Can Single Cell Respiration be Measured by Scanning Electrochemical Microscopy (SECM)?

Ultramicroelectrode (UME), or, equivalently, microelectrode, probes are increasingly used for single-cell measurements of cellular properties and processes, including physiological activity, such as metabolic fluxes and respiration rates. Major challenges for the sensitivity of such measurements include: (i) the relative magnitude of cellular and UME fluxes (manifested in the current); and (ii) issues around the stability of the UME response over time. To explore the extent to which these factors impact the precision of electrochemical cellular measurements, we undertake a systematic analysis of measurement conditions and experimental parameters for determining single cell respiration rates via the oxygen consumption rate (OCR) in single HeLa cells. Using scanning electrochemical microscopy (SECM), with a platinum UME as the probe, we employ a self-referencing measurement protocol, rarely employed in SECM, whereby the UME is repeatedly approached from bulk solution to a cell, and a short pulse to oxygen reduction reaction (ORR) potential is performed near the cell and in bulk solution. This approach enables the periodic tracking of the bulk UME response to which the near-cell response is repeatedly compared (referenced) and also ensures that the ORR near the cell is performed only briefly, minimizing the effect of the electrochemical process on the cell. SECM experiments are combined with a finite element method (FEM) modeling framework to simulate oxygen diffusion and the UME response. Taking a realistic range of single cell OCR to be 1 × 10–18 to 1 × 10–16 mol s–1, results from the combination of FEM simulations and self-referencing SECM measurements show that these OCR values are at, or below, the present detection sensitivity of the technique. We provide a set of model-based suggestions for improving these measurements in the future but highlight that extraordinary improvements in the stability and precision of SECM measurements will be required if single cell OCR measurements are to be realized.


SI-1 HeLa cell culture and media preparation
HeLa cells were obtained from Public Health England, supplied by the European Collection of Authenticated Cell Cultures (ECACC, catalogue number: 93021013).Cells were grown in Minimum Essential Medium Eagle (Sigma Aldrich, M2279) supplemented with Lglutamine (used in 100-fold dilution, Sigma Aldrich, G7513), penicillin and streptomycin (used in 1000-fold dilution, Sigma Aldrich, P4333), heat-inactivated fetal calf serum (10-fold dilution, HIFC, Sigma Aldrich, 12106C), and non-essential amino acids (100-fold dilution, Sigma Aldrich, M7145).Due to the sodium bicarbonate buffer used, cells were incubated at 37 °C in 5% CO2.Cell flasks were passaged when the confluency reached approximately 80% (no more than a week), and never exceeded a passage number of 10.
Individual sample dishes of cells were prepared on 50 mm WillCo Wells (Glass thickness No. 1.5, USE, HBST-5040) the evening prior to experiments, cultures were diluted to approximately 5´10 6 cell mL -1 .
For the purposes of all SECM measurements, the medium was replaced with a Minimum Essential Medium (Sigma Aldrich, 56416C, referred to here as M5) buffered with HEPES (100-fold dilution, Sigma Aldrich, H0887), warmed to 37 °C.

Fabrication of SECM probes
Disk shaped platinum UMEs with a radius, a, of 5 µm were prepared as described elsewhere. 1In short, this method involved encapsulating platinum wire (Goodfellow, 99.99 %, 10 µm diameter) in borosilicate glass capillaries (GC200-10, Clark Electromedical Instruments), which was then heated under vacuum and pulled to produce a tapered end using a PB-7 Narishige micropipette puller.The tapered end was then polished to create a smooth flat surface, and the glass sheath is receded through polishing the UME at an angle to create the desired ratio of active electrode surface (Pt) to glass (RG value) of 15.
The UME surface was platinised in a 0.1 M hexachloroplatinate solution (0.1 V/s deposition).Platinisation of the platinum probe creates a layer of amorphous platinum on the surface (often termed platinum black).This amorphous material increases the total surface area of available platinum, without significantly increasing the total radius of the UME.This increased platinum surface in turn allows greater sensitivity to detecting oxygen, without Supporting Information

S.4
needing to increasing the overall size of the UME.This was done, as several previous studies have reported increasing sensitivity through platinization for detection (reduction) of oxygen and peroxy species. 2,3 he amount of platinum deposited in the platinization was limited in order to ensure that the electrode geometry remained as a disk.A platinized probe disk surface is shown in Figure S-1, where the rough-textured platinum accumulates most at the edge between the disk and the glass sheath.approached to a glass slide (substrate) with the same feedback threshold as used for the cellular measurements.The normalized current fluctuates around a value of 0.85, demonstrating the extent of hindered diffusion at this tip-substrate distance (d), and demonstrating that it is highly similar to that found for the cell measurements (Figure 3).

Cell Staining
Some cells were stained with Tetramethylrhodamine, Methyl Ester (TMRM, Thermo Scientific, M20036), to observe the mitochondrial membrane potential using the confocal microscope.For staining, the media was removed and replaced with PBS, and supplemented with 50 nM TMRM (prepared from a stock solution of 7.5 µM in DMSO).Cells were then incubated with the dye for 30 minutes at 37 °C, before being washed with 3´1 mL applications Supporting Information S.6 of sterile PBS to remove of residual dye.Fresh M5 media was then added to the plate for scanning.

Confocal Laser Scanning Microscopy (CLSM) Imaging Conditions
A Leica TCS SP5 X CLSM was used for imaging HeLa cells stained with TMRM, located in the mitochondrial membrane.Cells were imaged using an Argon laser, set to a wavelength of 514 nm.Emission was detected between 590 and 605 nm.A ´40 oil immersion objective was used, with the confocal pinhole set to 67.97 µm giving a section thickness of approximately 0.967 µm.The 514 nm laser power under these conditions was recorded to be 0.05 mW at the sample plane.Cells were typically imaged at 30 second intervals for the duration of the SECM experiment.

FEM Simulations
All FEM simulations were performed with COMSOL Multiphysics (v 5.5.) using the Transport of Diluted Species module.A 2D axisymmetric simulation domain was used, as depicted in Figure 2 of the main manuscript.Mesh density (> than 35000 elements) and domain size (radial length and height of the domain > than 100a) were set so the simulation results are independent of both.All boundary/domain conditions are summarised in Table S

Figure 2
Figure 2 of the main manuscript.Only fluxes normal to the boundary were considered, noted by "".

Table S -1. Boundary
-1, and simulated parameters values in Table S-2./domain conditions implemented in the FEM simulations, as shown in

Table S -
2. Parameters used for the simulations, relating to the geometry and in other parts of the simulations.