Microfluidic integration of regeneratable electrochemical affinity-based biosensors for continual monitoring of organ-on-a-chip devices

Aleman, Julio; Kilic, Tugba; Mille, Luis S.; Shin, Su Ryon; Zhang, Yu Shrike

doi:10.1038/s41596-021-00511-7

Download PDF

Protocol
Published: 28 April 2021

Microfluidic integration of regeneratable electrochemical affinity-based biosensors for continual monitoring of organ-on-a-chip devices

Nature Protocols volume 16, pages 2564–2593 (2021)Cite this article

16k Accesses
81 Citations
17 Altmetric
Metrics details

Subjects

Abstract

Organs-on-chips have emerged as viable platforms for drug screening and personalized medicine. While a wide variety of human organ-on-a-chip models have been developed, rarely have there been reports on the inclusion of sensors, which are critical in continually measuring the microenvironmental parameters and the dynamic responses of the microtissues to pharmaceutical compounds over extended periods of time. In addition, automation capacity is strongly desired for chronological monitoring. To overcome this major hurdle, in this protocol we detail the fabrication of electrochemical affinity-based biosensors and their integration with microfluidic chips to achieve in-line microelectrode functionalization, biomarker detection and sensor regeneration, allowing continual, in situ and noninvasive quantification of soluble biomarkers on organ-on-a-chip platforms. This platform is almost universal and can be applied to in-line detection of a majority of biomarkers, can be connected with existing organ-on-a-chip devices and can be multiplexed for simultaneous measurement of multiple biomarkers. Specifically, this protocol begins with fabrication of the electrochemically competent microelectrodes and the associated microfluidic devices (~3 d). The integration of electrochemical biosensors with the chips and their further combination with the rest of the platform takes ~3 h. The functionalization and regeneration of the microelectrodes are subsequently described, which require ~7 h in total. One cycle of sampling and detection of up to three biomarkers accounts for ~1 h.

Digital colloid-enhanced Raman spectroscopy by single-molecule counting

Article 17 April 2024

Shell buckling for programmable metafluids

Article 03 April 2024

Best practices for single-cell analysis across modalities

Article 31 March 2023

Introduction

Organ-on-a-chip systems have emerged as an enabling in vitro platform for drug screening by providing the biomimetic 3D microenvironments and dynamic cues that are typically lacking in conventional simplified cell cultures¹. Individual organ models can be connected in the same manner in which they appear in the human body to further reproduce interactions among the multiple tissue types². Besides on-chip models of healthy tissues and organs, their diseased counterparts can also be generated to bridge the gap between animal models and humans, particularly for disease types that cannot be faithfully established in animals³. To date, almost all human tissues have been reproduced in organ-on-a-chip devices, ranging from cardiovascular⁴ and musculoskeletal⁵ systems to neuronal⁶, digestive⁷ and renal systems⁸, as well as those in various disease states such as cardiomyopathy⁹, progeria¹⁰ and cancer¹¹. These improved in vitro models of human physiology and pathophysiology have made it possible to screen pharmaceutical compounds at higher accuracy and to achieve precision medicine through personalizing chips using patient-derived materials¹².

Despite great success of the organ-on-a-chip systems for broad applications in medicine, it is not always straightforward to characterize microtissue behaviors and their interactions with drug molecules. For example, the levels of biomarkers secreted by the microtissues are strongly indicative of their status, which change dynamically as the microtissues form, grow and mature, and as they experience external stimuli and undergo damage^13,14,15. Often, the biomarkers are measured by off-chip methods such as enzyme-linked immunosorbent assay (ELISA), which necessitates the medium to be sampled at desired time points^16,17,18,19. The repeated operations might disturb the microtissue responses within the microenvironments, and the typically large sampling volumes (comparing to the sizes of miniaturized microfluidic chip devices) inevitably limit the ability to continuously monitor the alterations in biomarker levels at elevated frequency²⁰.

To this end, strategies to integrate biosensors with microfluidic chips that can be directly connected to organ-on-a-chip systems become critical to performing in situ, continual and low-volume biomarker analyses. In this protocol, we provide a detailed procedure describing our recently developed approach for the fabrication of electrochemical (EC) affinity-based biosensors and their integration with microfluidic chips^13,21,22.

Development of the approach

To achieve our strategy, EC affinity-based biosensors are a good candidate because of their unique advantages such as excellent limit of detection, wide linear detection range, label-free detection ability, scalability to monitor multiple biomarkers and relatively simple operations and fabrication processes compared with other existing biosensor systems^14,21,22,23. Importantly, electrodes—a major part of EC-based biosensor systems—are simple to miniaturize through microfabrication processes; hence, microfabricated electrodes facilitate integration into organ-on-a-chip systems via microfluidic platforms (Fig. 1). Also, EC biosensors can be easily expanded with an array of microelectrodes that allow for the monitoring of multiple biomarkers using small sample volumes, resulting in accurate and efficient monitoring of biological behaviors and responses of microtissues in organ-on-a-chip systems¹³.

**Fig. 1: Schematic representation of the electrode fabrication and functionalization.**

In an ideal situation, the integrated affinity-based biosensors in an organ-on-a-chip system should continually monitor biomarkers to evaluate the dynamic behaviors of the microtissue for an extended period of time. To achieve this strategy, one of the major drawbacks of existing affinity-based biosensors^24,25 should be solved: that is, a lack of multipoint detection ability using single affinity-based biosensors due to surface saturation of the sensor when bioreceptors (antibodies or aptamers) repeatedly capture target biomarkers via their strong binding interaction. Therefore, various surface-cleaning processes have been developed for the regeneration of saturated microelectrodes that can allow multipoint detection strategies to function in a continual manner without replacing saturated biosensors. However, most cleaning methods require harsh cleaning solutions (low- or high-pH solutions) and time-consuming procedures that can damage the biosensor surface. In particular, damage to bioreceptors (antibody or aptamer) during the regeneration processes can affect their ability to capture target biomarkers, resulting in decreased sensitivity and selectivity. To avoid biosensor surface damage in the regeneration process, robust biosensors are required when harsh cleaning processes are applied. In this protocol, we describe the fabrication of robust EC affinity-based biosensors along with an improved EC cleaning method that will maintain EC oxidation signals on bare microelectrodes for up to 25 regenerations. This two-step regeneration strategy can also remove all molecules from the microelectrode surface, allowing for subsequent reimmobilization of fresh bioreceptors, with reproducible sensitivity/selectivity of the same biosensor for up to five regeneration cycles (Figs. 1 and 2). The reusable EC affinity-based biosensors can be readily integrated into organ-on-a-chip platforms through microfluidic chips to enable continual and long-term measurements of secreted biomarkers from microtissues upon drug, chemical or other exposures.

**Fig. 2: Effect of Au electrode thickness on reproducibility, sensitivity and selectivity upon multiple regeneration cycles.**

Microfluidic integration of the EC affinity-based biosensors is key to their successful application in organ-on-a-chip platforms²⁰. This protocol further discusses the design of microfluidic chips that allow for seamless integration with microfabricated electrodes. Enabled by built-in microfluidic channels with digitally controlled pneumatic valves, automated liquid routing can be achieved to facilitate a series of events including microelectrode functionalization, medium sampling and microelectrode regeneration, all completed on-chip with minimum manual operations (Fig. 3). The microfluidic chip design enables direct integration of these EC affinity-based chips with most existing organ-on-a-chip platforms through simple coupler connections or a microfluidic breadboard (see the ‘Applications’ section below), making them almost universally adaptable to biomarker measurements from engineered microtissue models and organoids. EC affinity-based biosensing chips can also be designed to suit different purposes, such as single-electrode format for quantification of one biomarker (Fig. 3d), single-use (until saturation occurs) format without regeneration (Fig. 3e) and multielectrode format for multiplexed biomarker detections with regeneration capability (Fig. 3f). In this protocol we will only discuss the single-microelectrode format (Fig. 3d), as the other two systems can be readily derived from this format by reducing or increasing the number of microchannels built into the microfluidic chips.

**Fig. 3: Assembly and valve configuration of microfluidic EC affinity-based biosensor chips (see Supplementary Data 1 for designs).**

Besides the microfluidic chip that integrates the microelectrode for automated electrode functionalization, biomarker quantification and microelectrode regeneration, this protocol also discusses the design of a microfluidic breadboard system for convenient interfacing of the EC affinity-based biosensing chip with one or multiple microtissue models. The breadboard is a central liquid-routing system containing user-defined microfluidic channel arrangements and pneumatic valves, with inlets/outlets built into different segments to connect to side modules (Fig. 4). These side modules can include the EC affinity-based biosensing chips, bioreactors hosting engineered microtissue models or organoids, bubble traps and medium reservoirs. As such, the automated liquid routing taking place in the central breadboard will lead to well-defined flow sequence in the side modules and controlled perfusion of, for example, the EC affinity-based biosensing chip for medium-sampling at desired time points and intervals. The breadboard configuration enables flexible integration of the EC affinity-based biosensing chip with most of the standard and existing microfluidic units used in organ-on-a-chip platforms.

**Fig. 4: Diagrams of the assembly of the breadboard and its connections with side modules (see Supplementary Data 1 for designs).**

Overview of the procedure

In this protocol, we describe in detail the preparation of the microfabricated electrodes (Steps 1–10); microfluidic chip design and fabrication (Steps 11–64); the integration of the microelectrode with the microfluidic chip (Steps 65–95); integration with the breadboard and bioreactor (Steps 109–136); procedures for on-chip microelectrode functionalization (Steps 137–166); procedures for on-chip medium sampling and biomarker measurement (Steps 167–174); and finally procedures for on-chip microelectrode regeneration (Steps 175–180). We initially reported on an automatized antibody/aptamer-based single EC biosensor to detect cardiac biomarkers, with nonregeneration capabilities²². Following this work we developed another EC biosensor with regeneration capabilities for continuous detection of liver-associated secretomes via antibody affinity²¹. Finally, we reported on a multiplexed EC biosensor capable of continuous detection of hepatic and cardiac biomarkers from microbioreactor modules integrated into a microfluidic breadboard controller¹³. These publications summarize the conditions and detailed fabrication of our liver and cardiac tissues, as well as the implementation of the supplementary physical biosensors. This protocol presents the necessary steps for the fabrication, assembly and functionalization of an EC affinity-based biosensing chip, along with the necessary steps for automatization operations through the breadboard and other components. We recommend that users utilize a bench-top or customized incubator, whichever allows the peripheral electronical components and pneumatic tubing to be easily accessible.

Applications

The affinity-based microfluidic regeneratable EC biosensors described in this protocol can be used for the detection of cell secretomes from any cell culture medium, providing that corresponding bioreceptors are available. Monitoring protein biomarker levels with these biosensors can be useful for drug toxicity and drug screening studies as well as monitoring changes in cell culture microenvironments upon most types of external stimuli such as temperature, pH and osmolarity, among others, in both planar and 3D culture systems.

For example, isolation of single organs from a multiorgan tandem series has been established as a toxicology validation technique^16,26, as well as fixed parallel series of multitissue setups for cell migration studies^27,28. The automated platform allows for multibioreactor connectivity, with reversible bypass capabilities and programable flow parameters. In addition to flow control, the modular competence of the breadboard to intercalate microbioreactors and biosensor chips, plus the compliance of a wide range of molecules to the affinity-based biosensing platform, may allow for direct measurements of secretomes from one microbioreactor before the next one without systematic disruptions. These capacities boost monitoring of cell cultures, including organ-on-a-chip systems and data collection, to levels unattainable before.

