University of Groningen
C/EBPβ isoforms and the regulation of metabolism
Ackermann, Tobias
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Ackermann, T. (2018). C/EBPβ isoforms and the regulation of metabolism: A fine balance between health and disease. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
115
Chapter V
Stimulation of the malate aspartate shuttle by
C/EBPβ-LIP causes glycolysis addiction
Tobias Ackermann
1, Hidde R. Zuidhof
1, Gertrud Kortman
1, Mathilde
Broekhuis
1, Mohamad Amr Zaini
1, Götz Hartleben
1and Cornelis F.
Calkhoven
1 Manuscript in preparation1
European Research Institute for the Biology of Ageing (ERIBA), University Medical Centre Groningen, University of Groningen, 9700 AD Groningen, The Netherlands.
116
Abstract
Oncogenes reprogram cellular metabolism to support cell growth and thereby couple cell survival to certain metabolic pathways and nutrients. In order to synthesise amino acids and nucleotides de novo, glucose is oxidised in the cytoplasm. In this process NAD+ is used as the oxidizing agent, which is thereof reduced into NADH. To regenerate cytoplasmic NAD+ and maintain glycolysis cells transfer electrons from NADH to pyruvate reducing it into lactate, or transfer the electrons to the electron transport chain in the mitochondria via the malate aspartate shuttle (MAS) or the glycerol-3-phosphate shuttle (GPS). The NADH/NAD+ homeostasis plays a pivotal role in metabolic health, disease and ageing, but our knowledge about the cellular effects and the underlying mechanisms of disturbed homeostasis are still limited. Here, we show that the oncogenic transcription factor C/EBPβ-LIP induces the MAS and thereby addicts cells to glycolysis. Inhibition of glycolysis in C/EBPβ-LIP expressing cells results in low NADH/NAD+ ratios and apoptosis that can be rescued by inhibiting either the MAS or other NADH consuming processes. This study shows that low NADH/NAD+ ratio induces cell death upon inhibition of glycolysis by 2-deoxyglucose (2-DG). Further, it identifies C/EBPβ-LIP and high MAS activity as markers for 2-DG sensitivity and suggests that simultaneous inhibition of glycolysis and the lowering NADH/NAD+ may be considered to treat cancer.
117
Introduction
Cancer cells reprogram their metabolism to support de novo synthesis of macromolecules that are needed for cell growth and proliferation 1. Most prominently, cancer cells increase glucose uptake and metabolise glucose by aerobic glycolysis, which was first recognised by Otto Warburg (reviewed in 2). Later it was shown that cancer cells maintain both high rates of glycolysis and oxidative phosphorylation (OXPHOS) to meet the high demand of anabolic processes 3,4. The high glycolytic flux provides the cancer cell with macromolecules by uncoupling glycolysis from the mitochondrial tricarboxylic acid (TCA) cycle and diverting glucose carbon into biosynthetic pathways, including the pentose phosphate pathway, hexamine pathway and serine biosynthetic pathway 5. During glycolysis and serine biosynthesis NAD+ serves as an electron acceptor for reactions catalysed by glyceraldehyde-3-phospate dehydrogenase (GAPDH) and phosphoglycerate dehydrogenase (PHGDH), and is reduced to NADH. Therefore, to sustain a high glycolytic flux and allow serine biosynthesis NAD+ has to be regenerated. Cancer cells regenerate cytosolic NAD+ through reduction of pyruvate into lactate by the enzyme lactate dehydrogenase (LDH). However, LDH activity alone cannot provide sufficient cytosolic NAD+, and lactate is secreted and therefore glucose carbon is lost for biomolecule synthesis in the cell 6. An additional site of NADH re-oxidation is the mitochondria, where NAD+ is generated by complex I in the electron transport chain (ETC) 7. However, NAD+ and NADH cannot pass the mitochondrial membranes and therefore mitochondria cannot directly re-oxidise cytosolic NADH. Instead cells use substrate cycles that transport electrons derived from oxidation of cytosolic NADH into the mitochondria where the electrons contribute to oxidative respiration and ATP production. One shuttling mechanism is the glycerol-phosphate-shuttle. It transports the electrons from cytosolic NADH to mitochondrial FADH that enters the ETC at complex II using glycerol-phosphate as intermediate electron carrier. Alternatively, the electrons from cytosolic NADH are used to generate malate and transported to mitochondrial NADH that enters the ETC at complex I. This substrate cycle is called the malate-aspartate-shuttle (Fig.
S1A). In either way the carbon flux of the glycolysis is coupled to mitochondrial
function 6,8. It was already recognised very early that cancer cells depend on the MAS for biomolecule synthesis and proliferation, while this is less clear for the GPS 9,10.
