Toward Light‐Regulated Living Biomaterials

Abstract Living materials are an emergent material class, infused with the productive, adaptive, and regenerative properties of living organisms. Property regulation in living materials requires encoding responsive units in the living components to allow external manipulation of their function. Here, an optoregulated Escherichia coli (E. coli)‐based living biomaterial that can be externally addressed using light to interact with mammalian cells is demonstrated. This is achieved by using a photoactivatable inducer of gene expression and bacterial surface display technology to present an integrin‐specific miniprotein on the outer membrane of an endotoxin‐free E. coli strain. Hydrogel surfaces functionalized with the bacteria can expose cell adhesive molecules upon in situ light‐activation, and trigger cell adhesion. Surface immobilized bacteria are able to deliver a fluorescent protein to the mammalian cells with which they are interacting, indicating the potential of such a bacterial material to deliver molecules to cells in a targeted manner.


Construction of plasmids and bacterial strains
pETDuet-1 vector was purchased from Merckmillipore. pRsetB-TagRFP was a kind gift from Dr. Dorothee Wasserberg from the University of Twente. The TagRFP gene was amplified with NdeI and XhoI restriction sites with which it was inserted into the second MCS of pETDuet-1.
The eCPX gene was taken from plasmid pB33eCPX, which was a gift from Patrick Daugherty (Addgene plasmid # 23336). The mRGD gene was purchased as a synthesized oligonucleotide by Eurofins MWG Operon. BsrGI and NheI restriction sites were introduced between the OmpXss and eCPX gene in the pB33eCPX plasmid using the Q5 site directed mutagenesis kit from NEB. mRGD was amplified by PCR with the BsrGI and NheI restriction sites, which were then used to clone it between OmpXss and eCPX in pB33eCPX. OmpXss-mRGD-eCPX was then amplified with SacI and HindII sites with which they were cloned into the first MCS of pETDuet-1 to yield the plasmid named pD-mRGD-eCPX-RFP.
The scrambled mRDG version was constructed by site directed mutagenesis of the abovementioned plasmid using the Q5 site directed mutagenesis kit by NEB to yield pD-mRDG-eCPX-RFP. Similarly, His6 tag was added to the C-terminal of eCPX using site directed mutagenesis to yield pD-mRGD-eCPX-H6-RFP.
All plasmids were transformed by electroporation into ClearColi Bl21(DE3) cells from BioCat exactly as specified by the provider. Bio-Rad Micropulser TM Electroporator was used in combination with Bio-Rad 0.1 cm electroporation cuvettes (1652083).

Outer membrane and cytosolic proteins sub-fractionation
The bacterial cultures were grown in 50 ml LB Miller medium with 50 μg/ml Ampicillin (Roth #HP62) at 37°C, 250 rpm and appropriate cultures were induced at 0.5-0.8 OD with 0.1mM IPTG. The cultures were then harvested as spun-down pellets after 4h of growth (3300 x g, 20 min, 4°C, Hettich Rotanta 460 RS), quick-frozen in liquid Nitrogen and stored at -20°C.
The sub-fractionation was performed using a modified version of a previously reported strategy by Thein and coworkers (Method 1). [3] In short, pelleted cells were resuspended in 1 ml 0.2 M Tris-HCl pH 8, 1 Mbsucrose, 1 mM EDTA. 100 μL of lysozyme (Roth) (5 mg/mL in MQ) were added, vortexed and incubated for 5 min at RT. Four ml of MQ were added and incubated for 20 min at RT until spheroblast formation was observed under the microscope. Then 6 mL 50 mM Tris-HCl pH 8, 2% (w/v) Triton X-100, 10 mM MgCl2 and 12 μL Benzonase Nuclease (Novagen Merck, 300 U) were added and mixed until the suspension was clear. The mixture was centrifuged at 75 000× g, 60 min, 4 °C (Beckman Avanti J26XP JA 25.50 Rotor 10 ml PC vials).

SDS-PAGE and Western Blot
The E. coli strain carrying the plasmid pD-mRGD-eCPX-H6-RFP that results in the expression of a His6-tagged variant of mRGD-eCPX, named mRGD-eCPX-His6, was used in the following experiment. 20 μL sample (pellets and conc. supernatants) and 7 μL Laemmli 4x buffer were mixed and heated at 95°C for 5 min. 10 μl of each was loaded on 12% SDS gels and run at 120 V for 60 min. One gel was Biosafe Coomassie (Bio-Rad) stained, the second one was wet blotted to a PVDF membrane (Merck Millipore) with Mini Trans-Blot Cell (Bio-Rad) at 100V for 1h. Fluorescent antibody staining was performed with SNAP i.d. 2.0 (Millipore). As primary antibody Anti-His-TAG from mouse (Invitrogen/FisherScientific #10755503) was used and the secondary antibody was an Anti-mouse IgG(H+L), Cy 3 conjugated from goat (Dianova #115-165-146) both diluted 1:500 in PBST. Pictures were taken in RGB multiplex mode of Fluorchem Q imager (ProteinSimple).

