Healthy epithelial cells in the peripheral zone of the prostate have the unique capability to produce and secrete large amounts of citrate (Cit) in the lumina. Key in this production is Zn²⁺ binding to aconitase inhibiting the conversion of citrate into isocitrate.^1,2 Prostate cancer (PCa) cells are assumed to have lost this capability.¹ However, still little is known about the metabolic pathways supporting Cit production, although studies in rat prostate epithelial cells suggest that aspartate acts as four-carbon source for citrate production.³ Both healthy epithelial cells and the metastasis PCa cell line LNCaP possess high numbers of transporters for aspartate.⁴ LNCaP cells produce citrate in contrast to most other PCa derived cell lines (e.g. PC3, DU145).⁵

Aim: to determine which carbon sources are involved in citrate production of LNCaP to better understand the underlying mechanisms and to investigate if Cit metabolic features may serve as potential PCa biomarkers. We also included another human PCa metastatic cell line, VCaP, in our studies.

The cell lines LNCaP (8x10⁶-14x10⁶ cells) and VCaP (14x10⁶-17x10⁶ cells) were grown in medium (RPMI-1640 + Gln (2mM)) supplemented with different ¹³C-labeled substrates for 48h. LNCaP and VCaP were grown in medium supplemented with:

- [1,6-¹³C₂]glucose (11 mM) (aspartate conc. 150 µM)

- [U-¹³C₄]aspartate (2.0 mM) + pyruvate (6mM) (no glucose)

LNCaP was also grown in medium supplemented with:

- [U-¹³C₄]aspartate (1.2 mM) (glucose conc. 11 mM)

- [2-¹³C]pyruvate (7 mM) (no glucose)

- [5-13C]glutamine (2mM) (unlabeled glutamine removed)

The media were collected, 4-mL aliquots were lyophilized, dissolved in D₂O and pH was adjusted to 7.4 before analysis by high resolution water-suppressed ¹H-NMR spectroscopy (D₂O, 500 MHz, ns = 256-320) and ¹³C-NMR (D₂O, 500 MHz, ns = 5k–21k, WALTZ64 ¹H-decoupled).

Citrate was detected using ¹H-NMR in the growth medium of both cell lines in all experiments (data not shown). Providing ¹³C-labeled aspartate to the cells did not result in ¹³C-enrichment in citrate in neither LNCaP nor VCaP (Fig 1). ¹³C-labeled citrate was observed after 48h in the medium of LNCaP and VCaP supplemented with [1,6-¹³C]labeled glucose (Fig.2). Although the number of LNCaP cells was lower than those of VCaP cells, more labeled glucose remained in the medium after 48h. LNCaP metabolized about 4 times more ¹³C-labeled glucose than VCaP, resulting in less ¹³C-enriched citrate in VCaP. Overlapping resonances from single labeled [2/4-¹³C]citrate (2/4 means: C2 or C4, their chemical shift is identical) and doubly labeled [2,4-¹³C] or [2/4,3-¹³C]citrate result in a broad peak at 46.5 ppm (Fig 2D). Using [2-¹³C]pyruvate instead of labeled glucose in LNCaP cells results in ¹³C-labeled citrate at C1 and C3. The amount of ¹³C-label at position C1 is about ten-fold higher than at position C3 (Fig.3). Finally we tested whether glutamine could serve as a carbon source for citrate. [5-13C]glutamine is indeed converted into [1/5-¹³C]citrate (Fig. 2,4).We demonstrate that next to LNCaP also VCaP cells are able to produce citrate at NMR detectable levels. Both cells did not seem to use aspartate to produce Cit, indicating that this compound is not a major carbon source for Cit in these cells. These results are different from those obtained in healthy rat prostate epithelial cells³, which may Indicate a shift in citrate metabolism upon malignancy. From the significant generation of ¹³C-labeled citrate after supplementation of LNCaP and VCaP with ¹³C-labeled glucose, it follows that glucose is an important carbon source for citrate in these cell lines. The broad citrate resonance at 46.5 ppm is the result of ¹³C label completing at least one Krebs cycle round which shows that in these cells aconitase inhibition is not an absolute prerequisite for citrate production and secretion. Adding extra Zn²⁺ did not change this finding (results not shown). To test whether citrate is mainly formed via the Krebs cycle via condensation of acetyl-CoA with oxaloacetate or via pyruvate carboxylase LNCaP cells were provided with [2-¹³C]pyruvate. The 10-fold higher ¹³C-labeling at position C1 than at C3 of citrate indicates that the majority of citrate carbons arises via the Krebs cycle and not through pyruvate carboxylase. Labeling experiments with [5-¹³C]glutamine show that carbons for citrate in LNCaP can also originate from glutamine. The absence of ¹³C-labeled C6 in citrate suggests that glutamine carbons only end up in citrate via glutamate, α-ketoglutarate and isocitrate (“upstream” in the Krebs cycle or in the cytoplasm) and not via oxaloacetate. In conclusion, we detected several specific pathways for carbon supply to produce citrate in metastatic PCa cell lines, that may be valuable as biomarkers for transition from healthy prostate cells to malignant cells.