The same method can also be extended to beyond cell culture monitoring and essentially any measurements based on affinity, such as detection of viral materials, including those of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), in bodily fluids and environmental wastewater.

Alternative methods

Detection of protein biomarkers can also be achieved via ELISA or various other biosensing platforms with different transduction mechanisms, such as optical, EC and piezoelectric²⁹. Depending on the application, monitoring can either be on-line or endpoint. On-line monitoring requires continual or repeated operations over extended periods of time, which is a capacity lacking in most of the manual methods including ELISA as well as in most other automated biosensing strategies, making them more suitable for endpoint or short-term assays. Optical biosensors may be alternative tools for on-line monitoring. However, integration of biosensing platforms to organ-on-a-chip systems is another challenge since there are important constraints to take into account such as the minimum sampling volume and hardware/software automation. In this regard, the EC method provides advantages over other transduction mechanisms, because the microelectrodes are easily miniaturized while the design of portable EC workstations provides flexible solutions for integration purposes.

Limitations of the protocol

While the EC affinity-based biosensing chips can be almost universally interfaced with any existing organ-on-a-chip platforms, there are several limitations associated with the technology. In the current design, the regeneration can take up to 4 h, making it difficult to continuously perform biomarker measurements within this time frame using a single chip. Nevertheless, this limitation can be addressed by modifying the breadboard to integrate multiple EC affinity-based biosensing chips that can operate in tandem, that is, when one is under regeneration, the others are still functioning, ensuring continual biomarker monitoring at higher frequencies. Another limitation lies in the fact that the current microfluidic chips are still reliant on polydimethylsiloxane (PDMS), which might exert negative effects on biomarker adsorption and/or absorption. This challenge can be addressed by switching to whole-thermoplastic chips. We have recently reported on the development of whole-thermoplastic microfluidic devices containing various key microfluidic functional units relying also on pneumatic valves, such as normally open valves, normally closed valves, micro-check valves and micropumps^30,31. Additional recent efforts have further revealed the possibility of using 3D printing to produce Quake-style microvalves within thermoplastic-like materials³². The regeneration requires full removal of a layer of the material from the surface of a microelectrode, limiting the lifetime of the device and the number of detections up to 25 regeneration cycles for bare microelectrodes, or ~5 regeneration cycles for the functionalized biosensors. One solution is to start with a thicker microelectrode. In the case of a sudden, substantially elevated level of the target biomarker, the platform can immediately saturate, and the value above the detection range can be missed. In addition, as an affinity-based biosensing platform, we are limited by the number of available antibodies and their corresponding epitopes. Aptamers have become a great alternative²², and we have also described a method to implement them in our system.

Experimental design

Fabrication of microelectrodes

We typically use a three-electrode-configuration microelectrode with reference electrode (RE), counter electrode (CE) and working electrode (WE) as the basis of the affinity-based biosensor (Fig. 1a). The fabrication is done in two steps by electron-beam evaporation through 0.25 mm-thick shadow masks (see Supplementary Data 1 for the designs) for creating Au (WE, CE) and Ag (RE) patterns on a 25 × 25 mm² glass substrate precleaned by oxygen plasma. In the first step, 800 μm-diameter WE and 150 μm-wide CE are patterned in 500 nm-thick Au over a 20 nm-thick Pd, followed by a 20 nm-thick Ti. The second shadow mask enables the deposition of a 500 nm-thick Ag RE patterning over a similar 10 nm-thick Ti/Pd on the same glass substrate. The distance between WE and CE/RE is 200 μm. The microfabricated electrodes are enclosed in a furnace at 300 °C for 6 h to achieve annealing. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) are used for surface characterizations of the microfabricated electrodes (Fig. 1j–l).

Principle of microelectrode functionalization and EC detection

We provide a versatile functionalization strategy that incorporates biorecognition elements for capturing protein biomarkers secreted to the cell culture medium (Fig. 1c,d). Microfabricated Au WE surface is first coated with 11-mercaptoundecanoic acid (11-MUA) and then incubated with N-ethyl-N′-(3-dimethylaminopropyl) carbodiimide (EDC)/N-hydroxysuccinimide (NHS) crosslinker solution to convert the carboxylic end of 11-MUA to an NHS ester for coupling primary amines in streptavidin (SPV)- or amine-terminated aptamers. Upon creating uniform patterns of SPV molecules on the WE surface, the biotinylated capture probe is immobilized via strong biotin-SPV interaction that enables a flow-resistant bioreceptor surface. The selectivity of the biosensor surface is assured by a blocking step where nonspecific binding of nontarget proteins is eliminated through incubation of fresh cell culture medium that is used for the target microtissue culture and is rich in proteins such as bovine serum albumin (BSA) or other components.

Detection and quantification of protein biomarkers interacting with the biotinylated probe relies on a label-free EC method that enables monitoring the change of interfacial electron-transfer kinetics of the redox probe [Fe(CN)₆]^4−/3− upon biorecognition reaction. EC impedance spectroscopy (EIS) is employed to measure the impedance value after each sample incubation. The Randles equivalent circuit model is used to fit the raw EIS data collected over a wide frequency range and is comprised of electrolyte or solution resistance, R_s, between WE and RE, the double layer capacitance, C_dl, and the Faradaic impedance at the electrode–electrolyte interface that is subdivided into the charge transfer resistance, R_ct, and the Warburg impedance, W. The resulting data are presented by Nyquist plots in which real and imaginary parts of impedance are plotted to clearly differentiate charge transfer-controlled and diffusion-controlled processes that are related to semicircular and linear parts of the plots, respectively. Increased capture of antigens on the WE surface upon successive sample incubation results in gradual increase in R_ct values due to shielding of the redox probe by antigens. The increase in R_ct values continues until the point where the WE surface is saturated to the capture of more antigens and needs to be regenerated for further measurements. In addition to impedance measurements, AFM images taken at bare (Fig. 1j), as well as antibody- (Fig. 1k) and aptamer- (Fig. 1l) functionalized, microelectrodes also indicate the successful functionalization of the bioreceptors. According to AFM images, the mean surface roughness of the bare microelectrode (1.79 nm) is increased after antibody (2.75 nm) and aptamer (2.53 nm) functionalization.

Incorporation of different types of capture probes on EC affinity-based biosensors

Our versatile functionalization strategy enables incorporation of bioreceptor molecules in the form of proteins and nucleic acids such as antibodies and aptamers, respectively^21,22. Depending on the selected biomarker target, detection medium or duration of monitoring, the two capturing probes can be selected accordingly (Fig. 1f–h). Aptamers are widely employed when it comes to the matter of high selectivity and stability over time, as well as change in chemical composition, since they are meticulously designed by the systematic evolution of ligands by exponential enrichment (SELEX) method³³ so that they can bind to their targets with very high affinities and remarkable dissociation constants ranging from picomolar to nanomolar. Most of the aptamer-based biosensing platforms employ methylene blue-functionalized aptamers as molecular switches that go through a conformational change upon target binding and induce changes in EC signals. These molecular switch-based platforms hold great potential for monitoring in organ-on-a-chip systems because of their simplicity at the surface functionalization step^34,35. They are, however, not covered in our protocol because of their incompatibility with the generalized procedures described. On the other hand, aptamer synthesis is a tedious, time-consuming process requiring more resources and is sometimes hindered by the limited nucleobase repertoire.

Method for regeneration of EC affinity-based biosensors

Long-term monitoring of protein biomarkers in cell culture medium is achieved via successful regeneration of the biosensor surface upon saturation after several incubations. A two-step EC regeneration process is employed to remove all the molecules immobilized to the biosensor surface during functionalization and biomarker detection (Fig. 1b,e). The biosensor surface is first flushed with 10 mM H₂SO₄ between electrical sweep potentials of 0.0 V and +1.8 V at a scan rate of 100 mV s^–1, a sensitivity of 10^–2 A V^–1 and a 0.001 V sampling interval, then followed by a 50 mM K₃Fe(CN)₆ flow with another potential sweep between +1.0 V and 0.0 V at a scan rate of 200 mV s^–1, a sensitivity of 10^–4 A V^–1 and a sampling interval of 0.001 V. Continuous flows of the respective solutions over the courses of potential sweeps prevent readsorption of the detached molecules to the sensor surface and increase the efficiency of regeneration. Compared to nonflow-regeneration conditions, where the solution flow is stopped during potential sweep, the continuous flow regeneration is proven to create a smoother and more uniform surface that leads to a higher number of regenerations with a higher sensitivity for a longer monitoring period²¹. Another factor that is shown to substantially affect the performance of the regeneration is the thickness of the Au layer deposited on the Pd layer (Fig. 2a,b). Conductivity values of the regenerated and as-fabricated electrodes suggest higher discrepancy for the 150 nm Au layer with respect to the 500 nm Au layer (Fig. 2c), since the thickness (Fig. 2d—f) and the roughness (Fig. 2d,e,g) of the electrodes change through the regeneration cycles when the two-step regeneration protocol is applied. The bare Au electrode with 150 nm thickness could not measure the signals after the ~10th regeneration, while the bare Au electrode with 500 nm thickness could perform up to the 25th cycle of regeneration without substantial loss in the conductivity (Fig. 2c). Sensitivity of the functionalized, affinity-based biosensors seemed stable with ~5 cycles of repeated regeneration (Fig. 2h).

Strategies for microfluidic integration and automation

We provide the design of a microfluidic chip device that contains a microchamber to interface the microelectrode and a set of microchannels with built-in pneumatic valves to achieve digitally controlled liquid routing (Fig. 3a). This microfluidic chip features a three-layer configuration permanently bonded following plasma treatment. While the microfluidic channels in both the top pressure layer and the bottom fluid layer are fabricated using the standard soft lithography approach, the square microfluidic channels in the bottom layer of the chip are further converted to a convex hemicylindrical shape to facilitate complete pneumatic sealing (Fig. 3b)³⁶. Between the two layers, a thin PDMS membrane (20 μm in thickness) is sandwiched, functioning as the pneumatic valve³⁷. During an operational procedure, a pneumatic pressure is applied from the desired microchannel(s) in the valve layer of the device, which then pressurizes the thin membrane at the intersection to deform, blocking the microfluidic channel layer of the chip (Fig. 3c). The pneumatic microchannels are controlled by compressed nitrogen gas connected through polyvinyl chloride (PVC) tubing actuated via a digitized valve-controlling system (https://sites.google.com/site/rafaelsmicrofluidicspage/). The valve-controlling system is operated in full automation using a custom-defined MATLAB program (available via https://shrikezhang.com/publications/opensource; Electrochemical Biosensing, or in ref. ¹³) written to specify the intervals of application and release of the pneumatic pressures by opening/closure of the Festo valve array. Similarly, the fluid movement in the bottom layer of the chip is also controlled automatically using pneumatic pressures. Each inlet of the microfluidic channels is connected to a Teflon tube linked to the outlet of a sealed reagent reservoir (Fig. 3c), where the inlet of the reservoir when pressurized with compressed nitrogen gas drives the reagent to infuse into the fluidic channel of the chip. Coupled operations of pressure release on the reservoir and pressure application on the valve lead to flow cessation in the chosen microfluidic channel; on the contrary, sequential release of the pressure on the valve and application of the pressure on the reservoir restore the reagent flow. In our experiments, the valve control to seal the microchannels usually requires a high pneumatic pressure in the range of 10–20 PSI, while a lower range of pressures (1–5 PSI) is used to drive the reagent flows with desired flow rates.

