An Alternating Current Electroosmotic Flow‐Based Ultrasensitive Electrochemiluminescence Microfluidic System for Ultrafast Monitoring, Detection of Proteins/miRNAs in Unprocessed Samples

Abstract Early diagnosis of acute diseases is restricted by the sensitivity and complex process of sample treatment. Here, an ultrasensitive, rapid, and portable electrochemiluminescence‐microfluidic (ECL‐M) system is described via sandwich‐type immunoassay and surface plasmonic resonance (SPR) assay. Using a sandwich immunoreaction approach, the ECL‐M system employs cardiac troponin‐I antigen (cTnI) as a detection model with a Ru@SiO2 NPs labeled antibody as the signal probe. For miR‐499‐5p detection, gold nanoparticles generate SPR effects to enhance Ru(bpy)3 2+ ECL signals. The system based on alternating current (AC) electroosmotic flow achieves an LOD of 2 fg mL−1 for cTnI in 5 min and 10 aM for miRNAs in 10 min at room temperature. The point‐of‐care testing (POCT) device demonstrated 100% sensitivity and 98% specificity for cTnI detection in 123 clinical serum samples. For miR‐499‐5p, it exhibited 100% sensitivity and 97% specificity in 55 clinical serum samples. Continuous monitoring of these biomarkers in rats' saliva, urine, and interstitial fluid samples for 48 hours revealed observations rarely documented in biotic fluids. The ECL‐M POCT device stands as a top‐performing system for ECL analysis, offering immense potential for ultrasensitive, rapid, highly accurate, and facile detection and monitoring of acute diseases in POC settings.


Contents
Table S1 The sequences (5'-3') of miR-499-5p, capture and probe DNA in this work.Table S2 Summary of cost prices for cTnI test kit.Table S3 The cost of the components of ECL-M POCT device.Table S4 Summary of cost prices for chemiluminescence detection systems.Table S5 Parameters of COMSOL simulation.Table S6 Molecular interaction between cTnI antibody and cTnI antigen.Table S7 Sequences of other miRNA interferences.Table S8 Clinical information on patient samples.Table S9 Comparison of different methods for detection of cTnI in AMI diagnosis.Table S10 Comparison of different methods for amplification-free detection of miRNA in clinical diagnosis.Figure S1 Size of the designed ECL-M POCT device.Noteworthily, the height of the enclosed device was 135 mm and was 265 mm after opening the cover.Figure S4 AC voltage with a form of square wave was applied to the ECL-M chip which produced an alternating electric field between the two Ag electrodes.Figure S8 UV-vis adsorption spectra of AuNPs, Ru(bpy)3 2+ , Ru@SiO2 NPs. Figure S9 Zeta potential of Ru(bpy)3 2+ , Ru@SiO2 NPs, Ru@SiO2-NH2 and AuNPs.Figure S11 BCA analysis for Ab1 concentration in the ECL-M system.Figure S12 XPS spectroscopy of N 1s peak on AuNPs modified silicon wafer and Ab1 immobilized on AuNPs modified silicon wafer.Figure S13 Micrograph of dyed PBS separated by the engineered fluid containing 1% TX-100 in the capillary.The first injection was dyed yellow with potassium ferricyanide.The ECL reaction solution was dyed blue with methylene blue.Figure S14 ECL measurement of blank versus cTnI (concentration of 10 pg/mL) under three different conditions: without electric field, with DC-driven field and with ACdriven field.The working solution was PBS (pH 7.4) containing 0.1 M TPrA.Scan rate was 100 mV/s.Figure S15 COMSOL simulation model and boundary of ECL-M channel.Figure S16 Prediction of antigen-antibody interaction of cTnI.The X-ray crystal structures of a chicken anti-cardiac Troponin I scFv (4P48) and cardiac troponin C-troponin I complex (1MXL) were obtained from the Protein Data Bank.Multiple groups of residues were used to form hydrogen bonds between anti-cTnI antibody (purple one) and cTnI antigen (blue one), such as the hydrogen bond formed by Ser-199 of anti-cTnI and Lys-17 of cTnI.   Figure S22 EIS characterization of the ECL-M POCT biosensor corresponding with each modification step.a-h, CE (a), AuNPs/CE (b), Ab1/AuNPs/CE (c), cTnI/Ab1/AuNPs/CE (d), Ru@SiO2-Ab2/cTnI/Ab1/AuNPs/CE (e), capture DNA/AuNPs/CE (f), miR-499/capture DNA/AuNPs/CE (g), Ru(bpy)3 2+ -probe DNA/miR-499/capture DNA/AuNPs/CE (h).The concentration of cTnI was 1 pg/mL and miR-499-5p was 100 aM.The EIS working solution was 5 mM [Fe(CN)6] 3-/4- solution containing 0.1 M KCl.Scan rate was 100 mV/s.Figure S23 Synchronized electrochemical measurement of cTnI with (red curve) and without (blue curve) the AC voltage driven force under the voltage of 0~+2.0V.The concentration of cTnI was 1 pg/mL.The working solution was PBS (pH 7.4) containing 0.1 M TPrA.Scan rate was 100 mV/s.Figure S26 cTnI and miR-499-5p detection in different biological sample solutions (PBS, artificial saliva, urine, interstitial fluid, and serum).Significant difference between ECL signal of PBS and other biological samples was determined by one-way ANOVA followed by a t test.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. Blank; n.s., no significant.Figure S27 Time-course Infarct photograph of rat models.Once the blood collection was finished, rats were put to death after anesthetized with pentobarbital.Hearts were rapidly removed and sliced transversely into five sections.The hearts were soaked in 10% buffered formalin and frozen in -20 °C for 20 minutes.Five slices were incubated with 1% triphenyl tetrazolium chloride (TTC) at 37°C for 15 minutes under dark conditions to observe the infarction area.The normal heart tissues were stained with red color while infarct heart areas were white.

