ArticleJanuary 14, 2026

ChipNMR: Hyperpolarized NMR for Noninvasive Metabolic Flux Analysis in Perfused Microfluidic Chips
Click to copy article linkArticle link copied!

Thomas B. Wareham Mathiassen
Thomas B. Wareham Mathiassen
Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
More by Thomas B. Wareham Mathiassen
Juan D. Sánchez-Heredia
Juan D. Sánchez-Heredia
Department of Information Technologies and Communications, Technical University of Cartagena (UPCT), 302020 Cartagena, Spain
More by Juan D. Sánchez-Heredia
Ke-Chuan Wang
Ke-Chuan Wang
Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
More by Ke-Chuan Wang
Cajsa R. Haupt
Cajsa R. Haupt
Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
More by Cajsa R. Haupt
Magnus Karlsson
Magnus Karlsson
Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
More by Magnus Karlsson
Alexander Jönsson
Alexander Jönsson
Department of Micro and Nanotechnology, Technical University of Denmark, 2800 Kgs Lyngby, Denmark
More by Alexander Jönsson
Martin Dufva
Martin Dufva
Department of Micro and Nanotechnology, Technical University of Denmark, 2800 Kgs Lyngby, Denmark
More by Martin Dufva
https://orcid.org/0000-0001-5449-0189
Roland Thuenauer
Roland Thuenauer
Technology Platform Light Microscopy (TPLM), University of Hamburg (UHH); Advanced Light and Fluorescence Microscopy (ALFM) Facility, Centre for Structural Systems Biology (CSSB) Hamburg; Leibniz Institute of Virology (LIV), Notkestrasse 85, 22607 Hamburg, Germany
More by Roland Thuenauer
Pernille Rose Jensen*
Pernille Rose Jensen
Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
*Email: [email protected]
More by Pernille Rose Jensen
https://orcid.org/0000-0003-4359-848X

Open PDF Supporting Information (1)

Analytical Chemistry

Cite this: Anal. Chem. 2026, 98, 3, 1891–1900

Click to copy citationCitation copied!

https://doi.org/10.1021/acs.analchem.5c04058

Published January 14, 2026

Request reuse permissions

Abstract

Click to copy section linkSection link copied!

Dissolution dynamic nuclear polarization NMR Spectroscopy (dDNP-NMR) has become a transformative tool for metabolic studies by significantly enhancing signal sensitivity more than 3 orders of magnitude compared to traditional NMR. However, NMR detection probes are optimized for round narrow glass tubes typically 5 mm in diameter, which impose constraints on their utility for metabolic studies of adherent cells. Here, we present a novel NMR probe head integrated with a custom microfluidic chip that facilitates real-time monitoring of hyperpolarized substrate conversion from adherent cells. This system enables metabolic flux analysis in a controlled, in vitro environment, as demonstrated by tracking the conversion of [1-¹³C]pyruvate to [1-¹³C]lactate in HeLa cells over 48 h. The custom microfluidic chip design is modular and adaptable allowing expansion to two-chamber chips, demonstrating its potential in applications for more complex cellular models, such as organ-on-a-chip systems.

This publication is licensed for personal use by The American Chemical Society.

Subjects

Introduction

Click to copy section linkSection link copied!

The process of drug development is both resource-intensive and time-consuming, with high attrition rates often stemming from the limited predictive power of traditional preclinical models. (1) Historically, animal models have been the cornerstone of drug discovery, yet their translational relevance to human physiology remains suboptimal, leading to costly failures in clinical trials. This limitation underscores the necessity for more accurate and human-relevant in vitro models that better replicate the complexity of human disease. (2)

Conventional two-dimensional (2D) cell culture systems, while widely used, fail to recapitulate key aspects of the physiological microenvironment, including cell–cell interactions, extracellular matrix composition, and dynamic biochemical gradients. (3) Three-dimensional (3D) cultures provide improved structural and functional relevance, but they still lack precise control over perfusion, mechanical forces, and biochemical signaling present in vivo. (4) Organ-on-a-chip (OOC) systems address these limitations by integrating microfluidic systems that simulate organ-level functions, providing a physiologically relevant platform for studying cellular behaviors, disease mechanisms, and drug responses. (5−11)

Despite the advances in microfluidic models, a major challenge persists: the lack of methods to directly and continuously monitor metabolic activity within perfused systems in real time. Metabolic alterations often precede morphological and functional changes in cells, making metabolism a key early indicator of disease progression, therapeutic response, and cellular adaptation. (12−15) Current methods for metabolic assessment in microfluidic systems predominantly rely on indirect measurements, such as luminescence-based sensors for glucose consumption, oxygen utilization, and pH changes. (16−20) For example, electrochemical glucose and lactate biosensors operate via enzyme-mediated conversion to hydrogen peroxide, which is then quantified. (21) Electrochemical biosensors, while offering higher specificity, are also susceptible to signal drift, enzyme degradation, and the need for periodic calibration. (19,20,22) Several studies highlight the value of embedding sensors directly within chips to enhance the accuracy of metabolic measurements. (20,21,23,24) While these methods provide valuable metabolic insights, they lack the direct detection capability of real-time metabolic fluxes.

Nuclear magnetic resonance (NMR) spectroscopy is a powerful analytical tool for metabolic studies due to its noninvasive nature and ability to provide quantitative insights into metabolic pathways. In particular, dissolution dynamic nuclear polarization (dDNP) NMR significantly enhances signal sensitivity, enabling real-time monitoring of metabolic processes in living systems. (25) The potential synergies between dDNP-NMR and microfluidic culture systems have been recognized, yet practical implementations remain limited. (26) Previous work has demonstrated the feasibility of integrating ¹H NMR with microfluidic devices, as well as hyperpolarized ¹³C detection of signal from parahydrogen-induced polarization (PHIP). (27,28) However, no platform to date has fully integrated dDNP-NMR with a perfused microfluidic system for longitudinal metabolic assessment of adherent cells. In this study, we present a novel NMR probe integrated with a custom-designed microfluidic chip, enabling real-time metabolic analysis in perfused cell cultures. This platform allows for monitoring of hyperpolarized substrate conversion within a physiologically relevant microenvironment. We demonstrate the functionality of this system through longitudinal metabolic measurements of HeLa cells, tracking the conversion of hyperpolarized [1-¹³C]pyruvate to [1-¹³C]lactate over 48 h. Additionally, comparable kinetics are obtained in a more advanced chip with two chambers. For demonstration, we used HeLa cells, a robust and well-characterized cell line which has been used extensively as model system in dDNP-NMR research. (29,30) The stability of HeLa cells under multiday perfusion made them ideal for establishing feasibility. While the present work is not a full OOC study, the chip incorporates design elements common to OOC platforms (e.g., membrane separation, perfusion), which support future alignment with more advanced physiological models.

This study represents the first demonstration of hyperpolarized ¹³C NMR metabolic flux analysis in a perfused microfluidic chip seeded with an adherent mammalian monolayer, establishing feasibility for extension to more physiologically relevant culture formats.

Experimental Section

Click to copy section linkSection link copied!

All chemicals were purchased from Sigma-Aldrich unless otherwise stated.

RF Coil Design and Creation

The NMR probe head design includes two separate RF coils, and a support structure made of 3D printed resin (Formlabs, USA). It is made for a 400 MHz NMR spectrometer (Bruker Avance NEO). Coil 1 is a saddle-shaped coil, used exclusively for the ¹³C nucleus. The two rectangular loops are separated 4.5 mm (with the width of the chip being 4 mm), in order to maximize the filling factor. The coil is made of silver wire, with a wire diameter of 1 mm. This coil geometry provides an inductance of approximately 130 nH at the frequency of ¹³C (100.55 MHz). The return loss curves for the empty probe, as well as for the probe loaded with empty one- and two-chamber chips and those chips filled with 40 mM phosphate buffer, are shown in Figure S1 along with the corresponding Q-factors. The measured unloaded Q-factor was 165.9. Loading the probe with empty one- and two-chamber chips resulted in Q-factors of 165.6 and 165.1, respectively. When the chips were filled with 40 mM phosphate buffer, the Q-factors decreased to 156.1 for both one- and two-chamber chips. Overall, the presence of empty chips reduced the Q-factor by approximately 0.3%, while filled chips caused a reduction of about 6%. No discernible difference in Q-factor was observed between one- and two-chamber chip configurations. To benchmark the performance of the ¹³C coil, we measured both the line width and the signal-to-noise ratio (SNR) of a [¹³C]urea solution (2 M ¹³C urea and 8 mM gadodiamide in 40 mM phosphate buffer, pH 7.4) using two different probe setups. For a filled one-chamber chip (active volume 750 μL), the line width at half height was 48 Hz and the SNR, normalized to 1 mL solution, was 103. By comparison, a 10 mm room-temperature broadband probe (active volume 1.63 mL) produced a narrower line width of 15 Hz and an SNR of 662 (again normalized to 1 mL solution). As benchmark, the chip probe exhibited a 6-fold lower SNR and a broader line width than the 10 mm probe. Albeit not utilized in this study the probe was designed with a second double-tuned coil, to cover the ¹H and ²H nuclei (Coil 2). The coil has two opposed copper plates, acting similarly to a microstrip line. The separation between the two plates is 12 mm. This coil is made of copper and provides an inductance of approximately 49 nH at the ²H frequency, and 52 nH at the ¹H frequency (simulations of the B₁⁺ fields for both the ²H and ¹H coils are shown in Figure S2). An illustration of the full probe setup is shown in Figure 1.

Figure 1. RF Coil design. (A) exploded view of the NMR probe head, showing the two RF coils, the 3D-printed support, and the microfluidic chip. (B) Cross-section (Z = 0 plane) of the simulated B₁⁺ field distribution generated by Coil 1 at the ¹³C frequency (per 1 W accepted power). (C) Photograph of the fabricated NMR probe with the chip inserted.

Chip Fabrication

The fluidic chip was fabricated using the following materials and procedures (Figure 2). Depending on the experimental design, either a one-chamber or two-chamber fluidic chip was fabricated. A 1 mm sheet of PETG (Exolon group, Vivak) was covered with a 0.75 mm sheet of SEBS (Eden-Microfluidics, Flexym), and a 0.3 mm sheet of PETG was covered with a 0.25 mm sheet of SEBS. For the one-chamber chip, this bottom composite sheet was left uncut. These composite sheets were cut using a FLUX Beambox CO₂ laser operated through Beam Studios software. The laser’s intensity, power, and speed were optimized to cut through the SEBS layer without penetrating the PETG layer, allowing for precise removal of the channel shape with tweezers. Subsequently, the outer shape of the chip was cut from the composite sheets and the SEBS side of a composite sheet was adhered to a sheet of track-etched polycarbonate (ipCELLCULTURE Track-Etched Membrane Filter with an 8 μm pore diameter, a pore density 1 × 10⁵/cm² pore density, a thickness of 18 μm, and a porosity of 5%). The membrane was trimmed to fit using a scalpel, and the other composite sheet was pressed with its SEBS side on the opposite side of the membrane, forming a sandwich configuration as shown in Figure 2A. The assembled chip was then pressed at 2 Pa and 60 °C for 1 h using a hydraulic press. A top piece with four mini Luer injection ports was designed using SolidWorks and fabricated from BioMed Clear Resin on a Formlabs Form 3B printer with default settings and a layer height of 25 μm (Figure 2B). Postprinting, the inlets were washed with isopropanol and cured in a UV box at 70 °C for 20 min. After curing, the supports were trimmed, autoclaved in 1 L of demineralized water at 121 °C, and affixed to the chip using generic superglue.