Breadboard design and modular integration

The breadboard has a similar configuration with the microfluidic chip for microelectrode integration. It comprises three layers that allow for automation control of the valves and the fluid flows (Fig. 4a). The primary difference lies in the arrangement of the microchannels, which should be customized to accommodate the need for the peripheral modules. In the example (Fig. 4b–d), the breadboard features on-board microchannels that link to two bioreactors hosting two different engineered microtissues/organoids, a microfluidic EC affinity-based biosensing chip, a physical/chemical sensing chip and a medium reservoir. This configuration can be conveniently customized depending on the specific side modules to be included, where the breadboard and the side modules can be interfaced using either Teflon tubes or LEGO-type connections. Another primary difference is that, since the culture medium is perfused in this circulation through all modules in a closed loop, a peristaltic pump is utilized to achieve such medium movement rather than the unidirectional reagent injections enabled by pneumatic pressures.

Materials

Reagents

Caution

All reagents mentioned in this protocol are potentially harmful. It is necessary to use protective clothing (lab coats and gloves) throughout the procedure. Recycle or dispose of solid and liquid wastes according to local or institutional regulations

Critical

All reagents should be stored and prepared according to the manufacturers’ recommendations.

Dow SYLGARD 184 silicone encapsulant (Ellsworth, cat. no. 184 SIL ELAST KIT 0.5KG)
NANO SU-8 negative tone photoresists formulations 2050 (MICRO CHEM, cat. no. SU-8 2050)
Flow-mix 5min epoxy (Devcon, cat. no. 20445)
Silver conductive epoxy (MG Chemicals, cat. no. 8330S)
Potassium iodide (KI, Sigma-Aldrich, cat. no. 221945)
11-MUA (Sigma-Aldrich, cat. no. 450561)
SPV (Sigma-Aldrich, cat. no. 85878)
EDC (Sigma-Aldrich, cat. no. E6383)
NHS (Sigma-Aldrich, cat. no. 130672)
Potassium ferricyanide (K₃Fe(CN)₆, Sigma-Aldrich, cat. no. 702587)
Sulfuric acid (H₂SO₄, Sigma-Aldrich, cat. no. 320501)
HCl (37% (vol/vol), Sigma-Aldrich, cat. no. 320331-500ML)
NaOH (reagent grade, ≥98% (wt/vol)
Pellets, anhydrous (Sigma-Aldrich, cat. no.S5881)

Caution

H₂SO₄, HCl and NaOH are corrosive. Personal protective equipment (goggles, face shield, aprons, boots) should be worn when contact is possible.
Ethanol (C₂H₅OH, ≥99.8% (vol/vol) (GC), Sigma-Aldrich, cat. no. 24102)
Phosphate-buffered saline (PBS, Gibco cat. no. 14190-144)
HCM Hepatocyte Culture Medium BulletKit (Lonza, cat. no. CC-3198)
Penicillin–streptomycin (Gibco, cat. no. 15070063)
Roswell Park Memorial Institute (RPMI) 1640 medium (Thermo Fisher Scientific, cat. no. 11875093)
50× B-27 supplement (Thermo Fisher Scientific, cat. no. 17504044)
Albumin biotinylated antibody and full-length protein from Albumin ELISA Kit (Abcam cat. no. ab108788, RRID: AB_2876883)
Glutathione S-transferase-alpha (GST-α) biotinylated antibody and full-length protein from GST-α ELISA Kit (Abcam, cat. no. ab173835, RRID: AB_2876882)
Amine group-functionalized creatine kinase–myocardial band (CK–MB) aptamer (OTC biotech, customized)

Critical

For reproducibility of these results, the customized aptamer was designed to capture CK–MB1, with a 5′-amine.
CK–MB full-length protein from CK–MB ELISA kit (Abcam, ab193696; RRID: AB_2876884)
Acetone (ACS reagent, ≥99.5% (vol/vol), Sigma-Aldrich, cat. no. 179124)
Isopropyl alcohol (natural, ≥98% (vol/vol), FG), Sigma-Aldrich, cat. no. W292912
Deionized (DI) water
Red food coloring (Thermo Fisher Scientific, cat. no. S05371)
Green food coloring (Thermo Fisher Scientific, cat. no. S05376)
Milli-Q water (18 MΩ·cm; Millipore)
Sodium chloride (NaCl; Adamas, cat. no. 81793B)
Square glass substrates (25 × 25 mm²)
Nitrogen, ultra-high purity, 5.0 grade, size 200 cylinder, CGA-580 (Airgas, cat. no. NI UHP200)

Critical

Nitrogen gas is less prone to expansion compared with other gases, and is absent of moisture, assuring more consistent pressure controls. Nitrogen is nonflammable and does not cause rust or decay.
SU-8 developer (MicroChem, cat. no. Y020100)

Caution

Causes skin and serious eye irritation with some possible allergic skin reactions and respiratory irritation. Wear safety glasses, gloves and face protection. Use inside a ventilated chemical fume hood.
Trichloro(1H,1H,2H,2H-perfluorooctyl)silane (PFOTS; Sigma-Aldrich, cat. no. 448931)

Caution

Combustible, causes serious skin burns and eye damage. Wear safety glasses, gloves and face protection. Use inside a ventilated chemical fume hood.

Equipment

General

Automated valve controller (custom-assembled according to https://sites.google.com/site/rafaelsmicrofluidicspage/)
Oven (VWR, cat. no. 89511-406)
SEM (Zeiss, cat. no. EVO 55)
Stereo zoom microscope (Laxco, cat. no. MZS1MZ22)
AFM (Asylum Research, cat. no. Cypher ES)
Spin coater (Laurell, cat. no. WS-650-23B)
Benchtop plasma cleaner (Plasma Etch, cat. no. PE-50)
pH meter (Fisher Scientific, cat. no. 13-636-AB15PC)
High-capacity vacuum pump (Gast, cat. no. DOA-P504-BM)

Microfabrication

Clean room (Class 100)
Double-sided polyimide cleanroom tape with silicone adhesive (Ultratape, cat. no. 2159)
Nitrogen spray gun with hose assembly (Airgas, cat. no. GSEKPN-1)
Rapid thermal processor chamber (AccuThermo, cat. no. AW820)
Shadow masks for microelectrode fabrication (see Supplementary Data 1)
Microfluidic designs (see Supplementary Data 1 and Step 11)
Mask aligner (Karl-Süss, cat. no. MJB-55W)
Dicing saw machine (DISCO, cat. no. DAD3240)
E-beam evaporator (Denton, cat. no. EE-4)
Spin coating workstation (Reynolds Tech, cat. no. SB-3)
Rapid thermal processor (Modular Process Technology, cat. no. RTP-600xp)
Microbore tubing (0.012″ ID × 0.030″ OD; Cole-Parmer, cat. no. EW-06417-11)
Microbore tubing (0.020″ ID × 0.060″ OD; Cole-Parmer, cat. no. EW-06419-01)
Microbore tubing (0.030″ ID × 0.090″ OD; Cole-Parmer, cat. no. EW-06419-03)
Glass microscope slides (Thermo Fisher Scientific, cat. no. 12-550-A3)
Wafer tweezers (VWR, cat. no. OUTI42LB-6)
Scalpel (VWR, cat. no. 95039-117)

Hardware and software

Standard laptop or desktop running Windows 7 Intel core i5 processor and 4 GB RAM or higher
EC workstation (CHI Instruments, cat. no. CHI660e)
MATLAB (MathWorks, https://www.mathworks.com/products/matlab.html)
Custom-made MATLAB program for controlling the valve system (Electrochemical Biosensing; https://shrikezhang.com/publications/opensource)
CH Instruments software (Electrochemical Instrumentation CH Instruments, https://www.chinstruments.com/downloads.shtml)

Reagent setup

Self-assembled monolayer (SAM) solution

To make 10 mM SAM solution, dissolve 109.18 mg of MUA in 50 mL of ethanol. Seal the cap with parafilm to avoid evaporation, and store the solution at 4 °C. The SAM solution can be stored at 4 °C for ~2 months.

Citric buffer solution

To prepare 50 mM citric acid solution, dissolve 0.335 g of sodium citrate and 0.261 g of citric acid in 400 mL of distilled water. Adjust solution to pH 4.5 using 1 M HCl or 1 M NaOH, and fill the solution to a final volume of 500 mL. The citric buffer can be stored at 4 °C for at least 6 months.

EDC/NHS solution

To make equimolar (25 mM) EDC/NHS solution, prepare the 50 mM NHS solution by dissolving 5.75 mg of NHS in 1 mL of 50 mM citric buffer. Then prepare the 50 mM EDC solution by dissolving 9.55 mg of EDC in 1 mL of 50 mM citric buffer and mix the two solutions at a 1:1 ratio. The EDC/NHS solution should be prepared fresh before the experiment.

K₃Fe(CN)₆ stock solution

To make 50 mM K₃Fe(CN)₆ solution, dissolve 823.15 mg of K₃Fe(CN)₆ in 50 mL of DI water. The K₃Fe(CN)₆ solution can be stored at room temperature in the dark for ~2 months.

Caution

K₃Fe(CN)₆ is an irritant. Avoid contact with skin and eyes. Avoid breathing dust. Use in a fume hood.

Critical

K₃Fe(CN)₆ is light sensitive. K₃Fe(CN)₆ solution decomposes slowly on standing. Store in dark or opaque bottles and away from light.

SPV stock solution

To make 10 μg mL^–1 SPV, prepare SPV solution of 1 mg mL^–1 by dissolving 1 mg of SPV in 1 mL of PBS and then dilute to 10 μg mL^–1 by adding 140 μL of the 1 mg mL^–1 SPV solution into 13.88 mL of PBS. Store at –20 °C in 2 mL Eppendorf tubes. The SPV solution can be stored under these conditions for at least 3 months.

Critical

Avoid repeated freeze–thaw cycles.

H₂SO₄ stock solution

To prepare 10 mM H₂SO₄ solution, mix 500 μL of 1 M H₂SO₄ with 49.5 mL of distilled water. The H₂SO₄ solution can be stored in a chemical cabinet at room temperature (RT, 18–24 °C) for at least 6 months.

Caution

Exothermic and hazardous solution. Use in fume hood with protective equipment.

Dilution buffer

To prepare 2 mL of 1× dilution buffer, dilute 200 μL of 10× dilution buffer in the ELISA kit with 1,800 μL of distilled water. The dilution buffer should be prepared fresh before the experiment.

Biotinylated antibody solution

To prepare 10 μg mL^–1 biotinylated antibody solution, dilute the stock solution in PBS. The biotinylated antibody solution should be prepared fresh before the experiment.

Amine group-functionalized aptamer solution

To prepare 10 μg mL^–1 5′-end amine group-functionalized aptamer solution, dilute the stock solution in PBS. The amine group-functionalized aptamer solution should be prepared fresh before the experiment.

Washing buffer

Dilute 10× washing buffer in ELISA kit to 1× with distilled water. The washing buffer should be prepared fresh before the experiment.

Hepatocyte culture medium

Reconstitute according to manufacturer’s instructions. The medium should be prepared in aliquots in 50 or 15 mL conical tubes sealed with parafilm, and can be stored at 4 °C for ~1 month.