The transcription factor CCAAT/enhancer binding protein beta (C/EBPβ) is known to regulate organismal metabolism 11,12. The CEBPB-mRNA is translated into three protein isoforms; two transcriptional activators LAP1 and LAP2 (also named
118
LAP* and LAP) and the smaller transcriptional inhibitor LIP 13,14. LIP expression is tightly controlled by the mTORC1-4E-BP pathway and involves a cis-regulatory upstream open reading frame (uORF) in the CEBPB-mRNA leader sequence 13,15. Overexpression of C/EBPβ-LIP induces cellular transformation 13 and increases the tumour incidence in mice 16 while LIP deficiency reduces tumour incidence in mice
17
. Moreover, LIP is highly expressed in breast cancer, ovarian cancer, colorectal cancer and anaplastic large cell lymphoma 18-23. We recently showed that LIP induces cancer metabolism with increased glycolysis and mitochondrial respiration involving regulation of the let-7/LIN28B circuit (chapter IV).
Here, we show that high LIP expression addicts cells to glucose metabolism. This increased sensitivity is caused by an increased activity of the MAS in cells with high LIP levels. When glycolysis is inhibited in the higher MAS activity results in a low ratio of NADH/NAD+ and apoptosis, and cell viability is rescued by inhibition of the MAS or other NADH-oxidising processes (like LDH). This indicates that low NADH/NAD+ ratios are part of the 2-DG mediated toxicity and that metabolic reprogramming by LIP or otherwise increased MAS activity makes cancer cells vulnerable to glycolytic inhibitors.
Results
High expression of LIP addicts cells to glucose
We examined LIP- or LAP-induced changes in cellular metabolism by measuring the extracellular acidification rate (ECAR) as an indicator for glycolytic flux and the oxygen consumption rate (OCR) as a measure for mitochondrial metabolism (Seahorse XF96). In immortalised C/EBPβ-knockout (KO) MEFs ectopic expression of LIP was sufficient to induce both higher basal and maximal ECAR, while expression of LAP had no effect (Fig. 1A). Inhibition of glycolysis by treatment with 2-deoxyglucose (2-DG) showed that the LIP-induced ECAR is caused by glycolysis (Fig. S1B). In addition, LIP and to a lesser extend LAP increase the OCR in the cell (Fig. 1A and S1B). To investigate whether cells that express LIP or LAP depend on glucose and/or glutamine as alternative carbon sources we cultured the cells in glucose and/or glutamine free medium. Glucose deprivation results in a strong decrease in viable LIP-expressing cells measured after three days of culture compared to LIP-expressing cells cultured in the presence of glucose, while for LAP-expressing cells and empty vector (EV) control cells the viable cell numbers are not different between control (25 mM Glucose) and glucose deprived media (Fig. 1B). The induction of Caspase 3/7 activity in LIP-expressing cells shows that these cells
V
119
Figure 1. LIP overexpression addicts fibroblasts to Glucose consumption
A Western blot of cell extracts from, extracellular acidification rate (ECAR, n=6) and
oxygen consumption rate (OCR, n=6) of C/EBPβ ko fibroblasts and C/EBPβ ko
fibroblasts with ectopic LIP and LAP expression
B Relative cell number of C/EBPβ ko fibroblasts and C/EBPβ ko fibroblasts with ectopic LIP and LAP expression after 3 days of culture in medium with different glutamine and glucose composition (n=5)
C Relative Caspase3/7 activity of C/EBPβ ko fibroblasts and C/EBPβ ko fibroblasts with ectopic LIP and LAP expression after 3 days of culture in medium without glucose (n=5) D Representative pictures of control and 2-DG treated C/EBPβ ko fibroblasts and C/EBPβ
ko fibroblasts with ectopic LIP and LAP expression after one day of treatment
E relative Caspase3/7 activity of C/EBPβ ko fibroblasts and C/EBPβ ko fibroblasts with ectopic LIP and LAP expression after one day of treatment with 2-DG (n=5)
undergo apoptosis upon glucose starvation (Fig. 1C). Deprivation of glutamine alone or both glucose and glutamine resulted in a strong decrease in cell numbers in all
A B C EV LIP LAP 0 1 2 3 4
relative Caspase 3/7 activity upon Glucose starvation
** LIP β -actin C/EBPβko MEF LAP EV LIP LAP cont rol (25 mM Glu cose ) w/o Glu cose & Glu tam ine w/o Glu cose w/o Glu tam ine 0 200 400 600
800 relative cell number after 3 days culture
*** *** *** ** *** * flu o re s c e n t u n it s LIP LAP EV LIP LAP EV *** ** ***
relative Caspase 3/7 activity after 1 day of treatment
2-DG (10 mM) control D 15 10 5 0 LIP LAP EV EV LIP LAP c o n tr o l 2 -D G ( 1 0 m M ) E 200 µm C/EBPβko MEF basal maximal 0 50 100 150 ECAR m p H /m in ** *** LIP LAP EV basal maximal 0 50 100 150 200 250 OCR p m o l/ m in *** *** *** ***
V
β120
three cell lines (EV, LIP, LAP) showing a general requirement of glutamine for cell proliferation independent of LIP or LAP expression (Fig. 1B). Furthermore, treatment with the glycolytic inhibitor 2-DG drives LIP expressing cells into apoptosis as was revealed by a strong increase in Caspase 3/7 activity (Fig. 1D) and by microscopy (Fig. 1E), while LAP expressing cells survive under this condition. These data show that proliferation and survival of LIP expressing MEFs depends on glucose metabolism.
