Transport-exclusion pharmacology to localize lactate dehydrogenase activity within cells
Recent in vitro and in vivo work has shown that lactate provides an important source of carbon for metabolic reactions in cancer cell mitochondria. An interesting question is whether lactate is oxidized by lactate dehydrogenase (LDH) in the cytosol and/or in mitochondria. Since metabolic processes in the cytosol and mitochondria are affected by redox balance, the location of LDH may have important regulatory implications in cancer metabolism.
Within most mammalian cells, metabolic processes are physically separated by membrane-bound compartments. Our general understanding of this spatial organization and its role in cellular function, however, suffers from the limited number of techniques to localize enzymatic activities within a cell. Here, we describe an approach to assess metabolic compartmentalization by monitoring the activity of pharmacological inhibitors that cannot be transported into specific cellular compartments.
Oxamate, which chemically resembles pyruvate, is transported into mitochondria and inhibits LDH activity in purified mitochondria. GSK-2837808A, in contrast, is a competitive inhibitor of NAD, which cannot cross the inner mitochondrial membrane. GSK-2837808A did not inhibit the LDH activity of intact mitochondria, but GSK-2837808A did inhibit LDH activity after the inner mitochondrial membrane was disrupted.
Our results are consistent with some mitochondrial LDH that is accessible to oxamate, but inaccessible to GSK-2837808A until mitochondria are homogenized. This strategy of using inhibitors with selective access to subcellular compartments, which we refer to as transport-exclusion pharmacology, is broadly applicable to localize other metabolic reactions within cells.
KeywordsLactate Lactate dehydrogenase Transport-exclusion pharmacology Redox balance
Most mammalian cells contain organelles that are bounded by lipid membranes. The chemical reactions occurring in each of these compartments are sequestered from the rest of the cell, thereby providing an opportunity to specialize metabolism in support of specific organelle functions. Some examples include generating harmful metabolic byproducts in organelles where they can be neutralized (such as hydrogen peroxide in peroxisomes), adjusting chemical concentrations to drive reactions in a direction that they may not proceed in other parts of the cell (such as using a proton gradient to fuel ATP synthesis in mitochondria), and harboring anabolic and catabolic reactions in different compartments to limit unproductive futile cycling (such as fatty acid synthesis in the cytosol and fatty acid oxidation in mitochondria) [1, 2].
Despite increasing evidence that metabolic compartmentalization is essential to various cellular functions, the spatial organization of metabolism within a cell remains poorly understood due to the technical challenges of measuring subcellular location. In a typical metabolomic experiment, cell lysates are analyzed and the results therefore only provide average concentrations of metabolites from the entire cell. Although organelles can be efficiently purified for metabolic evaluation, co-purified contaminates are a considerable challenge . Metabolite interactions with the outer membrane leaflet or its associated proteins, for example, can complicate data analysis. For mitochondria, which are the focus of the current work, the intermembrane space creates additional difficulties. Proteomic analyses suffer from the same problems . High-resolution microscopy can be applied to image proteins within a cell, but this does not reflect protein activity. Functional assays from purified mitochondria can be insightful; however, it is difficult to confirm that protein activity occurs within the mitochondrial matrix. Further, an incomplete understanding of mitochondrial carrier systems has limited our ability to study compartmentalization by manipulating metabolite transport . Thus, strategies to localize metabolic transformations within subcellular compartments such as mitochondria are highly needed.
In this work, we were specifically interested in localizing the enzyme lactate dehydrogenase (LDH) within cells. Recent studies have shown that some cancer cells use lactate in vitro and in vivo as a primary carbon source for metabolic pathways in mitochondria, such as the tricarboxylic acid (TCA) cycle [6, 7, 8]. LDH is required to incorporate lactate carbon into TCA cycle intermediates. An interesting question is whether this LDH activity occurs in the cytosol and/or in mitochondria. When oxidizing lactate to pyruvate, LDH simultaneously reduces NAD+ to NADH. Neither NAD+ nor NADH can cross the inner mitochondrial membrane, and the ratio of NAD+ to NADH modulates numerous biological processes in both the cytosol and mitochondria. Thus, the location of LDH may selectively influence redox balance within subcellular compartments and therefore have important regulatory implications in cancer metabolism .
Cell culture and drug treatments
Unless otherwise noted, cells were cultured in high-glucose Dulbecco’s Modified Eagle Medium (DMEM, 4.5 g/L D-glucose) (Life Technologies) containing 10% Fetal Bovine Serum (FBS) (Life Technologies) and 1% penicillin/streptomycin (Life Technologies) at 37 °C with 5% CO2. In each drug experiment, either oxamate or GSK-2837808A (3-[[3-[(Cyclopropylamino) sulfonyl]-7-(2,4-dimethoxy-5-pyrimidinyl)-4-quinolinyl] amino]-5-(3,5-difluorophenoxy) benzoic acid, TOCRIS) was added into the assay buffer. To account for effects of DMSO, DMSO was added to the assay buffer in all experiments (including oxamate conditions and vehicle conditions). The final concentration of DMSO was 1%, unless otherwise stated. Three biological replicates were used for each condition tested.
