around themselves. The mild conditions for radical polymerizations provided a convenient approach for in situ bacterial encapsulation and differentiation. Download figure Download PowerPoint Introduction Living cells always gain nutrients from the environment to produce energy and necessary chemical building blocks for their activities. Metabolic and biosynthetic processes of the living cells are commonly governed by a vast library of enzymes and build the foundation for cells to perform recognition, secretion, signal transduction, among other functions.1 Elucidation of catalytic mechanisms enabled researchers to capitalize on the potential of biological systems to work as microscopic synthetic factories.2,3 Despite the complexities of living organisms, exploiting their capabilities for biocatalysis has been an intriguing research area. Unnatural reactions that can occur inside or on the surface of cells would prompt discoveries and applications to synthetic chemistry, chemical biology, and medicine.4 Early examples of cell-mediated abiotic transformations typically used a masked fluorophore as the substrate to demonstrate protecting group removal induced by an external catalyst,5,6 which could also catalyze bond formation reactions leading to intensely emissive products.6,7 Through delicate design, a major leap forward would be to synthesize bioactive molecules to impact their function or viability. For example, intracellular Cu(I) catalyst generated in situ can click together two fragment molecules to form a compound with excellent bactericidal efficiency.8 Nevertheless, rather than contributing to the reactions, these cells primarily function as bystander carriers. For living cell-mediated chemical reactions, the processes need to be inherently efficient under mild conditions such as room temperature, air exposure, aqueous solution, and preferably yielding functional products. Radical polymerization has been a popular way of chemical synthesis due to its facile procedure, fast reaction rate, functional group tolerance, as well as adjustable structural properties of polymer products.9,10 However, for biocatalytic purposes, the conditions for radical polymerization such as toxic catalysts and harsh degassing need to be circumvented. Radical polymerization was viable using various isolated natural catalysts in vitro, including DNA photolyase,11 bacteriochlorophylls,12 hemoglobin, and glucose oxidase.13 So far, only a few examples reported that living cells could handle the synthesis of acrylic polymers. By reducing Cu(II) to Cu(I), both Escherichia coli- and Pseudomonas aeruginosa-catalyzed atom transfer radical polymerization (ATRP) and their signature microbial surfaces also served as the template for polymer structures.2,14 The role of electron transport proteins on the bacterial membranes was further emphasized in ATRP utilizing the reductive capability of Shewanella oneidensis.15,16 Photoinduced radical polymerizations can occur on the membranes of and inside living cells,17,18 albeit forming polymers at a relatively miniature scale. In precedent works, polymerization reactions mediated by living cells under ambient conditions are limited by adding foreign transition-metal complex or photoredox catalysts. Thus, an alternative radical polymerization approach to significantly improve the harsh conditions is needed. Furthermore, the capability of living cells to produce polymers as well as bulk materials that establish biologically relevant functions also needs further exploration. Herein, we report an in situ photoinduced radical polymerizations of acrylamides using living bacteria in ordinary culture media [Luria-Bertani (LB) medium] under visible light irradiation (Scheme 1). We found that this photoinduced radical polymerization could be initiated without additives of foreign photosensitizer (PS). Electron spin resonance (ESR) spectra suggested that photosensitization in LB media converted the dissolved oxygen to hydroxyl radical (•OH) and initiated polymerization. As is well known, it is a significant challenge for radical polymerization to occur under ambient conditions because dissolved oxygen is an effective radical quencher that can inhibit the polymerization reaction. Fortunately, living facultative