Disolution DNP can increase the sensitivity of ¹³C MRS by at least four orders of magnitude and thus allows real time measurements of ¹³C- labelled substrates and the products of their metabolism. While there would be a sensitivity advantage in observing hyperpolarized protons, the fast relaxation times of ¹H resonances prevent in vivo applications of this kind. In this study we investigated whether hyperpolarized ¹H signals, obtained from hyperpolarized ¹³C via spontaneous ¹³C → ¹H polarisation transfer¹, could be used to carry out dynamic measurements of enzymatic activity in vitro. A substantial enhancement of the ¹H resonance of [1-¹H, 2,2,2-²H₃,1-¹³C] acetaldehyde, produced in situ by solutions containing purified yeast Pyruvate Decarboxylase (yPDC), could thus be obtained starting from ¹³C-hyperpolarized [U-²H₃,2-¹³C] pyruvate as the substrate.

An ammonium sulphate suspension of yeast Pyruvate Decarboxylase (yPDC; the reaction catalyzed is shown in figure 1) was purchased from Sigma-Aldrich. A sample containing [U-²H₃,2-¹³C] pyruvic acid was hyperpolarized² and rapidly dissolved in 0.1 M phosphate buffer containing NaOH. Approx. 3 sec after dissolution, a 500 µL sample of this material, containing 6 or 50 mM [U-²H₃,2-¹³C] pyruvate at pH 7.0, was forcibly injected into a phosphate buffer solution of yPDC (1 unit of activity†) in a 5 mm NMR tube placed inside of a 500 MHz vertical bore spectrometer (Varian). ¹H MR spectra were recorded starting immediately from the time of injection, using a 30°/800Hz bandwidth, 3-lobe sinc pulse centred at 9.6 ppm with a TR of 4 s. In the INEPT experiment, a reverse refocused pulse sequence was used to transfer polarization from the hyperpolarized ¹³C to the aldehyde ¹H of acetaldehyde using a J (¹³C-¹H) of 177 Hz. ¹H spectra were acquired using a frequency-selective excitation pulse with a TR of 5 s.

† One unit of PDC activity converts 1 µmole of pyruvate to acetaldehyde per minute in the presence of cofactors, at pH 6.0 and 25°C.

A substantially enhanced, spontaneously polarized signal from [1-¹H, 2,2,2-²H₃,1-¹³C] acetaldehyde, was observed in arrays of ¹H spectra acquired after the addition of hyperpolarized [U-²H₃,2-¹³C] pyruvate to a solution containing yPDC (fig 2). This resonance was not observed on this time scale when using thermally polarized substrate (data not shown). In the INEPT experiment, a substantially higher signal-to-noise ratio (SNR) was obtained (fig 3), when compared with an equivalent experiment using conventional ¹H acquisition (SNR of 195 and 15, respectively), yet its single-shot nature prevented dynamic studies. The maximum intensity of the acetaldehyde signal with saturating pyruvate concentration³ (50 mM), was observed approximately 8 seconds earlier than the respective maximum recorded with 6 mM pyruvate (fig 4a). Kinetic analysis of the evolution of the acetaldehyde signal (details not shown) with these two different substrate concentrations revealed a threefold decrease in the k_PDC, the apparent initial rate of the conversion from pyruvate to acetaldehyde, at the lower pyruvate concentration (fig 4b).We observed a substantial enhancement in the ¹H resonance intensity of [1-¹H, 2,2,2-²H₃,1-¹³C] acetaldehyde produced by yPDC from ¹³C - hyperpolarized [U-²H₃,2-¹³C] pyruvate. This may help resolve a complication arising when trying to follow this reaction by hyperpolarised ¹³C NMR in vivo, owing to the close proximity of the [1-¹³C] acetaldehyde and [2-¹³C] pyruvate resonances⁴. Although a substantially higher SNR can be obtained in the reversed INEPT experiment, the multi-shot capability of the spontaneous ¹³C → ¹H polarisation transfer allows the relevant information, including enzyme kinetics, to be derived from dynamic arrays of ¹H spectra acquired using a simple single-resonance MR system. We are currently exploring the potential applications of this phenomenon for the investigation of cell metabolism in vivo.