Cardiomyocyte culture medium

Add 10% (vol/vol) FBS, 1% (vol/vol) penicillin–streptomycin and 1× B-27 supplement to RPMI 1640. The medium should be prepared in aliquots in 50 or 15 mL conical tubes sealed with parafilm and can be stored at 4 °C for ~1 month.

Common medium

Mix hepatocyte culture medium and cardiomyocyte culture medium at 1:1 volume ratio.

Critical

Preparation of culture medium and common medium should be adjusted according to the cell type(s) used in the project.

PDMS

In a disposable container, load desired amount of elastomer base and curing agent at a 10:1 mass ratio. Using a glass stirring rod, mix the two components thoroughly. Place the container into a vacuum desiccator for 30 min to remove all bubbles. Pour volume according to molding needs. The PDMS can be stored at 4 °C for ~1 month or prepared fresh before each fabrication.

Equipment setup

Setup of the EC cell

In this protocol, all experiments are carried out by the three-electrode system fabricated on the microelectrodes Au WE, Ag RE and Au CE. We have used cyclic voltammetry (CV) for cleaning purposes and EIS for signal measurements. Randles circuit, R_s(C_dl[R_ctW]) with solution resistance, R_s, double-layer capacitance, C_dl, Warburg element, W, and charge transfer resistance, R_ct, are used to fit the EIS data³⁸.

Procedure

Microelectrode fabrication

Timing 5 h

1
Wash square glass substrates in acetone, and dry with compressed air gun in a cleanroom.
2
Attach the first shadow mask (0.25 mm in thickness) for WE and CE to the glass substrates using double-sided tape, and place them in the electron-beam evaporator.
3
Wait until the chamber pressure reaches at least 10^–6 level. This step takes ~45–60 min.
4
Deposit 20 nm Ti layer, 20 nm Pd layer and 500 nm Au layers for both WE and CE (Fig. 1a).
5
Attach the second shadow mask for RE on the Au-deposited glass substrates using double-sided tape, and place in the electron-beam evaporator.

Critical step

Check all the microelectrodes one by one to make sure there is good alignment for all the REs and WEs/CEs.
6
Wait until the chamber pressure reaches at least 10^–6 level. This step takes ~45–60 min.
7
Deposit 20 nm Ti layer, 20 nm Pd layer and 500 nm Ag layers for RE (Fig. 1a).
8
After the deposition step, remove the samples from the electron-beam evaporator.
9
Anneal the fabricated microelectrodes in a nitrogen atmosphere using a rapid thermal processor chamber (300 °C, 2 h).

Caution

Wait until temperature reaches RT before removing the samples.

Critical step

Handle the microelectrodes carefully, making sure they do not scratch.
10
Cut the fabricated microelectrode substrates (25 × 25 mm²) using a dicing saw machine.

Pause point

The microelectrodes can be stored indefinitely at RT away from dust.

Photolithographic fabrication of microfluidic controlling devices

Timing 18 h

Critical

Steps 13–31 should be carried out inside a clean room facility. Dust and light exposure to resin and related materials decrease microfabrication precisions. Exposure of SU-8 to white light causes partial or complete crosslinking, disrupting the resin’s optical transparency.

11
Use AutoCad to design the following masks: (i) the fluidic layer of the breadboard, (ii) the valve layer of the breadboard, (iii) the fluidic layer of the EC detection chip and (iv) the valve layer of the EC detection chip. See Supplementary Data 1 for our example designs.
12
Print the photomasks through a commercial source (e.g., CAD/Art Services, https://www.outputcity.com/). The photomasks are in the form of transparent polymer films, where the patterns are left transparent.
13
Cut the photomasks individually.

Critical step

Handle photomasks with gloves at all times. Skin grease and impurities can diffract light during photolithography.
14
Place a 4″ silicon wafer on the spinner chuck of a spin coater and apply nitrogen vacuum.
15
To clean the wafer surface, spin the wafer at 3,000 r.p.m. for 1 min at room temperature and in sequence apply acetone, methanol and isopropyl alcohol.
16
Remove solvents by placing the wafer on a hot plate at 200 °C for 10 min.
17
Place wafer on a clean and even surface, and allow it to cool for 2 min.

Caution

Handle with wafer tweezers to avoid melting of nitrile gloves and/or skin burns.
18
On the center of the wafer, pour 1 mL of SU-8 2050 photoresist per inch of surface.

Critical step

Pour photoresist slowly and carefully to prevent the formation of air bubbles.
19
With the wafer tweezers, pick up the wafer and place it centered back onto the spin coater. Fix it in place by turning on the vacuum. Let it sit for 5 min.
20
Set up the spin coater and run the wafer according to the following coating conditions:
- Acceleration: 100 r.p.m. s^–1
- Spread: 500 r.p.m. for 15 s
- Acceleration: 200 r.p.m. s^–1
- Spin cycle: 1,700 r.p.m. for 30 s
- Break cycle: 500 r.p.m. for 10 s
21
Carefully transfer the covered wafer into a leveled hot plate at 65 °C for 5 min, for prebake.
22
Cautiously transfer the prebaked wafer into an oven at 95 °C for 20 min, for soft bake.
23
Place the wafer on an even surface at RT to let cool for 10 min.
24
Prepare mask aligner by loading the photomask (either the corresponding valve layer or fluidic layer mask) and the wafer.
25
Set the mask and the wafer into conformal contact, and expose to UV light for 26 s at 9.5 mW cm^–2.
26
Immediately, perform a postexposure bake by placing the exposed wafer on the hot plate at 65 °C for 4 min, and then into an oven for 9 min at 95 °C.
27
Develop the wafer by immersing the wafer in SU-8 developer for 15 min.
28
Pick up the wafer with tweezers, and rinse off undeveloped resin by spraying with SU-8 developer.
29
Stop the developer reaction by spraying the wafer with isopropyl alcohol.
30
Dry the wafer with pressurized, filtered nitrogen gas.
31
In a glass container that can fit the wafer and a small liquid container, place the wafer upside down, held on the edges. In the liquid container, place three drops (~60 μL) of PFOTS. Leave for overnight silanization through chemical vapor deposition.

Caution

Wear safety glasses, gloves and face protection at all times when handling silane. Perform salinization under a ventilated chemical fume hood.

Pause point

Store the silanized master wafers at RT until needed. The master molds can be stored indefinitely away from dust.

PDMS membrane fabrication

Timing 1.5 h

32
Place the lid of a 100 mm Petri dish facing down in the spin-coater vacuum chuck. Pour 1 g of degassed 10:1 PDMS pre-polymer.
33
Set up the spin coater and run the Petri dish according to the following coating conditions:
- Acceleration: 300 r.p.m. s^–1
- Spread: 500 r.p.m. for 10 s
- Acceleration: 500 r.p.m. s^–1
- Spin cycle: 1,000 r.p.m. for 30 s
- Break cycle: 300 r.p.m. for 10 s
34
Cover the surface with the bottom of the Petri dish facing down, to protect the newly generated membrane.

Critical step

When placing the bottom cover and handling the coated lid, avoid any disturbance that might disrupt the PDMS layer.
35
Cure PDMS in an oven at 80 °C for 1 h.

Critical step

Make sure the coated Petri dish is level inside the oven during curing. Avoid stacking the Petri dishes on top of each other.
36
Remove the coated Petri dish from the oven, and allow to cool at RT for 20 min.

Pause point

Store PDMS-coated dishes covered by the bottoms of the Petri dishes. Do not remove the covers from the membranes until ready to use. The PDMS membranes can be stored indefinitely at RT.

Fluidic layer fabrication

Timing 3 h

37
Use pressurized nitrogen to blow dust particles from the silicone master wafer mold of the fluidic layer from Step 31.
38
Using double-sided tape, attach the corresponding silicone master of the fluidic layer to enough aluminum foil to encase it in a circle of 5 mm-high walls.
39
Pour enough degassed 10:1 PDMS pre-polymer onto the wafer to generate a 3 mm-thick PDMS layer.
40
Place the PDMS pre-polymer-covered wafer in a vacuum chamber for 30 min, for further degassing.
41
Transfer the wafer to an oven, and cure PDMS at 80 °C for at least 2 h.

Critical step

Make sure the wafer is level inside the oven during curing.
42
Remove from the oven, and allow it to cool at RT for 20 min.
43
By using a scalpel, carefully remove the negatively casted cured PDMS from the wafer.

Caution

Be careful when using the scalpel, to avoid scratches to the wafer and/or user injury.

Pause point

Store the PDMS-cast fluidic layer, protecting it from dust by placing it inside a clean Petri dish. The PDMS fluidic layer can be stored indefinitely at RT.

Hemicylindrical microchannel fabrication

Timing 16 h

44
Remove the protective lid from a fabricated membrane (from Step 35), and place the uncovered membrane inside the plasma chamber.
45
Remove the PDMS replica of the fluidic layer fabricated in Step 42 from the clean Petri dish. Use a 20 G punch to punch out a single inlet of the corresponding valve layer.

Critical step

Make sure the PDMS ‘worm’ is fully removed and no lacerations occur in the PDMS. This is the port for the generation of a tight negative pressure that will homogeneously deform the thin PDMS membrane, and a smooth and tight grip of the hole is needed (Fig. 3b).
46
Place the negative channels facing up inside the plasma chamber.
47
Treat simultaneously both the PDMS membrane and the PDMS replica with plasma for ~60–90 s.
48
Place the negative feature side of the PDMS replica facing the membrane. Bond the PDMS replica to the membrane immediately following plasma treatment.
49
Place the bonded piece into an oven at 80 °C for another 30 min to secure bonding.

Critical step

Handle all materials with gloves to avoid oil or dust contamination. We also recommend the use of surgical masks to avoid any particle contamination of the surfaces.
50
Place a 127 × 102 mm² glass slide on a spin coater and apply vacuum.
51
On the center of the glass slide, pour 1 mL of SU-8 2050 photoresist per inch of surface.
52
Set up the spin coater, and run the glass slide according to the following coating conditions:
- Acceleration: 100 r.p.m. s^–1
- Spread: 500 r.p.m. for 15 s
- Acceleration: 200 r.p.m. s^–1
- Spin cycle: 1,700 r.p.m. for 30 s
- Break cycle: 500 r.p.m. for 10 s
53
Carefully transfer the SU-8-covered glass slide onto a level surface.

Critical step

When handling the SU-8-covered glass slide, avoid any disturbance that might disrupt the SU-8 layer.
54
Remove the bonded PDMS from the oven, and allow it to cool at RT for 20 min.
55
By using a scalpel, carefully cut the edges of the bonded PDMS membrane and replica from the Petri dish lid.

Caution

Be careful when using the scalpel, to avoid scratches to the PDMS or user injury.
56
Insert a 27 G metal connector into the inlet on the featureless side of the slab.
57
Connect the metal connector into Teflon tubing, and further extend the connection to a vacuum pump.
58
Set vacuum at 10 mmHg, and apply negative pressure by turning on the vacuum.

Critical step

At this point, all the negative microchannels should have the membrane deformed into the grooves (Fig. 3b). Make sure all tubing and connectors are tightly sealed into the PDMS to avoid uneven distribution of vacuum in the microchannels.