Figure 2. Fluidic chip production and assembly. (A) visual representation of the chip components and their assembly process. The stacked components are shown in order from top to bottom: 1 mm PETG, 0.75 mm SEBS, track-etched polycarbonate membrane (8 μm pore size, 18 μm thickness, 5% porosity), 0.25 mm SEBS, and 0.3 mm PETG. (B) 3D-printed inlets/injection ports, designed using SolidWorks, shown with female mini Luer connectors made from BioMed clear resin. (C) Fully assembled microfluidic chip.

Chip Sterilization and Treatment

To ensure minimal leeching from the adhesives, the chip was injected with sterile H₂O and placed in a water bath for 24 h. Silicone tubes were attached to the chip with minileur connectors (ChipShop product number 100000094) and 70% ETOH was perfused for 1 h at 500 μL/min using a peristaltic pump, the setup of which is outlined in previous work. (31) Following that, sterile water was flushed into the entire system for an additional 2 h. Prior to seeding, the chips were coated with a solution containing 300 μg/mL Collagen I solution (Rat Tail Collagen Type I, Gibco) in DMEM (Biowest L0103), and placed at 37 °C for 1 h.

Cell Culturing and Chip Seeding

The HeLa human cervical cancer cell line (ATCC CCL-2, American Type Culture Collection, Manassas, VA) was routinely cultured in DMEM (Biowest L0103), which contains 4.5 g/L d-glucose, stable l-glutamine, and sodium pyruvate. The medium was supplemented with 10% fetal bovine serum (FBS; Biowest S1810) and 1% penicillin/streptomycin (Biowest L0022). Cells were maintained at 37 °C in a humidified 5% CO₂ incubator.

The cell medium was refreshed 2 to 3 times per week, and the cells were cultured until they reached 80% confluence, at which point they were harvested using a 0.25% Trypsin-EDTA solution (Biowest X0930). For the dDNP-NMR experiments, 5 million HeLa cells were seeded 4–5 days prior to the experiment in a T175 flask. On the day of seeding the chips the cells were harvested by trypsinization, followed by centrifugation and washing in DMEM containing 10% FBS. Subsequently, the cells were washed in PBS with Ca²⁺/Mg²⁺ (HyClone, SH30264.01) and resuspended in DMEM with 10% FBS to a final concentration of 4 × 10⁶ cells/mL.

The membrane surface area of each chip was 13 cm², and chips were seeded at a density of 2.5 × 10⁵ cells/cm², resulting in a total of 3 × 10⁶ cells per chip. Homogeneous cell suspensions were injected into a vertically placed chips via the narrow channel using a 2 mL syringe fitted with a male mini Luer tip.

The seeding density (3 × 10⁶ cells per chip) was determined by automated cell counting with Trypan Blue exclusion prior to seeding. After cell adherence in the chip, cell numbers were quantified following trypsinization using the same method. This recovery-based quantification likely underestimates absolute values, as not all adherent cells are fully recovered.

Following seeding, the chips were incubated horizontally at 37 °C in a humidified 5% CO₂ atmosphere for 3 h to allow cell adhesion, which was confirmed via visual inspection. Once adhesion was confirmed, a continuous flow of DMEM at 200 μL/min was initiated and maintained until the dDNP-NMR experiment commenced at 3, 24, or 48 h. For two-chamber chips the top chamber was seeded with a HeLa cell using the same seeding protocol as for the one-chamber chip.

For each time point (3, 24, and 48 h), separate but identically prepared chips were used to avoid perturbations from repeated handling. During NMR acquisition, the chips were outside the incubator for approximately 5–10 min without CO₂ and gas exchange. Handling was brief, minimizing potential impact on culture conditions. This period accounted for preparation steps (medium exchange, chip mounting, dissolution of hyperpolarized [1-¹³C]pyruvate, injection and data acquisition).

Cell Staining and Microscopy

The chips were washed twice with approximately 800 μL of PBS, then stained with a solution of 2 μg/mL Calcein AM (Invitrogen C1430) for 30 min at 37 °C, protected from light. Following staining, the chips were washed again with PBS to remove excess dye. Fluorescence microscopy was performed using a Leica DMI3000 B microscope system (Leica, Wetzlar, Germany) with an excitation wavelength of 485 nm and an emission wavelength of 530 nm to visualize live cells.

dDNP-NMR Experiments

Substrate stock solutions were prepared using [1-¹³C]pyruvic acid (Sigma-Aldrich), doped with 30 mM trityl radical AH111501 (GE Healthcare). Approximately 4.5 mg of the pyruvic acid stock solution was hyperpolarized using a SpinAligner 6.7 T polarizer (PolarizeTM) for ∼1 h until equilibrium polarization was achieved. The sample was rapidly dissolved in 5 mL of phosphate buffer (pH 7.4, 40 mM), with 5 μL of 10 M NaOH added to stabilize the pH, maintaining a final temperature of ∼310 K. The dissolved pyruvate concentration was ∼10 mM, with 45 ± 3% liquid-state polarization and a stable pH in hyperpolarized [1-¹³C]pyruvate solution (pH = 7.4 ± 0.2).

Pyruvate is imported into cells via the MCT1 transporter, which should be saturated but not oversaturated for optimal experimental conditions. Here, we used a 10 mM [1-¹³C]pyruvate concentration based on the Km for pyruvate, which has been reported to be 2 mM in cancer cells, in a study using hyperpolarized [1-¹³C]pyruvate (32) and consistent with other previous hyperpolarized cell studies. (29,30,33,34)

In the beginning of each experimental day shimming was performed manually on the FID from a chip filled with a [¹³C]urea solution matching the salt concentration of the culture medium. Reproducible line widths were observed replacing the chip. This strategy was used to minimize the time between inserting cell laden chips and dDNP-NMR measurement. The hyperpolarized [1-¹³C]pyruvate solution was collected in a 50 mL receiver, and 3 mL was drawn into a 5 mL syringe; 1.5 mL was then injected into the empty chip positioned in the NMR spectrometer.

At designated time points, a chip was fitted with PEEK tubing (1/32″ ID, 0.5 mm, Microlab) for the inlet and silicone tubing (1 mm ID, Hounisen) for the outlet, mounted in the NMR probe, and inserted into a Bruker 400 MHz spectrometer. The chip was mounted in a vertical orientation, which facilitated bubble removal and thereby improved substrate delivery. The physical transfer of hyperpolarized [1-¹³C]pyruvate from the polarizer to the chip took approximately 10–15 s. Just before injection of the hyperpolarized [1-¹³C]pyruvate the culture medium was withdrawn from the chip to avoid dilution of the bolus. For two-chamber chips, the bottom chamber was filled with PBS and sealed immediately prior to inserting the chip into the probe. The measured metabolic signal therefore originated exclusively from the top chamber.

¹³C NMR spectra were acquired over 256 s as a series of 128 1D ¹³C spectra each using a 20° pulse angle and a 2 s delay between pulses. The data acquisition was started manually upon start of the injection of hyperpolarized pyruvate into the chip.

Cell Counting

Following the dDNP-NMR experiment, the chip was immediately removed from the spectrometer, cells were washed twice with PBS, followed by trypsinization for 10 min. The outflow was mixed with equal parts DMEM + 10% FBS to neutralize the trypsin and spun down at 300 G before resuspending in 10 mL PBS. Cells were counted using an EVE Automated Cell Counter (nanoEntek) with 0.4% trypan blue stain.

Data Analysis

The NMR data analysis was performed using MestReNova software (Mestrelab Research). After phase and baseline correction, the signals of the substrate ([1-¹³C]pyruvate) and the product ([1-¹³C]lactate) in the time series were integrated. A Python-based model was then applied to the integrals using a numerical solution of the ordinary differential equations (ODEs) that describe the forward reaction rate (k). In this model, the flip angle (α) was fixed, while the T₁ values (T_1S, T_1P) were allowed to vary. Across experiments, the fitted relaxation times were consistent, with T_1s = 52 ± 14 s and T_1p = 6 ± 2 s (mean ± SD, n = 8). The ODEs used in the model were as follows

\frac{d S}{d t} = - k \cdot S (t) - \frac{1}{T_{1 s}} \cdot S (t) - (1 - \cos {(α)}^{n}) \cdot S (t)

\frac{d P}{d t} = k \cdot S (t) - \frac{1}{T_{1 p}} \cdot P (t) - (1 - \cos {(α)}^{n}) \cdot P (t)

As it took 8 s to fill the chip with hyperpolarized pyruvate, the time curves for pyruvate were extrapolated back to the beginning of injection to obtain an exponentially decreasing substrate curve over the full course of the experiment. This approach enabled kinetic modeling of the reaction rate constant by incorporating the entire substrate exposure period, thereby capturing the initial build-up phase of the lactate curve, which is important for robust rate constant determination.

Statistical Analysis

All statistical analysis was performed in GraphPad prism. p values were calculated with one-way ANOVA and values less than 0.05 were considered statistically significant.

Results and Discussion

Click to copy section linkSection link copied!

NMR Coil Design and Functionality

The NMR probe head was designed to accommodate a flat microfluidic chip with a surface area as large as possible in order to allow for more than one million adherent cells. The microfluidic chips were fabricated, leak-tested, and validated before experimentation. The corresponding chip included Luer-compatible injection ports, enabling the injection of hyperpolarized substrate while mounted in the probe.

To validate the probe’s functionality, hyperpolarized [1-¹³C]pyruvate was injected into an empty chip, and the decay profile due to T₁ relaxation was monitored by acquiring a time series of 1D ¹³C NMR spectra using a 20° pulse angle and a 2 s delay between pulses. A monoexponential T₁ decay of 52 ± 2 s for pyruvate indicates that the chip was filled smoothly and homogeneously, without trapped air bubbles or other disturbances (Figure 3A). The area under the curve (AUC), for three repetitions was 24 ± 0.6, with a coefficient of variation (CV) of 2.5%, demonstrating high reproducibility of the chip system. This CV was calculated from the AUC of the raw pyruvate signals, without normalization. It thus reflects the reproducibility of the full experimental procedure including solid-state polarization level, dissolution and injection steps. The signal-to-noise ratio (SNR) of the highest pyruvate peak spectrum was approximately 5000, confirming high coil sensitivity.

