Click-electrochemistry for the rapid labeling of virus, bacteria and cell surfaces

Methods for direct covalent ligation of microorganism surfaces remain poorly reported, and mostly based on metabolic engineering for bacteria and cells functionalization. While effective, a faster method avoiding the bio-incorporation step would be highly complementary. Here, we used N-methylluminol (NML), a fully tyrosine-selective protein anchoring group after one-electron oxidation, to label the surface of viruses, living bacteria and cells. The functionalization was performed electrochemically and in situ by applying an electric potential to aqueous buffered solutions of tagged NML containing the viruses, bacteria or cells. The broad applicability of the click-electrochemistry method was explored on recombinant adeno-associated viruses (rAAV2), Escherichia coli (Gram-) and Staphyloccocus epidermidis (Gram + ) bacterial strains, and HEK293 and HeLa eukaryotic cell lines. Surface electro-conjugation was achieved in minutes to yield functionalized rAAV2 that conserved both structural integrity and infectivity properties, and living bacteria and cell lines that were still alive and able to divide.

Ionization sources were performed with the available methods (ESI + , ESI -, ASAP + , ASAP -). A tolerance of 5 ppm was applied between calculated and experimental values.
The three cell lines HEK293, HUH7 and HeLa are not recorded in the misidentified cell lines list. They were cultured based on ATCC protocol. The HEK293 cell line comes from our Working Cell Bank (WCB) number 3 batch 14/03/2016 used routinely for AAV vector production in our vector core (https://umr1089.univ-nantes.fr/en/facilities-cores/cpv). They are regularly characterized by qPCR and RT-qPCR for E1A gene expression and they are tested for mycoplasma free and sterility. The HUH7 and HeLa cell lines were kindly provided by academic laboratories and are regularly tested for mycoplasma free and sterility. These regular characterizations are performed under our ISO 9001 quality management system (LRQA certificate number 10395603, expiry date 30 September 2024).

Supplementary Figure 1. Most likely path for eY-click with NML species
The precise oxidative radical coupling with tyrosines remains elusive, but is probably based on radical addition of the oxidized NML specie onto Y, followed by further oxidation of the eY-clicked product (by different pathways) and its spontaneous rearomatisation (path a).

Supplementary Figure 2. Other possible paths for eY-click with NML species
No evidences definitively exclude tyrosine co-oxidation and radical co-ligation with oxidized NML (path b) or a two-electrons electro-oxidation of NML followed by ene-type addition of tyrosine in a similar manner than with PTAD (path c

mL scale setup
Dimension of anode and cathode used was 4.5 cm x 0.3 cm x 0.10 cm (accessible surface = 3.63 cm²).
Considering a 1 mL solution with 40-50 % of electrodes dipping, the surface/volume ratio is estimated to be ~1.5 cm²/mL. As described in Figure S3, electrodes were connected to the electrodes holder part (A), the low-binding vial containing the solution of biological target and luminol derivative (B) was inserted into electrasyn-adapted 5 mL glass-vial (C) and the latter was screwed to the electrodes holder (ABC). Potentiostat was linked through electrical connections of the electrodes holder part and the resulting (ABC) vial was placed on the vial support on top of the orbital shaking device.

I. Synthesis and characterization of luminol anchors
➢ Synthesis

Compound 1
Azide derivative was prepared in 5 steps as we previously reported. 1

Compound S11
Propargylamide derivative S11 was prepared as described in literature from commercially available carboxylic acid S10. 2

II. eY-click on AAV2: Procedures and characterizations
In a 2 mL low-binding vial, 100 L of AAV2-GFP (1 E 12 vg in dPBS pH 7.4, produced by the vector production center-CPV of Nantes UMR1089 gene therapy laboratory) were added to 900 L dPBS pH 7.4 containing the appropriate quantity of carbohydrate luminol derivative 2 or 3 (final conc. 0.5 mM or 5 mM). The 1 mL scale setup was assembled ( Figure S3) and 750 mV vs Ag/AgCl were applied during the studied time at room temperature under gentle orbital shaking. After modification, the excess of unreacted luminol anchor was removed by dialysis in four successive rounds against dPBS pH 7.4 (+0.001% Poloxamer) in a 10 kDa MWCO cassette.

Titration of viral genomes (vg)
To determine the titer (

Dot Blot, Western Blot and Silver Staining
For dot blot analysis, nitrocellulose membrane was soaked briefly in PBS prior to assembling the dot blot manifold (BioRad), then AAV vectors (2.10 10 vg) were loaded. The obtained nitrocellulose membrane was then treated for the appropriate characterization (see below: capsid integrity or carbohydrate detection).
For silver nitrate or western blot, all AAV vectors (2.10 10 vg) were denatured at 100 °C for 5 min using Laemmli sample buffer (5 L) and separated by SDS-PAGE on 10% Tris-glycine polyacrylamide gels (Life Technologies

In vitro transduction
The infectivity of each sample was measured as follows. HUH7 or HEK293 cells were seeded in DMEM with 10% FBS serum and 1% penicillin-streptomycin in 6-well culture plates at a density of 10 6 cells/well.
Cells were then incubated overnight at 37 °C with 5% CO2 to reach 50% confluence. Then, AAV samples were prepared by serial dilution considering the studied multiplicity of infection (MOI = n virus / n cells, varying from 10 3 to 10 4 ) and 2 L of the samples were added to separate wells in the 6-well plates. The latter were incubated at 37 °C for 24 h. AAV-GFP-infected cells were detected and quantified by fluorescence microscopy and flow cytometry on a BD-LSRII Flow Cytometer (BD Bioscience). All data S16 were processed by FlowJo (V10, Flowjo LLC, Ashland, OR). Data are shown as mean ± SD (standard deviation). Distinct samples were measured (n=3). Each sample was cultured (n=3) and OD600 was regularly measured. Source data are provided as a source data file