Troubleshooting
59
With the pump on, place the edge of the PDMS-bonded slab in a 45° contact with the SU-8 cover glass slide. Make sure the negatively bent membrane features are facing the SU-8.
60
In a steady and single movement, place the slab in full contact with the SU-8. Let the arrangement sit for 10 min, allowing SU-8 to fill the entirety of the grooves.

Critical step

Avoid the formation of bubbles or the application of forces that will spread the SU-8 unevenly.
61
Place the PDMS/SU-8 arrangement under UV light at 850 mW for 3 min at a 5 mm distance from the glass side.
62
Remove the arrangement from the UV light. Turn off the vacuum, and carefully remove the PDMS slab from the baked SU-8 positive mold.
63
In a glass container that can fit the secondary mold with hemicylindrical features and a small liquid container, place the mold upside down, held on the edges. In the liquid container, place three drops (~60 μL) of PFOTS. Leave for overnight silanization through chemical vapor deposition.

Caution

Wear safety glasses, gloves and face protection at all times when handling silane. Perform salinization under a ventilated chemical fume hood.
64
Repeat Steps 36–42 to generate the PDMS replica of the fluidic layer containing hemicylindrical microchannels.
65
Use a scalpel to cut off the out-of-plane edges of the fluidic layer slab to facilitate its bonding with the valve layer.

Pause point

Store the PDMS-cast valve layer containing hemicylindrical microchannels, protecting it from dust by placing it inside a clean Petri dish. The PDMS fluidic layer with hemicylindrical microchannels can be stored indefinitely at RT.

EC detection chip assembly

Timing 20 h

Critical step

Steps 65–86 should be carried out utilizing gloves to handle all materials to avoid oil or dust contamination. We also recommend the use of surgical masks to avoid any particle contamination of the surfaces.

66
Use pressurized nitrogen to blow dust particles away from the silicone master wafer mold for the valve layer of the EC detection chip.
67
Repeat Steps 36–42 to generate the PDMS replica of the valve layer.
68
Clean the cured PDMS replica of the valve layer from dust and impurities by firmly placing Scotch tape on its surface, and in one single and rapid movement peel it off.
69
Use a 20G punch to punch out the inlets of the valve layer.

Critical step

Make sure the PDMS ‘worms’ are fully removed and no lacerations occur in the PDMS. These are the pressure ports of the system, and a smooth and tight grip of the holes is needed.

Caution

When using the punch, be careful to avoid damage to the PDMS or user injury.
70
Place the PDMS valve layer replica into the plasma chamber, with the microchannels facing up.
71
Remove the protective lid from a fabricated PDMS membrane, and place the uncovered membrane inside the plasma chamber.
72
Treat both PDMS membrane and the replica simultaneously with nitrogen plasma for 60–90 s.
73
Bond the PDMS replica to the membrane, by placing the feature side of the PDMS slab in contact with the membrane immediately after plasma treatment.
74
Place the bonded piece into an oven at 80 °C for 30 min to further secure bonding.
75
Remove the bonded PDMS from the oven, and allow it to cool at RT for 20 min.
76
By using a scalpel, carefully cut the edges of the bonded membrane and PDMS slab, and peel the piece from the Petri dish lid.

Critical step

Remove the bonded membrane slowly and carefully to avoid fissure or wreckage of the membrane.
77
With a scalpel, evenly cut off the edges of the valve layer slab to facilitate its bonding with the fluidic layer.

Caution

Be careful when using the scalpel, to avoid scratches to the PDMS or self-harm.
78
Use a 20G punch to punch out the inlets and outlets through the PDMS replica of the valve layer, in marked positions (red dots in Fig. 3a) corresponding to the microchannel inlets in the fluidic layer.

Critical step

Make sure the PDMS ‘worms’ are fully removed and no lacerations occur in the PDMS. These are the flow ports of the system, and a smooth and tight grip of the holes is needed.
79
Use a 3 mm punch to punch out the detection chambers through a PDMS replica of the fluidic layer, prepared in Step 64 (Fig. 3a).

Caution

Be careful when using the punch, to avoid scratches to the PDMS or user injury.
80
Clean the PDMS replica of the fluidic layer and the valve layer from dust and impurities by firmly placing Scotch tape on the surface, and in one single and rapid movement peeling it off.
81
Place both the PDMS replica of the fluidic layer and the membrane-bond valve layer with feature sides facing up in the plasma chamber.
82
Treat both layers with nitrogen plasma simultaneously for 60–90 s.
83
Remove the plasma-activated PDMS slabs, and fully cover both with 200 proof ethanol.
84
Place the fluidic layer with hemicylindrical channels on top of the membrane-bond valve layer, by facing the negative feature side of the fluidic layer with the membrane side of the valve layer (Fig. 3a).

Critical step

Although the surface-active groups on PDMS can be maintained in ethanol for extended periods of time, perform the bonding as soon as possible to avoid any possible deterioration of the functional groups.
85
With the help of a stereoscope, as needed, align all the valve points with their corresponding microchannels.

Critical

When the two layers are aligned, make sure to hold them for enough time to allow evaporation of the excess ethanol, avoiding additional sliding between the two layers.
86
Carefully transfer the aligned chip assembly into a Petri dish with a Kimwipe placed at the bottom, and cover with the lid. Leave the assembly at RT until complete evaporation of ethanol. Then, place it into an oven at 80 °C overnight.
87
Remove the bonded PDMS from the oven, and allow it to cool at RT for 20 min.

Pause point

Store and protect the EC detection chips from dust by placing them inside a Petri dish until use. The EC detection chips can be stored indefinitely at RT.

EC detection chip and sensor incorporation

Timing 1 h

Critical step

Steps 87–95 should be carried out utilizing gloves to handle all materials to avoid oil or dust contamination. We also recommend the use of surgical masks to avoid any particle contamination of the surfaces.

88
Remove an assembled EC detection chip from the Petri dish. Clean dust and impurities from the surface using Scotch tape.
89
Clean three microelectrodes from dust and impurities by blowing with a compressed nitrogen gun.
90
Place the microelectrodes facing upwards, and the EC detection chip with bottom facing up, into the plasma chamber.

Critical step

The open detection chambers in the fluidic layer of the PDMS chip should face up.
91
Treat the microelectrodes and the chip simultaneously with the nitrogen plasma for 30–60 s.

Critical step

Oxidation of the RE, manifested by turning black, can cause corrosive resistance. We recommend terminating the microelectrode if this happens.
92
Bond the microelectrodes to the PDMS-based EC sensing chip, with the detection areas aligning with the corresponding chambers (Fig. 1b & 3a).
93
Use silver epoxy to ensure electrical contact in connecting the three leads of each microelectrode and individual copper wires. Place in an oven at 80 °C for 30 min.

Critical step

After curing, check electrical connections by a multimeter.
94
Use 5 min epoxy to further secure the connection points between the copper wires and the microelectrode leads to prevent any breakage during usage.
95
Use 5 min epoxy to cover also the connection between the microelectrodes and the PDMS chip to prevent any leakage during usage.
96
Insert Teflon tubing into the valve and fluidic layer inlets and outlets for gas and liquid flows. With 5-min epoxy, secure all tubing/PDMS connections to prevent any leakage during usage.

Caution

Use personal protective equipment throughout the procedure. Epoxy can be an irritant to skin and damaging to clothes.

Critical step

Use a P-10 microtip to place the epoxy around the tubing, to avoid disrupting the tubing or introducing extra epoxy.

Pause point

Place assembled EC detection chips individually in Petri dishes and keep away from dust at RT until ready for integration. The assembled EC detection chips can be stored indefinitely under these conditions, avoiding humidity.

Reservoir fabrication

Timing 1.5 h

97
With an 18G needle, punch two holes into the lid of a cryogenic vial, or a conical tube, depending on the amount of reagent needed.
98
Use a scalpel to cut enough Teflon tubing.

Caution

Be careful when using the scalpel and needle, to avoid vial scratching or user injury.
99
Through one of the punched holes, insert one Teflon tube all the way into the bottom of the vial (outlet).
100
Through the second hole, partially insert a second Teflon tube to just underneath the lid (inlet) (Fig. 3c-ii).
101
Use 5 min epoxy to fix and seal the tubing in place on the vial lid.
102
Place the vial in a vertical position in an assembled reservoir holder. Let epoxy completely cure.
103
Inside a laminar flow hood, wash both interiors and exteriors of the reservoir and the tubing using 70% (vol/vol) ethanol followed by rinsing with sterile PBS.
104
Place the reservoir under UV light, and further sterilize for 30 min.

Pause point

The sterile reservoirs can be stored at RT indefinitely until ready for integration. If the reservoirs are used for reagents and have no direct contact with cell culture, they might not need to be sterilized.

Breadboard assembly

Timing 20 h

Critical step

Steps 104–108 should be carried out utilizing gloves to handle all materials to avoid oil or dust contamination. We also recommend the use of surgical masks to avoid any particle contamination of the surfaces.

105
Use pressurized nitrogen to blow dust particles away from the silicone master molds of the breadboard valve layer and hemicylindrical flow layer.
106
Repeat Steps 36–42 to generate the PDMS replicas of both layers.
107
Use a 20G punch to punch out the inlets of the valve layer, and the inlets and outlets of the flow layer.

Critical step

Make sure the PDMS ‘worms’ are fully removed and no lacerations occur in the PDMS. These are the ports for gas and fluid flows, and a smooth and tight grip of the holes is needed.

Caution

Be careful when using the needle, to avoid damage to the PDMS or user injury.
108
Repeat Steps 69–86 to assemble the membrane and the breadboard, except for Step 78, as the breadboard has no detection chambers (Fig. 4a).
109
Repeat Step 95 to connect the tubing to all inlets and outlets.

Pause point

Place assembled breadboards individually in Petri dishes, and keep away from dust at RT until ready for integration. The assembled breadboards can be stored indefinitely under these conditions.

Automation operational system preparation

Timing 1.5 h

110
Turn on the nitrogen gas flows and regulators for valve and fluid controls.
111
Turn on WAGO controller for Festo solenoid valves, and connect the system to the computer running the custom-coded MATLAB program (https://shrikezhang.com/publications/opensource or ref. ¹³).

Critical step

Steps 111–120 should be carried out utilizing gloves to handle all materials to avoid oil, dust or biological contamination. We also recommend the use of surgical masks to avoid any further contamination.
112
Remove one EC sensing chip and one breadboard from their respective Petri dishes.
113
Carefully fill tubing connecting to all the valves with food dye prepared in water. Food dye is not a necessity, although it will aid the visualization.
114
Connect the solenoid system to gas flow control. Individually connect each solenoid to its corresponding Teflon tubing linked to the valves on the chips, through metal connectors.
115
Allow for the valve microchannels on the chips to be filled by opening the solenoids and applying 5–15 PSI of gas pressure for ~30 min.

Caution

Constantly monitor each valve as it gets filled, since overly high pressures can burst the membrane layers.

Troubleshooting
116
On the EC sensing chip, leave the solenoids open to keep the valves on the chip closed to seal the flow microchannels (Fig. 3c-iii).
117
Fill the reservoirs with their corresponding reagents (Fig. 3d–f).
118
Individually connect the remaining solenoids to their corresponding reservoirs through the inlet tubing.
119
Set the pneumatic pressure at 1–5 PSI to drive each specific reagent through the reservoir’s outlet tubing (Fig. 3c-ii).