Figure 3. Performance evaluation. (A) single pyruvate spectrum, SNR of the highest pyruvate peak was 5000, (B) substrate integral over time (n = 3) and (C) comparative substrate integrals (AUC = 24 ± 0.6) following injection of 10 mM hyperpolarized [1-¹³C] pyruvate, CV for the AUC was 2.5%. T₁ of pyruvate was 52 ± 2 s.

Compared to conventional 5 mm NMR probes, which are generally incompatible with adherent cell cultures, our probe provides a tailored solution for microfluidic environments that support adherent cell cultures. This capability enables real-time monitoring of dynamic biological processes, offering a robust platform for advanced metabolic tracking and analysis. Unlike earlier NMR compatible microfluidic devices, (35,36) our setup has compromised on magnetic field homogeneity over the sample in exchange for larger cell surface area. This complementary strategy achieves a good balance between cell number, ease of use, and SNR. The large surface area of the chip allows the system to be operated via a simple peristaltic pump, in contrast to the more complex fluidic systems required for smaller (2 μL) microfluidic chip systems. (27,28) While this sacrifice in B1 homogeneity, leading to decreased peak separation ability, is not considered a significant issue for X-nuclei, it may present a bigger problem for ¹H spectroscopy.

Microfluidic Chip as a Cell Culturing Device

A custom microfluidic chip was developed in-house, specifically designed to optimize cell density within the NMR coil’s detection range. The chip was made transparent to allow for easy microscopic observation and to support fluorescence staining. Additionally, the chip was engineered to be biocompatible, flow-compatible, leak-proof, and resistant to material leaching, ensuring a stable and reliable environment for cell culture. Depending on the experimental design, either a one-chamber or two-chamber fluidic chip was used. Both configurations were equipped with female mini Luer injection ports on one end, allowing them to slide seamlessly into the dDNP-NMR probe without obstruction. The inlets were designed with protruding channels that extended into the chip, acting as a physical barrier to prevent affixing glue from entering the flow path.

To assess the chip’s ability to sustain cell culture, HeLa cell morphology and proliferation were monitored over 48 h of culture. A total of 3 × 10⁶ HeLa cells were seeded onto the one-chamber chips and incubated at 37 °C with 5% CO₂. After a 3 h adhesion period, medium perfusion was initiated at a flow rate of 200 μL/min using a peristaltic pump. When necessary, flow was temporarily halted, and the chips were examined under a bright-field microscope before being returned to the incubator. At designated time points, cells were stained with the fluorescence stain Calcein AM for live cell imaging or detached using trypsin and counted with a cell counter employing trypan blue exclusion.

Cells remained stably adhered throughout the experiment, forming a highly confluent monolayer (Figure 4A), and the population approximately doubled over 48 h (Figure 4B), demonstrating the chip’s suitability for maintaining viable, proliferating cells under continuous perfusion. The doubling time of ∼24 h matches the typical growth rate of HeLa cells in conventional flask culture, (37) indicating that the perfused microfluidic environment preserved normal proliferation rates.

Figure 4. One-chamber chip cell growth. (A) one-chamber chip seeded with 3 × 10⁶ HeLa cells, cultured for 3 h, and imaged with bright-field microscopy (bottom left, BF 20×) and fluorescence microscopy of cells stained with Calcein AM (bottom right, Calcein AM 20×). Scale bar: 100 μm. (B) Cell counts in chips after 3, 24, and 48 h. The mean ± SD cell counts (in millions) were 1.90 ± 0.26 at 3 h, 2.70 ± 0.61 at 24 h, and 4.46 ± 0.69 at 48 h. Statistical analysis (one-way ANOVA) showed a significant difference (p < 0.01) in cell count between 3 and 48 h. N = 3 for each time point.

Detection of Metabolic Activity via dDNP-NMR

Click to copy section linkSection link copied!

After validating the probe’s signal detection capabilities and confirming the microfluidic chip’s effectiveness as a cell culture platform, we assessed its ability to monitor metabolic activity in real time. Hyperpolarized [1-¹³C] pyruvate was injected into the mounted chip laden with HeLa cells, achieving a final concentration of 10 mM [1-¹³C] pyruvate. Metabolic conversion was tracked by monitoring lactate production from hyperpolarized [1-¹³C] pyruvate at 3, 24, and 48 h after seeding into the chip. Figure 5A shows a representative spectrum from a time series 18 s after start of the injection where lactate has its maximum. The SNR of the lactate peak in this spectrum was approximately 50. The full dynamic curves showed a steady increase in metabolic activity over time (Figure 5B). The forward rate constants (k) per million cells were fitted to a set of differential equations using the integrals of the dynamic curves normalized to the initial pyruvate signal in each experiment (Figure 5C). The normalized rate constants were 4.12 ± 0.53 × 10^–4 s^–1 at 3 h, 3.76 ± 0.64 × 10^–4 s^–1 at 24 h, and 4.78 ± 0.53 × 10^–4 s^–1 at 48 h. The corresponding CV values were 13% at 3 h, 17% at 24 h, and 11% at 48 h, with an average CV of 14%, indicating moderate variability in metabolic activity over the 48 h period. The observed variation fell within expected experimental error and was not statistically significant, indicating stable per-cell metabolic turnover across time points. This stability suggests that the perfusion environment maintained metabolic function even as total cell number doubled. Normalizing the rate constants to the increasing cell number between 3 and 48 h resulted in the same turnover per cell throughout the experiment underscoring the stability of the system performance regarding metabolic flux determination (Figure 5C). For comparison, the unnormalized rate constants, together with the corresponding cell counts at each time point, are shown in Figure S3.

Figure 5. One chamber chip hyperpolarized ¹³C NMR. (A) spectrum with the highest lactate signal (spectrum 9), showing pyruvate C2 at 205 ppm, lactate C1 at 182.3 ppm, pyruvate hydrate C1 at 178.4 ppm, and pyruvate C1 at 171 ppm. (SNR of highest lactate peak was approximately 50) (B) time-series spectra displaying the production of [1-¹³C] lactate in one-chamber chips at 3, 24, and 48 h postseeding, following the injection of 10 mM hyperpolarized [1-¹³C] pyruvate. (C) Rate constants normalized to cell counts, yielding mean ± SD values (s^–1) of (4.12 ± 0.53) × 10^–4 at 3 h, (3.76 ± 0.64) × 10^–4 at 24 h, and (4.78 ± 0.53) × 10^–4 at 48 h. Statistical analysis (one-way ANOVA) showed no significant difference (NS) between normalized rate constants. N = 3 for each time point.

An increasing total rate constant was observed as total cell numbers increased (Figure S3), whereas the rate constant normalized to cell number was stable (Figure 5). This observation shows the stability of the system where circulating medium removes dead cells. It is not possible to directly compare if the adherent cells have higher turnover compared to harvested cells because the suspended cells would be washed out of the chip upon injection of the hyperpolarized substrate. Nevertheless, in previous studies, we calculated the rate constants of pyruvate-to-lactate conversion in 2 million HeLa cells to approximately k = 2 × 10^–4 s^–1 equivalent to 1 × 10^–4 s^–1 per million cells. (30) Our current results show comparable normalized rate constants (4 × 10^–4 s^–1 per million cells) consistent with the previous study. Because these values were obtained under different experimental conditions (suspension vs adherent chip), the comparison is meant only as an approximate reference rather than a statistical test of significance. Although differences in experimental setup complicate direct comparison, these results are consistent with the notion that detaching cells can alter their metabolic activity, (38−42) and that measuring metabolism in adherent cells provides more physiologically relevant data.

The literature also reflects variability across models: some studies found no metabolic difference between suspension and scaffold-grown HeLa cells, (29) others reported increased glycolytic activity in multicellular tumor spheroids compared to 2D culture, (43) and in primary endothelial progenitor cell suspension culture even enhanced metabolic turnover relative to adhesion. (44) Taken together, these findings underscore the importance of culture format and cell type in shaping metabolic readouts, and highlight the value of our chip for directly probing adherent monolayers.

High-quality flux data could be obtained on 2 × 10⁶ adherent cells. This sensitivity is comparable to data obtained from standard 5 mm probes on suspensions of cancer cells, where typically 2–20 × 10⁶ cells are used. (30,45−47) To the best of our knowledge, this is the first demonstration of cell metabolism from an adherent monolayer of mammalian cells in combination with hyperpolarized NMR. In the chip system, a maximum SNR of approximately 50 was measured on lactate. While direct comparison to the performance of a conventional RT probe is challenging, it provides useful context. Translating SNR from a previously published experiment using 2 × 10⁶ HeLa cells and 10 mM hyperpolarized [1-¹³C]pyruvate to a 400 MHz RT probe yields a maximum lactate SNR of about 480. (30) Thus, the current chip probe produces only a 10-fold lower SNR compared to an RT 5 mm probe at equivalent field strength, which is similar to the probe performance estimated from a [¹³C] urea phantom (6-fold). Considering the unique experimental advantages of the chip system for supporting adhering cells, this initial SNR performance is very encouraging. The CV of our metabolic measurements over the 48 h was 14 ± 2.5%. While this reproducibility is lower than our previous studies using HeLa cells in a 5 mm probe (CV 2.4 ± 0.8%), (30) it is important to note that we have established entirely new protocols for seeding, injection, and testing of the chip. The obtained CV corresponds well with typical values obtained in the literature, (33) where the CV across six studies ranged from 3% to 46%, with a mean CV of 23% ± 15% underscoring the impact of experimental conditions on reproducibility. (33) In this context, the current average CV of 14 ± 2% is significantly below the mean, indicating good reproducibility and measurement consistency. This performance is noteworthy given the complexity of integrating microfluidic systems with NMR technology. The chip is compatible with histological staining and optical microscopy, such as fluorescence and brightfield imaging (Figure 4), which is important for integrated use and benchmarking against other microfluidic platforms. Combining dDNP-NMR with fluorescence imaging provides complementary modalities, integrating functional metabolic measurements with static morphological insights.

Comparative Analysis of One-Chamber and Two-Chamber Chips

Many microfluidic models utilize two-chamber chips, where the chambers are separated by a membrane onto which cells adhere. (11) This two-chamber design enhances physiological relevance and versatility, enabling more complex experimental setups and improving the functionality of microfluidic models. A key advantage of this configuration is the ability to implement differential flow between the top and bottom chambers, allowing cells in the top chamber to adhere undisturbed while nutrients are supplied through the bottom chamber. This is particularly important for growing epithelial monolayers of cells, which naturally form a barrier layer and need to be supplied with nutrients from the basolateral side of the monolayer to replicate physiological conditions. (48,49)

Using the same base design, we adapted our system for metabolic measurements in a two-chamber chip. Here the top chamber was seeded with a HeLa monolayer using the same cell seeding protocol as for the one-chamber chip. This adaptation achieved similar cell densities as obtained in the one-chamber chip, with a confluent HeLa cell layer of approximately 4 × 10⁶ cells per chip 24 h of culture (Figure 6A). To evaluate the system’s adaptability, we measured lactate production in the two-chamber chip. This design, commonly employed for barrier models, demonstrated metabolic activity comparable to the one-chamber chip. The two-chamber chip exhibited similar lactate production kinetics, with a calculated rate constant of 2.55 ± 0.28 × 10^–4 s^–1 (Figure 6B).

