Supporting Information

Transcription

1 Electronic Supplementary Material (ESI) for Analyst. This journal is The Royal Society of Chemistry 2016 Supporting Information Development of a Rechargeable Optical Hydrogen Peroxide Sensor Sensor Design and Biological Application Klaus Koren a *, Peter Ø. Jensen b, Michael Kühl ac * a: Marine Biological Section, Department of Biology, University of Copenhagen, 3000 Helsingør, Denmark klaus.koren@bio.ku.dk mkuhl@bio.ku.dk b: Department of Clinical Microbiology, Rigshospitalet Copenhagen, 2100 Copenhagen Denmark. c: Plant Functional Biology and Climate Change Cluster, University of Technology Sydney, Ultimo, New South Wales 2007 Australia. Contents List of used Chemicals:...2 Stability and Reproducibility of the Optical Sensor:...2 Cross Sensitivity of the Optical Sensor:...4 PMN Measurements:...6 Biofilm Measurements:...8 Photograph of the Sensor Measuring Setup...8 From Sensor Signal to Concentration Profile...9 References:...9 1

2 List of used Chemicals: Prussian blue (soluble for microscopy), ethanol, L-Ascorbic acid, Citric acid monohydrate, Sodium phosphate, 2-amino-2-hydroxymethyl-propane-1,3-diol (TRIS), D-Glucose, (30% solution), phorbol- 12-myristate-13-acetate (PMA), K, FeCl 2, NaOCl solution (10-15%) and PBS tablets were obtained from Sigma-Aldrich (sigmaaldrich.com). The enzymes Glucose oxidase from Aspergillus niger (100, ,000 units/mg) and Catalase from bovine liver (2,000-5,000 units/mg) were also obtained from Sigma-Aldrich. Molecular biology grade Agarose was obtained from Calbiochem (merckmillipore.com). All solvents were obtained from commercial sources. If not stated otherwise, Milli Q water was used. Stability and Reproducibility of the Optical Sensor: The novel sensor is based on the use of Prussian blue (PB) as sensitive element. The stability of PB during measurements is essential. To study this, an oxidized sensor (PB state) was kept in ph 9 buffer for >17 hours with continuous sensor readout and illumination. Figure S1 shows that the sensor signal was extremely stable, indicating that both PB and Cr-YAB are not susceptible to bleaching and/or leaching within this time period. Upon the addition of ascorbic acid (AA) the signal increased immediately as the PB got reduced to PW normalized sensor signal time (h) addition of AA Figure S1: Sensor stability at ph 9. The stability of the optical sensor (PB state, oxidized state) at high ph was evaluated for >17 hours, where the sensor signal did not change significantly. Upon addition of ascorbic acid (AA) the signal increased immediately as the PB got reduced to PW. Figures S2 and S3 visualize the sensor stability upon continuous measurement and compare normal readout with differential readout. 2

3 signal Figure S2: Sensor signal over 1 hour of measurement in 50 µm. Over a period of 1 hour the sensor was repeatedly dipped into 50 µm and recharged. A slight decrease in the sensor signal was observed. In one of the cycles the sensor was kept in the 50 µm solution for an extended time period. All in all, the sensor stability is satisfactory, but a slight drift can be observed S t Figure S3: Differential sensor readout over 1 hour of measurement in 50 µm. In contrast to the pure intensity readout, the differential readout was less prone to signal drift. In the short measurement period a fast signal drop was followed by a short (1-3 second) stable plateau, during which the sensor reading was done before sensor recharging. Upon extended exposure in the 50 µm solution, the signal change over time decreased due to diffusion of towards the sensor and a decreasing availability of unreacted PW. 3

4 0.2 0 µm 10 µm 20 µm 0.1 S/ t Average signal: RSD: 2.4% Average signal: RSD: 5.0% Average signal: RSD: 4.5% Figure S4: Repeatability test; 10 repeated measurements at 3 known concentrations (0, 10 and 20 µm) were performed. From the obtained signal (ΔS/Δt) the average signal and RSD were calculated. For all 3 concentrations the RSD was smaller than 5%. The main factor affecting sensor reproducibility is the dip coating of the fiber tip. It is difficult to generate a similarly thick sensor coating when preparing several fiber sensors, although we aimed after obtaining a similar initial signal when coating the individual sensor fibers. Nevertheless, the sensitivity of each sensor differed. Each sensor was calibrated before and after use (at least 4 points). In all cases, a highly linear calibration curve (R 2 >0.995) was obtained and the calibration curve did not change throughout experiments. Every experiment was performed with one sensor. Cross Sensitivity of the Optical Sensor: A major concern when evaluating optical sensor performance is the potential cross sensitivity to other analytes. PB is known to be sensitive to ph 1. The effect of ph and also variation on the sensor sensitivity (slope of the calibration curve) is shown in figures S5 and S6. 4