High expression of C/EBPβ-LIP has been described in particular aggressive breast cancer types 18. Immunoblot analysis of a panel of breast cancer cell lines revealed high endogenous LIP expression (high LIP/LAP ratio) in triple negative breast cancer (TNBC) subtype, while low LIP expression with lower LIP/LAP ratios were found in luminal A subtype breast cancer cell lines (Fig. 2A). To determine whether the high LIP expression in TNBC cell lines is associated with higher sensitivity for 2-DG compared to the low LIP expressing Luminal A cell lines we generated dose-response curves for 2-DG treatment and cell multiplication. Of the TNBC type cells BT20 cells with the highest LIP levels were most sensitivity to 2-DG (IC50=0.6 mM), followed by BT549 (IC50=2.4 mM) and MDA-MB-231 (IC50=4.1 mM) (Fig. 2B). The luminal A type cells with low LIP expression show a poor response to 2-DG treatment (IC50>20 mM) (Fig. 2B). Furthermore, the sensitivity to 2-DG correlates with increased Caspase3/7 activity as a measurement for apoptosis in BT20 cells and to lesser extend in BT549 and MDA-MB-231 cells, while Luminal A breast cancer cells show a slight decrease in Caspase3/7 activity (Fig. 2C). To address whether the high sensitivity for 2-DG of BT20 cells depends on C/EBPβ expression we generated Cebpb knockout BT20 cells by CRISPR/Cas9 knockout. In three independent knockout clones (Fig. 2D and S2A) the sensitivity to 2-DG was reduced as measured by cell multiplication (Fig 2E) and a significant reduction in Caspase 3/7 activation, although to various extend (Fig. 2F). Next we investigated whether ectopic expression of LIP in the low LIP-expressing T47D and MCF-7 Luminal A cell lines results in increased sensitivity to 2-DG, as measured by cell multiplication. The higher LIP expression increased the sensitivity to 2-DG by 5-fold for T47D and 2.3-fold for MCF7 (Fig. 2G,H and S2B, C). However, LIP overexpression does not induce Caspase 3/7 activity in T47D cells, suggesting that in addition to LIP other factors contribute to apoptosis induction through 2-DG treatment in the TNBC cells (Fig.
S2D). Taken together, these data show that high levels of LIP expression renders
TNBC cells dependent on glycolysis for cell proliferation and survival.