Figure 6. Two-chamber chip cell growth and lactate production. (A) two-chamber chip seeded with 3 × 10⁶ HeLa cells, using a flow rate of 0.200 mL/min in the lower chamber, and no flow in the top chamber, for 24 h at 37 °C and 5% CO₂. Bright-field microscopy image (bottom left, BF 20×) and fluorescence microscopy image of cells stained with Calcein AM (bottom right, Calcein AM 20×). Scale bar: 100 μm. (B) Time-series spectra showing [1-¹³C] lactate accumulation in the two-chamber chip seeded with 3 × 10⁶ HeLa cells, following injection of 10 mM hyperpolarized [1-¹³C] pyruvate. The mean ± SD rate constant for lactate production was 2.55 ± 0.28 × 10^–4 (n = 2).

Beyond the demonstrated validation, the two-chamber chip design allows independent perfusion or injection into each chamber, enabling the development of OOC models. For example, gut-on-chip systems benefit from flow on the apical side to enhance apical-basal polarity. (50) This design also supports studies of directional transport (31) and cocultures. (51) These capabilities extend beyond those possible in single-compartment systems and position the platform for advanced investigations of barrier permeability and cross–compartment interactions.

These results underscore the robustness and flexibility of our system across different chip configurations, supporting its future use in barrier models and other complex applications.

Our microfluidic chip successfully supports longitudinal cell culture while maintaining stable metabolic function (Figure 4). Unlike conventional designs, which typically accommodate 10³ to 10⁵ cells, our chip can support up to 4 million cells in a monolayer. (11,28) As a proof of concept, we conducted metabolic assessments over 48 h, which was chosen as a practical validation window. Future studies should extend this time frame to up to 4 weeks, which is necessary for more complex cocultures. (52) Long-term OOC cultures can develop into 4–8 cell layers, at full differentiation. This would correspond to 12–36 million cells in the current chip. This increased cell density would significantly enhance sensitivity, possibly allowing for the detection of lower-concentration metabolic intermediates.

While the results are promising, several limitations and areas for future improvement should be noted. First, the current study focused on HeLa cells, a well-characterized cancer cell line. Future work should explore the system’s applicability to other cell types, including primary cells and cocultures, to better replicate the complexity of human tissues and organs. The system is highly adaptable and can be extended to alternative cell models. For example, intestinal differentiation can be achieved in Caco-2 cells by asymmetrically culturing them with no fetal bovine serum (FBS) on the apical channel and with FBS on the basolateral side. (53) This protocol can be implemented on our two-chamber chip to analyze the metabolism of more complex and differentiated tissues, allowing for studies on epithelial barriers. Other cell types could also be used to model specific disease states or drug responses, further expanding the system’s applicability. In this proof-of-concept study, we focused on demonstrating NMR compatibility of the two-chamber architecture using HeLa monolayers. This establishes the technical foundation for future studies exploring flow-dependent effects and cross–compartment interactions. The NMR probe demonstrated excellent signal stability, ensuring reproducibility across different experimental setups. Ensuring consistency across operators will be crucial for widespread adoption. Developing standardized protocols and calibration methods will be essential to address this challenge.

Conclusion

Click to copy section linkSection link copied!

Advanced microfluidic culture models offer significant potential to reduce the need for animal experiments and to improve drug development workflows. As these technologies continue to mature and demonstrate their utility, the demand for physiologically relevant in vitro systems is expected to grow substantially. (5) Currently, options for noninvasive metabolic monitoring in such platforms are limited. Integrating metabolic measurements with microfluidic chips enhances their physiological relevance, predictive power, and overall utility. NMR, and in particular hyperpolarized approaches such as dDNP-NMR, offers real-time process monitoring in biological systems.

This study presents a platform for noninvasive, real-time metabolic analysis in perfused microfluidic cell cultures. The integration of dDNP-NMR with custom-designed microfluidic chips enables dynamic monitoring of metabolic fluxes, offering a powerful tool for drug development, disease modeling, and personalized medicine. The system’s adaptability to different chip configurations and its ability to support long-term cell culture makes it a versatile addition to current advanced in vitro modeling strategies. This work was validated only with HeLa cells, but the chip supports other adherent types, such as hepatocyte-derived HepG2 cells relevant for metabolic and toxicological assays, (54) or epithelial models (e.g., Caco-2). (50) Future studies will extend to these applications to broaden the system’s applicability. While not a full organ-on-chip model, the platform incorporates several features common to OOC systems, providing a foundation for future adaptation toward more complex and physiologically relevant tissue models. By bridging the gap between advanced NMR spectroscopy and microfluidic technology, this work paves the way for more accurate and human-relevant in vitro models, ultimately contributing to the development of safer and more effective therapies.

Data Availability

Click to copy section linkSection link copied!

All data are available from the corresponding author upon request.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.5c04058.

Return loss curves for the chip probe, Simulated B1+ fields of ²H and ¹H coils, Rate constants from dDNP-NMR measurements in microfluidic chips not normalized to cell count (PDF)

ac5c04058_si_001.pdf (668.42 kb)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!

Corresponding Author
- Pernille Rose Jensen - Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark; https://orcid.org/0000-0003-4359-848X; Email: [email protected]
Authors
- Thomas B. Wareham Mathiassen - Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
- Juan D. Sánchez-Heredia - Department of Information Technologies and Communications, Technical University of Cartagena (UPCT), 302020 Cartagena, Spain
- Ke-Chuan Wang - Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
- Cajsa R. Haupt - Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
- Magnus Karlsson - Center for Hyperpolarization in Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Ørsteds Plads 349, 2800 Kgs. Lyngby, Denmark
- Alexander Jönsson - Department of Micro and Nanotechnology, Technical University of Denmark, 2800 Kgs Lyngby, Denmark
- Martin Dufva - Department of Micro and Nanotechnology, Technical University of Denmark, 2800 Kgs Lyngby, Denmark; https://orcid.org/0000-0001-5449-0189
- Roland Thuenauer - Technology Platform Light Microscopy (TPLM), University of Hamburg (UHH); Advanced Light and Fluorescence Microscopy (ALFM) Facility, Centre for Structural Systems Biology (CSSB) Hamburg; Leibniz Institute of Virology (LIV), Notkestrasse 85, 22607 Hamburg, Germany
Notes
The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

This research was funded by the Danish National Research Foundation (grant DNRF124), EU Interreg Öresund-Kattegat-Skagerrak project “Hanseatic Life Science Research Infrastructure Consortium” (HALRIC) and the Novo Nordisk Foundation (infrastructure grant NNF19OC0055825). JDSH contribution to this work was supported by Fundación Séneca (22401/SF/23).

References

Click to copy section linkSection link copied!

This article references 54 other publications.