5 S/ t k=-2205 k=-2774 k=-2597 ph 9 ph 4 ph Figure S5: Calibration curve of a sensor at 3 different ph values (ph 9, ph 7.4 and ph 4). A decrease in ph increases the sensitivity of the sensor system. At ph >9, the PB exhibits a decreased stability 2. Pure -0.1 Pure N 2 S/ t Figure S6: Effect of on sensor calibration. Sensor sensitivity increased with increased levels (increases by ~18%). This might be due to a destabilization of PW by. Besides ph and, other ROS could interfere with the sensor. Although PB is known to be very selective for and is often referred to as an artificial peroxidase 3 we tested the cross sensitivity to ROS. Figure S7 shows the response of the sensor to the hydroxyl radical, hypochlorite and superoxide anion. Hydroxyl radical was prepared by reacting with Fe 2+ ions. NaOCl was obtained as a solution and diluted. Superoxide was obtained by adding solid K. 5

6 S/ t -0.4 O PBS OH. OCl- O catalase Figure S7: Response of the sensor to pure buffer (PBS), 50 µm, 50 µm OH, 50 µm HOCl, and 50 µm of -. While the sensor showed a strong response to, it was insensitive to the hydroxyl radical and hypochlorite. Exposure to superoxide anion lead to a change in sensor signal of ~50% of the induced signal change. When catalase was added to the superoxide containing solution the change in signal was similar to pure buffer. This leads to the conclusion that superoxide dismutates spontaneously towards and causes a response by the sensor. Removing the produced by catalase resulted in a loss in signal. Even when more superoxide is added (in form of K ) to the catalase containing solution no change in signal is observed. The columns in the graph show means ±SD (n=3-6). PMN Measurements: Experimental details: As defined by the Danish Act on Research Ethics Review of Health Research Projects Section 2, the project does not constitute a health research project and was thus initiated without special approval from The Committees on Health Research Ethics in the Capital Region of Denmark. Therefore, verbal informed consent was obtained using waiver of documentation of consent. Human blood samples were obtained from normal healthy volunteers by venous puncture and collected in BD vacutainers coated with mm Na-citrate (Becton-Dickinson, ). The blood was mixed with dextran (T-500) 1:5 and the erythrocytes were sedimented for 1 hour. The supernatant was applied to Lymphoprep (Axis-Shield PoC) and centrifuged at 300g for 15 min at 5 C. The supernatant was discarded and the neutrophils were treated with 2 ml 0.2% NaCl in order to lyse remaining erythrocytes. Lysis was terminated by adding 2 ml 1.6% NaCl and 6 ml Eagle-MEM (Bie & Berntsen, Denmark). The cells were centrifuged at 300 g for 10 min at 5 C, the supernatant was discarded, and the PMNs were resuspended in Krebs-ringer buffer with 10% glucose to a density of 5x10 6 ml -1. Measurements were performed in a climate room set to 37 C, and the and (Pyro Science GmbH, Gaermany) sensors were calibrated at this temperature. The measuring setup is shown in figure S8. 6

7 Figure S8: Measurement setup for measurements on PMNs. The and sensors were both dipped into a stirred PMN suspension in Krebs-Ringer Buffer. The measurements were performed in a climate room set to 37 C. Figure S9 shows a replicate of the experiment shown in the main text Addition of PMA Figure S9: Production of by stimulated PMNs. Very similar changes in and concentrations over time were observed in a similar experiment shown in figure 4C in the main article text. A time lag of ~2 min was observed between the onset of PMN stimulation upon PMA addition and the onset of consumption and production. The levels peaked at ~10 µm. 7

8 Biofilm Measurements: Figure S10: Close-up photograph of the optical sensor positioned at the surface of a natural biofilm. The recharging gel was placed ~5 mm above the biofilm. Photograph of the Sensor Measuring Setup Figure S11: Measuring setup for measurements with the new sensor. The optical sensor is lowered with the help of a computer-controlled micromanipulator through the recharging gel into the sample. The recharging gel is filled into a 15 ml centrifuge tube that is cut open at the bottom. 8

9 From Sensor Signal to Concentration Profile In order to visualize the way from raw data to concentration profiles, figure S12 shows the raw sensor luminescence intensity trace (black), the differentiated sensor intensity trace (blue) and the final concentration profile (red) of the enzyme reaction on top of each other. When the sensor is dipped into the sample a decrease in signal is observed (negative ΔS/Δt) and when the sensor is retracted into the recharger gel the signal increases (positive ΔS/Δt). After every measurement, i. e., after each instance of dipping the sensor into the sample, the ΔS/Δt value was read out from the differentiated sensor intensity trace and correlated to the previously obtained sensor calibration, yielding the final concentration profile of the enzyme reaction was obtained. signal S/ t GOX Catalase Catalase GOX + Glucose Figure S12: Comparison of raw sensor luminescence intensity trace (black line, upper panel), the differentiated sensor intensity trace (blue line, middle panel), and the final concentration profile (red symbols and line, lower panel) along with the dynamics (black line, lower panel) during the enzyme reaction shown in figure 4a in the main text. References: 1. Koncki, R. & Wolfbeis, O. S. Optical chemical sensing based on thin films of Prussian Blue. Sensors Actuators B Chem. 51, (1998). 2. Koncki, R. & Wolfbeis, O. S. Composite Films of Prussian Blue and N-Substituted Polypyrroles: Fabrication and Application to Optical Determination of ph. Anal. Chem. 70, (1998). 3. Karyakin, A. A. & Karyakina, E. E. Prussian Blue-based `artificial peroxidase as a transducer for hydrogen peroxide detection. Application to biosensors. Sensors Actuators B Chem. 57, (1999). 9