anaerobic bacteria can deplete the unwanted oxygen and allow the radical propagation to proceed. Thus, the combination of LB medium and bacteria regulated the level of dissolved oxygen that allowed radical initiation and propagation. Remarkably, we found that the livingness of bacteria, rather than their secretions, was a key factor that enabled the radical chain propagation. Further, in the presence of a cross-linker monomer, bacteria were able to produce bulk hydrogel materials and encapsulated themselves inside the hydrogels simultaneously, maintaining high cell viability.19–21 This whole process resembles silkworms' behavior of spinning cocoons around themselves. More importantly, bacteria species could be differentiated based on their gelation efficiency and behaviors under various monomer concentrations. Scheme 1 | Bacteria-mediated photopolymerization of acrylamides. Acrylamide monomers undergo photoinduced radical polymerization mediated by living bacteria in LB medium under visible light irradiation. In the presence of a bisacrylamide cross-linker, polymer hydrogel can be formed, simultaneously encapsulating the bacteria in situ with potential capabilities of bacterial differentiation. Download figure Download PowerPoint Experimental Methods Reagents and instruments N,N-dimethylacrylamide (DMA; Alfa Aesar, Shanghai, China) was filtered through silica to remove the mequinol inhibitor. N,N'-Methylenebisacrylamide (BIS; J&K Chemicals, Beijing, China), deuterium oxide (D2O; Innochem, Beijing, China), 5,5-dimethyl-1-pyrroline N-oxide (DMPO; Dojindo, Kyushu, Japan), resazurin (Innochem), phosphate-buffered saline (PBS; Multicell, without Ca2+ or Mg2+), LIVE/DEAD™ Bac Light™ Bacterial Viability Kit (Thermo Fisher, Shanghai, China) were purchased and used as received. LB medium for bacteria culture was prepared by dissolving 10 g tryptone, 5 g yeast extract, and 10 g NaCl in 950 mL sterile water, followed by further sterilization in an autoclave set at 121 °C for 15 min. Proton nuclear magnetic resonance (1H NMR) spectra were obtained on a Bruker Avance III 400 HD (Rheinstetten, Germany) operated at 400 MHz at room temperature. Gel permeation chromatography (GPC) results were obtained on an Agilent 1260 system (Palo Alto, CA) and calibrated with poly(ethylene glycol) standards in water. X-band continuous-wave ESR spectra were collected on a Bruker ELEXSYS-II E500 (Rheinstetten, Germany) operated at 9.8 GHz at room temperature at the Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. Scanning electron microscopy (SEM) images were obtained on a Phenom ProX microscope (Hillsboro, OR) (10 kV). Atomic force microscopy (AFM) images were obtained on a Bruker MultiMode 8 microscope (Santa Barbara, CA) under the ScanAsyst mode. The p-values were obtained by the two-sample Student's t-test using the hypothesis testing functions in Origin 2016. Confocal laser scanning microscopy (CLSM) images were captured on an Olympus FV1200 (Hataya, Japan) microscope. Rheological properties were tested on a Discovery HR-1 hybrid rheometer manufactured by TA Instruments (New Castle, DE). General procedure for bacteria-mediated polymerization A single colony of ampicillin-resistant (Ampr) E. coli on a solid LB agar plate was transferred to a 50 mL centrifuge tube containing 10 mL of fresh liquid LB culture medium supplemented with 50 μg/mL ampicillin. The mixture was cultured overnight in a constant temperature incubator set at 37 °C with continuous shaking at 180 rpm. The resulting E. coli suspension was diluted by fresh liquid LB culture medium to an optical density (OD) of 1.0 at 600 nm using fresh liquid LB culture medium as the reference. (An OD600 = 1.0 corresponded to a bacterial concentration of 1 × 108E. coli cells per mL.) The supernatant of the E. coli suspension was removed by centrifugation (7100 rpm, 3 min), and the cell precipitate was resuspended in a fresh LB medium before the addition of DMA to bring the final E. coli concentration to the predetermined value of 2, 4, 8 × 108 cells per mL. Exactly 300 μL of this reaction mixture was placed in a 2 mL clear glass vial (with screw cap and septum) and irradiated from the bottom of the vial. A UV filter that allowed the transmittance of λ > 420 nm was applied to a xenon light