Lactate production assay
Approximately 7 × 105 HeLa cells were seeded in a 12-well plate and allowed to attach overnight. Cells were then washed and supplemented with FBS-free, low-glucose media (1 g/L D-glucose) and treated with oxamate, GSK-2837808A, or DMSO alone (vehicle). After 6 h, the culture media were collected and extracted as described previously and detailed below . Samples were analyzed by liquid chromatography/mass spectrometry (LC/MS) in negative ion mode with a triple quadrupole mass spectrometer (6460, Agilent Technologies). Samples were separated with a Luna Aminopropyl column (3 μm, 150 mm × 1.0 mm I.D., Phenomenex) coupled to an Agilent 1260 LC system. A flow rate of 50 μL/min was used. The mobile phases and linear gradient were A = 95% water, 5% acetonitrile (ACN), 20 mM ammonium hydroxide (NH4OH), 20 mM ammonium acetate (NH4Ac); B = 100% ACN; 85% B from 0 to 3 min, 85% to 50% B from 3 to 7 min, 50% to 5% B from 7 to 11 min, and 5% B from 11 to 13 min.
Purification of mitochondria
Mitochondria were purified as described previously . Briefly, cells were harvested, pelleted, and re-suspended in cold mitochondrial isolation media (MIM) (300 mM sucrose, 10 mM HEPES, 0.2 mM EDTA, and 1 mg/mL bovine serum albumin (BSA), pH 7.4) and then homogenized with a glass-Teflon potter. Next, samples were centrifuged at 700×g (4 °C) for 7 min to separate mitochondria from the remaining cellular material. The supernatant was decanted after centrifugation and set aside. The remaining pellets were homogenized again in MIM to recover more mitochondria. The supernatant was then pooled with the supernatant from above and centrifuged at 10,000×g (4 °C) for 10 min to obtain mitochondrial pellets. Mitochondrial pellets were washed and quantified by performing a Bradford assay, unless otherwise noted.
LDH activity assay
LDH activity was assessed in a 96-well plate. First, mitochondria were purified from ~ 6 × 107 HeLa cells as above. Mitochondrial pellets were then lysed with 1% triton X-100/50 mM Tris (pH 7.4). The mitochondrial lysates were treated with oxamate, GSK-2837808A, or DMSO alone (vehicle). The 1% triton X-100/50 mM Tris solution was used as a negative control (blank). A standard mixture was prepared containing phenazine methosulphate (360 μg/mL), p-iodonitrotetrazolium violet (1.3 mg/mL), and NAD+ (340 μg/mL). A 50 μL aliquot of the standard mixture, 200 mM Tris (pH 8), and 50 mM lactate were added to each well before adding 50 μL of sample (38 μg of mitochondrial protein/well). The final concentration of DMSO was 0.4% in all three conditions. The kinetic assay was run at 490 nm with a Cytation 5 microplate reader (BioTek), and LDH activity was determined by the maximum slope.
Labeling whole cells with U-13C lactate
HeLa cells were grown to ~ 35% confluency in 100-mm culture dishes. The culture media were then changed to fresh low-glucose (5 mM) media supplemented with 3 mM uniformly 13C-labeled lactate (U-13C lactate, Cambridge Isotope Laboratories). Cells were treated with 50 mM oxamate, 75 µM GSK-2837808A, or DMSO alone (vehicle) for 24 h. The final concentration of DMSO was 0.3% in all three conditions. After 24 h, cells were washed with phosphate-buffered saline (PBS) and HPLC-grade water, quenched with 1 mL cold HPLC-grade methanol, scraped from the plate, and pelleted. Pellets were dried on a SpeedVac (Thermo Fisher Scientific) and subsequently lyophilized (Labconco). Dried samples were weighed and extracted by using the protocol described below . Experiments were performed with n = 3 cultures per sample group.