Fluorescence microscopy and flow cytometry
Electrobioconjugation efficiency was visualized by fluorescence microscopy (FITC excitation conditions) using Nikon Eclipse NI-E microscope (data treated with NIS software) and quantified by flow cytometry using CYTOFLEX cytometer from Beckman Coulter® -Life Sciences (data treated with FlowJo software). Quantification of proteins migrated in lanes "total", "inner" and "outer" after the fractionation protocol and electrophoresis. Quantification using a LI-COR imaging system. Experiments were performed in duplicate. Source data are provided as a source data file S21

IV. eY-click on living cells: Procedures and characterizations
Cells (HEK293 or HeLa) were cultured with 10% FBS serum and 1% penicillin-streptomycin at 37 °C with 5% CO2. The cells were trypsinized and harvested in PBS pH 7.4 at a concentration of 6.10 6 cells/mL.
Then, in a 2 mL low-binding vial, 500 L of the cell solution (final conc. 3.10 6 cells/mL) were added to 500 L of a 2 mM solution of 1 in PBS pH 7.4 (final conc. 1 mM). The 1 mL scale setup was assembled ( Figure S3) and 750 mV vs Ag/AgCl were applied during the studied time at room temperature under gentle orbital shaking. After modification, the excess of unreacted luminol anchor was removed by performing three times the following sequence: i) centrifugation ( HeLa cells were cultured with 10% FBS serum and 1% penicillin-streptomycin at 37 °C with 5% CO2. The cells were trypsinized and harvested in PBS pH 7.4 at a concentration of 6.10 6 cells/mL. Then, in a 2 mL low-binding vial, 500 L of the cell solution (final conc. 3.10 6 cells/mL) were added, centrifuged to remove supernatant and 1 mL of 1 mM solution of 4 in PBS pH 7.4 (final conc. 1 mM) was added (4 has poor solubility in aqueous media and it was helped with ultrasonic bath). The 1 mL scale setup was assembled ( Figure S2) and 750 mV vs Ag/AgCl were applied during the studied time at room temperature under gentle orbital shaking. After modification, the excess of unreacted luminol anchor was removed by performing three times the following sequence: i) centrifugation (1200g during 2 min), ii) supernatant withdrawal, iii) cells resuspension in 1 mL PBS. At the end of 3 rd sequence, cells were resuspended in 50 L of a Streptavidin-FAM solution (obtained from ThermoFisher®) and incubated at 37 °C during 1h under moderate orbital shaking. Then, non-complexed Streptavidin was removed by performing three times the previous centrifugation/removal/washings with 1 mL PBS. At the end of 3 rd sequence, modified cells were resuspended in the appropriate volumes for characterizations.

Cell viability
Cell culture: Cell viability after modification was evaluated by cell culture. Unmodified, controls and conjugated cells were seeded in DMEM with 10% FBS serum and 1% penicillin-streptomycin in a 24-well culture plate and incubated at 37 °C with 5% CO2. Growth ability and confluences were evaluated, quantified and compared by microscopy and using Vi-CELL XR after 24 h, 48 h and 72 h.

Supplementary Figure 16. Cell growth ability after eY-click
Cells quantification by Vi-CELL XR after 24h or 48h culture of samples. The ability to multiply was conserved for all samples. Source data are provided as a source data file.
Viability test with trypan blue: Approx. 5.10 5 cells in 600 L PBS were subjected to automatized trypan blue viability test using Vi-CELL XR (from Beckman Coulter, Life Sciences). % of viable cells is calculated from the ratio of viable cells and total cells. Viability was also evaluated after 24h and 48h culture of electro-labelled cells.

Supplementary Figure 17. Cell viability after eY-click
Automatized Trypan Blue viability test (Vi-CELL XR). All samples were found viable. Data are presented as mean values +/-SD of three independent electro-conjugation experiments for viability tests right after eY-click and as the obtained value (n=1) after 24h and 48h cell culture. Source data are provided as a source data file.

S23
Viability test with Zombie Yellow and Annexin V: Approx. 2.10 6 electro-conjugated cells were resuspended in 1 mL of freshly prepared Zombie Yellow staining buffer (100 L Zombie Yellow BV605 from Biolegend® diluted in 1 mL with Brilliant Violet Stain Buffer from BD Biosciences®) and incubated at RT in the dark. Centrifugation/supernatant withdrawal/PBS washing sequence was performed twice and the cells were resuspended in 30 L of Annexin V 1X buffer and 1.5 L of PE-CF594 Annexin V (from Fischer Scientific®) were added. After 15 min incubation at RT in the dark, samples were diluted with 400 L of Annexin V 1X buffer and analyzed by flow cytometry to evaluate proportions of dead cells and apoptotic cells.

Supplementary Figure 18. Zombie Yellow and Annexin V viability tests.
Zombie Yellow test led to fully conserved viability. For Annexin V assay, untreated cells labelled with annexin V were used as reference. The electro-bioconjugation protocol didn't compromise cells viability, as an overall comparison with experiments without electrical current (eY-click OFF) or without NML derivative (Washing sequence). HEK cell line seems more stressed by the protocol, probably because these are adherent cells. Source data are provided as a source data file.