source, and the lamp power was adjusted to maintain a light intensity of 50 mW/cm2 at the bottom of the vial. To sample the reaction mixture for NMR, 30 μL of the reaction liquor was withdrawn using a micropipette and dissolved with 500 μL D2O. This NMR sample was recovered and freeze-dried to remove water and unconverted monomer. The white solid obtained was dissolved in water for GPC characterizations. General procedure for bacteria-mediated hydrogel formation For rheological property measurements, a mixture of bacteria (final OD600 = 4), DMA (1 mol/L), BIS (0.05 mol/L) in LB culture medium was pipetted into a 12-well plate (1.5 mL in each well) and irradiated under white light for 1 h. The hydrogel was soaked in PBS before a gentle removal from the well. For bacterial differentiation, 300 μL of the mixture (with variable DMA concentration was placed in a 2 mL clear glass vial and irradiated under white light for 1 h. The vials were placed upside down for taking digital photos with the bottom facing the camera. Results and Discussion Bacteria-mediated polymerization of DMA We initially discovered that the polymerization of DMA occurred in E. coli-inoculated LB medium under white light irradiation for 2 h (Figure 1a), evidenced by emerging proton peaks of the poly(N,N-dimethylacrylamide) (PDMA) at 1.1–1.7 ppm (a'/b'), 2.3–2.6 ppm (c'), and 2.7–3.0 ppm (d'/e') in 1H NMR spectra (Figure 1b). To identify the key components for this reaction, we designed a series of control experiments to eliminate one factor at a time (Figure 1c and Supporting Information Table S1), as follows: setting up the reaction with (1) LB medium, but without bacteria, (2) the supernatant part of E. coli in LB suspension, (3) dead E. coli (heat-deactivated) in LB medium, and (4) living E. coli suspended in PBS. We observed that DMA polymerization did not occur in sterile LB under light irradiation (Figure 1d). Failure to produce PDMA by the supernatant suggested that the bacteria, rather than their secretions, played an essential role in DMA polymerization. Living activities of bacteria proved vital to the reaction, as the monomer conversion reached 15.8% with living E. coli in LB but not detected with dead bacteria in the same medium. Moreover, no DMA conversion was observed when the experiment was performed in LB medium in the dark, suggesting light irradiation as a crucial factor for polymerization. Figure 1 | E. coli-mediated polymerization of DMA. (a and b) Protons of interest are labeled in the reaction scheme and 1H NMR spectra of the reaction mixture. (c) The schematic diagram for the preparation of the [E. coli + LB] reaction medium and relevant control groups. (d) Control experiments revealed that living bacteria under proper nutrients and light were essential for the polymerization to proceed. DMA conversion was studied under variable concentrations of E. coli (e) and monomer (f). Conversion rose rapidly in the first 2 h of the reaction (g). Results are presented as mean ± SD from at least three replicated experiments. (h) Results of bacteria-mediated polymerizations under variable reaction time and bacterial species. Reaction vessels were irradiated under a xenon lamp with a λ > 420 nm filter at an intensity of 50 mW/cm2. Monomer conversions were calculated from 1H NMR spectra. Number- and weight-average molecular weights (Mn and Mw) were obtained from GPC calibrated with poly(ethylene glycol) standards. PDI = Mw/Mn. Download figure Download PowerPoint Concentrations of E. coli and DMA dictated the monomer conversion in different manners. In LB medium, E. coli diluted to 1 × 108 per mL22 appeared insufficient for the reaction (Figure 1e). Concentrating E. coli dramatically improved the monomer conversion, but the effect diminished beyond 4 × 108 per mL. In contrast, an optimal DMA concentration existed near 1 mol/L (Figure 1f). While the chain propagation rate favored higher monomer concentrations, excessive DMA posed severe toxicity to E. coli, thereby killing the bacteria with the subsequent termination of the reaction. Under our optimized conditions (4 × 108E. coli per mL, 1 mol/L DMA), polymerization of DMA proceeded most rapidly in the first 2 h with monomer conversion reaching 15.8% and further increased to 33.6% after 8 h (Figure 