V
121
Figure 2. high LIP/LAP ratios make breast cancer cells sensitive to 2-deoxyglucose A Western blot of cell extracts and quantified LIP/LAP ratios from BT20, BT549,
MDA-MB-231, MCF7, T47D and ZR-75-1 breast cancer cell lines
B Dose-response-curve of six breast cancer cell lines after 3 days of 2DG treatment (n=5) C Relative Caspase 3/7 activity of six breast cancer cell lines after 3 days of 2-DG
treatment (n=5)
D Western blot of extracts from wt and C/EBPβ ko BT20 breast cancer cells
E Dose-response-curve of wt and C/EBPβ ko BT20 breast cancer cells after 3 days of 2-DG treatment (n=5)
F Relative Caspase 3/7 activity of wt and C/EBPβ ko BT20 breast cancer cells after 3 days of 2-DG treatment (n=5) A B C BT2 0 MD A-M B-2 31 BT5 49 MC F7 T47DZR -7 5-1 LIP LAP Luminal A TNBC αα-Tubulin 0 1 2 3 4 5 LIP/LAP ratio BT2 0 MD A-M B-2 31 BT4 59 MC F7 T47D ZR-7 5-1 D -5 -4 -3 -2 -1 0.0 0.5 1.0 1.5 2-DG sensitivity log c [M] re la ti v e c e ll n u m b e r BT20 MDA-MB-231 BT549 T47D MCF7 ZR-75-1 BT5 49 BT2 0 MD A-M B-2 31 T47D MC F7 ZR-75-1 0 2 4 6 8
relative Caspase3/7 activity after 3 days 2-DG treatment
*** *** *** *** *** G H -4 -3 -2 -1 0 0.0 0.5 1.0 1.5 2-DG sensitivity log c [M] re la tiv e c e ll n u m b e r EV LIP T47D LIP LAP EV LIP ββ-actin -5 -4 -3 -2 -1 0.0 0.5 1.0 1.5 2-DG sensitivity log c [M] wt CBko #1 CBko #2 CBko #3 re la ti v e c e ll n u m b e r control 2-DG (5 mM) 0 2 4 6 8
relative Caspase 3/7 activity after 3 days of 2-DG treatment wt CBko #1 CBko #2 CBko #3 *** *** *** *** *** E 2-DG (5 mM) control *** LIP LAP BT20 ββ-actin
wt ko1 ko2 ko3 F
0.6 mM IC50 2.4 mM 21.4 mM 59.1 mM 38.7 mM 4.1 mM 0.6 mM IC50 1.0 mM 1.4 mM 1.3 mM 64.8 mM IC50 13.6 mM 0.995 0.966 0.973 0.977 0.967 0.99 0.991 0.991 0.991 0.976 0.986 0.963 R² R² R²
V
β β α122
G Western blot of extracts from EV and LIP overexpressing T47D breast cancer cells H Dose-response-curve of EV and LIP overexpressing T47D breast cancer cells after 3
days of 2DG treatment (n=5)
LIP increases the use of glycolysis-derived NADH for mitochondrial respiration
Since ATP promotes cell survival and depletion of ATP results in apoptosis we first examined LIP-induced changes in ATP/ADP ratios to address differences in energy homeostasis 24. In T47D cells the ATP/ADP ratio increases with 46% by ectopic expression of LIP (T47D-LIP cells) compared to the control T47D-EV cells, revealing a higher energy state of the T47D-LIP cells (Fig. 3A). One source of ATP production in cancer cells is the high glycolytic flux, which is accompanied by the secretion of lactate (the Warburg effect). Inhibition of glycolysis with 2-DG reduced the lactate production, measured as ECAR, to the same extend in T47D-LIP cells and control T47D cells (Fig. S3). The expectation was that blocking glycolysis would also result in a comparable degree of reduction of ATP/ADP ratios due to the similar reduction in aerobic glycolysis. However, 2-DG treatment more strongly reduces the ATP/ADP ratio in the T47D-LIP cells with 70% compared to 51% in T47D-EV cells (Fig. 3A). Thus, in addition to the ATP produced during aerobic glycolysis other glucose-dependent pathway(s) contribute for 19% to the production of cellular ATP in T47D-LIP cells, which if prevented compromises cell viability. We reasoned that this could only be related to mitochondrial respiration since most of the ATP is generated in the electron transport chain (ETC).
