Hepatic Gluconeogenesis Influences ¹³C Enrichment in Lactate in Human Brain Tumors During Metabolism of [1,2-¹³C]acetate

Abstract

¹³C-enriched compounds are readily metabolized in human malignancies. Fragments of the tumor, acquired by biopsy or surgical resection, may be acid-extracted and ¹³C NMR spectroscopy of metabolites such as glutamate, glutamine, 2-hydroxyglutarate, lactate and others provide a rich source of information about tumor metabolism in situ. Recently we observed ¹³C-¹³C spin-spin coupling in ¹³C NMR spectra of lactate in brain tumors removed from patients who were infused with [1,2-¹³C]acetate prior to the surgery. We found, in four patients, that infusion of ¹³C-enriched acetate was associated with synthesis of ¹³C-enriched glucose, detectable in plasma. ¹³C labeled glucose derived from [1,2-¹³C]acetate metabolism in the liver and the brain pyruvate recycling in the tumor together lead to the production of the ¹³C labeled lactate pool in the brain tumor. Their combined contribution to acetate metabolism in the brain tumors was less than 4.0%, significantly lower than the direct oxidation of acetate in the citric acid cycle in tumors.

1. Introduction

Metabolic pathways in human tumors in situ may be examined by infusion of ¹³C-enriched compounds followed by ¹³C NMR spectroscopy of an aqueous extract of the tissue. This approach is readily incorporated into the work flow of the operating room because of the absence of ionizing radiation and simple substrates involved, such as acetate or glucose. Recently we found that glioblastomas and brain metastases have the capacity to oxidize acetate in the citric acid cycle based on an analysis of ¹³C-¹³C spin-spin coupling in glutamate [1]. ¹³C-¹³C coupling due do J_1,2 and J_2,3 was also observed in the ¹³C NMR spectra of lactate in the tumors. This finding was unexpected because there is no simple pathway for [1,2-¹³C]acetate to enter the lactate or pyruvate pool [1-6]. There have been a few reports that discuss the possibility of ¹³C-enriched pyruvate or lactate arising from ¹³C-enriched acetate through ¹³C-labeled acetyl-CoA contributing to ¹³C enrichment in oxaloacetate (OAA) followed by decarboxylation and generation of ¹³C-enriched pyruvate, known as pyruvate recycling [3-8]. However, systemic acetate may also enter the citric acid cycle of the liver. Although net synthesis of glucose from acetyl groups does not occur in mammalian liver, ¹³C from acetate may mix in the oxaloacetate pool and enter gluconeogenesis. Exported ¹³C-glucose may be metabolized in the brain tumor. Lactate could also be produced due to the metabolism of acetate elsewhere in the body and made available to the tumor through plasma. Here, we demonstrate that ¹H and ¹³C NMR of plasma is a simple and powerful method to study hepatic gluconeogenesis during infusion of [1,2-¹³C]acetate.

2. Material and methods

Two patients with glioblastomas and two patients with brain metastases (breast cancer and lung cancer) were enrolled in a protocol approved by the University of Texas Southwestern Institutional Research Board. Subjects were infused with [1,2-¹³C]acetate at 6 mg/kg/min for 5 minutes, followed by 3 mg/kg/min for 2 hours. Methods for sampling tumor tissue and plasma for NMR spectroscopy have been described previously [1,2,9]. Proton decoupled ¹³C spectra of tumor extracts and plasma were acquired at 150 MHz (for ¹³C) on an Avance Bruker NMR Spectrometer equipped with 10-mm broadband cryogenically-cooled probe (Bruker Biospin, Billerica, MA). The lactate C3 carbon signal at 20.8 ppm was used as internal chemical shift reference. Examples of ¹³C NMR spectra of these brain tumors have been published previously [1]. Signal areas were measured using ACD (Advanced Chemistry Development, Toronto, Canada) [1,9,10]. Amounts of ¹³C-¹³C doublet peak areas are reported as a fraction of its total carbon resonance area.