1
Preclinical CRO Market Size, Share & Growth | Report [2030]. https://www.fortunebusinessinsights.com/preclinical-cro-market-106939 (accessed 2024–07–22).
Google Scholar
There is no corresponding record for this reference.
2
Helmlinger, G.; Al-Huniti, N.; Aksenov, S.; Peskov, K.; Hallow, K. M.; Chu, L.; Boulton, D.; Eriksson, U.; Hamrén, B.; Lambert, C.; Masson, E.; Tomkinson, H.; Stanski, D. Drug–disease modeling in the pharmaceutical industry─Where mechanistic systems pharmacology and statistical pharmacometrics meet. Eur. J. Pharm. Sci. 2017, 109, S39– S46, DOI: 10.1016/j.ejps.2017.05.028
Google Scholar
There is no corresponding record for this reference.
3
Kapałczyńska, M.; Kolenda, T.; Przybyła, W.; Zajączkowska, M.; Teresiak, A.; Filas, V.; Ibbs, M.; Bliźniak, R.; Łuczewski, Ł.; Lamperska, K. 2D and 3D cell cultures─A comparison of different types of cancer cell cultures. Arch. Med. Sci. 2018, 14 (4), 910– 919, DOI: 10.5114/aoms.2016.63743
Google Scholar
There is no corresponding record for this reference.
4
Zingales, V.; Esposito, M. R.; Torriero, N.; Taroncher, M.; Cimetta, E.; Ruiz, M. J. The growing importance of three-dimensional models and microphysiological systems in the assessment of mycotoxin toxicity. Toxins 2023, 15 (7), 422, DOI: 10.3390/toxins15070422
Google Scholar
There is no corresponding record for this reference.
5
Singh, D.; Mathur, A.; Arora, S.; Roy, S.; Mahindroo, N. Journey of organ-on-a-chip technology and its role in future healthcare scenario. Appl. Surf. Sci. Adv. 2022, 9, 100246, DOI: 10.1016/j.apsadv.2022.100246
Google Scholar
There is no corresponding record for this reference.
6
Moro, L. G.; Guarnier, L. P.; Azevedo, M. F.; Fracasso, J. A. R.; Lucio, M. A.; Castro, M. V. d.; Dias, M. L.; Lívero, F. A. d. R.; Ribeiro-Paes, J. T. A brief history of cell culture: From Harrison to organs-on-a-chip. Cells 2024, 13 (24), 2068, DOI: 10.3390/cells13242068
Google Scholar
There is no corresponding record for this reference.
7
Huh, D.; Matthews, B. D.; Mammoto, A.; Montoya-Zavala, M.; Hsin, H. Y.; Ingber, D. E. Reconstituting organ-level lung functions on a chip. Science 2010, 328 (5986), 1662– 1668, DOI: 10.1126/science.1188302
Google Scholar
There is no corresponding record for this reference.
8
Yan, J.; Li, Z.; Guo, J.; Liu, S.; Guo, J. Organ-on-a-chip: A new tool for in vitro research. Biosens. Bioelectron. 2022, 216, 114626, DOI: 10.1016/j.bios.2022.114626
Google Scholar
There is no corresponding record for this reference.
9
Ingber, D. E. Human organs-on-chips for disease modelling, drug development and personalized medicine. Nat. Rev. Genet. 2022, 23 (8), 467– 491, DOI: 10.1038/s41576-022-00466-9
Google Scholar
There is no corresponding record for this reference.
10
Ma, C.; Peng, Y.; Li, H.; Chen, W. Organ-on-a-chip: A new paradigm for drug development. Trends Pharmacol. Sci. 2021, 42 (2), 119– 133, DOI: 10.1016/j.tips.2020.11.009
Google Scholar
There is no corresponding record for this reference.
11
Leung, C. M.; de Haan, P.; Ronaldson-Bouchard, K.; Kim, G. A.; Ko, J.; Rho, H. S.; Chen, Z.; Habibovic, P.; Jeon, N. L.; Takayama, S.; Shuler, M. L.; Vunjak-Novakovic, G.; Frey, O.; Verpoorte, E.; Toh, Y. C. A guide to the organ-on-a-chip. Nat. Rev. Methods Primers 2022, 2, 33, DOI: 10.1038/s43586-022-00118-6
Google Scholar
There is no corresponding record for this reference.
12
Dorst, J.; Weydt, P.; Brenner, D. Metabolic alterations precede neurofilament changes in presymptomatic ALS gene carriers. EBioMedicine 2023, 90, 104521, DOI: 10.1016/j.ebiom.2023.104521
Google Scholar
There is no corresponding record for this reference.
13
Tang, C. C.; Poston, K. L.; Dhawan, V.; Eidelberg, D. Abnormalities in metabolic network activity precede the onset of motor symptoms in Parkinson’s disease. J. Neurosci. 2010, 30 (3), 1049– 1056, DOI: 10.1523/JNEUROSCI.4188-09.2010
Google Scholar
There is no corresponding record for this reference.
14
Duarte, J. M. N.; Schuck, P. F.; Wenk, G. L.; Ferreira, G. C. Metabolic disturbances in diseases with neurological involvement. Aging Dis 2014, 5 (4), 238– 255, DOI: 10.14336/AD.2014.0500238
Google Scholar
There is no corresponding record for this reference.
15
Hu, S.; Balakrishnan, A.; Bok, R. A. 13C-pyruvate imaging reveals alterations in glycolysis that precede c-Myc-induced tumor formation and regression. Cell Metab. 2011, 14 (1), 131– 142, DOI: 10.1016/j.cmet.2011.04.012
Google Scholar
There is no corresponding record for this reference.
16
Sánchez-Salazar, M. G. Continuous inline monitoring of glucose in an organ-on-chip using FreeStyle Libre glucometers. Front. Lab Chip Technol. 2024, 3, 1289437, DOI: 10.3389/frlct.2024.1289437
Google Scholar
There is no corresponding record for this reference.
17
Zirath, H. Every breath you take: Non-invasive real-time oxygen biosensing in two- and three-dimensional microfluidic cell models. Front. Physiol. 2018, 9, 815, DOI: 10.3389/fphys.2018.00815
Google Scholar
There is no corresponding record for this reference.
18
Prill, S. Real-time monitoring of oxygen uptake in hepatic bioreactors shows mitochondrial toxicity. Arch. Toxicol. 2016, 90 (5), 1181– 1191, DOI: 10.1007/s00204-015-1537-2
Google Scholar
There is no corresponding record for this reference.
19
Weltin, A. Cell culture monitoring for drug screening and cancer research. Lab Chip 2014, 14 (1), 138– 146, DOI: 10.1039/C3LC50759A
Google Scholar
There is no corresponding record for this reference.
20
Bavli, D. Real-time monitoring of metabolic function in liver-on-chip microdevices. Proc. Natl. Acad. Sci. U.S.A. 2016, 113 (16), E2231– E2240, DOI: 10.1073/pnas.1522556113
Google Scholar
There is no corresponding record for this reference.
21
Dornhof, J.; Kieninger, J.; Muralidharan, H.; Maurer, J.; Urban, G. A.; Weltin, A. Microfluidic organ-on-chip system for multi-analyte monitoring of metabolites in 3D cell cultures. Lab Chip 2022, 22 (2), 225– 239, DOI: 10.1039/D1LC00689D
Google Scholar
There is no corresponding record for this reference.
22
Ferrari, E.; Palma, C.; Vesentini, S.; Occhetta, P.; Rasponi, M. Integrating biosensors in organs-on-chip devices: A perspective on current strategies to monitor microphysiological systems. Biosensors 2020, 10 (9), 110, DOI: 10.3390/bios10090110
Google Scholar
There is no corresponding record for this reference.
23
Izadifar, Z.; Charrez, B.; Almeida, M.; Robben, S.; Pilobello, K.; van der Graaf-Mas, J.; Marquez, S. L.; Ferrante, T. C.; Shcherbina, K.; Gould, R.; LoGrande, N. T.; Sesay, A. M.; Ingber, D. E. Organ chips with integrated multifunctional sensors enable continuous metabolic monitoring at controlled oxygen levels. Biosens. Bioelectron. 2024, 265, 116683, DOI: 10.1016/j.bios.2024.116683
Google Scholar
There is no corresponding record for this reference.
24
Henry, O. Y. F.; Villenave, R.; Cronce, M. J.; Leineweber, W. D.; Benz, M. A.; Ingber, D. E. Organs-on-chips with integrated electrodes for transepithelial electrical resistance measurements of human epithelial barrier function. Lab Chip 2017, 17 (13), 2264– 2271, DOI: 10.1039/C7LC00155J
Google Scholar
There is no corresponding record for this reference.
25
Ardenkjær-Larsen, J. H.; Fridlund, B.; Gram, A.; Hansson, G.; Hansson, L.; Lerche, M. H.; Servin, R.; Thaning, M.; Golman, K. Increase in signal-to-noise ratio of > 10,000 times in liquid-state NMR. Proc. Natl. Acad. Sci. U.S.A. 2003, 100 (18), 10158– 10163, DOI: 10.1073/pnas.1733835100
Google Scholar
There is no corresponding record for this reference.
26
Eills, J.; Hale, W.; Utz, M. Synergies between hyperpolarized NMR and microfluidics: A review. Prog. Nucl. Magn. Reson. Spectrosc. 2022, 128, 44– 69, DOI: 10.1016/j.pnmrs.2021.09.001
Google Scholar
There is no corresponding record for this reference.
27
Barker, S. J.; Dagys, L.; Levitt, M. H.; Utz, M. Efficient parahydrogen-induced ̂13C hyperpolarization on a microfluidic device. J. Am. Chem. Soc. 2024, 146 (27), 18379– 18386, DOI: 10.1021/jacs.4c03271
Google Scholar
There is no corresponding record for this reference.
28
Rogers, G.; Barker, S.; Sharma, M.; Khakoo, S.; Utz, M. Operando NMR metabolomics of a microfluidic cell culture. J. Magn. Reson. 2023, 349, 107405, DOI: 10.1016/j.jmr.2023.107405
Google Scholar
There is no corresponding record for this reference.
29