Mitochondrial metabolism is fuelled by several pathways, most prominently by, pyruvate and NADH from glycolysis, Acyl-CoA from fatty acid metabolism, and glutamate from glutaminolysis. The question was which of the pathways/metabolites is involved in LIP-induced mitochondrial respiration and sensitivity to 2-DG. The engagement of the individual pathways can be studied by measuring changes in oxygen consumption rate (OCR; Seahorse XF96) upon treatment with specific pathway inhibitors (Fig. 3B). The products of glycolysis, pyruvate and electrons from NADH, are imported into the mitochondria to fuel respiration. Inhibition of the mitochondrial pyruvate carrier with the drug UK5099 reduces the OCR in T47D-LIP cells to a much lesser extend compared to control T47D cells, suggesting that pyruvate is not a critical metabolite to fuel respiration in T47D-LIP cells (Fig. 3C). To examine whether the higher OCR in T47D-LIP cells is caused by an increase in usage of palmitate or glutamine, we blocked these pathways with Etomoxir and BPTES
V
123
Figure 3. LIP overexpression induces the usage of electrons from cytoplasmic NADH in the mitochondrion A B C D Glucose Pyruvate Acetyl-CoA Fatty acids Palmitate NADH Glutamine Glutamate OAA OAA Asp Asp Mal Mal NADH NAD+ αKG EtomoxirBPTES 2-DG UK5099 mitochondrion cytoplasma 0 50 100 20 40 60 80 100 120 Time (minutes) O C R ( % ) EV LIP UK5099 BPTES/ Etomoxir 2-DG Oligomycin UK5099 BPTES Etomoxir + 2-DG 0 20 40 60 80 OCR [%] *** *** T47D 0 20 40 60 80 100 0 100 200 300 400 Time (minutes) O C R [ p mo l/ mi n ] EV LIP Saponin Glutamate Malate Oligomycin H LIP EV G UK5099 BPTES Etomoxir + Rotenone 0 20 40 60 80 OCR [%] *** LIP EV F UK5099 BPTES/ Etomoxir Rotenone Oligomycin 0 50 100 20 40 60 80 100 120 Time (minutes) O C R [ %] EV LIP T47D E Glucose Pyruvate NADH OAA OAA Asp Asp Mal Mal NADH NAD+ 2-DG NADH NAD+ NAD+ NADH ETC complex 1 Rotenone DHA-P G-3-P FAD FADH 2 ETC complex 2 G-3-P DHA-P mitocho ndrion malate-aspertate-shuttle glycerol-3- phosphate-shuttle cytoplasma 50 mM 2-DG ctr. ctr. 50 mM 2-DG 0.0 0.5 1.0 1.5 2.0
relative ATP/ADP ratio
*** *** *** LIP -51% +46% -70% EV
V
124
A relative ATP/ADP ratios of control and DG treated EV and LIP overexpressing T47D breast cancer cells (n=5)
B Schematic representation of main nutrients used for ETC within the mitochondrion and their inhibitors
C Oxygen consumption rate of control EV and LIP overexpressing T47D breast cancer cells at basal condition and after injection of UK5099, BPTES+etomoxir, 2-DG and oligomycin (n=6)
D Oxygen consumption rate of control EV and LIP overexpressing T47D breast cancer cells after injection of UK5099, BPTES and etomoxir and after the addition of 2-DG (n=6) E Schematic representation of process which transport NADH from cytoplasm to the
mitochondrion
F Oxygen consumption rate of control EV and LIP overexpressing T47D breast cancer cells at basal condition and after injection of UK5099, BPTES+etomoxir, rotenone and oligomycin (n=6)
G Oxygen consumption rate of control EV and LIP overexpressing T47D breast cancer cells at basal condition and after injection of saponin (permeabilization of cell membrane), glutamate, malate and oligomycin (n=6)
H Oxygen consumption rate of control EV and LIP overexpressing T47D breast cancer cells after injection of UK5099, BPTES and etomoxir and after the addition of rotenone (n=6)
(Bis-2-5-phenylacetamido-1,3,4-thiadiazol-2-ylethyl sulphide). With this treatment T47D-LIP cells maintain their higher OCR compared to the T47D-EV cells (Fig. 3D).
Thus, although metabolism of pyruvate, Acyl-CoA and glutamine is blocked the high-LIP T47D-LIP cells still have a higher respiratory capacity compared to low-LIP T47D-EV cells. One possible metabolite that after UK5099/Etomoxir/BPTES treatment can fuel mitochondrial respiration is the NADH generated by glycolysis. Therefore, we used 2-DG to inhibit glycolysis on top of the other three drugs and found the OCR of T47D-LIP cells decreased to similar extend than control T47D cells (Fig. 3C and D). These data show that LIP increases the usage of cytosolic glycolysis-derived NADH for promoting mitochondrial respiration.