3. Results and Discussion

The simple metabolic pathway for metabolism of [1,2-¹³C]acetate in a brain tumor is its conversion to acetyl-CoA and oxidation to CO₂ in the citric acid cycle, and thus providing no direct mechanism for ¹³C entry into the lactate pool of the tumor. Nevertheless, ¹³C NMR spectra of tumors demonstrated both [1,2-¹³C]- and [2,3-¹³C]lactate through the appearance of ¹³C-¹³C doublets C2D12 and C3D23 respectively (Figure 1). To test whether ¹³C label originating from infused [1,2-¹³C]acetate could enter hepatic gluconeogenesis, plasma from the patient was obtained at the time of tumor excision. A typical ¹³C NMR spectrum of the plasma is shown in Figure 2. [1,2-¹³C] and [2,3-¹³C] isotopomers of glucose in the plasma would generate [2,3-¹³C]pyruvate and [1,2-¹³C]pyruvate, respectively, in the tumor which would yield the same ¹³C labeled isotopomers of lactate. Amounts of [1,2-¹³C] (C2D12), [2,3-¹³C] (C3D23) isotopomers of lactate in the tumor (n=4) are 0.30±0.07, 0.17±0.02 respectively and [2,3-¹³C] (C2D23), [1,2-¹³C] (C1D12) in plasma glucose are 0.23±0.05, 0.17±0.04 respectively. Similar values of C3D23, C2D12 of tumor lactate and C1D12, C2D23 plasma glucose suggest that [1,2-¹³C]acetate leads to the production of ¹³C enriched glucose generated in the liver through gluconeogenesis, which then enters the brain to generate corresponding ¹³C enriched lactate isotopomers through glycolysis. C3D23 tumor lactate value (0.17±0.02) is same to the value of C1D12 of plasma glucose (0.17±0.04), which suggests that ¹³C labeled glucose from [1,2-¹³C]acetate metabolism in the liver is the main source of [2,3-¹³C]lactate in the brain tumor. However, C1D12 tumor lactate value (0.30±0.07) is slightly higher than C2D23 of plasma glucose value of (0.23±0.05). This small difference in the [1,2-¹³C] isotopomer in the tumor lactate pool could be due to the effect of other metabolic pathways within the tumor. Pyruvate recycling has been reported in rat brain following infusion of [1,2-¹³C]acetate [3,7]. As shown in Figure 3, pyruvate recycling produces equimolar mixture of [1,2-¹³C] and [3-¹³C] pyruvate isotopomers from [1,2-¹³C] and [3,4-¹³C] OAA isotopomers through the activity of either malic enzyme or combined activity of the enzymes phosphoenolpyruvate carboxykinase (PEPCK) and pyruvate kinase (PK). [1,2-¹³C]pyruvate isotopomer produced from the pyruvate recycling pathway could also contribute to the production of [1,2-¹³C]lactate in the brain tumor lactate pool. Also, [1,2-¹³C] pyruvate isotopomers could produce mono-labeled [1-¹³C]acetyl-CoA (Fc1; fraction of acetyl-CoA ¹³C enriched only in its carbonyl carbon, C1) through the enzymatic action of pyruvate dehydrogenase (PDH), which was determined from C5 glutamate ¹³C resonance signal through the ratio between C5 glutamate singlet and doublet C5D multiplied by the total [1,2-¹³C]acetyl-CoA, which in this case was 3.1%±0.9%. Without the contribution from pyruvate recycling and the hepatic gluconeogenesis contributions, Fc1 should be at the ¹³C isotope natural abundance value of 1.1%. ¹³C fractional enrichment of plasma glucose is 4.0%±1.0% (from ¹H NMR spectra of plasma glucose, data not shown) and its contribution to the total Fc1 cannot exceed more than 0.8%. Therefore, the rest of the Fc1 contributions (~2%) are likely due to the activity of pyruvate recycling in the brain tumor. Based on the C1D12 value of plasma glucose of 0.17±0.04, its contribution to the [1,2-¹³C]acetyl-CoA pool in the brain tumor from peripheral metabolism of acetate cannot be more than 0.7%, which further confirms our earlier report of [1,2-¹³C]acetate being directly oxidized in human GBM and brain metastasis, contributing up to 48%±4% of [1,2-¹³C]acetyl-CoA production in the tumor [1]. Additionally, ¹³C-labeled lactate isotopomers (C2D12 and C3D23) were also detected in plasma, which could be a source of the ¹³C-labeling observed in the lactate pool in the tumor [11]. Similar values of ¹³C fractional enrichments of plasma glucose (C1 carbon) and C3 lactate in the tumor observed from ¹H NMR spectroscopy may suggest that ¹³C labeled plasma glucose will be one of the main source for the lactate ¹³C labeling detected in the tumors reported in this study.

Figure 1.

Tumor lactate ¹³C-labeling in a patient with GBM (A and C) and a patient with lung metastasis to brain (B, D). ¹³C-¹³C doublets C2D12 and C3D23 correspond to [1,2-¹³C] and [2,3-¹³C] isotopomers of ¹³C labeled lactate respectively. C2S and C3S are singlet signals from C2 and C3 carbons of lactate respectively.

Figure 2.

Plasma glucose ¹³C-labeling from the patient presented in Figure 1A, C. ¹³C-¹³C doublets C1D12 and C2D23 are due to [1,2-¹³C] and [2,3-¹³C] isotopomers of ¹³C labeled glucose respectively. C1S and C2S are singlet ¹³C signals from C1 and C2 carbons of glucose respectively.

Figure 3.

Schematic diagram showing the sources of ¹³C labeling in the tumor lactate pool during the [1,2-¹³C]acetate as infusion substrate. Filled (green) circles represent ¹³C labeled carbons. Infused [1,2-¹³C]acetate can only lead to the production of [1,2-¹³C]acetyl-CoA. Numbers represent carbon atom positions. Abbreviations: LDH, lactate dehydrogenease; PDH, pyruvate dehydrogenease; PEPCK, phosphoenolpyruvate carboxykinase; PK, pyruvate kinase; α-KG, α-ketoglutarate; OAA, oxaloacetate.

4. Conclusion

In summary, we have presented ¹³C isotopomer analysis of contributions from peripheral metabolism and brain pyruvate recycling in human GBM and brain metastasis patients infused with [1,2-¹³C]acetate. Together, they account for less than 4.0% to the ¹³C enriched acetyl-CoA pool (both [1-¹³C] and [1,2-¹³C]acetyl-CoA) in the brain tumor. The present analysis will be applicable to study [1,2-¹³C] acetate metabolism in vivo in other organ systems.