Critical step

Constantly monitor each reservoir as the solution flows through the outlet tubing. Stop each solenoid as the corresponding solution reaches the end of the tubing (Fig. 3c-iv). Steps 117–118 can also be done manually as needed.
120
Connect each outlet tubing on the reservoir to its corresponding flow layer inlet on the chip through a metal connector.
121
Turn the solenoids off to open the valves on the chip, and reinitiate the pneumatic pressures driving the reagents into their corresponding flow microchannels, until all the microchannels are filled with their desired reagents and are bubble free.

Critical step

Each flow microchannel should be individually filled and monitored as it gets filled, to avoid contamination of reagents across the microchannels. Also, turning off the solenoids controlling the valves on the chip and initiating the pneumatic pressures driving the reagent flows can be programmed a few seconds apart in the MATLAB codes, to avoid abrupt changes in pressures during operations that might disturb the flows.

Pause point

The EC sensing chip and breadboard are now ready to integrate with the microbioreactors. Ideally these individual components should be used immediately following preparation, but if necessary they can be stored in a parafilm-sealed humidified box at 4 °C for up to 1 month.

Microbioreactor and breadboard fluidic integration

Timing 1.5 h

Critical step

Steps 121–136 should be carried out utilizing gloves to handle all materials to avoid biological contamination. We also recommend the use of surgical masks to avoid any further contamination.

122
Insert the inlet tubing of a sterile reservoir into an MP² peripump PVC tube.
123
Place the MP² peripump PVC tubing into one of the grooves of an ESI MP² peristaltic pump.
124
Connect the outlet tubing of the reservoir with a metal connector into tubing inserted into the medium inlet on the breadboard (Fig. 4b–d).
125
Fill the reservoir with desired amount of desired medium (e.g., 5 mL), and tightly seal it.
126
Set up the peristaltic pump at desired flow rate (e.g., 200 μL h^–1). The increased pressure inside the reservoir will be released by the flow of medium through the reservoir outlet tubing into the flow microchannel of the breadboard.

Critical step

Medium should flow through the tubing and microchannels smoothly with no back flow.

Troubleshooting
127
Stop the peristaltic pump once medium reaches end of the tubing inserted in the organ-1 inlet on the breadboard.
128
Remove one assembled microbioreactor from the incubator.
129
Connect the tubing coming from organ-1 inlet to the inlet tubing of the microbioreactor through a metal connector (Fig. 4b–d).
130
Connect the outlet tubing of the microbioreactor to the organ-1 outlet tubing on the breadboard.
131
Start the micropump again.

Critical step

Medium should flow through the tubing, microbioreactor and microchannels smoothly with no back flow.
132
Stop the peristaltic pump once medium reaches the end of the tubing inserted in the EC sensor inlet on the breadboard.
133
Gather one assembled EC sensing chip prepared in Step 120.
134
Connect the tubing inserted at the microbioreactor inlet in the EC sensing chip (Fig. 3d–f) to the corresponding tubing coming from the EC sensor-inlet on the breadboard with a metal connector (Fig. 4b–d). Repeat with the outlet tubing.
135
Re-start the peristaltic pump.

Critical step

All the valves of the EC sensing chip should remain closed except for the two controlling the microbioreactor flow in/out. Medium should flow smoothly through the tubing and the desired microchannels in the EC detection chip and microbioreactors with no back flow.
136
Repeat Steps 126–130 to connect the second microbioreactor through the organ-2 inlet/outlet to the breadboard flow.
137
Connect the outlet tubing of the medium reservoir to the free side of the MP² peripump PVC tubing with a metal connector.

Pause point

The EC sensing chip, breadboard and bioreactors are now sharing a constant medium stream, and the system is now ready to run EC functionalization and detection of the released biomarkers. Once assembled, the integral system should be used immediately.

Functionalization of the microfluidic EC sensing chip

Timing 6 h

138
Flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.

139

Perform the first step of EC cleaning by flowing the 10 mM H₂SO₄ solution for 7 min at 1.0 mL h^–1 followed by additional 7 min flow under the CV potential sweep with the CV parameters of:

Parameter	Value
Initial E (V)	0.0
High E (V)	1.8
Low E (V)	0.0
Final E (V)	0.0
Initial scan polarity	Positive
Scan rate (V s^–1)	0.1
Sweep segments	8
Sample interval (V)	0.001
Quiet time (s)	2
Sensitivity (A V^–1)	1.0 × 10^–0.002
Communication port speed	Fast
Line frequency	60 Hz
Potential axis	Positive left

Critical step

Do not stop the H₂SO₄ flow during EC cleaning.

140
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.

141

Perform the second step of EC cleaning by flowing the 50 mM K₃Fe(CN)₆ solution for 7 min at 1.0 mL h^–1 followed by additional 7 min flow under the CV potential sweep with the CV parameters of:

Parameter	Value
Initial E (V)	1.0
High E (V)	1.2
Low E (V)	–1.2
Final E (V)	0.0
Initial scan polarity	Negative
Scan rate (V s^–1)	0.2
Sweep segments	6
Sample interval (V)	0.002
Quiet time (s)	0
Sensitivity (A V^–1)	1.0 × 10^–0.004
Communication port speed	Fast
Line frequency	60 Hz
Potential axis	Positive left

Critical step

Do not stop the K₃Fe(CN)₆ flow during EC cleaning.

142

Measure the bare microelectrode signal via CV with the parameters as follows:

Parameter	Value
Initial E (V)	0.0
High E (V)	0.6
Low E (V)	–0.6
Final E (V)	0.0
Initial scan polarity	Negative
Scan rate (V s^–1)	0.075
Sweep segments	6
Sample interval (V)	0.001
Quiet time (s)	0
Sensitivity (A V^–1)	1.0 × 10^–0.004
Communication port speed	Fast
Line frequency	60 Hz
Potential axis	Positive left

Critical step

Stop the K₃Fe(CN)₆ flow, and close the inlet and outlet valves of the detection chamber before starting the CV procedure.

143
Measure the bare microelectrode signal via EIS with the parameters as follows:

Parameter
Value
Initial E (V)
0.0
High frequency (Hz)
100,000
Low frequency (Hz)
0.1
Amplitude (V)
0.05
Quiet time (s)
0
Sensitivity (A V^–1)
1.0 × 10^–0.004
Bias DC current during run
<1 Hz
144
Open the ethanol valve, and flow absolute ethanol for 7 min at 1 mL h^–1.

Critical step

Bubble formation might occur in this step because of the change in polarity of reagents from K₃Fe(CN)₆ solution to ethanol flowing through the main microchannel. If bubble formation occurs, increase the flow time for ethanol until bubbles disappear.
145
Open the SAM valve, and flow 10 mM SAM for 7 min at 1 mL h^–1.

Critical step

Close all other valves except the SAM valve to ensure that the detection chamber is only filled with SAM without any possible contamination by other reagents. While this is true for all operations in this section, it is particularly important for the SAM functionalization step.
146
Incubate SAM for 1 h at RT.
147
Open the ethanol valve, and flow absolute ethanol for 7 min at 1 mL h^–1.
148
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.

Critical step

Bubble formation might occur in this step because of the change in polarity of reagents from ethanol to PBS flowing through the main microchannel. If bubble formation occurs, increase the flow time for PBS until bubbles disappear.
149
Open the EDC/NHS valve, and flow 25 mM equimolar solution of EDC/NHS for 7 min at 1.0 mL h^–1.
150
Incubate EDC/NHS solution for 30 min at RT.
151
Open the SPV valve, and flow 10 μg mL^–1 SPV for 7 min at 1.0 mL h^–1.
152
Incubate SPV for 1 h at RT.
153
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.
154
Open the K₃Fe(CN)₆ valve, and flow 50 mM K₃Fe(CN)₆ solution for 7 min at 1.0 mL h^–1.
155
Measure the SPV-modified microelectrode signal via EIS with the parameters as listed in Step 142.

Critical step

Stop the K₃Fe(CN)₆ flow, and close inlet and outlet valves of the detection chamber before starting the EIS procedure.

Troubleshooting
156
Open the bioreceptor valve, and flow the bioreceptor solution for 7 min at 1.0 mL h^–1.
157
Incubate bioreceptor for 1 h at RT.

Critical step

There are two possible strategies for functionalization of the microelectrode surface for biorecognition:
- Aptamer-based functionalization: open aptamer bioreceptor solution valve, and flow 10 μg mL^–1 of amine-functionalized aptamer solution for 7 min at 1.0 mL h^–1. We suggest using this strategy for longer shelf life and sensitivity, if desired aptamers are available.
- Antibody-based functionalization: open antibody solution valve, and flow 10 μg mL^–1 of biotinylated antibody solution for 7 min at 1.0 mL h^–1.
The antibodies and aptamers presented in Table 1, with their corresponding concentrations and incubation times listed, are examples from prior studies with hepatic and cardiac microtissue-on-a-chip cultures^13,21,22. In general, functionalization with antibodies or aptamers of interest (concentrations and times) needs to be optimized in a case-by-case manner.
158
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.
159
Open the K₃Fe(CN)₆ valve, and flow 50 mM K₃Fe(CN)₆ solution for 7 min at 1.0 mL h^–1.
160
Measure the bioreceptor-modified microelectrode signal via EIS with the parameters listed in Step 142.

Critical step

Stop the K₃Fe(CN)₆ flow, and close inlet and outlet valves of the detection chamber before starting the EIS procedure.

Troubleshooting
161
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.
162
Open the medium valve, and flow cell culture medium for 7 min at 1.0 mL h^–1. Exemplary medium types are provided in the Reagents list and their reconstitution/storage in Reagent Setup. In reality, the medium should match that used for the specific cell culture of desire.
163
Incubate cell culture medium for 1 h at RT to ensure blocking of the microelectrode surface to eliminate nonspecific binding.

Critical step

During blocking step via incubation of cell culture medium, close inlet and outlet valves of detection chamber.
164
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.
165
Open the K₃Fe(CN)₆ valve, and flow 50 mM K₃Fe(CN)₆ solution for 7 min at 1.0 mL h^–1.
166
Measure the microelectrode signal after blocking via EIS with the parameters listed in Step 142.

Critical step

Stop the K₃Fe(CN)₆ flow, and close inlet and outlet valves of the detection chamber before starting the EIS procedure.
167
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.

Pause point

Leave the detection chamber filled with PBS. The functionalized EC sensing chip is ready for biomarker sampling and quantification (Fig. 1f–i). The functionalized EC sensing chip can be immediately used (if functionalized together with the integrated platform), or can be stored in a parafilm-sealed humidified chamber at 4 °C for up to 1 month (if functionalized separately from the integrated platform).

Table 1 Exemplary on-chip microelectrode functionalization procedures for EC measurements and differences between those using aptamers and antibodies

Full size table

On-chip medium sampling and biomarker quantification

Timing 1.5 h

168
Open bioreactor sampling valves while shutting down the medium return valve to allow sampling for 5–10 min at the medium flow rate.
169
Incubate the bioreactor medium sample for 1 h at RT.

Critical step

During incubation of the bioreactor medium sample, close inlet and outlet valves of the detection chamber to avoid any potential interference.
170
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.
171
Open the K₃Fe(CN)₆ valve, and flow 50 mM K₃Fe(CN)₆ solution for 7 min at 1.0 mL h^–1.
172
Measure the captured biomarker signal via EIS with the parameters as listed in Step 142.

Critical step

Stop the K₃Fe(CN)₆ flow, and close inlet and outlet valves of the detection chamber before starting the EIS procedure.