Mangas-Florencio, L.; Herrero-Gómez, A.; Eills, J.; Azagra, M.; Batlló-Rius, M.; Marco-Rius, I. A DIY bioreactor for in situ metabolic tracking in 3D cell models via hyperpolarized 13C NMR spectroscopy. Anal. Chem. 2025, 97 (3), 1594– 1602, DOI: 10.1021/acs.analchem.4c04183
Google Scholar
There is no corresponding record for this reference.
30
Mathiassen, T. B. W.; Høgh, A. E.; Karlsson, M.; Katsikis, S.; Wang, K.; Pennestri, M.; Ardenkjær-Larsen, J. H.; Jensen, P. R. Hyperpolarized 13C NMR for longitudinal in-cell metabolism using a mobile 3D cell culture system. J. Magn. Reson. Open 2023, 16–17, 100131, DOI: 10.1016/j.jmro.2023.100131
Google Scholar
There is no corresponding record for this reference.
31
Tan, H. Y.; Trier, S.; Rahbek, U. L.; Dufva, M.; Kutter, J. P.; Andresen, T. L. A multi-chamber microfluidic intestinal barrier model using Caco-2 cells for drug transport studies. PLoS One 2018, 13 (5), e0197101 DOI: 10.1371/journal.pone.0197101
Google Scholar
There is no corresponding record for this reference.
32
Harris, T.; Eliyahu, G.; Frydman, L.; Degani, H. Kinetics of hyperpolarized ̂13C₁-pyruvate transport and metabolism in living human breast cancer cells. Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (43), 18131– 18136, DOI: 10.1073/pnas.0909049106
Google Scholar
There is no corresponding record for this reference.
33
Christensen, N. V.; Holm, R.; Sanchez, J. D.; Hansen, E. S. S.; Lerche, M. H.; Ardenkjær-Larsen, J. H.; Laustsen, C.; Bertelsen, L. B. A continuous-flow bioreactor system for high-throughput hyperpolarized metabolic flux analysis. NMR Biomed 2024, 37 (5), e5107 DOI: 10.1002/nbm.5107
Google Scholar
There is no corresponding record for this reference.
34
Christensen, N. V.; Laustsen, C.; Bertelsen, L. B. Differentiating leukemia subtypes based on metabolic signatures using hyperpolarized ̂13C NMR. NMR Biomed 2024, 37 (12), e5264 DOI: 10.1002/nbm.5264
Google Scholar
There is no corresponding record for this reference.
35
Massin, C.; Vincent, F.; Homsy, A.; Ehrmann, K.; Boero, G.; Besse, P. A.; Daridon, A.; Verpoorte, E.; de Rooij, N. F.; Popovic, R. S. Planar microcoil-based microfluidic NMR probes. J. Magn. Reson. 2003, 164 (2), 242– 255, DOI: 10.1016/S1090-7807(03)00151-4
Google Scholar
There is no corresponding record for this reference.
36
Wensink, H.; Benito-Lopez, F.; Hermes, D. C.; Verboom, W.; Gardeniers, H. J. G. E.; Reinhoudt, D. N.; van den Berg, A. Measuring reaction kinetics in a lab-on-a-chip by microcoil NMR. Lab Chip 2005, 5 (3), 280– 284, DOI: 10.1039/b414832k
Google Scholar
There is no corresponding record for this reference.
37
Liu, Y.; Mi, Y.; Mueller, T.; Kreibich, S.; Williams, E. G.; van Drogen, A.; Borel, C.; Frank, M.; Germain, P. L.; Bludau, I. Multi-omic measurements of heterogeneity in HeLa cells across laboratories. Nat. Biotechnol. 2019, 37 (3), 314– 322, DOI: 10.1038/s41587-019-0037-y
Google Scholar
There is no corresponding record for this reference.
38
Mason, J. A.; Davison-Versagli, C. A.; Leliaert, A. K.; Pape, D. J.; McCallister, C.; Zuo, J.; Durbin, S. M.; Buchheit, C. L.; Zhang, S.; Schafer, Z. T. Oncogenic Ras differentially regulates metabolism and anoikis in extracellular matrix-detached cells. Cell Death Differ. 2016, 23 (8), 1271– 1282, DOI: 10.1038/cdd.2016.15
Google Scholar
There is no corresponding record for this reference.
39
Buchheit, C. L.; Rayavarapu, R. R.; Schafer, Z. T. The regulation of cancer cell death and metabolism by extracellular matrix attachment. Semin. Cell Dev. Biol. 2012, 23 (4), 402– 411, DOI: 10.1016/j.semcdb.2012.04.007
Google Scholar
There is no corresponding record for this reference.
40
Schafer, Z. T.; Grassian, A. R.; Song, L.; Jiang, Z.; Gerhart-Hines, Z.; Irie, H. Y.; Gao, S.; Puigserver, P.; Brugge, J. S. Antioxidant and oncogene rescue of metabolic defects caused by loss of matrix attachment. Nature 2009, 461 (7260), 109– 113, DOI: 10.1038/nature08268
Google Scholar
There is no corresponding record for this reference.
41
Avivar-Valderas, A.; Salas, E.; Bobrovnikova-Marjon, E.; Diehl, J. A.; Nagi, C.; Debnath, J.; Aguirre-Ghiso, J. A. PERK integrates autophagy and oxidative stress responses to promote survival during extracellular matrix detachment. Mol. Cell. Biol. 2011, 31 (17), 3616– 3629, DOI: 10.1128/MCB.05164-11
Google Scholar
There is no corresponding record for this reference.
42
Hawk, M. A.; Schafer, Z. T. Mechanisms of redox metabolism and cancer cell survival during extracellular matrix detachment. J. Biol. Chem. 2018, 293 (20), 7531– 7538, DOI: 10.1074/jbc.TM117.000260
Google Scholar
There is no corresponding record for this reference.
43
Takakusagi, Y.; Takakusagi, K.; Inoue, K.; Saito, K.; Homma, Y.; Ichikawa, K. Hyperpolarized [1–13C]pyruvate NMR spectroscopy reveals transition of tumor energy metabolism in microscale multicellular spheroids. Sci. Rep. 2025, 15 (1), 3454, DOI: 10.1038/s41598-025-03454-1
Google Scholar
There is no corresponding record for this reference.
44
Nielsen, N.; Laustsen, C.; Bertelsen, L. B. ̂13C dynamic nuclear polarization for measuring metabolic flux in endothelial progenitor cells. Exp. Cell Res. 2016, 349 (1), 95– 100, DOI: 10.1016/j.yexcr.2016.10.002
Google Scholar
There is no corresponding record for this reference.
45
Harris, T.; Degani, H.; Frydman, L. Hyperpolarized 13C NMR studies of glucose metabolism in living breast cancer cell cultures. NMR Biomed 2013, 26 (12), 1831– 1843, DOI: 10.1002/nbm.3024
Google Scholar
There is no corresponding record for this reference.
46
Sriram, R.; Nguyen, J.; Santos, J. D. L.; Nguyen, L.; Sun, J.; Vigneron, S.; van Criekinge, M.; Kurhanewicz, J.; MacKenzie, J. D. Molecular detection of inflammation in cell models using hyperpolarized ̂13C-pyruvate. Theranostics 2018, 8 (12), 3400– 3411, DOI: 10.7150/thno.24322
Google Scholar
There is no corresponding record for this reference.
47
Keshari, K. R.; Kurhanewicz, J.; Jeffries, R. E.; Wilson, D. M.; Dewar, B. J.; van Criekinge, M.; Zierhut, M.; Vigneron, D. B.; Macdonald, J. M. Hyperpolarized ̂13C spectroscopy and an NMR-compatible bioreactor system for the investigation of real-time cellular metabolism. Magn. Reson. Med. 2010, 63 (2), 322– 329, DOI: 10.1002/mrm.22225
Google Scholar
There is no corresponding record for this reference.
48
Thuenauer, R.; Nicklaus, S.; Frensch, M.; Troendle, K.; Madl, J.; Römer, W. A microfluidic biochip for locally confined stimulation of cells within an epithelial monolayer. RSC Adv. 2018, 8 (14), 7839– 7846, DOI: 10.1039/C7RA11943G
Google Scholar
There is no corresponding record for this reference.
49
Thuenauer, R.; Rodriguez-Boulan, E.; Römer, W. Microfluidic approaches for epithelial cell layer culture and characterisation. Analyst 2014, 139 (13), 3206– 3218, DOI: 10.1039/C4AN00056K
Google Scholar
There is no corresponding record for this reference.
50
Kim, H. J.; Ingber, D. E. Gut-on-a-chip microenvironment induces human intestinal cells to undergo villus differentiation. Integr. Biol. 2013, 5 (9), 1130– 1140, DOI: 10.1039/c3ib40126j
Google Scholar
There is no corresponding record for this reference.
51
Chung, H. H.; Mireles, M.; Kwarta, B. J.; Gaborski, T. R. Use of porous membranes in tissue barrier and co-culture models. Lab Chip 2018, 18 (12), 1671– 1689, DOI: 10.1039/C7LC01248A
Google Scholar
There is no corresponding record for this reference.
52
Maschmeyer, I.; Lorenz, A. K.; Schimek, K.; Hasenberg, T.; Ramme, A. P.; Hübner, J.; Lindner, M.; Drewell, C.; Bauer, S.; Thomas, A.; Sambo, N. S.; Sonntag, F.; Lauster, R.; Marx, U. A four-organ-chip for interconnected long-term co-culture of human intestine, liver, skin, and kidney equivalents. Lab Chip 2015, 15 (12), 2688– 2699, DOI: 10.1039/C5LC00392J
Google Scholar
There is no corresponding record for this reference.
53
Ferruzza, S.; Rossi, C.; Scarino, M. L.; Sambuy, Y. A protocol for differentiation of human intestinal Caco-2 cells in asymmetric serum-containing medium. Toxicol. In Vitro 2012, 26 (8), 1252– 1255, DOI: 10.1016/j.tiv.2012.01.008
Google Scholar
There is no corresponding record for this reference.
54
Kopp, B.; Khawam, A.; Di Perna, K.; Lenart, D.; Vinette, M.; Silva, R.; Zanoni, T. B.; Rore, C.; Guenigault, G.; Richardson, E.; Kostrzewski, T.; Boswell, A.; Van, P.; Valentine, C.; Salk, J.; Hamel, A. Liver-on-chip model and application in predictive genotoxicity and mutagenicity of drugs. Mutat. Res. Genet. Toxicol. Environ. Mutagen. 2024, 896, 503762, DOI: 10.1016/j.mrgentox.2024.503762
Google Scholar
There is no corresponding record for this reference.

