The application of electrochemical techniques for providing insight into biological processes has become common practice over the last several decades. The study of cellular metabolism, largely ignored by the fields of molecular biology and toxicology until now [1, 2], is a vast field exploring how cellular energetics respond to internal or external influences, as well as how the metabolic state of the cell influences cellular regulation . Traditionally, metabolism has been studied by following the uptake of radioactive metabolites [4, 5] or by quantifying consumption or production of analytes in flasks containing millions of cells [6–8]. Designing electrochemical experiments to study cellular metabolism requires extensive considerations of the cellular environment and sensor design. Many biosensors designed today are intended for point-of-care measures, where a sample of blood, urine, or cerebral spinal fluid is obtained and diluted in a buffer that allows for optimal electrode performance and a single measurement performed. When studying the real-time metabolism of living cells, no alteration of the cellular environment can be performed; thus the sensor must be capable of operating in these environments. A number of electrochemical methods have been developed and applied to advance the study of metabolism [7, 9–13]. One of these methods, multi-analyte microphysiometry, will be presented as an example of a successful method for investigations of cellular metabolism. This review will discuss current techniques employed in microphysiometry, including the application of the instrument to the study of ischemic neurons , the contributions of insulin and glucose toward cellular energetic states  and the investigation of the metabolic effects of protein toxins .
Amperometric glucose, lactate and oxygen detection is performed with a multi-chamber bipotentiostat and related LabVIEW software designed by the Vanderbilt Institute for Integrative Biosystems Research and Education (VIIBRE) which enables monitoring of multiple analytes in four to eight chambers simultaneously, a feat not currently possible with commercial potentiostats. In the MAMP, oxygen is detected at −0.45 V versus Ag/AgCl (2 M KCl) through direct reduction at a 127 μm diameter platinum electrode modified with Nafion. Two 0.5 mm diameter Pt electrodes are coated with glucose oxidase (GOx) and lactate oxidase (LOx) films for indirect detection through the oxidation of hydrogen peroxide produced by their respective immobilized oxidase enzymes. A low-buffered (1 mM PO4 3−) and low-glucose (5 mM) RPMI 1640 is used in most experiments, where low buffering allows for detectable changes in extracellular pH, and low glucose reduces the background against which consumption must be measured, as well as conforms to physiologically relevant glucose conditions in vivo.
To further enhance the sensitivity of the MAMP, as well as enable calculation of metabolic rates, a flow/stop-flow pattern (80 s on, 40 s off) is employed to measure real-time consumption and accumulation of the cellular metabolites. During each flow period, the current or potential at each electrode reaches a steady state. When flow is halted, depletion of glucose and oxygen in the chamber is observed as a decrease in current magnitude and accumulation of lactate is measured as an increase in current magnitude (Fig. 1b). Accumulation of extracellular acid is measured as a decrease in open circuit potential over time, and the extracellular acidification rate (ECAR) is calculated as -μV/s. Rates are calculated for each stop-flow period, allowing for monitoring of changes in metabolic rate in response to stimuli over the course of several hours. By accounting for the microfluidic volume, the number of cells present and the length of the stop-flow period, metabolic rates for each stop-flow period can be calculated in terms of mol•cell−1•s−1 and can be compared to rates calculated using traditional methods [6, 16]. Further detail of this process , as well as mathematical modelling , has been previously discussed.
Discrimination of Select List Agents
One of the early investigations of cellular metabolism in the MAMP concerned the discrimination of select list agents based on resulting changes in cellular metabolism upon exposure. The metabolic profiles of neuroblastoma, macrophage, and kidney cell lines exposed to ricin, botulinum neurotoxin, cholera toxin (CTx), muscarine, alamethicin and anthrax protective antigen (PA) were performed , and each presented a different metabolic profile. For example, macrophages treated with alamethicin and PA, both pore-forming agents, exhibited distinguishable changes in metabolism. Treatment with alamethicin leads to rapid loss of lacate and acid release related to cellular death and PA treatment reflects uncoupling of lactate and acid production, possibly from the intracellular sequestration of protons in lysosomes. The individual metabolic profiles of each agent suggest that the MAMP could be a useful tool in discrimination of select list agents in unknown samples.
Metabolic Response of Neurons to Cholera Toxin Exposure
The metabolic effects of CTx were further explored in the MAMP to determine the metabolic response of several cell types to CTx . Although CTx exposure is typically studied with intestinal cell lines, the largest metabolic response was found using a neuronal cell line, PC-12. Previous electrophysiology studies of intestinal cell lines exposed to CTx revealed a secretion of chloride ions from the cell beginning roughly 30 min to an hour after exposure, with the lag-time attributed to the time required for the toxin to transverse the cell and activate continuous cAMP production . Using the MAMP to investigate CTx exposure in PC-12 neurons, significant increases of extracellular acidification and lactate production, as well as decreases in oxygen consumption, were seen in as little as 10 min. Additional use of inhibitors and other effectors which target specific cellular functions relavant to CTx exposure demonstrated that the observed metabolic response was due to toxin binding and cAMP activation instead of some downsteam effect of toxin entry or transport within the cell. Additionally, these experiments show that the MAMP is a useful tool for determining the acute metabolic effects of biological toxins in real-time and for exploring the signaling pathways triggered by exposure to a toxin or other agents to determine the root cause for metabolic disruption.