Bacterial surface preparation and Cell culture
The Nexterion coverslip was divided in 12 wells (0.56 cm 2 per well) by placing a silicone gasket on the top. 50 μL Poly-D-lysine solutions (2 mg/ml in PBS) were incubated in the wells for 60 min, allowing covalent immobilization of the molecules by reaction of their amine groups with the activated carboxylic acids at the Nexterion surface. Substrates were blocked by immersing in 50mM ethanolamine in PBS for 60 min and rinsed with water 3 times. Before the cell experiment, the substrates were sterilized by incubating in 70% ethanol for 5 min and rinsed with sterile PBS 3 times.
To immobilize bacteria on these surfaces, bacteria were inoculated from glycerol stocks in 5 mL LB broth containing 50 μg/mL ampicillin and grown overnight at 30 o C, 250 rpm. The bacterial densities typically reached ~0.5 O.D. 600nm by the morning. The cultures were then spun down at 3500 rcf for 10 mins and the pellets were resuspended in in sterile PBS with O.D. 600nm 1.0. 100 μL of these bacterial solutions were incubated for 30 min in each well. These surfaces were then washed 3 times with sterile PBS by vigorous pipetting.
WT eGFP-Vinculin mouse embryonic fibroblast lines (MEF-vincGFP) were cultured in DMEM with 10% fetal bovine serum, 1% Nonessential Amino Acids, 1% Sodium Pyruvate and 4 mM L-Glutamine (all of them purchase from GIBCO) in a humidified incubator at 37 o C/5%CO 2 . For every experiment, the cells were trypsinized and the cell density was determined a TC20™ Automated Cell Counter (Bio-Rad). The cells were directly seeded in the wells at a density of 2.5 x 10 4 cells / well. For all experiments with bacteria, the cellculture medium was supplemented with 50 μg/mL ampicillin.

Bacterial growth control assay
The E. coli+ bacterial surfaces were incubated in the DMEM-based cell culture medium containing 50 μg/mL ampicillin. Appropriate amounts of tetracycline and 0.1 mM IPTG were added when necessary as mentioned. Live-cell imaging was performed using the Nikon Ti-Eclipse microscope with a 20x objective. Images were taken every 20 min for 22 h in both phase contrast and red fluorescence channels. After 22 h, all samples were gently washed 3 times with PBS and DMEM-based cell culture medium along with appropriate additions were added. Live-cell imaging was then repeated for another 7 h. Bacterial surface density analysis was performed by quantifying the area of the background, where bacterial cells had not grown. Fluorescence intensity values were determined by quantifying the mean-grey values of the fluorescing bacteria in the red channel minus the background value.

Induction of protein expression
Three strategies were used to induce protein expression in the surface-immobilized bacteria using PA-IPTG and IPTG. PA-IPTG stock solution was prepared in DMSO at a concentration of 500 mM and IPTG stock solution was prepared in water at a concentration of 100 mM.
(i) PA-IPTG was diluted in the appropriate medium at a final concentration of 500 nM and this solution was exposed to 360 nm irradiation using a Lumos 43 illuminator (Atlas photonics) for 1 min. This medium was then added on the bacterial surfaces. (ii) PA-IPTG was diluted in the appropriate medium at a final concentration of 500 nM and added on the bacterial surfaces in the dark. When required, 360 nm light was irradiated from the Zeiss Axio Observer Z1 microscope using a Colibri 2/365 nm LED module light source at 50% intensity and an EC-Plan NeoFluar 20x objective for 2 min. When only bacteria were present, the objective was scanned over the whole well and when MEF-vincGFP cells were also present, irradiation was done only in one corner of the well. (iii) IPTG was added to the medium at a final concentration of 100 nM

Sample fixation and staining
For high magnification and confocal imaging, samples were fixed by washing 3 times with sterile PBS then incubating for 15 min with 4% paraformaldehyde (PFA) solution in PBS then washing again 3 times with sterile PBS. The cell nuclei were stained using DAPI and mounted using standard protocols.

Sample Preparation for Scanning Electron Microscopy and Image acquisition
The samples were washed with PBS, fixed for 30 min with 2 wt% glutaraldehyde in 0.1 M Cacodylat (214 g/mol), and then washed 3 times with sterile PBS. In order to dehydrate, the samples were immersed into ethanol with increasing concentrations, 30% for 10 min, 50% for 10 min, 70% for 10 min, 80% for 10 min, 90% for 10 min, 96% for 10 min, 100% for 2x15 min. Subsequently, the sample was incubated in 100% ethanol mixed at a 1:1 volume ratio with hexamethyldisilazan (HMDS) for 15 min. After that, samples were immersed into pure HMDS for 15 min twice. At the end, the samples were dried in an exhaust fume hood. The surface of the cell membrane was observed using the scanning electron microscope (SEM) (JSM-7500F; JEOL).