Name
NPs.

Figure S9
Zeta potential of Ru(bpy) 3 2+ , Ru@SiO 2 NPs, Ru@SiO 2 -NH 2 , Ru@SiO 2 NPs-Ab 1 and AuNPs.The mixture was then centrifuged at 12000 rpm for 10 min, followed by redispersed in 15 μL PBS solution (pH 7.4).Then the mixture solution was analyzed following the quantification of the protein concentration by the BCA method (Figure S11).The final concentration of Ab 1 was estimated as 79.2 μg/mL, showing the effective loading efficiency of 31.7 % in the protein analysis system.
To explain the load efficiency of capture DNA loading on AuNPs, we tested the concentration of capture DNA by a K5600 ultra-micro spectrophotometer.
20 μL of AuNPs were mixed with 15 μL of 100 μM capture DNA at 4 ℃ for a whole night.The mixture was then centrifuged at 12000 rpm for Figure S2 (a) Overall dimension diagram of microfluidic chip.(b) Schematic diagram of position and size of flow channel and electrodes.
Figure S3 (a) Front and (b) back photograph of the ECL-M chip.

Figure S5
COMSOL simulation of electrical field under the voltage of -4.5 V (a) and +4.5 V (b).

Figure S7
Statistical analyses of the size distributions of AuNPs (a) and Ru@SiO2 NPs (b).
Figure S10 AFM image (in 1× TM buffer) of Au electrode (a) and antibody immobilized on Au electrode (b) at 4 ℃ for one night.The color bar indicates the height of the scanned surface.

Figure S17
Photograph of the sample tested solution viscosity.Figure S18 Temperature field of the ECL-M channel under ACET effect.

Figure S19
Velocity distribution of ACET-induced Antigen solution in ECL-M channels.Figure S20 Photograph of the electrical conductivity of antigen solution.

Figure S21
Velocity distribution of ACEO-induced Antigen solution in ECL-M channels.

Figure S24
Optimization of experimental conditions.Optimization of Vpp of AC (a), AC-driven incubation time of cTnI (b), AC-driven incubation time of miR-499-5p (c).
Figure S28 (a) Comparison of ECL-M POCT device (this work) with detection of cTnI with other methods.Incubation time and LOD were estimated and compared with other methods.(b) Comparison of ECL-M POCT device (this work) with detection of miR-499-5p by other methods.Incubation time, reaction temperature, and LOD were estimated and compared with other methods.

Sequence ( 5 'Figure S1
Figure S1 Size of the designed ECL-M POCT device.Noteworthily, the

Figure
Figure S2 (a) Overall dimension diagram of microfluidic chip.(b)

Figure
Figure S3 (a) Front and (b) back photograph of the ECL-M chip.

Figure S4
Figure S4AC voltage with a form of square wave was applied to the ECL-

Figure S5
Figure S5 COMSOL simulation of electrical field under the voltage of -

Figure S10
Figure S10 AFM image (in 1× TM buffer) of Au electrode (a) and antibody

Figure S11
Figure S11 BCA analysis for Ab 1 concentration in the ECL-M system.
10 min, followed by redispersed in 15 μL PBS solution (pH 7.4).Then the mixture solution was analyzed following the quantification of the protein concentration by the BCA method (FigureS11).The final concentration of capture DNA was estimated as 92.4 μM, showing the effective loading efficiency of 92.4 % in the protein analysis system.

Figure S12
Figure S12 XPS spectroscopy of N 1s peak on AuNPs modified silicon

Figure S13
Figure S13 Micrograph of dyed PBS separated by the engineered fluid

Figure S14
Figure S14 ECL measurement of blank versus cTnI (concentration of 10

Figure S15
Figure S15 COMSOL simulation model and boundary of ECL-M channel.

Figure S16
Figure S16Prediction of antigen-antibody interaction of cTnI.The X-ray

Figure S17
Figure S17 Photograph of the sample tested solution viscosity.

Figure S18
Figure S18Temperature field of the ECL-M channel under ACET effect.

Figure S19
Figure S19 Velocity distribution of ACET-induced Antigen solution in

Figure S20
Figure S20 Photograph of the electrical conductivity of antigen solution.

Figure S21
Figure S21 Velocity distribution of ACEO-induced Antigen solution in

Figure S22
Figure S22 EIS characterization of the ECL-M POCT biosensor

Figure S23
Figure S23 Synchronized electrochemical measurement of cTnI with (red

Figure S24
Figure S24 Optimization of experimental conditions.Optimization of V pp

Figure S25 Figure
Figure S25 Optimization of experimental conditions.a, b, c, Optimization

Figure S27
Figure S27 Time-course Infarct photograph of rat models.Once the blood

Figure
Figure S28 (a) Comparison of ECL-M POCT device (this work) with

Table S2
Summary of Cost Prices for cTnI test kit. *The

price in the table is for one-person test. Table S3
The cost of the components of ECL-M POCT device.

Table S4
Summary of Cost Prices for Chemiluminescence DetectionSystems.
*The equipment prices in the table are cost prices.

Table S5
Parameters of COMSOL simulation.

Table S6
Molecular interaction between cTnI antibody and cTnI antigen.

Table S8
Clinical information on patient samples.

Table S9
Comparison of different methods for detection of cTnI in AMI diagnosis.

Table S10
Comparison of different methods for amplification-free detection of miRNAs in clinical diagnosis.