First hyperpolarized [2-13C]pyruvate MR studies of human brain metabolism

Abstract

We developed methods for the preparation of hyperpolarized (HP) sterile [2-¹³C]pyruvate to test its feasibility in first-ever human NMR studies following FDA-IND & IRB approval. Spectral results using this MR stable-isotope imaging approach demonstrated the feasibility of investigating human cerebral energy metabolism by measuring the dynamic conversion of HP [2-¹³C]pyruvate to [2-¹³C]lactate and [5-¹³C]glutamate in the brain of four healthy volunteers. Metabolite kinetics, signal-to-noise (SNR) and area-under-curve (AUC) ratios, and calculated [2-¹³C]pyruvate to [2-¹³C]lactate conversion rates (k_PL) were measured and showed similar but not identical inter-subject values. The k_PL measurements were equivalent with prior human HP [1-¹³C]pyruvate measurements.

Keywords: Brain metabolism; Hyperpolarized C13; Metabolic imaging.

Fig. 1.

Diagram showing [2-¹³C]pyruvate metabolism investigated in this hyperpolarized NMR spectroscopy study of the human brain.

Fig. 2.

Representative Carbon-13 NMR summed spectrum from the brain of a healthy volunteer acquired with a 32-channel head coil following an injection of 1.43 mL/kg of 250 mM [2-¹³C]pyruvate. Peak identification was assigned following those by Park et al. from studies of HP [2-¹³C]pyruvate in the murine brain: (A) [2-¹³C]pyruvate, (B) [5-¹³C]glutamate, (C) [1-¹³C]citrate and/or [5-¹³C]glutamine, (D) [1-¹³C]pyruvate (natural abundance doublet), (E) [2-¹³C]pyruvate-hydrate, (F) [2-¹³C]lactate doublet [17].

Fig. 3.

Flip angle plot of the RF excitation pulse sequence with the parameters used for this study. Note the decreased excitation of the upfield [2-¹³C]lactate resonance versus the downfield by approximately one half.

Fig. 4.

Spectra for all four volunteers at a single timepoint 16 sec post-injection. Similar levels of [5-¹³C]glutamate and [2-¹³C]lactate reflect the underlying biochemistry of the healthy human brain of similar rates of conversion of [2-¹³C]pyruvate to [2-¹³C]lactate catalyzed by LDH as [2-¹³C]pyruvate to [5-¹³C]glutamate catalyzed by PDH.

Fig. 5.

Dynamic plots of metabolite kinetics for each of the four volunteers. Results were consistent noting minor differences in intensity scale. As shown in the corresponding Fig. 4, the rates of conversion of [2-¹³C]pyruvate to [2-¹³C]lactate and [5-¹³C]glutamate are similar in the normal human brain.

Fig. 6.

Plots showing k_PL analysis demonstrated similar results between a previously acquired [1-¹³C]pyruvate dataset from a volunteer (left) and [2-¹³C]pyruvate volunteer dataset acquired in this study with the mean ± standard error for all 4 volunteers (right).

Fig. 7.

Images acquired using a metabolite-specific flip angle schedule and echo planar imaging (EPI) readout. From left to right ¹H proton reference image, overlaid [2-¹³C] pyruvate, and [5-¹³C]glutamate images of a volunteer’s brain are shown. The single-shot HP ¹³C EPI images were acquired with: in-plane resolution = 2.5 × 2.5 cm², slice thickness = 5 cm, bandwidth = 8.06 kHz, TR = 100 msec, temporal resolution = 3 sec, TE = 19 msec, and flip angles θ_Pyr = 10°, θ_Glu = 60°. Average SNR of the pyruvate signal = 682 and glutamate signal = 31.1.