RFP Secretion test
5 mL bacterial cultures were grown from glycerol stocks to an O.D. 600nm of ~0.5 as already described. The cultures were split in 2 parts then spun down and resuspended either in LB medium or DMEM-based cell culture medium with 50 μg/mL ampicillin. These cultures were further each split in 2 parts. In one part 0.1 mM IPTG was added and in the other nothing was added. The cultures were incubated at 37 o C 250 rpm. After 3 h of incubation, 100 μL of the cultures were taken and spun down at 6000 rcf for 5 min and 50 μL of the medium was pipetted out from the top and incubated with 10 μL of Ni-NTA agarose beads (Qiagen) for 5 min. The beads were then spun down at 1000 rcf for 30s and washed 2 times with PBS. The bead solutions were placed in 96 well-plate wells and imaged using the Nikon Ti-Eclipse microscope. After 16h of incubation, the cultures were spun down and 20 μL of the medium was pipetted out from the top and used for SDS-PAGE analysis, performed as already described.

Sample Preparation for Membrane Labeling and Image acquisition
L929 fibroblast cells were seeded on E. coli+ or E. coli-bacterial surfaces and allowed to interact for 18 h in the presence of 100mM IPTG and 2 µg/mL of Tetracycline. After that, the samples were washed with Hank's balanced salt solution (HBSS). Wheat germ agglutinin conjugates Oregon Green® 488 (WGA, invitrogen ) was diluted to 5 µg /ml by HBSS to get the working concentration. This labeling solution was added to cover cells and was incubated for 10 minutes at 37°C. When labeling was complete, the labeling solution was removed, and cells were washed twice with 200 µL HBSS. Then samples were mounted in pre-warmed HBSS buffer for imaging.
The fluorescence of the L929 cell membrane and RFP were observed using a confocal microscope (Zeiss LSM 880) controlled by ZEN black software. For the quantification and comparison of the fluorescence intensity, all the Z-stacks images were acquired under the same conditions. To assess the fluorescence intensity, a line scan cross the cell membrane was performed and the plot profile of fluorescence was measured. The xy, xz and yz optical sections of Z-stacks were generated by orthogonal view of ZEN black software.

Control of bacterial growth using tetracycline
Addition of tetracycline at concentrations spanning 1 -10 µg/mL slowed down bacterial growth, with nearly no growth at 10 µg/mL (Figure 1b). When tetracycline was removed and IPTG added, the bacterial surfaces developed red fluorescence at similar rates indicating that no permanent damage was imparted to the bacterial cells' ability to produce heterologous proteins (Figure 1c). Bacterial surfaces exposed to 10 and 5 µg/mL of tetracycline showed a slightly faster rate of production and higher intensity of red fluorescence compared to the other surfaces, most likely since the bacteria did not overgrow into a possible overcrowded saturation phase. In all cases, red fluorescence reached detectable levels after approximately 90 mins, fitting well with previously reported half-time for the maturation of the protein. [36] In the presence of both tetracycline and IPTG, it was seen that protein expression slowed down progressively with increasing tetracycline concentrations ( Figure 1e). Notably, protein expression, induced by IPTG, also delayed bacterial growth even in the absence of tetracycline and a lower tetracycline concentration was required to arrest the growth over several hours (Figure 1d). Hence in further experiments, 10 µg/mL tetracycline was used before induction of gene expression and either 2 µg/mL or 0 µg/mL was used after induction when the duration of the experiment was longer or shorter than 10 h respectively.  Figure S2. Two-step light-responsive activation mechanism of PA-IPTG. On irradiation with 360 nm light, PA-IPTG rapidly forms regioisomeric NP-nitrosocarbonyl esters as intermediates that then get hydrolyzed to active IPTG by intracellular esterases.

Mammalian cell response to IPTG-induced bacterial surfaces
MEF-vincGFP cells were seeded on bacterial surfaces after which 0.1 mM IPTG was added into the medium to induce bacterial protein expression. The cells recognized and rapidly pulled the E. coli+ bacteria off the surface. This seemed to happen in all directions around the MEF-vincGFP cells and resulted in the cells migrating over the surface and accumulating E. coli+ cells under them ( Figure 2a). As expected, the cells showed on interaction with the activated E. coli-cells (Figure 2b). Z-stack images of MEF-vincGFP cells on E. coli+ surfaces, obtained using laser-scanning confocal microscopy, show focal adhesions completely enveloping the bacteria, indicating that the cells used some sort of gripping mechanism to pull the bacteria off the surface (Figure 2c), while no such interaction was seen with E. coli- (Figure 2d). SEM images further revealed cell protrusions completely engulfing the bacteria (Figure 2e,f). The bacteria were seen both below and above the cell extensions and some even emerging from underneath through pores.

Supporting information videos:
SI Video S1: Phase-contrast and Red fluorescence merged channel time-lapse imaging of in situ light-activated E. coli+ in the presence of MEF-vincGFP cells. Time counter format is in hh:mm and starts 1h after light-activation.