Cited By

Click to copy section linkSection link copied!

This article has not yet been cited by other publications.

Get e-Alerts

Analytical Chemistry

Cite this: Anal. Chem. 2026, 98, 3, 1891–1900

Click to copy citationCitation copied!

https://doi.org/10.1021/acs.analchem.5c04058

Published January 14, 2026

Request reuse permissions

Article Views

Altmetric

Citations

Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

Abstract
High Resolution Image
Download MS PowerPoint Slide
Figure 1
Figure 1. RF Coil design. (A) exploded view of the NMR probe head, showing the two RF coils, the 3D-printed support, and the microfluidic chip. (B) Cross-section (Z = 0 plane) of the simulated B₁⁺ field distribution generated by Coil 1 at the ¹³C frequency (per 1 W accepted power). (C) Photograph of the fabricated NMR probe with the chip inserted.
High Resolution Image
Download MS PowerPoint Slide
Figure 2
Figure 2. Fluidic chip production and assembly. (A) visual representation of the chip components and their assembly process. The stacked components are shown in order from top to bottom: 1 mm PETG, 0.75 mm SEBS, track-etched polycarbonate membrane (8 μm pore size, 18 μm thickness, 5% porosity), 0.25 mm SEBS, and 0.3 mm PETG. (B) 3D-printed inlets/injection ports, designed using SolidWorks, shown with female mini Luer connectors made from BioMed clear resin. (C) Fully assembled microfluidic chip.
High Resolution Image
Download MS PowerPoint Slide
Figure 3
Figure 3. Performance evaluation. (A) single pyruvate spectrum, SNR of the highest pyruvate peak was 5000, (B) substrate integral over time (n = 3) and (C) comparative substrate integrals (AUC = 24 ± 0.6) following injection of 10 mM hyperpolarized [1-¹³C] pyruvate, CV for the AUC was 2.5%. T₁ of pyruvate was 52 ± 2 s.
High Resolution Image
Download MS PowerPoint Slide
Figure 4
Figure 4. One-chamber chip cell growth. (A) one-chamber chip seeded with 3 × 10⁶ HeLa cells, cultured for 3 h, and imaged with bright-field microscopy (bottom left, BF 20×) and fluorescence microscopy of cells stained with Calcein AM (bottom right, Calcein AM 20×). Scale bar: 100 μm. (B) Cell counts in chips after 3, 24, and 48 h. The mean ± SD cell counts (in millions) were 1.90 ± 0.26 at 3 h, 2.70 ± 0.61 at 24 h, and 4.46 ± 0.69 at 48 h. Statistical analysis (one-way ANOVA) showed a significant difference (p < 0.01) in cell count between 3 and 48 h. N = 3 for each time point.
High Resolution Image
Download MS PowerPoint Slide
Figure 5
Figure 5. One chamber chip hyperpolarized ¹³C NMR. (A) spectrum with the highest lactate signal (spectrum 9), showing pyruvate C2 at 205 ppm, lactate C1 at 182.3 ppm, pyruvate hydrate C1 at 178.4 ppm, and pyruvate C1 at 171 ppm. (SNR of highest lactate peak was approximately 50) (B) time-series spectra displaying the production of [1-¹³C] lactate in one-chamber chips at 3, 24, and 48 h postseeding, following the injection of 10 mM hyperpolarized [1-¹³C] pyruvate. (C) Rate constants normalized to cell counts, yielding mean ± SD values (s^–1) of (4.12 ± 0.53) × 10^–4 at 3 h, (3.76 ± 0.64) × 10^–4 at 24 h, and (4.78 ± 0.53) × 10^–4 at 48 h. Statistical analysis (one-way ANOVA) showed no significant difference (NS) between normalized rate constants. N = 3 for each time point.
High Resolution Image
Download MS PowerPoint Slide
Figure 6
Figure 6. Two-chamber chip cell growth and lactate production. (A) two-chamber chip seeded with 3 × 10⁶ HeLa cells, using a flow rate of 0.200 mL/min in the lower chamber, and no flow in the top chamber, for 24 h at 37 °C and 5% CO₂. Bright-field microscopy image (bottom left, BF 20×) and fluorescence microscopy image of cells stained with Calcein AM (bottom right, Calcein AM 20×). Scale bar: 100 μm. (B) Time-series spectra showing [1-¹³C] lactate accumulation in the two-chamber chip seeded with 3 × 10⁶ HeLa cells, following injection of 10 mM hyperpolarized [1-¹³C] pyruvate. The mean ± SD rate constant for lactate production was 2.55 ± 0.28 × 10^–4 (n = 2).
High Resolution Image
Download MS PowerPoint Slide
References
This article references 54 other publications.
1. 1
  Preclinical CRO Market Size, Share & Growth | Report [2030]. https://www.fortunebusinessinsights.com/preclinical-cro-market-106939 (accessed 2024–07–22).
  There is no corresponding record for this reference.
2. 2
  Helmlinger, G.; Al-Huniti, N.; Aksenov, S.; Peskov, K.; Hallow, K. M.; Chu, L.; Boulton, D.; Eriksson, U.; Hamrén, B.; Lambert, C.; Masson, E.; Tomkinson, H.; Stanski, D. Drug–disease modeling in the pharmaceutical industry─Where mechanistic systems pharmacology and statistical pharmacometrics meet. Eur. J. Pharm. Sci. 2017, 109, S39– S46, DOI: 10.1016/j.ejps.2017.05.028
  There is no corresponding record for this reference.
3. 3
  Kapałczyńska, M.; Kolenda, T.; Przybyła, W.; Zajączkowska, M.; Teresiak, A.; Filas, V.; Ibbs, M.; Bliźniak, R.; Łuczewski, Ł.; Lamperska, K. 2D and 3D cell cultures─A comparison of different types of cancer cell cultures. Arch. Med. Sci. 2018, 14 (4), 910– 919, DOI: 10.5114/aoms.2016.63743
  There is no corresponding record for this reference.
4. 4
  Zingales, V.; Esposito, M. R.; Torriero, N.; Taroncher, M.; Cimetta, E.; Ruiz, M. J. The growing importance of three-dimensional models and microphysiological systems in the assessment of mycotoxin toxicity. Toxins 2023, 15 (7), 422, DOI: 10.3390/toxins15070422
  There is no corresponding record for this reference.
5. 5
  Singh, D.; Mathur, A.; Arora, S.; Roy, S.; Mahindroo, N. Journey of organ-on-a-chip technology and its role in future healthcare scenario. Appl. Surf. Sci. Adv. 2022, 9, 100246, DOI: 10.1016/j.apsadv.2022.100246
  There is no corresponding record for this reference.
6. 6
  Moro, L. G.; Guarnier, L. P.; Azevedo, M. F.; Fracasso, J. A. R.; Lucio, M. A.; Castro, M. V. d.; Dias, M. L.; Lívero, F. A. d. R.; Ribeiro-Paes, J. T. A brief history of cell culture: From Harrison to organs-on-a-chip. Cells 2024, 13 (24), 2068, DOI: 10.3390/cells13242068
  There is no corresponding record for this reference.
7. 7
  Huh, D.; Matthews, B. D.; Mammoto, A.; Montoya-Zavala, M.; Hsin, H. Y.; Ingber, D. E. Reconstituting organ-level lung functions on a chip. Science 2010, 328 (5986), 1662– 1668, DOI: 10.1126/science.1188302
  There is no corresponding record for this reference.
8. 8
  Yan, J.; Li, Z.; Guo, J.; Liu, S.; Guo, J. Organ-on-a-chip: A new tool for in vitro research. Biosens. Bioelectron. 2022, 216, 114626, DOI: 10.1016/j.bios.2022.114626
  There is no corresponding record for this reference.
9. 9
  Ingber, D. E. Human organs-on-chips for disease modelling, drug development and personalized medicine. Nat. Rev. Genet. 2022, 23 (8), 467– 491, DOI: 10.1038/s41576-022-00466-9
  There is no corresponding record for this reference.
10. 10
  Ma, C.; Peng, Y.; Li, H.; Chen, W. Organ-on-a-chip: A new paradigm for drug development. Trends Pharmacol. Sci. 2021, 42 (2), 119– 133, DOI: 10.1016/j.tips.2020.11.009
  There is no corresponding record for this reference.
11. 11
  Leung, C. M.; de Haan, P.; Ronaldson-Bouchard, K.; Kim, G. A.; Ko, J.; Rho, H. S.; Chen, Z.; Habibovic, P.; Jeon, N. L.; Takayama, S.; Shuler, M. L.; Vunjak-Novakovic, G.; Frey, O.; Verpoorte, E.; Toh, Y. C. A guide to the organ-on-a-chip. Nat. Rev. Methods Primers 2022, 2, 33, DOI: 10.1038/s43586-022-00118-6
  There is no corresponding record for this reference.
12. 12
  Dorst, J.; Weydt, P.; Brenner, D. Metabolic alterations precede neurofilament changes in presymptomatic ALS gene carriers. EBioMedicine 2023, 90, 104521, DOI: 10.1016/j.ebiom.2023.104521
  There is no corresponding record for this reference.
13. 13
  Tang, C. C.; Poston, K. L.; Dhawan, V.; Eidelberg, D. Abnormalities in metabolic network activity precede the onset of motor symptoms in Parkinson’s disease. J. Neurosci. 2010, 30 (3), 1049– 1056, DOI: 10.1523/JNEUROSCI.4188-09.2010
  There is no corresponding record for this reference.
14. 14
  Duarte, J. M. N.; Schuck, P. F.; Wenk, G. L.; Ferreira, G. C. Metabolic disturbances in diseases with neurological involvement. Aging Dis 2014, 5 (4), 238– 255, DOI: 10.14336/AD.2014.0500238
  There is no corresponding record for this reference.
15. 15
  Hu, S.; Balakrishnan, A.; Bok, R. A. 13C-pyruvate imaging reveals alterations in glycolysis that precede c-Myc-induced tumor formation and regression. Cell Metab. 2011, 14 (1), 131– 142, DOI: 10.1016/j.cmet.2011.04.012
  There is no corresponding record for this reference.
16. 16
  Sánchez-Salazar, M. G. Continuous inline monitoring of glucose in an organ-on-chip using FreeStyle Libre glucometers. Front. Lab Chip Technol. 2024, 3, 1289437, DOI: 10.3389/frlct.2024.1289437
  There is no corresponding record for this reference.
17. 17
  Zirath, H. Every breath you take: Non-invasive real-time oxygen biosensing in two- and three-dimensional microfluidic cell models. Front. Physiol. 2018, 9, 815, DOI: 10.3389/fphys.2018.00815
  There is no corresponding record for this reference.
18. 18
  Prill, S. Real-time monitoring of oxygen uptake in hepatic bioreactors shows mitochondrial toxicity. Arch. Toxicol. 2016, 90 (5), 1181– 1191, DOI: 10.1007/s00204-015-1537-2
  There is no corresponding record for this reference.
19. 19
  Weltin, A. Cell culture monitoring for drug screening and cancer research. Lab Chip 2014, 14 (1), 138– 146, DOI: 10.1039/C3LC50759A
  There is no corresponding record for this reference.
20. 20
  Bavli, D. Real-time monitoring of metabolic function in liver-on-chip microdevices. Proc. Natl. Acad. Sci. U.S.A. 2016, 113 (16), E2231– E2240, DOI: 10.1073/pnas.1522556113
  There is no corresponding record for this reference.
21. 21
  Dornhof, J.; Kieninger, J.; Muralidharan, H.; Maurer, J.; Urban, G. A.; Weltin, A. Microfluidic organ-on-chip system for multi-analyte monitoring of metabolites in 3D cell cultures. Lab Chip 2022, 22 (2), 225– 239, DOI: 10.1039/D1LC00689D
  There is no corresponding record for this reference.
22. 22
  Ferrari, E.; Palma, C.; Vesentini, S.; Occhetta, P.; Rasponi, M. Integrating biosensors in organs-on-chip devices: A perspective on current strategies to monitor microphysiological systems. Biosensors 2020, 10 (9), 110, DOI: 10.3390/bios10090110
  There is no corresponding record for this reference.
23. 23
  Izadifar, Z.; Charrez, B.; Almeida, M.; Robben, S.; Pilobello, K.; van der Graaf-Mas, J.; Marquez, S. L.; Ferrante, T. C.; Shcherbina, K.; Gould, R.; LoGrande, N. T.; Sesay, A. M.; Ingber, D. E. Organ chips with integrated multifunctional sensors enable continuous metabolic monitoring at controlled oxygen levels. Biosens. Bioelectron. 2024, 265, 116683, DOI: 10.1016/j.bios.2024.116683
  There is no corresponding record for this reference.
24. 24
  Henry, O. Y. F.; Villenave, R.; Cronce, M. J.; Leineweber, W. D.; Benz, M. A.; Ingber, D. E. Organs-on-chips with integrated electrodes for transepithelial electrical resistance measurements of human epithelial barrier function. Lab Chip 2017, 17 (13), 2264– 2271, DOI: 10.1039/C7LC00155J
  There is no corresponding record for this reference.
