Introduction

Metabolic imaging of the conversion of hyperpolarized (HP) [1-¹³C]-pyruvate into lactate has demonstrated strong clinical potential for characterization of tumor metabolism and response to therapy^1-5. Pharmacokinetic (PK) modeling and analysis⁶ can be used to determine k_PL, the apparent rate constant for conversion of HP pyruvate into lactate. However, further validation is needed to demonstrate the potential of this metabolic imaging biomarker to guide clinical care. HP pyruvate is administered by intravenous injection and must extravasate from vasculature and cross the cell membrane before interacting with intracellular enzymes that mediate chemical conversion to lactate. Due to the difficulty distinguishing between signals arising from intravascular, extracellular/extravascular, and intracellular compartments, simplified precursor-product models are often used to analyze HP MRI data; however, these models can be biased by changes in vascular function or membrane transport, and generally under-estimate the true intracellular chemical conversion rate. Improved understanding of the intracellular conversion rate could lead to more robust analyses of clinical and in vivo HP MRI data. Therefore, we sought to characterize intracellular k_PL in an intuitive closed model system involving cancer cell suspensions.

Methods

HP NMR: HP [1-¹³C]-pyruvate was prepared using a HyperSense dissolution dynamic nuclear polarization system (Oxford Instruments, Abingdon, UK) as previously described⁷, and all NMR experiments were conducted using a Spectrospin DPX-300 NMR spectrometer equipped with a broadband 10 mm NMR probe (Bruker, Billerica, MA, USA). Approximately 1×10⁷ cells (Hth-83 anaplastic thyroid cancer; LNCap and PC3 prostate cancer) were suspended in 900 μL of cell media and added to a 10 mm Shigemi tube, along with 100 μL of D₂O. Approximately 181 µL of 20 mM HP [1-¹³C]-pyruvate was delivered to cell samples in the NMR, resulting in a final HP pyruvate concentration of approximately 3.07 mM/L⁸. A pulse-acquire sequence with a 10° excitation angle and a 2 s repetition time was used to acquire dynamic ¹³C spectra.
Ion Chromatography Mass Spectrometry: A secondary method for assessing uptake and conversion of pyruvate was carried out using IC-MS following exposure to [U-¹³C₃]-pyruvate in media. Cells were expanded and cultured in standard R10 media. At the start of each measurement, the media was discarded and cells were exposed to fresh R10 media that contained 1.0 mM of [U-¹³C₃]-pyruvate. Plates of cells were subjected to U-Pyr exposure times of 0, 5, 15, 30, 45, and 60 s. Following U-Pyr exposure, cells were washed with ice-cold buffer, frozen with liquid N₂, and prepared for targeted tracer analysis using an Orbitrap Fusion Tribrid mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA).
Pharmacokinetic Modeling: A closed PK model with two physical compartments (intracellular, extracellular) and two chemical pools (pyruvate, lactate)⁶ was used to analyze HP NMR and IC-MS data. The following differential equations relate exchange between pyruvate and lactate in extracellular media (e) and within cells (ic):

$$\frac{\partial Pyr_{e}(t)}{\partial t}=\frac{k_{ecp}}{v_{e}}Pyr_{ic}(t)-(\frac{k_{ecp}}{v_{e}}+R_{Pyr})Pyr_{e}(t)$$
$$\frac{\partial Lac_{e}(t)}{\partial t}=\frac{k_{ecl}}{v_{e}}Lac_{ic}(t)-(\frac{k_{ecl}}{v_{e}}+R_{Lac})Lac_{e}(t)$$
$$\frac{\partial Pyr_{ic}(t)}{\partial t}=k_{LP}Lac_{ic}(t)+\frac{k_{ecp}}{v_{ic}}Pyr_{e}(t)-(k_{PL}+\frac{k_{ecp}}{v_{ic}}+R_{Pyr}+k_{PX})Pyr_{ic}(t)$$
$$\frac{\partial Lac_{ic}(t)}{\partial t}=k_{PL}Pyr_{ic}(t)+\frac{k_{ecl}}{v_{ic}}Lac_{e}(t)-(k_{LP}+\frac{k_{ecl}}{v_{ic}}+R_{Lac}+k_{LX})Lac_{ic}(t)$$

Here, k_ecp and k_ecl represent the transport of pyruvate and lactate across the cell membrane. v_e denotes the extracellular volume fraction and v_ic denotes the intracellular volume fraction. The apparent rate constant for chemical conversion of HP pyruvate to lactate is given by k_PL, and the reverse reaction rate is given by k_LP. R_Pyr and R_Lac denote losses due to T₁ relaxation, when appropriate for analyzing NMR data, and k_PX and k_LX represent conversion of pyruvate and lactate to chemical endpoints outside of the model (for IC-MS data). Excitation losses were modeled as instantaneous following each excitation, and the total observed pyruvate and lactate signals were calculated as the sum of compartmental signals weighted by their respective volume fractions. To assess the uniqueness of the solutions, nuisance variables were fit to the data while k_PL was varied across a supraphysiological range of values.

Results

IC-MS data revealed surprisingly rapid uptake and conversion of M+3 labeled pyruvate, which was reproducible across replicates and cell lines. Representative data in Figure 1 shows mass labels in intracellular pyruvate and lactate at ~70% of their equilibrium value after only 5 s exposure to 1 mM [U-¹³C₃]-pyruvate in normal media. Analysis of HP NMR data shows a range of virtually equivalent solutions beginning at a threshold k_PL value (Figure 2); above this point, nuisance variables (and k_LP in particular) counteract effects of higher k_PL values to yield indistinguishable signal curves. Both modalities indicate intracellular k_PL > 0.3 s^-1 for Hth-83 cells.

Discussion and Conclusion

PK analysis of HP NMR and IC-MS data shows good agreement in the apparent rate constant for intracellular conversion of HP [1-¹³C]-pyruvate and [U-¹³C₃]-pyruvate into lactate. The values for k_PL observed in this study are higher than those observed using traditional precursor-product models because conversion is only permitted within the intracellular volume fraction, which was carefully estimated by counting cells and measuring average cell diameter. More accurate understanding of intracellular k_PL will enhance our ability to characterize changes in HP MRI signal evolution in vivo.

Acknowledgements

This work was supported by funding from the National Cancer Institute (R01CA211150) and the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK105346) of the National Institutes of Health, and the Cancer Prevention and Research Institute of Texas (RP170366). The content is solely the responsibility of the authors and does not necessarily represent the official views of the sponsors.