LIP stimulates the malate aspartate shuttle (MAS)
The two pathways that transport electrons derived from cytosolic NADH into the mitochondria are the malate-aspartate-shuttle (MAS) and the glycerol-phosphate-shuttle (GPS) (Fig. 3E and Fig. S1A). Because electrons in the MAS are used to generate NADH that enters complex I of the ETC and electrons in the GPS are used to generate FADH that enters the ETC at complex II, we can discriminate between those two by inhibition of complex I with rotenone (Fig 3E). Following inhibition of
V
125
mitochondrial pyruvate (UK5099), palmitate (Etomoxir) and glutamine (BPTES) metabolism, rotenone abrogates the difference in OCR caused by LIP expression in T47D cells, indicating that MAS is the critical pathway involved in enhancing mitochondrial respiration (Fig. 3F and G). Next, we examined the capacity of the MAS in T47D-LIP versus control T47D cells by experimentally applying exogenous malate after cell membrane permeabilisation using saponin. Permeabilisation of the cell membrane resulted in a strong reduction of the OCR for both cell lines, because substrates for mitochondrial respiration are leaking out (Fig. 3H). The subsequent supply of glutamate slightly increases the OCR in both cell lines while the supply of the MAS substrate malate results in a strong increase in OCR in the T47D-LIP cells and to a much lesser extend in the control T47D cells (Fig. 3H). These data show LIP stimulates MAS to transport electrons from cytoplasmic NADH into the mitochondria to stimulate mitochondrial respiration.Altered NADH usage causes apoptosis in LIP overexpressing cells
Next we asked whether the LIP-induced increase in MAS makes the cells sensitive to inhibition of glycolysis. Hypothetically, inhibition of glycolysis while LIP-driven MAS continues to convert cytosolic NADH into NAD+ may lead to lower NADH/NAD+ ratios in cells with potential detrimental effects on cell viability. Cytosolic NADH/NAD+ ratios can be experimentally increased by treatment with aminooxyacetic acid (AOA) that broadly inhibits transaminases including the aspartate aminotransferase of the MAS (Fig. 4A). Alternatively, inhibition of LDH with oxamate prevents the usage of NADH for reduction of pyruvate into lactate and thereby increases the NADH/NAD+ ratio (Fig. 4A). First we examined BT20 cells that are the TNBC cells with the highest LIP expression in our study (Fig. 2A) and most sensitive to DG (Fig. 2B and C). Inhibition of glycolysis in BT20 cells by 2-DG in resulted in a strong decrease in NADH/NAD+ ratio (Fig. 4B) and in a strong induction of apoptosis (Caspase 3/7 activity) (Fig. 4C). Subsequent inhibition of the MAS with AOA fully restored the cellular NADH/NAD ratio (Fig. 4B) in 2-DG treated cells and results in a significant although not full decrease in apoptosis (Fig.
4C). Notably, the ATP/ADP ratio in the 2-DG treated cell does not increase upon
additional AOA treatment (Fig. D) and therefore cannot contribute to the increase in cell viability. The incomplete rescue from apoptosis may be due to AOA being a “promiscuous” drug, inhibiting all transaminases in the cell with potential affecting cell survival 25. In the TNBC cell lines BT549 and MDA-MB-231 that have lower LIP expression 2-DG treatment likewise lowers the NADH/NAD+, although to a lesser
126
extent then in BT20 cells (Fig. S4A, S4D). Also in these cell lines subsequent inhibition of the MAS with AOA restores the cellular NADH/NAD+ ratio (Fig. S4A,
Figure 4. Blockage of NADH consuming processes in the cytoplasm reduces toxicity of 2-DG
A Schematic representation of cytoplasmic NADH consuming processes and their inhibitors
B Relative NADH/NAD+ ratios in BT20 breast cancer cells after 1 day of treatment with
solvent, 2-DG, AOA, or 2-DG and AOA (n=4)
C Relative caspase3/7 activity of BT20 breast cancer cells after 3 days of treatment with solvent, 2-DG, AOA or 2-DG and AOA (n=4)
BT20 B control 5 mM 2-DG * *** F A Glucose Pyruvate Acetyl-CoA NADH OAA OAA Asp Asp Mal Mal NADH NAD+ KG 2-DG Oxamate NADH NAD+ Lactate AOA Lactate cytoplasm mitochondrium control 5 mM 2-DG 0 2 4 6 8 10 *** * * 1.5 mM AOA control 50 mM oxamate control caspase 3/7 activity caspase 3/7 activity C D control 5 mM 2-DG 0.0 0.5 1.0 1.5 NADH/NAD+ ratio NADH/NAD+ ratio *** * *** control 5 mM 2-DG 0.0 0.5 1.0 1.5 ATP/ADP ratio ** ** * control 5 mM 2-DG 0 2 4 6 *** *** *** E BT20 G control 5 mM 2-DG 0.0 0.5 1.0 1.5 ATP/ADP ratio * * ***
V
127
D Relative ATP/ADP ratios in BT20 breast cancer cells after 1 day of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=4)
E Relative NADH/NAD+ ratios in BT20 breast cancer cells after 1 day of treatment with solvent, 2-DG, oxamate or 2-DG and oxamate (n=4)
F Relative caspase3/7 activity of BT20 breast cancer cells after 3 days of treatment with solvent, 2-DG, oxamate or 2-DG and oxamate (n=4)
G Relative ATP/ADP ratios in BT20 breast cancer cells after 1 day of treatment with solvent, 2-DG, oxamate or 2-DG and oxamate (n=4)
S4D), results in a significant decrease in apoptosis (Fig. S4B, S4E), and does not
alter the ATP/ADP ratio in the 2-DG treated cells (Fig. S4C, S4F).