25. 25
  Ardenkjær-Larsen, J. H.; Fridlund, B.; Gram, A.; Hansson, G.; Hansson, L.; Lerche, M. H.; Servin, R.; Thaning, M.; Golman, K. Increase in signal-to-noise ratio of > 10,000 times in liquid-state NMR. Proc. Natl. Acad. Sci. U.S.A. 2003, 100 (18), 10158– 10163, DOI: 10.1073/pnas.1733835100
  There is no corresponding record for this reference.
26. 26
  Eills, J.; Hale, W.; Utz, M. Synergies between hyperpolarized NMR and microfluidics: A review. Prog. Nucl. Magn. Reson. Spectrosc. 2022, 128, 44– 69, DOI: 10.1016/j.pnmrs.2021.09.001
  There is no corresponding record for this reference.
27. 27
  Barker, S. J.; Dagys, L.; Levitt, M. H.; Utz, M. Efficient parahydrogen-induced ̂13C hyperpolarization on a microfluidic device. J. Am. Chem. Soc. 2024, 146 (27), 18379– 18386, DOI: 10.1021/jacs.4c03271
  There is no corresponding record for this reference.
28. 28
  Rogers, G.; Barker, S.; Sharma, M.; Khakoo, S.; Utz, M. Operando NMR metabolomics of a microfluidic cell culture. J. Magn. Reson. 2023, 349, 107405, DOI: 10.1016/j.jmr.2023.107405
  There is no corresponding record for this reference.
29. 29
  Mangas-Florencio, L.; Herrero-Gómez, A.; Eills, J.; Azagra, M.; Batlló-Rius, M.; Marco-Rius, I. A DIY bioreactor for in situ metabolic tracking in 3D cell models via hyperpolarized 13C NMR spectroscopy. Anal. Chem. 2025, 97 (3), 1594– 1602, DOI: 10.1021/acs.analchem.4c04183
  There is no corresponding record for this reference.
30. 30
  Mathiassen, T. B. W.; Høgh, A. E.; Karlsson, M.; Katsikis, S.; Wang, K.; Pennestri, M.; Ardenkjær-Larsen, J. H.; Jensen, P. R. Hyperpolarized 13C NMR for longitudinal in-cell metabolism using a mobile 3D cell culture system. J. Magn. Reson. Open 2023, 16–17, 100131, DOI: 10.1016/j.jmro.2023.100131
  There is no corresponding record for this reference.
31. 31
  Tan, H. Y.; Trier, S.; Rahbek, U. L.; Dufva, M.; Kutter, J. P.; Andresen, T. L. A multi-chamber microfluidic intestinal barrier model using Caco-2 cells for drug transport studies. PLoS One 2018, 13 (5), e0197101 DOI: 10.1371/journal.pone.0197101
  There is no corresponding record for this reference.
32. 32
  Harris, T.; Eliyahu, G.; Frydman, L.; Degani, H. Kinetics of hyperpolarized ̂13C₁-pyruvate transport and metabolism in living human breast cancer cells. Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (43), 18131– 18136, DOI: 10.1073/pnas.0909049106
  There is no corresponding record for this reference.
33. 33
  Christensen, N. V.; Holm, R.; Sanchez, J. D.; Hansen, E. S. S.; Lerche, M. H.; Ardenkjær-Larsen, J. H.; Laustsen, C.; Bertelsen, L. B. A continuous-flow bioreactor system for high-throughput hyperpolarized metabolic flux analysis. NMR Biomed 2024, 37 (5), e5107 DOI: 10.1002/nbm.5107
  There is no corresponding record for this reference.
34. 34
  Christensen, N. V.; Laustsen, C.; Bertelsen, L. B. Differentiating leukemia subtypes based on metabolic signatures using hyperpolarized ̂13C NMR. NMR Biomed 2024, 37 (12), e5264 DOI: 10.1002/nbm.5264
  There is no corresponding record for this reference.
35. 35
  Massin, C.; Vincent, F.; Homsy, A.; Ehrmann, K.; Boero, G.; Besse, P. A.; Daridon, A.; Verpoorte, E.; de Rooij, N. F.; Popovic, R. S. Planar microcoil-based microfluidic NMR probes. J. Magn. Reson. 2003, 164 (2), 242– 255, DOI: 10.1016/S1090-7807(03)00151-4
  There is no corresponding record for this reference.
36. 36
  Wensink, H.; Benito-Lopez, F.; Hermes, D. C.; Verboom, W.; Gardeniers, H. J. G. E.; Reinhoudt, D. N.; van den Berg, A. Measuring reaction kinetics in a lab-on-a-chip by microcoil NMR. Lab Chip 2005, 5 (3), 280– 284, DOI: 10.1039/b414832k
  There is no corresponding record for this reference.
37. 37
  Liu, Y.; Mi, Y.; Mueller, T.; Kreibich, S.; Williams, E. G.; van Drogen, A.; Borel, C.; Frank, M.; Germain, P. L.; Bludau, I. Multi-omic measurements of heterogeneity in HeLa cells across laboratories. Nat. Biotechnol. 2019, 37 (3), 314– 322, DOI: 10.1038/s41587-019-0037-y
  There is no corresponding record for this reference.
38. 38
  Mason, J. A.; Davison-Versagli, C. A.; Leliaert, A. K.; Pape, D. J.; McCallister, C.; Zuo, J.; Durbin, S. M.; Buchheit, C. L.; Zhang, S.; Schafer, Z. T. Oncogenic Ras differentially regulates metabolism and anoikis in extracellular matrix-detached cells. Cell Death Differ. 2016, 23 (8), 1271– 1282, DOI: 10.1038/cdd.2016.15
  There is no corresponding record for this reference.
39. 39
  Buchheit, C. L.; Rayavarapu, R. R.; Schafer, Z. T. The regulation of cancer cell death and metabolism by extracellular matrix attachment. Semin. Cell Dev. Biol. 2012, 23 (4), 402– 411, DOI: 10.1016/j.semcdb.2012.04.007
  There is no corresponding record for this reference.
40. 40
  Schafer, Z. T.; Grassian, A. R.; Song, L.; Jiang, Z.; Gerhart-Hines, Z.; Irie, H. Y.; Gao, S.; Puigserver, P.; Brugge, J. S. Antioxidant and oncogene rescue of metabolic defects caused by loss of matrix attachment. Nature 2009, 461 (7260), 109– 113, DOI: 10.1038/nature08268
  There is no corresponding record for this reference.
41. 41
  Avivar-Valderas, A.; Salas, E.; Bobrovnikova-Marjon, E.; Diehl, J. A.; Nagi, C.; Debnath, J.; Aguirre-Ghiso, J. A. PERK integrates autophagy and oxidative stress responses to promote survival during extracellular matrix detachment. Mol. Cell. Biol. 2011, 31 (17), 3616– 3629, DOI: 10.1128/MCB.05164-11
  There is no corresponding record for this reference.
42. 42
  Hawk, M. A.; Schafer, Z. T. Mechanisms of redox metabolism and cancer cell survival during extracellular matrix detachment. J. Biol. Chem. 2018, 293 (20), 7531– 7538, DOI: 10.1074/jbc.TM117.000260
  There is no corresponding record for this reference.
43. 43
  Takakusagi, Y.; Takakusagi, K.; Inoue, K.; Saito, K.; Homma, Y.; Ichikawa, K. Hyperpolarized [1–13C]pyruvate NMR spectroscopy reveals transition of tumor energy metabolism in microscale multicellular spheroids. Sci. Rep. 2025, 15 (1), 3454, DOI: 10.1038/s41598-025-03454-1
  There is no corresponding record for this reference.
44. 44
  Nielsen, N.; Laustsen, C.; Bertelsen, L. B. ̂13C dynamic nuclear polarization for measuring metabolic flux in endothelial progenitor cells. Exp. Cell Res. 2016, 349 (1), 95– 100, DOI: 10.1016/j.yexcr.2016.10.002
  There is no corresponding record for this reference.
45. 45
  Harris, T.; Degani, H.; Frydman, L. Hyperpolarized 13C NMR studies of glucose metabolism in living breast cancer cell cultures. NMR Biomed 2013, 26 (12), 1831– 1843, DOI: 10.1002/nbm.3024
  There is no corresponding record for this reference.
46. 46
  Sriram, R.; Nguyen, J.; Santos, J. D. L.; Nguyen, L.; Sun, J.; Vigneron, S.; van Criekinge, M.; Kurhanewicz, J.; MacKenzie, J. D. Molecular detection of inflammation in cell models using hyperpolarized ̂13C-pyruvate. Theranostics 2018, 8 (12), 3400– 3411, DOI: 10.7150/thno.24322
  There is no corresponding record for this reference.
47. 47
  Keshari, K. R.; Kurhanewicz, J.; Jeffries, R. E.; Wilson, D. M.; Dewar, B. J.; van Criekinge, M.; Zierhut, M.; Vigneron, D. B.; Macdonald, J. M. Hyperpolarized ̂13C spectroscopy and an NMR-compatible bioreactor system for the investigation of real-time cellular metabolism. Magn. Reson. Med. 2010, 63 (2), 322– 329, DOI: 10.1002/mrm.22225
  There is no corresponding record for this reference.
48. 48
  Thuenauer, R.; Nicklaus, S.; Frensch, M.; Troendle, K.; Madl, J.; Römer, W. A microfluidic biochip for locally confined stimulation of cells within an epithelial monolayer. RSC Adv. 2018, 8 (14), 7839– 7846, DOI: 10.1039/C7RA11943G
  There is no corresponding record for this reference.
49. 49
  Thuenauer, R.; Rodriguez-Boulan, E.; Römer, W. Microfluidic approaches for epithelial cell layer culture and characterisation. Analyst 2014, 139 (13), 3206– 3218, DOI: 10.1039/C4AN00056K
  There is no corresponding record for this reference.
50. 50
  Kim, H. J.; Ingber, D. E. Gut-on-a-chip microenvironment induces human intestinal cells to undergo villus differentiation. Integr. Biol. 2013, 5 (9), 1130– 1140, DOI: 10.1039/c3ib40126j
  There is no corresponding record for this reference.
51. 51
  Chung, H. H.; Mireles, M.; Kwarta, B. J.; Gaborski, T. R. Use of porous membranes in tissue barrier and co-culture models. Lab Chip 2018, 18 (12), 1671– 1689, DOI: 10.1039/C7LC01248A
  There is no corresponding record for this reference.
52. 52
  Maschmeyer, I.; Lorenz, A. K.; Schimek, K.; Hasenberg, T.; Ramme, A. P.; Hübner, J.; Lindner, M.; Drewell, C.; Bauer, S.; Thomas, A.; Sambo, N. S.; Sonntag, F.; Lauster, R.; Marx, U. A four-organ-chip for interconnected long-term co-culture of human intestine, liver, skin, and kidney equivalents. Lab Chip 2015, 15 (12), 2688– 2699, DOI: 10.1039/C5LC00392J
  There is no corresponding record for this reference.
53. 53
  Ferruzza, S.; Rossi, C.; Scarino, M. L.; Sambuy, Y. A protocol for differentiation of human intestinal Caco-2 cells in asymmetric serum-containing medium. Toxicol. In Vitro 2012, 26 (8), 1252– 1255, DOI: 10.1016/j.tiv.2012.01.008
  There is no corresponding record for this reference.
54. 54
  Kopp, B.; Khawam, A.; Di Perna, K.; Lenart, D.; Vinette, M.; Silva, R.; Zanoni, T. B.; Rore, C.; Guenigault, G.; Richardson, E.; Kostrzewski, T.; Boswell, A.; Van, P.; Valentine, C.; Salk, J.; Hamel, A. Liver-on-chip model and application in predictive genotoxicity and mutagenicity of drugs. Mutat. Res. Genet. Toxicol. Environ. Mutagen. 2024, 896, 503762, DOI: 10.1016/j.mrgentox.2024.503762
  There is no corresponding record for this reference.
Supporting Information
Supporting Information
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.5c04058.
- Return loss curves for the chip probe, Simulated B1+ fields of ²H and ¹H coils, Rate constants from dDNP-NMR measurements in microfluidic chips not normalized to cell count (PDF)
- ac5c04058_si_001.pdf (668.42 kb)
Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

ChipNMR: Hyperpolarized NMR for Noninvasive Metabolic Flux Analysis in Perfused Microfluidic ChipsClick to copy article linkArticle link copied!

Analytical Chemistry

Abstract

Introduction

Experimental Section

RF Coil Design and Creation

Figure 1

Chip Fabrication

Figure 2

Chip Sterilization and Treatment

Cell Culturing and Chip Seeding

Cell Staining and Microscopy

dDNP-NMR Experiments

Cell Counting

Data Analysis

Statistical Analysis

Results and Discussion

NMR Coil Design and Functionality

Figure 3

Microfluidic Chip as a Cell Culturing Device

Figure 4

Detection of Metabolic Activity via dDNP-NMR

Figure 5

Comparative Analysis of One-Chamber and Two-Chamber Chips

Figure 6

Conclusion

Data Availability

Supporting Information

Terms & Conditions

Author Information

Acknowledgments

References

Cited By

Analytical Chemistry

Article Views

Altmetric

Citations

Recommended Articles

Abstract

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

References

Supporting Information

Supporting Information

Terms & Conditions

ChipNMR: Hyperpolarized NMR for Noninvasive Metabolic Flux Analysis in Perfused Microfluidic Chips
Click to copy article linkArticle link copied!