Real-time measurement of small molecules directly in awake, ambulatory animals

Abstract

The development of a technology capable of tracking the levels of drugs, metabolites, and biomarkers in the body continuously and in real time would advance our understanding of health and our ability to detect and treat disease. It would, for example, enable therapies guided by high-resolution, patient-specific pharmacokinetics (including feedback-controlled drug delivery), opening new dimensions in personalized medicine. In response, we demonstrate here the ability of electrochemical aptamer-based (E-AB) sensors to support continuous, real-time, multihour measurements when emplaced directly in the circulatory systems of living animals. Specifically, we have used E-AB sensors to perform the multihour, real-time measurement of four drugs in the bloodstream of even awake, ambulatory rats, achieving precise molecular measurements at clinically relevant detection limits and high (3 s) temporal resolution, attributes suggesting that the approach could provide an important window into the study of physiology and pharmacokinetics.

Keywords: E-DNA; aptamer; in vivo; precision medicine; square-wave voltammetry.

Fig. 1.

Real-time, continuous measurement of specific drugs directly in the living body. (A) The E-AB sensing platform, in which the binding-induced folding of an electrode-bound, redox-reporter-modified aptamer leads to a change in electron transfer easily detected using square wave voltammetry. (B) A microporous (0.2 µm) polysulfone membrane protects the sensor from fouling by blood cells. (C) The resultant device is small enough to emplace in one of the external jugulars of a rat using an 18-gauge catheter (the cartoon overlay illustrates sensor location). (D) To correct the drift seen in vivo, we record data at two square wave frequencies (here 30 and 240 Hz; optimal values depend on the aptamer used). At one frequency, the sensor’s voltammetric signal increases upon target binding, whereas at the other, it is reduced; taking the difference between the two eliminates drift and enhances signal-to-noise (26). (E) Using drift-corrected E-AB sensors, we have monitored the in vivo concentrations of multiple drugs continuously and in real-time over the course of many hours in measurements that achieve clinically relevant precision and few-second time resolution. Shown here, for example, is the measurement of the antibiotic tobramycin in the blood of an anesthetized rat after two serial injections into the opposite external jugular. (F) At 3 s per measurement, the time resolution of these measurements is sufficient to monitor both the injection itself and the subsequent distribution of the drug within the circulatory system and reflects an orders of magnitude improvement over the resolution of traditional pharmacokinetic methods (45).

Fig. 2.

Continuous molecular measurements in vivo. We have successfully measured multiple drugs using E-AB sensors emplaced in the jugulars of anesthetized rats. (A) Shown here are five i.v. injections of 2 mg/m² of the cancer chemotherapeutic DOX, a dose more than 25 times lower than typical human doses (46). (B–D) To illustrate the generality of the approach, we have also used an aminoglycoside-detecting E-AB sensor to monitor in vivo levels of the antibiotics kanamycin, gentamicin, and tobramycin at the indicated doses. The kanamycin doses used here span the 10–30 mg/kg therapeutic range used in humans (34). For gentamicin, we focus here on two sequential i.v. injections of the drug, separated by a 2-h interval. For tobramycin, we present an overlay of data collected after sequential i.m. (thigh) and i.v. (the external jugular opposite the sensor) injections carried out in a single rat.

Fig. 3.

High-precision pharmacokinetics. Shown are high-resolution pharmacokinetic profiles for the drugs DOX (A) and gentamicin (B) upon i.v. injection of 50 mg/m² and 20 mg/kg doses, respectively. As is easily seen, the resolution of in vivo E-AB sensors is sufficient not only to define the slower β phases of these drugs (red dots) but also to define their much more rapid α phases (blue dots) with excellent statistical significance. These measurements constitute a precise determination of the i.v. distribution phase of a small-molecule drug.

Fig. 4.

The measurement of inter- and intraanimal pharmacokinetic variability. The precision of E-AB measurements is sufficient to measure not only interanimal pharmacokinetic variability but also variability within an individual animal over time. Shown are the pharmacokinetic profiles of the drug tobramycin following two sequential 20 mg/kg i.v. injections in three different rats (A, B, and C). These high-precision measurements reveal a decrease in the rate of drug elimination kinetics (β phase) in the second injection with respect to the first in all three animals, an effect that presumably arises due to changes in the animal’s blood pressure and/or hydration after several hours under anesthesia. The bold black lines represent the mathematical fit of each injection dataset to a two-compartment pharmacokinetic model.

Fig. 5.

Continuous, in vivo molecular measurements on awake, ambulatory animals. (A) The small size and physical robustness of E-AB sensors renders it possible to use them in animals as they eat, drink, and explore their cage (Movie S1). This robustness, in turn, enables the measurement of specific molecules in the blood of animals as they undertake their normal daily routine, conditions perhaps more relevant to human health than those traditionally used for the collection of metabolic and pharmacokinetic data. Shown are blood levels of the drug tobramycin after a 25 mg/kg i.m. injection (thigh) (B) or sequential 40 mg/kg i.v. (jugular vein) injections (C) in awake, freely moving animals.