Independently of the MAS, inhibition of LDH with oxamate strongly increases the cytoplasmic NADH/NAD+ ratio in 2-DG treated and untreated BT20 cells (Fig. 4A, E). Notably, the strong induction of apoptosis by 2-DG treatment is almost completely reverted by oxamate treatment (Fig. 4F). In this case the ATP/ADP ratio slightly increases in the 2-DG and oxamate treated cells, but does not reach control levels (Fig. 4G). This increased energy level upon combined oxamate and 2-DG treatment might be caused by an increased availability of NADH for the MAS and OXPHOS which again suggest the MAS as an important energy source in BT20 cells (Fig. S4B, C and G). Together, the data indicate that the increased usage MAS drains NADH from the cytoplasm, which contributes to the toxicity of 2-DG in high-LIP expressing cancer cells. In further support of this hypothesis, treatment with 2-DG resulted in a stronger reduction (71%) in NADH/NAD+ ratios in T47D-LIP cells compared to the reduction (47%) in parent T47D cells (Fig. S4G). Inhibition of the MAS by AOA in addition to 2-DG treatment restores the NADH/NAD+ ratio with 15% in T47D-LIP but not in T47D-EV cells.
Duroquinone makes cells sensitive to 2-DG
NAD(P)H dehydrogenase (quinone) 1 (NQO1) is a cytoplasmic detoxification enzyme which uses NAD(P)H to reduce different free quinone species in the cell to prevent ROS production caused by free quinones. Duroquinone (DQ) is a known NQO1 substrate. We used duroquinone (DQ) treatment to experimentally reduce the cytoplasmic NADH/NAD+ ratio. Thereby we wanted to investigate whether a reduction of the NADH/NAD+ ratio is sufficient to increase the sensitivity to 2-DG treatment and thus can mimic the effect of LIP-induced MAS. Treatment with DQ lowers the NADH/NAD+ ratio and increases the ATP/ADP ratio in MCF7 cells (Fig.
5A and B). On the contrary, 2-DG treatment has no effect on the cellular
128
NADH/NAD+ ratio but significantly lowers the ATP/ADP ratio in MCF7 cells (Fig.
5A and B). Double treatment of MCF7 cells with 2-DG and DQ lowers NADH/NAD+
Figure 5. Pharmacological reduction of NADH level makes cells sensitive to DG A Relative NADH/NAD+ ratios in MCF7 breast cancer cells after 1 day of treatment with
solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
B Relative ATP/ADP ratios in MCF7 breast cancer cells after 1 day of treatment with solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
C Relative cell numbers of MCF7 breast cancer cells after 3 days of treatment with solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
D Relative NADH/NAD+ ratios in BT20 breast cancer cells after 1 day of treatment with solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
E Relative ATP/ADP ratios in BT20 breast cancer cells after 1 day of treatment with solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
F Relative cell numbers of BT20 breast cancer cells after 3 days of treatment with solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
G Western blot of extracts from BT20, Hepa1-6 and HEK293T cancer cell lines
A B C control 5 mM DG 0.0 0.5 1.0 1.5 cell number * * * control 100 µM DQ MCF7 control 5 mM DG 0.0 0.5 1.0 1.5 NADH/NAD+ ratio *** *** control 5 mM DG 0.0 0.5 1.0 1.5 2.0 ATP/ADP ratio *** ** * D control 5 mM DG 0.0 0.5 1.0 1.5 cell number control 100 µM DQ ** *** *** Hepa1-6 control 5 mM DG 0.0 0.5 1.0 1.5 cell number *** *** *** HEK293T E control 5 mM DG 0.0 0.5 1.0 1.5 cell number *** *** ns control 100 µM DQ BT20 control 5 mM DG 0.0 0.5 1.0 1.5 NADH/NAD+ ratio * *** ** control 5 mM DG 0.0 0.5 1.0 1.5 ATP/ADP ratio ** * F H I G mouse LAP LIP human LAP ββ-actin BT20 Hepa 1-6 HE K293 T
V
β129
H Relative cell numbers of HEK293T cancer cells after 3 days of treatment with solvent, 2-DG, duroquinone, or 2-DG and DQ (n=4)
I Relative cell numbers of Hepa1-6 cancer cells after 3 days of treatment with solvent,
2-DG, duroquinone, or 2-DG and DQ (n=4)
and ATP/ADP ratios similar to 2-DG in cancer cells with high LIP level (BT20) in the absence of DQ (Fig. 5A and B, compare to Fig. 5E, G). Caspase activity is not increased in MCF7 or BT20 cells upon 2-DG and DQ double treatment (Fig. S5A and
B). However, combined treatment of 2-DG and DQ in cells with low LIP level
(MCF7) leads to a lower cell number after three days of proliferation compared to DQ or 2-DG treatment alone (Fig. 5C, H and I). In contrast, cells with high level of LIP (BT20) don’t show the synergistic effects of the double treatment on cell numbers after three days of proliferation (Fig. 5F). This indicates that low NADH/NAD+ ratios are partly responsible for the toxicity of 2-DG.