Oxygen consumption rate
Respiration of intact mitochondria was measured with an XFp analyzer (Seahorse Bioscience) or a high-resolution OROBOROS Oxygraph-2k respirometer (Oroboros Instruments). For the Seahorse experiments, after purifying mitochondria from ~ 2 × 107 HeLa cells, approximately 8 μg of mitochondrial pellets were re-suspended in cold mitochondria assay solution (MAS, 70 mM sucrose, 220 mM mannitol, 10 mM KH2PO4, 5 mM MgCl2, 2 mM HEPES, 1 mM EGTA, and 0.2% (w/v) fatty acid-free BSA, pH 7.2) with 10 mM lactate and 5 mM malate. Before measuring respiration, mitochondria were brought to room temperature. ADP (4 mM) was added to induce respiration. The oxygen consumption rate (OCR) of HeLa mitochondria was monitored under three different conditions: oxamate, GSK-2837808A, or DMSO alone (vehicle). To remove background contributions, the OCR value before the addition of ADP was subtracted from the OCR value after the addition of ADP (Fig. 3b). For assessing mitochondrial function, no drugs or DMSO were added. Sample sizes were used that produced OCR numbers within the recommended range of the vendor. For the Oroboros experiments, mitochondrial respiration media were used. Approximately the same number of HeLa cell mitochondria (300 μg mitochondrial protein) was added to each chamber followed by metabolic substrates and inhibitors.
Labeling whole cells with 2-2H lactate prior to mitochondrial purification
Cells were cultured in a T-150 flask until reaching 90% confluency. Cells were then transferred to glucose-free media for 4 h. After 4 h, cells were supplemented with 10 mM 2-2H lactate for 45 min prior to being washed, harvested, and pelleted. For mitochondrial purification, the cell pellets were re-suspended in 500 μL of cold MIM with 100 mM oxamate. The isolated mitochondrial pellets were lyophilized and subsequently treated with a methanol/acetonitrile/water (2:2:1) solution prior to being reconstituted in 40 μL acetonitrile/water (1:1) per milligram dry weight. LC/MS analysis was performed as described below.
Labeling purified mitochondria with U-13C lactate
Approximately 2 × 108 HeLa cells were harvested at 90% confluence. Mitochondria were purified as above. Purified mitochondria were split into wells (170 μg of mitochondrial protein/well) and incubated in 1 mL MAS buffer with 5 mM malate and 5 mM lactate. Samples were treated with oxamate, GSK-2837808A, or DMSO alone (vehicle) for 10 min. The final concentration of DMSO was 0.3% in all three conditions. After 10 min, 10 mM U-13C lactate was added to the MAS buffer for 20 min before harvesting. Mitochondrial pellets were washed, collected, and snap frozen in liquid nitrogen prior to extraction.
Metabolite extraction and LC/MS analysis
Cell pellets or purified mitochondria were extracted and analyzed by LC/MS as described before [6, 10]. Cell pellets were treated with a methanol/acetonitrile/water (2:2:1) solution and reconstituted in 40 μL acetonitrile/water (1:1) per milligram dry weight. Mitochondrial pellets were treated with a methanol/acetonitrile/water (2:2:1) solution and reconstituted in 50 μL acetonitrile/water (1:1) per 170 μg of mitochondrial protein, unless otherwise noted. Samples were separated with a Luna Aminopropyl column (3 μm, 150 mm × 1.0 mm I.D., Phenomenex) coupled to a Dionex UltiMate® 3000 RSLCnano LC system. MS detection was performed on a Thermo Q Exactive Plus mass spectrometer (Thermo Fischer Scientific) in negative ion mode at 140,000 resolving power. The column was used in hydrophilic interaction mode with a flow rate of 50 μL/min. The following mobile phases and linear gradient were used: A = 95% water, 5% ACN, 20 mM NH4OH, 20 mM NH4Ac; B = 95% ACN, 5% water; 100% B from 0 to 3 min, 100% B to 0% B from 3 to 40 min, and 0% B from 40 to 45 min.
Data are reported as means ± SD. Dataset comparisons were performed with a Student’s unpaired, two-tailed t test.
Functional LDH is a homo- or heterotetramer made up of LDHA and LDHB subunits . We considered two compounds (oxamate and GSK-29837808A) known to inhibit both LDHA and LDHB subunits at the concentrations we used [12, 13, 14, 15, 16]. We note that because our inhibition experiments are not specific to enzyme subtype, we cannot distinguish between LDHA and LDHB in the analyses. It has been shown previously, however, that LDHB is concentrated in HeLa cell mitochondria .
As a second method to assess LDH activity in purified HeLa cell mitochondria, we also used stable isotope tracers and LC/MS. After purifying HeLa mitochondria, we incubated them in the same buffer as above, but we added U-13C lactate for 20 min. We then used the incorporation of two 13C labels into citrate as an indicator of LDH activity. The M + 2 isotopologue of citrate results from the sequential actions of LDH, the pyruvate dehydrogenase complex, and citrate synthase. Consistent with the OCR data, we found that oxamate reduced citrate labeling while GSK-29837808A treatment led to no statistically significant change relative to vehicle controls (Fig. 3c). Together, these data are consistent with LDH localization to the mitochondrial matrix, where it is accessible to oxamate but not GSK-29837808A (rather than localization to the outer membrane or intermembrane space).
