Electronic communication of cells with a surface mediated by boronic acid saccharide interactions

Gold surfaces were molecularly tailored with a saccharide binding motif capable of covalently adhering cells. This facilitated communication via the macrophage membrane with implications for understanding mammalian cell signalling.

°C in a 5% CO 2 atmosphere reaching an approximate confluence of 80%. The DMEM was aspirated off and 3 mL of fresh DMEM was placed in each flask and cells were detached using a cell scraper.
The cells were then counted using a haemocytometer and diluted to give a cell suspension of 2 x 10 6 cells/mL. Prior to assay cells were centrifuged at 1000 rpm and the DMEM aspirated off leaving a cell pellet. The pellet was re-suspended in 50 mM PBS containing 100 mM KCl to give an equivalent cell concentration.  washed in PBS to remove loosely bound cells. Each chip was then placed into a well of a 12 well plate containing 1 ml of PBS per well. Chips were exposed to 10 µg/ml calcein and incubated for 30 minutes and then images were taken.

ATR-FTIR:
The spectra were recorded at a resolution of 16 cm -1 with a spectral range of 650-4000 cm -1 .The absorption spectra were the result of 64 scans and were recorded at room temperature. The spectra are then recorded using MicroLab FTIR software in a HP laptop connected to the spectrophotometer.

S-4
Gold surfaces were modified with carboxyl acid-terminated SAMs (COOH-SAMs) and then subsequently functionalised with 3-aminophenyl boronic acid (BA-SAMs). To confirm these modifications water contact angle and ellipsometry measurements were performed on the COOH-SAMs and BA-SAMs (Table 1). The advancing contact angles for COOH-SAMs and BA-SAMs are in good agreement with the literature. [1], [2] Previous literature reported a contact angle of 31.5° for a surface modified with boronic acid-terminated SAM that is in good agreement with the value obtained for our BA-SAM, 38.9°. The ellipsometric thickness of the COOH-SAM was determined to be 0.69nm, which is similar to theoretical values in Table 1 and literature values recorded by Barriet et al [1] .
Following modification with 3-aminophenylboronic acid, the thickness increased to 1.08 nm. The increase in surface thickness of 0.67 nm to 1.08 nm on coupling of the boronic acid to the SAM is significantly different (a paired t-test, p-value < 0.01) confirming the surface modification. To gather more evidence of surface functionalization of the COOH-SAM with BA-modified surfaces, ATR-FTIR spectroscopy was carried out. In Figure 1S, the presence of B-O peak (1342cm -1 ) [3] shown for BAmodified surfaces was present but absent from COOH-SAM, indicating the presence of the BA.  We can rule out that the redox behaviour observed in Fig 2 arises due to dilution effects, which may occur when cells are trapped on surfaces and release molecules, as with just hundreds of cells on the surface we observe redox behaviour in the cyclic voltammograms (Fig 3S). It would be envisaged that with cells adhered at surface compared to an equal number of cells in solutions in which the filtrate would be appraised, the concentration of any excreted molecules at the surface would be significantly higher than the filtrate. Consequently, with the filtrate a false negative may be possible. However, in our studies to ensure this was not the case we used an large number of cells (20 × 10 6 ) compared to the hundreds for surface studies (Figure 3S) , and the increased length of time in filtrate studies to 48 hours incubation compared to 3 hours for surface studies. Therefore, dilution effects would be negligible and data obtained from cyclic voltammograms suggests that the electrochemical observations of direct electron transfer proceed via the cell surface. Additionally, It is well known that one function of macrophage in defending against infection is to increase oxidative stress via increasing reactive oxygen species (ROS) production [5], [6], [7] . ROS have been shown to interact with boronic acids by converting them to phenol which, is an irreversible process [8] . If this were happening then we could expect a reduction in binding as only H-bonding would be possible as seen similarly with the COOH-SAM surface. However, one can note from the SPR results ( Figure 2S) for cells adhering to boronic acid surfaces that no decrease in binding is observed. Additionally the process is observed to be chemically reversible as demonstrated by the redox couple in the CV. Therefore, if ROS were converting the boronic acid to phenol which is irreversible, the redox couple in the cyclic voltammograms would not be reversible. Consequently we can therefore eliminate that the origin of the signal has anything to do with reactive oxygen species that are sequestered by the cell  SAMs in which a value of 7.7 x 10 -10 mol/cm 2 [9] . Thus this indicates that we have a monolayer coverage and the surfaces are stable.