1g). According to the GPC results, the number-average molecular weights (Mn) of the isolated PDMA samples were measured in the range of 160–180 kDa (Figure 1h). When a supplementary DMA monomer was added to the bacteria (S. aureus), it led to negligible changes to the monomer conversion. Besides, we isolated the "spent" bacteria by repeated rinsing and centrifugation and reused them in the polymerization reaction under the same experimental conditions. We found that the monomer conversion reached 35.1% with the "recycled" bacteria, compared with 37.8% attained with "fresh" bacteria. Directing the DMA polymerization proved not to be an exclusive capability of E. coli cells alone but a broad range of other living bacteria. Interestingly, the polymerization progressed more efficiently with facultative anaerobic bacteria than aerobic bacteria (Figure 1h). As the NMR results suggested ( Supporting Information Figure S1), polymerizations using facultative anaerobes Staphylococcus aureus, Enterococcus faecalis, and Bacillus subtilis yielded DMA conversions of 38.3%, 32.9%, and 34.8%, respectively, in 8 h, while those of the aerobic P. aeruginosa and Saccharomyces cerevisiae produced low conversions of only 15.6% and 21.8%, respectively, under the same reaction conditions. Based on our results, we proposed that microbial aerotolerance and reductive capabilities are the two most important reaction efficiency determining factors between the bacterial strains. Anaerobes can consume oxygen for the polymerization to proceed and still thrive to provide sufficient reducing power. However, aerobes need constant supply oxygen for living activities, which, in turn, inhibited the polymerization process, thereby producing lower monomer conversion values. For the isolated PDMA obtained with various bacteria, Mn scattered in the range of 163–335 kDa without an apparent trend of bacteria-dependent polydispersity. Polymerizations were carried out using different culture media, including LB, Mueller–Hinton (MH), and yeast extract peptone dextrose (YPD). Although most of the bacterium-medium combinations produced the polymer, medium-dependent monomer conversion was observed as YPD > LB > MH during the conversion ( Supporting Information Table S3), most apparently in S. cerevisiae. Since YPD is the recommended culture medium for S. cerevisiae growth, it presumably, facilitated the reducing power of the microbes. The polymerization mechanism To understand the mechanism of this bacteria-mediated photopolymerization, we investigated the availability of radical species in this system using ESR spectroscopy. Transient radicals originating from biological samples were difficult to detect directly by ESR due to their short lifetimes. Therefore, a spin trap such as DMPO was used to capture the radicals of interest (X•) and form adducts (DMPO-X•) which were relatively long-lived nitroxide radicals. Both high radical concentrations and fast-trapping rates were critical in producing ESR-active adducts, as the first-order kinetics suggested in eq 1: d [ DMPO − X • ] d t = k T [ DMPO ] [ X • ] (1)where kT is the spin-trapping rate constant of the X• (in M−1s−1).23 In these adducts, the substituent at β-position affects the local magnetic environment, and thus, the hyperfine splitting constant (a) by the adjacent N (aN) and by the Hβ (aHβ) atoms revealed the identity of radicals captured by DMPO. We probed the presence and identity of radicals in multiple experiments utilizing the DMPO spin trap with LB alone, LB with DMA monomer, LB with bacteria S. aureus, and the LB with both bacteria and DMA (Figures 2a–2d). The ESR signals were comparable with background noise in all groups without irradiation but differed markedly after visible light irradiation (detailed parameters in Supporting Information Table S2). LB alone generated radicals under light irradiation (Figure 2a), and a similar splitting pattern was displayed with additional DMA monomer (Figure 2b). hyperfine splitting = = the presence of (Figure Interestingly, LB with increased bacteria to produce ESR signals that could be as a (Figure When both bacteria and DMA were added to LB to the polymerization conditions, no a ESR signal appeared with hyperfine splitting of = and = (Figure a ESR pattern of