Increasing evidence supports that lactate is not only a prominent fuel in cancer cells, but also a major source of carbon for anabolic processes such as lipid synthesis [6, 7, 8, 25]. Utilization of lactate in this capacity requires that lactate be oxidized by LDH, which could potentially occur in the cytosolic and/or mitochondrial compartments of a cancer cell. While the difference between cytosolic and mitochondrial LDH activity only changes the location of two electrons, the regulatory implications of a compartmentalized shift in redox homeostasis are potentially significant. Many metabolic enzymes are regulated by the ratio of NAD+ to NADH, such as glyceraldehyde 3-phosphate dehydrogenase in the cytosol and isocitrate dehydrogenase 3 in mitochondria . The site of LDH activity may therefore have an important effect on metabolic fluxes. Oxidation of lactate by cytosolic LDH, for example, may slow glycolytic flux and glucose consumption, while oxidation of lactate by mitochondrial LDH may promote lipid synthesis via the accumulation of citrate [6, 9].
Although the possibility of a mitochondrial LDH has been considered for several decades, disagreements persist about its precise location and its biochemical role, as has recently been reviewed in detail [27, 28, 29]. Progress in the field has been complicated by the technical limitations of localizing lactate metabolism within different subcellular compartments. Mitochondrial LDH cannot be assessed by tracing labels from 13C-lactate because of cytosolic lactate-pyruvate exchange . Imaging approaches have generally provided limited resolution to localize LDH to the mitochondrial matrix. Functional assays examining whether mammalian mitochondrial preparations oxidize lactate have provided mixed results. Thus, alternative experimental approaches to provide additional information about mitochondrial lactate metabolism are needed.
We have described a strategy that exploits unique mitochondrial transport properties of different small-molecule inhibitors (oxamate and GSK-29837808A) to assess LDH location within cells. Interestingly, when evaluating intact purified mitochondria, oxamate decreased LDH activity but GSK-29837808A did not. Yet, both drugs had similar effects on LDH activity once mitochondria were disrupted, indicating differential access to LDH in previous experiments. Even though we have applied the approach to LDH here, the same approach can be broadly applied to investigate the spatial location of other enzymes. These data will be highly complementary to metabolomic and proteomic experiments performed on purified mitochondria and may provide key insights as we seek to understand the ways in which mitochondria are functionally integrated within the cell. We note that employing transport-exclusion pharmacology requires having inhibitors with unique biochemical properties to mediate subcellular distribution, which are not available for many enzymes. However, we are optimistic that the rapidly growing interest in mitochondrial biology will inspire the development of such drugs not only for these types of experiments but also for their potential therapeutic significance in selectively targeting disease processes within specific subcellular compartments.
Mitochondria isolated from HeLa cells have the capacity to oxidize lactate, suggesting the presence of a mitochondrial LDH. Of particular interest, however, is the location of the enzyme within mitochondria. Its association with the outer mitochondrial membrane, the intermembrane space, or the outer leaflet of the inner mitochondrial membrane affects cytosolic redox balance. Its association with the inner leaflet of the inner membrane or the matrix, in contrast, affects mitochondrial redox balance. Intact purified mitochondria given oxamate, a competitive inhibitor of pyruvate, show decreased LDH activity. GSK-29837808A, which is a competitive inhibitor of NAD, only affected LDH activity in mitochondrial homogenates when the inner mitochondrial membrane had been disrupted. These data are consistent with some mitochondrial LDH that is accessible to oxamate but not GSK-29837808A. Our approach of using small-molecule inhibitors with different mitochondrial transport properties to localize protein activity is broadly applicable to the study of other enzymes.
Financial support was received for this work from NIH grants R35ES028365 (GJP), U01CA235482 (GJP), and R24OD024624 (GJP) as well as the Pew Scholars Program in the Biomedical Sciences (GJP) and the Edward Mallinckrodt, Jr., Foundation (GJP).
Availability of data and materials
The datasets used during the current study are available from the corresponding author on reasonable request.
XN and Y-JC contributed equally to this work. All authors analyzed and interpreted data. All authors contributed to manuscript drafting and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
GJP is a scientific advisory board member for Cambridge Isotope Laboratories and a recipient of the 2017 Agilent Early Career Professor Award. The remaining authors have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 9.Kanow MA, Giarmarco MM, Jankowski CS, Tsantilas K, Engel AL, Du J, Linton JD, Farnsworth CC, Sloat SR, Rountree A, et al. Biochemical adaptations of the retina and retinal pigment epithelium support a metabolic ecosystem in the vertebrate eye. Elife. 2017;6:e28899.Google Scholar
- 16.Xie H, Hanai J, Ren JG, Kats L, Burgess K, Bhargava P, Signoretti S, Billiard J, Duffy KJ, Grant A, et al. Targeting lactate dehydrogenase--a inhibits tumorigenesis and tumor progression in mouse models of lung cancer and impacts tumor-initiating cells. Cell Metab. 2014;19(5):795–809.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.