Troubleshooting
173
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.
174
Repeat Steps 167–171 for repeated biomarker measurements until the biosensor surface saturates.

Troubleshooting
175
Regenerate the biosensor surface by a two-step cleaning procedure. Follow Steps 175–177 listed below.

On-chip EC affinity-based biosensor regeneration

Timing 40 min

176

Perform the first step of EC cleaning by opening the H₂SO₄ valve and flowing the 10 mM H₂SO₄ solution for 7 min at 1.0 mL h^–1, followed by an additional 7 min flow under the CV potential sweep with the CV parameters of:

Parameter	Value
Initial E (V)	0.0
High E (V)	1.8
Low E (V)	0.0
Final E (V)	0.0
Initial scan polarity	Positive
Scan rate (V s^–1)	0.1
Sweep segments	8
Sample interval (V)	0.001
Quiet time (sec)	2
Sensitivity (A V^–1)	1.0 × 10^–0.002
Communication port speed	Fast
Line frequency	60 Hz
Potential axis	Positive left

Critical step

Do not stop the H₂SO₄ flow during EC cleaning.

177
Open the PBS valve, and flow PBS from the main microchannel for 7 min at 1.0 mL h^–1.

178

Perform the second step of EC cleaning by opening the K₃Fe(CN)₆ valve and flowing the 50 mM K₃Fe(CN)₆ solution for 7 min at 1.0 mL h^–1, followed by additional 7 min flow under the CV potential sweep with the CV parameters of:

Parameter	Value
Initial E (V)	1.0
High E (V)	1.2
Low E (V)	–1.2
Final E (V)	0.0
Initial scan polarity	Negative
Scan rate (V s^–1)	0.2
Sweep segments	6
Sample interval (V)	0.002
Quiet time (sec)	0
Sensitivity (A V^–1)	1.0 × 10^–0.004
Communication port speed	Fast
Line frequency	60 Hz
Potential axis	Positive left

Critical step

Do not stop the K₃Fe(CN)₆ flow during EC cleaning.

179

Measure the bare microelectrode signal via CV with the parameters as follows:

Parameter	Value
Initial E (V)	0.0
High E (V)	0.6
Low E (V)	–0.6
Final E (V)	0.0
Initial scan polarity	Negative
Scan rate (V s^–1)	0.075
Sweep segments	6
Sample interval (V)	0.001
Quiet time (sec)	0
Sensitivity (A V^–1)	1.0 × 10^–0.004
Communication port speed	Fast
Line frequency	60 Hz
Potential axis	Positive left

Critical step

Stop the K₃Fe(CN)₆ flow, and close inlet and outlet valves of the detection chamber before starting the CV procedure.

180
Measure the bare microelectrode signal via EIS with the parameters as follows:

Parameter
Value
Initial E (V)
0.0
High frequency (Hz)
100,000
Low frequency (Hz)
0.1
Amplitude (V)
0.05
Quiet time (sec)
0
Sensitivity (A V^–1)
1.0 × 10^–0.004
Bias DC current during run
Below 1 Hz
181
Once regenerated, refunctionalization of the microelectrode for biomarker detections can be performed by iterating Steps 143–174.

Critical step

Regenerations and detections will be limited by the thickness of the microelectrodes (Fig. 2) and the total volumes of detection solutions available. For continued biosensing when EC sensing chips can no longer be used, they can be modularly replaced, while the breadboard and microbioreactors can carry on working.

Troubleshooting

Troubleshooting

Troubleshooting advice can be found in Table 2.

Table 2 Troubleshooting table

Full size table

Timing

Steps 1–10, Microelectrode fabrication: 5 h
Steps 11–31, Photolithographic fabrication of microfluidic controlling devices: 18 h
Steps 31–35, PDMS membrane fabrication: 1.5 h
Steps 36–42, Fluidic layer fabrication: 3 h
Steps 43–64, Hemicylindrical channels fabrication: 16 h
Steps 65–86, EC detection chip assembly: 20 h
Steps 87–95, EC detection chip and sensor incorporation: 1 h
Steps 96–103, Reservoir fabrication: 1.5 h
Steps 104–108, Breadboard assembly: 20 h
Steps 109–120, Automation operational system preparation: 1.5 h
Steps 121–136, Microbioreactors and breadboard fluidics integration: 1.5 h
Steps 137–166, Functionalization of the microfluidic EC sensing chip: 6 h
Steps 167–174, On-chip medium sampling and biomarker quantification: 1.5 h
Steps 175–180, On-chip affinity-based biosensor regeneration: 40 min

Anticipated results

Integration of aptamer-based and antibody-based microfluidic EC affinity-based biosensors into organ-on-a-chip platforms enables quantification of protein biomarker secretion and, hence, monitoring not only of the effect of drug toxicity but also of how cell fate and viability are affected by various parameters in the cell culture microenvironment. Thanks to specific functionalization of the microelectrode surface and the two-step regeneration strategy, nanomolar-range multiplexed detection of biomarkers can be achieved for lengthy periods with high selectivity (Fig. 5)^13,15. The label-free detection strategy based on charge transfer resistance measurement is versatile and could be applied to all biomarkers, providing that commercial bioreceptor molecules exist, or that aptamers can be synthesized via the SELEX process. After successful functionalization of the microelectrode, R_ct values increase by ~2 orders of magnitude and continue to increase after each successive step (Fig. 1c,d,f,h) until the electrode surface is saturated, where the two-step regeneration is applied to revert back to the bare microelectrode signal (Fig. 1e, Fig. 2). The cell culture medium blocking step is an essential step for the calibration of the affinity-based biosensors and, depending on the medium composition, that is, the amount and type of soluble proteins and other biological agents, different R_ct values may be obtained. Hence, normalization of the measured R_ct values for target biomarkers with respect to the R_ct recorded in the medium blocking step is necessary.

**Fig. 5: Examples of EC affinity-based biosensor detection capabilities in multi-organ-on-a-chip configurations.**

In a few practical examples, we have previously shown that our EC affinity-based biosensors could be conveniently integrated with dual-organ-on-chips platforms while showing high selectivity and no crosstalk for biomarker detection (Fig. 5a–c). Pharmaceutical compound-induced chronic hepatotoxicity and minimum impact on cardiac functions were readily picked up in a human heart-liver-on-chips configuration (Fig. 5d–g). Similarly, anticancer drug treatment efficacy on and simultaneous side cardiotoxicity were successfully monitored in a human heart–liver cancer-on-chips configuration (Fig. 5h–j). It is worth noting that, due to the noninvasiveness of our EC affinity-based biosensing mechanism, the data shown in Fig. 5e–g,i–k were based on their respective individual dual-organ platforms, continually monitored by dynamically sampling the media at predefined time points. This ability to monitor an individual organ pair over extended periods is critical, especially for future personalized medicine, since with conventional endpoint assays average values will have to be taken from multiple samples that might either bias the results or require larger sample numbers that are not always feasible. Nevertheless, results derived from parallel systems might still be averaged to understand the collective statistical effects of the pharmaceutical compounds applied to the organ models, as also shown in our previous work¹³.

Antibody- and aptamer-based functionalization strategies have their own advantages and disadvantages. Aptamers provide better sensitivity than antibodies in complex biological media containing large amounts of nonspecific biomolecules since they do not bind to serum proteins and have higher binding affinities to their target molecules. Moreover, aptamers enable longer shelf lives than antibodies because of their stability at varying temperatures, while antibodies denature at elevated temperatures. However, the development of specific aptamers for a wide range of targets still represents a challenge, whereas antibody production is a routine protocol and thus has much better availability. In addition, aptamers are prone to degradation of nucleases if not modified, which might hamper the use of aptamers for detection in blood or media containing nucleases^39,40.

Reporting Summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

Most data associated with this protocol can be found in previous publications^13,21,22. Additional datasets that support this protocol are available from the corresponding authors upon reasonable request. All requests for raw and analyzed data and materials will be promptly reviewed by the Brigham and Women’s Hospital to verify whether the request is subject to any intellectual property or confidentiality obligations. Any data and materials that can be shared will be released via a Material Transfer Agreement.

Code availability

The MATLAB codes can be accessed through previous publications^13,21,22 and the Zhang Lab website (Electrochemical Biosensing; https://shrikezhang.com/publications/opensource). The running of these codes is standardized, and additional instructions can be found in the readme file in the package.

