Overproduction of Native and Click-able Colanic Acid Slime from Engineered Escherichia coli

The fundamental biology and application of bacterial exopolysaccharides is gaining increasing attention. However, current synthetic biology efforts to produce the major component of Escherichia sp. slime, colanic acid, and functional derivatives thereof have been limited. Herein, we report the overproduction of colanic acid (up to 1.32 g/L) from d-glucose in an engineered strain of Escherichia coli JM109. Furthermore, we report that chemically synthesized l-fucose analogues containing an azide motif can be metabolically incorporated into the slime layer via a heterologous fucose salvage pathway from Bacteroides sp. and used in a click reaction to attach an organic cargo to the cell surface. This molecular-engineered biopolymer has potential as a new tool for use in chemical, biological, and materials research.


General materials and methods
Unless otherwise stated, starting materials and reagents were obtained from commercial suppliers and were used without further purification. All water used experimentally was purified with a Suez Select purification system (18 MΩ.cm, 0.2 µM filter).

NMR spectroscopy
Proton nuclear magnetic resonance spectra ( 1 H NMR) were recorded using a Bruker AVA400, AVA500, Pro500 or AVA600 NMR spectrometer at the specified frequency at 298 K. Proton chemical shifts are expressed in parts per million (ppm, δ scale) and are referenced to residual protium in the NMR solvent (DMSO-d6 = 2.50 ppm). Carbon nuclear magnetic resonance spectra ( 13 C NMR) were recorded using a Bruker AVA400, AVA500, Pro500 or AVA600 NMR spectrometer at the specified frequency at 298 K.

Media recipes Lysogeny Broth (LB) Medium
Bacto-tryptone (10 g/L), yeast extract (5 g/L) and NaCl (10 g/L) were dissolved in Milli-Q water. LB was autoclaved at 121 ˚C for 20 minutes, cooled and stored at room temperature. LB agar was prepared using the same recipe but with the addition of agar (15 g/L).

M9 CA minimal medium
M9 minimal medium was prepared as described above, with the addition of 25 g/L casamino acids being added to the 5X M9 stock solution, for a final concentration of 5 g/L casamino acids.

YESCA medium
0.5 g/L yeast extract and 5 g/L casamino acids were dissolved in Milli-Q water and autoclaved at 121 °C for 20 minutes then cooled to room temperature.

Molecular biology
The fkp gene (Table 1) was codon-optimized for E. coli BL21(DE3) and synthesized using GeneArt TM (Thermo Scientific). Oligonucleotide primers were synthesized by Integrated DNA Technologies.
Recombinant plasmid DNA was purified with a Miniprep Kit (Qiagen). E. coli strain JM109_pRcsA was kindly provided by the laboratory of Prof. French (University of Edinburgh). pRSFDuet-1 was purchased from Novagen. All restriction enzymes were purchased from Thermo Fisher as FastDigest TM enzymes.
All restriction enzyme digests were carried out at 37 °C using FastDigest TM Green buffer. All plasmids were sequenced by Sanger sequencing at Edinburgh Genomics (Edinburgh, UK).
OneTaq 2X premix (New England Biolabs) was used for all colony PCR reactions, which contained 0.5 µM forward and reverse primers and water to a final volume of 12. 5 Table S1 using an annealing temperature of 60 ˚C under standard PCR conditions. Knockout colonies were confirmed via colony PCR using primers gmd-fcl_A-D or waaF_A-D (Table S1) as appropriate using an annealing temperature of 60 ˚C under standard colony PCR conditions. Table S1. Primers used for the preparation of JM109(DE3)∆gmd-fcl and JM109∆waaF knockout strains.

Colanic acid quantification using plate reader
The colanic acid quantification assay was successfully miniaturized for high throughput analysis using to account for differences in incubation times. A representative standard curve is shown in Figure S3.

Quantitative analysis of total carbohydrate content
The total carbohydrate content of the samples was quantified by the anthrone-sulfuric acid assay based on a protocol from Rondel and coworkers 5 . The purified EPS samples were diluted in Milli-Q water depending on their expected concentration to fit within the standard curve, and 400 μL aliquots were added to a glass vial for each sample. To this was added 800 μL of a freshly prepared anthrone solution (2% w/v in 96% aq. H2SO4). The mixture was heated at 60 °C for 30 minutes then cooled to room temperature. The absorbance of the resulting solution at 620 nm was measured and correlated to glucose concentration using a standard curve ranging from 5 μg/mL to 100 μg/mL. A representative standard curve is shown in Figure S4.

Procedure for preparation of fluorescence microscopy samples
Cultures of JM109(DE3)Δgmd-fcl_pRcsA_pFkp were grown as described with Fuc-N3 to produce azidelabelled colanic acid, or with fucose to produce non-labelled colanic acid as the control culture. Both cultures were treated identically throughout. Fluorescent labelling was performed as described, and the finished reactions dialyzed for 50 hours in water in 3.5 kDa molecular weight cut-off dialysis cassettes, changing the dialysis water every 8-16 hours.
For fluorescence microscopy, the dialyzed cultures were diluted with water to OD600 = 0.2 and 100 µL was transferred onto an air-dried glass slide coated with 0.05 % (w/v) poly-D-lysine and allowed to bind for 30 minutes before washing with water. The slide was air-dried and 50 µL ProLong TM Diamond Antifade (Thermo Fisher Scientific) was dropped onto the sample before covering with a coverslip.
Images were taken using a Nikon Eclipse Ti2 microscope (Nikon Europe, Amstelveen, Netherlands) equipped with a x100 objective and Prime 95B sCMOS camera (Teledyne Photometrics, Birmingham UK). Fluorescence was detected using a 488 nm excitation filter, Sedat Quad dichroic and 520 nm emission filter (Chroma Technology Corp, Bellows Falls). All images were analyzed post acquisition using Fiji 6 .