Real-time measurement of hyperpolarized lactate production and efflux as a biomarker of tumor aggressiveness in an MR compatible 3D cell culture bioreactor

Abstract

We have developed a 3D cell/tissue culture bioreactor compatible with hyperpolarized (HP) (13)C MR and interrogated HP [1-(13)C]lactate production and efflux in human renal cell carcinoma (RCC) cells. This platform is capable of resolving intracellular and extracellular HP lactate pools, allowing the kinetic measurement of lactate production and efflux in the context of cancer aggressiveness and response to therapy. HP (13)C MR studies were performed on three immortalized human renal cell lines: HK2, a normal renal proximal tubule cell line from which a majority of RCCs arise, UMRC6, a cell line derived from a localized RCC, and UOK262, an aggressive and metastatic RCC. The intra- (Lacin ) and extracellular (Lacex ) HP lactate signals were robustly resolved in dynamic (13)C spectra of the cell lines due to a very small but reproducible chemical shift difference (0.031 ± 0.0005 ppm). Following HP [1-(13)C]pyruvate delivery, the ratio of HP Lacin /Lacex was significantly lower for UOK262 cells compared with both UMRC6 and HK2 cells due to a significant (p < 0.05) increase in the Lacex pool size. Lacin /Lacex correlated with the MCT4 mRNA expression of the cell lines, and inhibition of MCT4 transport using DIDS resulted in a significant reduction in the HP Lacex pool size. The extension of these studies to living patient-derived RCC tissue slices using HP [1,2-(13)C2]pyruvate demonstrated a similarly split lactate doublet with a high Lacex pool fraction; in contrast, only a single NMR resonance is noted for HP [5-(13)C]glutamate, consistent with intracellular localization. These studies support the importance of lactate efflux as a biomarker of cancer aggressiveness and metastatic potential, and the utility of the MR compatible 3D cell/tissue culture bioreactor to study not only cellular metabolism but also transport. Additionally, this platform offers a sophisticated way to follow therapeutic interventions and screen novel therapies that target lactate export.

Keywords: aerobic glycolysis; cancer aggressiveness; dynamic nuclear polarization (DNP); hyperpolarized 13C magnetic resonance (HP 13C MR); lactate; lactate efflux; pyruvate; renal cell carcinoma (RCC).

Figure 1

Bioreactor set-up and the viability assessment. (A) Graphical representation of the 5mm MR compatible bioreactor optimized for maximal homogeneity. (B) Representative ³¹P spectra of showing unaltered bioenergetics of the UOK262 cells under study before and after hyperpolarized carbon-13 experiment.

Figure 2

Differential compartmentalization of HP [1-¹³C] lactate. (A) Lower panel shows the spectrum in alginate microspheres devoid of cells, infused with copolarized [1-¹³C] lactate and [1-¹³C] pyruvate. Only one peak was observed for the [1-¹³C]lactate signal. While two peaks were observed in the alginate microspheres with UOK262 cells when infused with HP [1-¹³C] pyruvate (upper panel). The inset (2.5x) clearly shows the well resolved peaks of lactate, where the chemical shift of the downfield peak coincides with that of the signal of lactate in empty alginate microspheres. (B) Lorentzian decomposition of the 2 peaks clearly reveals the wider line width of the Lac_in peak compared to the Lac_ex peak (downfield). (C) Plot of Lac_in (blue diamonds) as a function of cell density. The x-axis represents the β-NTP concentration, which was used to quantify the viable UOK262 cells within the sensitive coil of the bioreactor.

Figure 3

Differential modulation of the lactate pools. (A) Representative ¹³C MR spectra of hyperpolarized [1-¹³C]lactate acquired from untreated (top) and DIDs treated UOK262 cells acquired at the time of maximal lactate production (indicated by the dotted line in B). (B) Dynamic measurement of HP Lac_ex (filled circles) and Lac_in (open circles) without (top panel) and with DIDS treatment (bottom panel). (C) Bar graph of the HP [1-¹³C]lactate peaks of UOK262 cells under different flow and treatment conditions relative to control (flow at 0.5ml/min) normalized to 100%. The bar filled with black slanted lines represent the Lac_in/Lac_ex ratio of UOK262 cells pretreated with DIDS inhibitor for 45 minutes (n=3). The change in the ratio is significantly different (p=0.016). The grey bar denotes the Lac_in/Lac_ex measurements after flow was stopped 15 secs post HP [1-¹³C] pyruvate infusion for a period of 4.25 minutes (n=3) and has a p of 0.055 as measured by students’s t-test when compared to control.

Figure 4

Rate of lactate efflux inhibition of UOK262 cells using non-HP MR measurements. Bar graph of [3-¹³C]lactate efflux rate of untreated (n = 3) and DIDS treated (slanted lines, n = 3) UOK262 cells grown in [3-¹³C]pyruvate containing media. The significantly decreased efflux rate (p ≤ 0.05) with DIDS treatment relative to the untreated controls is consistent with hindered lactate efflux due to MCT4 inhibition.

Figure 5

Estimation of extracellular HP lactate pools in renal cells of varying phenotypes. (A) Graph of Lac_in/Lac_ex for HK2, UMRC6 and UOK262 cell lines. The Lac_in/Lac_ex increases going from UOK262, to UMRC6 to HK-2 (n=3 each). The metastatic UOK262 cell line had significantly (p<0.05) lower HP Lac_in/Lac_ex ratio than both HK2 and UMRC6 cell lines. (B) Graph of HP Lac_in/Lac_ex as a function of MCT4 mRNA expression. A negative linear correlation exists between the measured HP Lac_in/Lac_ex ratio and the MCT4 mRNA expression in the 3 cell lines studied. All data are represented as mean ± std.error.

Figure 6

¹³C spectrum of HP [1,2-¹³C₂]pyruvate metabolism of tissue slice culture. 4 tissue slices of a high-grade urothelial carcinoma were perfused in the 5mm MR compatible bioreactor. The C-1 doublet of HP [1,2-¹³C₂]lactate from metabolism of [1,2-¹³C₂]pyruvate, each split into two peaks represents the two cellular compartments. And only a singlet denoting the intracellular compartment is observed for the [5-¹³C]glutamate is seen in the spectral inset.