In one study, CHO cells were exposed to chromate, a potent environmental toxin which enters the cytoplasm and interferes with redox enzymes . Ultimately, CHO cells exhibited a rapid and drastic decrease in lactate production, followed by a slower decrease in glucose consumption and acid production. This result could be due to an immediate inhibition of lactate dehydrogenase, halting anaerobic respiration and lowering lactate production, followed by irreversible cell damage which reduced the metabolic rate of glucose consumption and acid production. Preliminary work has also be performed to investigate the metabolic response of cells to pesticides, such as parathion. The MAMP can be used to probe the metabolic disruption caused by exposure to environmental toxins.
Metabolic Compensation of Ischemic Neurons
Stroke is the second leading cause of death and the largest cause of long-term adult disability, with a majority of strokes being ischemic in nature, where a loss of oxygenated blood flow results in neuronal death . The MAMP was utilized in the development of a new model of transient ischemic attack (TIA), also described as “mini-strokes,” where primary neurons given short 5 min deprivations of oxygen and glucose in the days leading up to an otherwise lethal 90 min deprivation achieved increased survival rates over those receiving only the lethal deprivation . Traditional toxicology performed in this study revealed an increase in cellular ATP 24 h after each nonlethal 5 min deprivation and a total loss of ATP production in neurons deprived for 90 min, indicating that cellular metabolism may play a role in the protective mechanisms occurring in TIA. This was confirmed in the MAMP where neuronal oxygen consumption was significantly increased less than 1 h after 5 min glucose deprivation but permanently decreased in neurons receiving the lethal deprivation. Lactate production in nonlethally deprived neurons recovered to control levels within 30 min, while those receiving the lethal deprivation failed to fully recover. The increase in aerobic respiration in neurons receiving nonlethal stress, as indicated by increased oxygen consumption and ATP production, suggests that the protective pathways upregulated in this ischemic model include immediate increased production of energy stores through aerobic respiration, which may help to prevent cellular starvation upon subsequent stress.
Mamp Modification for Insulin Detection for Islet Analysis
Detection of insulin secretion is crucial to the study of diabetes mellitus, particularly for pancreatic islet metabolism. The MAMP sensor head was modified to allow for direct and sensitive detection of insulin released from islets . As discussed in Snider et al., several electrochemical sensors for direct detection of insulin had been created, but due to film thickness and adherence of these films under flow conditions, a new insulin-sensitive electrode was devised and used to obtain a dynamic profile of insulin release from pancreatic islets undergoing glucose stimulation. The typical MAMP sensor head was modified with a 1 mm glassy carbon rod, which was then coated in multi-walled carbon nanotubes dispersed in 3,4,-dihydro-2H-pyran (DHP), resulting in a thin-film insulin sensor capable of measuring changes in insulin secretion through electrochemical oxidation of the analyte with a response time on the order of 1 s. Numerous pancreatic islets (75–125) were immobilized within the cellular insert and insulin was measured in the MAMP as islets were perfused with 2.8 mM glucose in Hank’s balanced salt solution (HBSS) and then stimulated with 16.7 mM glucose. The stimulation resulted in a rapid increase of insulin secretion of 228 ± 1 % compared to control islets and demonstrated the feasibility of performing direct, real-time metabolic analysis of islets in the MAMP. Modification of the MAMP for specialized purposes, such as insulin detection, is achievable due the versatility of the sensor head and instrumentation, providing a wealth of opportunities for future work with this instrument.
New MAMP Platform for Detection of Immunoresponse
A novel screen-printed platinum electrode (SPE) containing four modifiable working electrodes was designed and paired with a PDMS microfluidic chamber to create a new MAMP platform. The SPE was initially used in the development of a novel sensor for superoxide (SO), a molecule released by macrophages during an immune response known as oxidative burst . SO has a very short lifetime and is typically detected as its downstream products, such as hydrogen peroxide. By coupling MAMP methods with a new platform and sensor, it was possible to detect real-time superoxide release from RAW macrophages treated with phorbol myristate acetate (PMA), a known promoter of oxidative burst. Upon treatment, a 46 ± 19 % increase in current was measured over a 30 min time period, or an increase of 8.8 ± 3.2 attomoles O2/cell/s over control levels, demonstrating successful detection of sustained macrophage oxidative burst. The SPE MAMP platform can further be modified for the detection of glucose, lactate, oxygen, and acid to allow for detection of oxidative burst in conjunction with cellular immune responses.
The potential applications of multianalyte microphysiometry are numerous, including topics not discussed herein, including further investigations of the metabolic compensation of nutrient-challenged neurons , the investigation of macrophage response to bacterial infections [28–29], as well as the metabolic action of drug targets . The VIIBRE potentiostat has also been modified to add potentiometric detection simultaneously with amperometric detection. This advancement will allow for expansion beyond the LAPS currently used for acid detection, including new sensing platforms capable of microphysiometry techniques such as the SPE. This improvement will also allow for addition of other potentiometric sensors, including ions key to cellular metabolism, such as H+, K+, Ca2+, and Na+, which are in constant flux in cellular physiology . As the sensitivity and durability of electrochemical sensors continues to improve and evolve, these sensors can be applied to increasingly complex studies of cellular metabolism, advancing our understanding of metabolic contributions to cellular physiology.