the radical Figure Figure also the only under which DMA would that the radical formation was crucial to bacteria-mediated radical polymerization. Figure 2 | the mechanism with ESR and the resazurin ESR spectra collected in the and after irradiation of (a) LB alone, LB + (c) LB + bacteria (S. aureus), (d) LB + bacteria + all captured by the DMPO spin labeled were irradiated for 3 under light conditions to the polymerization. (e) The ESR signal pattern in (a) and from the with hyperfine splitting of = = The pattern in (d) suggested the = and = The kT was the rate constant for spin trapping by DMPO (in to Reaction scheme of of the resazurin to and to changes that the LB medium alone could only under light without However, without bacteria S. in LB efficiently the and the dissolved air into the aqueous suspension the and rather than Bacterial occurred at a lower efficiency in as suggested by the compared with DMA was resazurin and Download figure Download PowerPoint Based on the ESR results and from the resazurin (Figure we proposed a mechanism of the photopolymerization to identify the role of LB and bacteria. Since LB was able to produce and resazurin only under irradiation (Figure it a that could an electron would the dissolved oxygen to radical of ESR signals of the was due to its trapping rate constant and short in aqueous of and by DMPO to ESR signals M−1s−1).23 In of DMA polymerization, these led to the of the initiation as reported In an additional the mixture containing all necessary components bacteria, and the of an vial and was first placed in the for 1 h to deplete the dissolved oxygen by resazurin in a light irradiation for 2 h, no polymer of DMA in this further the from oxygen to in the initiation Figure 3 | mechanism of bacteria-mediated photopolymerization of DMA. Under the in LB could oxygen to hydroxyl radical (•OH) that DMA polymerization. In the of bacteria dissolved oxygen can inhibit the polymerization by forming the radicals for chain propagation. With bacteria oxygen for chain propagation to proceed. The and were rate for chain propagation and oxygen respectively, to Download figure Download PowerPoint The addition of to the first DMA monomer transferred radical from oxygen to the atom in the ESR pattern of However, the of in Figure radical under aerobic conditions. bacteria, the radical would with oxygen to form radical species which was acrylic thereby the radical chain propagation (Figure radicals would be in a between the oxygen and the chain propagation as follows: d [ • ] d t = k [ 2 ] [ • ] (2) d [ − DMA • ] d t = k [ DMA ] [ • ] is the rate constant than the propagation rate constant The could not be by the high monomer concentration = 1 thus, oxygen under aerobic conditions. The of the ESR signal from the was also due to its rate constant M−1s−1).23 In contrast, reductive bacteria removed oxygen through and the formation of thus, the radical chain to (Figure S. and E. coli are both facultative anaerobic bacteria that could aerobic and anaerobic conditions. bacteria are for their leading to a of and to an anaerobic environment bacteria could an additional of electron to the radical of the and the LB a of oxygen necessary at the initiation to produce but oxygen the chain propagation and to be removed by the bacteria. between bacteria and in PDMA Although most PDMA in aqueous solution, we probed polymer also on the bacterial Both and images revealed in E. coli before and after polymerization. To remove components from the culture medium, the E. coli was with the mixture before images of E. coli before polymerization clear of without on the ( Supporting Information Figure S2). In contrast, E. coli after polymerization and appeared with a of foreign the cells and the in situ by E. In these of E. coli cells could to an of for Therefore, we used E. coli samples for to the and of images on also captured the of E. coli before polymerization and their after polymerization (Figures and and Supporting Information and of nm in and 1 in were visible in the samples after polymerization. To the data of the were collected the of E. coli cells (Figures and and as a (Figures and and Supporting Information Figure Although occurred in the of the E. coli cell (Figure the from ± to ± when with the (Figure Since