References

Ahadian, S. et al. Organ-on-a-chip platforms: a convergence of advanced materials, cells, and microscale technologies. Adv. Healthc. Mater. https://doi.org/10.1002/adhm.201800734 (2018).
Wang, Y. I., Carmona, C., Hickman, J. J. & Shuler, M. L. Multiorgan microphysiological systems for drug development: strategies, advances, and challenges. Adv. Healthc. Mater. 7, 1701000 (2018).
Article Google Scholar
Ronaldson-Bouchard, K. & Vunjak-Novakovic, G. Organs-on-a-chip: a fast track for engineered human tissues in drug development. Cell Stem Cell 22, 310–324 (2018).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y. S. et al. From cardiac tissue engineering to heart-on-a-chip: beating challenges. Biomed. Mater. 10, 034006 (2015).
Article PubMed PubMed Central Google Scholar
Arrigoni, C., Lopa, S., Candrian, C. & Moretti, M. Organs-on-a-chip as model systems for multifactorial musculoskeletal diseases. Curr. Opin. Biotechnol. 63, 79–88 (2020).
Article CAS PubMed Google Scholar
Mobini, S., Song, Y. H., McCrary, M. W. & Schmidt, C. E. Advances in ex vivo models and lab-on-a-chip devices for neural tissue engineering. Biomaterials 198, 146–166 (2019).
Article CAS PubMed Google Scholar
Bein, A. et al. Microfluidic organ-on-a-chip models of human intestine. Cell. Mol. Gastroenterol. Hepatol. 5, 659–668 (2018).
Article PubMed PubMed Central Google Scholar
Ashammakhi, N., Wesseling-Perry, K., Hasan, A., Elkhammas, E. & Zhang, Y. S. Kidney-on-a-chip: untapped opportunities. Kidney Int. 94, 1073–1086 (2018).
Article CAS PubMed PubMed Central Google Scholar
Conant, G. et al. High-content assessment of cardiac function using heart-on-a-chip devices as drug screening model. Stem Cell Rev. Rep. 13, 335–346 (2017).
Article PubMed Google Scholar
Ribas, J. et al. Biomechanical strain exacerbates inflammation on a progeria-on-a-chip model. Small https://doi.org/10.1002/smll.201603737 (2017).
Zhang, Y. S., Zhang, Y. N. & Zhang, W. Cancer-on-a-chip systems at the frontier of nanomedicine. Drug Discov. Today 22, 1392–1399 (2017).
Article CAS PubMed PubMed Central Google Scholar
Skardal, A., Shupe, T. & Atala, A. Organoid-on-a-chip and body-on-a-chip systems for drug screening and disease modeling. Drug Discov. Today 21, 1399–1411 (2016).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y. S. et al. Multisensor-integrated organs-on-chips platform for automated and continual in situ monitoring of organoid behaviors. Proc. Natl Acad. Sci. USA 114, E2293–E2302 (2017).
Article CAS PubMed PubMed Central Google Scholar
Riahi, R. et al. Automated microfluidic platform of bead-based electrochemical immunosensor integrated with bioreactor for continual monitoring of cell secreted biomarkers. Sci. Rep. 6, 24598 (2016).
Article CAS PubMed PubMed Central Google Scholar
Skardal, A. et al. Multi-tissue interactions in an integrated three-tissue organ-on-a-chip platform. Sci. Rep. 7, 8837 (2017).
Article PubMed PubMed Central Google Scholar
Skardal, A. et al. Drug compound screening in single and integrated multi-organoid body-on-a-chip systems. Biofabrication 12, 025017 (2020).
Article CAS PubMed Google Scholar
Edington, C. D. et al. Interconnected microphysiological systems for quantitative biology and pharmacology studies. Sci. Rep. 8, 4530 (2018).
Article PubMed PubMed Central Google Scholar
Oleaga, C. et al. Multi-organ toxicity demonstration in a functional human in vitro system composed of four organs. Sci. Rep. 6, 20030 (2016).
Article CAS PubMed PubMed Central Google Scholar
Novak, R. et al. Robotic fluidic coupling and interrogation of multiple vascularized organ chips. Nat. Biomed. Eng. 4, 407–420 (2020).
Article PubMed PubMed Central Google Scholar
Kratz, S. R. A., Holl, G., Schuller, P., Ertl, P. & Rothbauer, M. Latest trends in biosensing for microphysiological organs-on-a-chip and body-on-a-chip systems. Biosensors (Basel) https://doi.org/10.3390/bios9030110 (2019).
Shin, S. R. et al. Label-free and regenerative electrochemical microfluidic biosensors for continual monitoring of cell secretomes. Adv. Sci. (Weinh.) 4, 1600522 (2017).
Google Scholar
Shin, S. R. et al. Aptamer-based microfluidic electrochemical biosensor for monitoring cell-secreted trace cardiac biomarkers. Anal. Chem. 88, 10019–10027 (2016).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y. S. et al. Google Glass-directed monitoring and control of microfluidic biosensors and actuators. Sci. Rep. 6, 22237 (2016).
Article CAS PubMed PubMed Central Google Scholar
Rusling, J. F. Multiplexed electrochemical protein detection and translation to personalized cancer diagnostics. Anal. Chem. 85, 5304–5310 (2013).
Article CAS PubMed PubMed Central Google Scholar
Goode, J. A., Rushworth, J. V. H. & Millner, P. A. Biosensor regeneration: a review of common techniques and outcomes. Langmuir 31, 6267–6276 (2015).
Article CAS PubMed Google Scholar
Rajan, S. A. P. et al. Probing prodrug metabolism and reciprocal toxicity with an integrated and humanized multi-tissue organ-on-a-chip platform. Acta Biomater. 106, 124–135 (2020).
Article CAS PubMed Google Scholar
Aleman, J. & Skardal, A. A multi-site metastasis-on-a-chip microphysiological system for assessing metastatic preference of cancer cells. Biotechnol. Bioeng. 116, 936–944 (2019).
Article CAS PubMed Google Scholar
Aleman, J. et al. Deconstructed microfluidic bone marrow on-a-chip to study normal and malignant hemopoietic cell-niche interactions. Small 15, e1902971 (2019).
Article PubMed PubMed Central Google Scholar
Soucy, J. R., Bindas, A. J., Koppes, A. N. & Koppes, R. A. Instrumented microphysiological systems for real-time measurement and manipulation of cellular electrochemical processes. IScience 21, 521–548 (2019).
Article PubMed PubMed Central Google Scholar
Pourmand, A. et al. Fabrication of whole-thermoplastic normally closed microvalve, micro check valve, and micropump. Sens. Actuat. B-Chem. 262, 625–636 (2018).
Article CAS Google Scholar
Shaegh, S. A. M. et al. Rapid prototyping of whole-thermoplastic microfluidics with built-in microvalves using laser ablation and thermal fusion bonding. Sens. Actuat. B-Chem. 255, 100–109 (2018).
Article CAS Google Scholar
Lee, Y.-S., Bhattacharjee, N. & Folch, A. 3D-printed Quake-style microvalves and micropumps. Lab Chip 18, 1207–1214 (2018).
Article CAS PubMed PubMed Central Google Scholar
Ellington, A. D. & Szostak, J. W. In vitro selection of RNA molecules that bind specific ligands. Nature 346, 818–822 (1990).
Article CAS PubMed Google Scholar
Dauphin-Ducharme, P., Arroyo-Currás, N. Y. & Plaxco, K. W. High-precision electrochemical measurements of the guanine-, mismatch-, and length-dependence of electron transfer from electrode-bound DNA are consistent with a contact-mediated mechanism. J. Am. Chem. Soc. 141, 1304–1311 (2019).
Article CAS PubMed PubMed Central Google Scholar
Kurnik, M., Pang, E. Z. & Plaxco, K. W. An electrochemical biosensor architecture based on protein folding supports direct real‐time measurements in whole blood. Angew. Chem. (2020).
Kang, E., Shin, S.-J., Lee, K. H. & Lee, S.-H. Novel PDMS cylindrical channels that generate coaxial flow, and application to fabrication of microfibers and particles. Lab Chip 10, 1856–1861 (2010).
Article CAS PubMed Google Scholar
Kang, E. et al. Digitally tunable physicochemical coding of material composition and topography in continuous microfibres. Nat. Mater. 10, 877–883 (2011).
Article CAS PubMed Google Scholar
MacDonald, M. A. & Andreas, H. A. Method for equivalent circuit determination for electrochemical impedance spectroscopy data of protein adsorption on solid surfaces. Electrochim. Acta 129, 290–299 (2014).
Article CAS Google Scholar
McCauley, T. G., Hamaguchi, N. & Stanton, M. Aptamer-based biosensor arrays for detection and quantification of biological macromolecules. Anal. Biochem. 319, 244–250 (2003).
Article CAS PubMed Google Scholar
Kara, P. et al. Aptamers based electrochemical biosensor for protein detection using carbon nanotubes platforms. Biosens. Bioelectron. 26, 1715–1718 (2010).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

The authors acknowledge support by the National Institutes of Health (R00CA201603 and UG3TR003274) and the Brigham Research Institute. This work was also partially sponsored by the Office of the Secretary of Defense through the Advanced Regenerative Manufacturing Institute (ARMI|BioFabUSA) and was accomplished under Agreement Number W911NF-17-3-003. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Office of the Secretary of Defense or the US Government. The US Government is authorized to reproduce and distribute reprints for government purposes notwithstanding any copyright notation here. T.K. is grateful to the Swiss National Science Foundation for the Postdoc Mobility fellowship (no. P400PM_180788/1).

Author information

These authors contributed equally: Julio Aleman, Tugba Kilic.

Authors and Affiliations

Department of Bioengineering, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, PA, USA
Julio Aleman
Center for Systems Biology and Department of Radiology, Massachusetts General Hospital Research Institute, Harvard Medical School, Boston, MA, USA
Tugba Kilic
Division of Engineering in Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Cambridge, MA, USA
Luis S. Mille, Su Ryon Shin & Yu Shrike Zhang

Authors

Julio Aleman
View author publications
You can also search for this author in PubMed Google Scholar
Tugba Kilic
View author publications
You can also search for this author in PubMed Google Scholar
Luis S. Mille
View author publications
You can also search for this author in PubMed Google Scholar
Su Ryon Shin
View author publications
You can also search for this author in PubMed Google Scholar
Yu Shrike Zhang
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Y.S.Z. and S.R.S. conceived the project; J.A., T.K. and Y.S.Z. drafted the initial manuscript; L.S.M. created the figures; J.A., T.K., S.R.S. and Y.S.Z. revised the manuscript. All authors approved the manuscript.

Corresponding authors

Correspondence to Su Ryon Shin or Yu Shrike Zhang.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Protocols thanks Xingyu Jiang, Arben Merkoçi and José Pingarrón for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Reporting Summary

Supplementary Data 1

AutoCAD designs of breadboard and EC sensing chip with regular, simplified and multiplexed channel configurations

Rights and permissions

Reprints and permissions

About this article

Cite this article

Aleman, J., Kilic, T., Mille, L.S. et al. Microfluidic integration of regeneratable electrochemical affinity-based biosensors for continual monitoring of organ-on-a-chip devices. Nat Protoc 16, 2564–2593 (2021). https://doi.org/10.1038/s41596-021-00511-7

Download citation

Received: 03 September 2020
Accepted: 22 January 2021
Published: 28 April 2021
Issue Date: May 2021
DOI: https://doi.org/10.1038/s41596-021-00511-7

This article is cited by

Flexible hemline-shaped microfibers for liquid transport
- Chaoyu Yang
- Yunru Yu
- Yuanjin Zhao
Nature Chemical Engineering (2024)
Neuropathogenesis-on-chips for neurodegenerative diseases
- Sarnai Amartumur
- Huong Nguyen
- Chaejeong Heo
Nature Communications (2024)
Humanized brain organoids-on-chip integrated with sensors for screening neuronal activity and neurotoxicity
- Pelin Saglam-Metiner
- Ender Yildirim
- Ozlem Yesil-Celiktas
Microchimica Acta (2024)
Open-Source System for Real-Time Functional Assessment of In Vitro Filtration Barriers
- Tess K. Fallon
- Merve Zuvin
- Evren U. Azeloglu
Annals of Biomedical Engineering (2024)
Wearable flexible microfluidic sensing technologies
- Shuwen Chen
- Zheng Qiao
- Chwee Teck Lim
Nature Reviews Bioengineering (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Parameter	Value
Initial E (V)	0.0
High frequency (Hz)	100,000
Low frequency (Hz)	0.1
Amplitude (V)	0.05
Quiet time (s)	0
Sensitivity (A V^–1)	1.0 × 10^–0.004
Bias DC current during run	<1 Hz

Subjects

Abstract

Similar content being viewed by others

Digital colloid-enhanced Raman spectroscopy by single-molecule counting

Shell buckling for programmable metafluids

Best practices for single-cell analysis across modalities

Introduction

Development of the approach

Overview of the procedure

Applications

Alternative methods

Limitations of the protocol

Experimental design

Fabrication of microelectrodes

Principle of microelectrode functionalization and EC detection

Incorporation of different types of capture probes on EC affinity-based biosensors

Method for regeneration of EC affinity-based biosensors

Strategies for microfluidic integration and automation

Breadboard design and modular integration

Materials

Reagents

Caution

Critical

Caution

Critical

Critical

Caution

Caution

Equipment

General

Microfabrication

Hardware and software

Reagent setup

Self-assembled monolayer (SAM) solution

Citric buffer solution

EDC/NHS solution

K3Fe(CN)6 stock solution

Caution

Critical

SPV stock solution

Critical

H2SO4 stock solution

Caution

Dilution buffer

Biotinylated antibody solution

Amine group-functionalized aptamer solution

Washing buffer

Hepatocyte culture medium

Cardiomyocyte culture medium

Common medium

Critical

PDMS

Equipment setup

Setup of the EC cell

Procedure

Microelectrode fabrication

Critical step

Caution

Critical step

Pause point

Photolithographic fabrication of microfluidic controlling devices

Critical

Critical step

Caution

Critical step

Caution

Pause point

PDMS membrane fabrication

Critical step

Critical step

Pause point

Fluidic layer fabrication

Critical step

Caution

Pause point

Hemicylindrical microchannel fabrication

Critical step

Critical step

Critical step

Caution

K₃Fe(CN)₆ stock solution

H₂SO₄ stock solution