Discussion
Here, we show that C/EBPβ-LIP addicts cells to aerobic glycolysis by inducing the malate aspartate shuttle (MAS) for the transfer of electrons from cytoplasmic NADH into the mitochondria for ATP production at the electron transport chain (ETC). Depletion of glucose from the medium or inhibition of glycolysis by 2-deoxyglucose (2-DG) induces apoptosis in cells with high expression of LIP, while elimination of LIP or inhibition of the MAS reduces the toxicity of 2-DG in these cells. Our data suggest that the significant decrease in NADH/NAD+ ratio upon 2-DG treatment contributes to the cell toxicity, since increasing the NADH/NAD+ ratio by inhibition of NADH oxidising processes in the cytoplasm (MAS or LDH) alleviates the toxicity of 2-DG.
The MAS is a key process in the cell to connect metabolic pathways in the mitochondria and the cytoplasm and has been linked to cancer metabolism. For example, pancreatic cancer cells show an increased activity of MAS to stimulate glutamine metabolism and NADPH production 26. Another study found that amplification of MDH1 and enhanced MAS activity in non-small lung carcinoma is required as an alternative to LDH-catalysed NAD+ generation, since glucose carbons are shuttled in biosynthetic pathways and therefore the carbon supply to LDH is not sufficient to regenerate NAD+ to maintain the high glycolytic flux 10. We show in this manuscript that this activation of the MAS can become a burden for cancer cells. We show that upon glucose starvation or inhibition of glycolysis an active MAS creates low NADH/NAD+ ratios that are toxic to cells.
130
NAD+ is an essential co-factor for metabolic reactions and a substrate for reactions in cell signalling pathways, including sirtuins and poly-adenosine ribose-polymerase (PARP) 27,28. In addition, the ratio between NADH and NAD+ reflects the redox state of the cell 29. So far no mechanism is known by which NADH supresses or NAD+ induces apoptosis. However, recently it was shown that NADH is able to bind to apoptosis inducing factor (AIF), inducing its dimerization and thereby maintaining the AIF-dimers in the mitochondria as the apoptosis-inactive form 30. Low levels of NADH results in AIF monomerization and the AIF monomers can leave the mitochondria and translocate into the nucleus to induce apoptosis. In addition to the NADH level, AIF translocation is regulated by mitochondrial membrane integrity and protein maturation by cleavage. In order for AIF to be apoptogenic, the mitochondria targeting signal and membrane associated domain has to be cleaved off the AIF precursor protein by calpain or another protease. Therefore, multiple pathways are required to activate AIF and induce apoptosis 31 and alterations in these AIF translocation pathways might influence 2-DG sensitivity similar to LIP.
We analysed available transcriptome and proteome data of Cebpb knockout fibroblast with ectopic LIP overexpression (data not shown) and we found no regulation of parts of the MAS at the RNA or protein level by LIP. However, the activity of MAS seems to be mainly regulated by post-transcriptional modifications of MDH1. For example, the protein arginine N-methyltransferase CARM1 methylates and inhibits MDH1 by disrupting its dimerization, required for its activation. MDH1 methylation supresses MAS activity and inhibits mitochondrial respiration of pancreatic cancer cells 26. Intriguingly, LAP but not LIP interacts with CARM1 32. LAP therefore might influence the activity or the interaction between CARM1-targets and CARM1, which may be counteracted by LIP interaction with the same CARM1-targets.
So far, cancer therapy with 2-DG is only tested for a few specific cancer types or in combination with other chemotherapeutic drugs 33. Our study suggests that high LIP expression may be used as a biomarker for effectiveness of 2-DG or other glycolysis inhibiting drugs in cancer treatment. Further work is needed to evaluate the predictive power of LIP expression for 2-DG treatment success. Furthermore, we identified the NADH/NAD+ ratio as an important mediator of 2-DG toxicity in vitro. More experiments are required to evaluate whether 2-DG lowers the NADH/NAD+ ratio in tumours or whether artificial oxidation of NADH with small compounds (e.g. duroquinone) will make cancer cells sensitive to 2-DG treatment. In addition, overexpression of LIP or duroquinone treatment significantly lowers NADH/NAD+
V
131
ratios in luminal A breast cancer cells and makes them more sensitive to 2-DG. However, duroquinone treatment fails to induce apoptosis in these cells. Therefore additional factor(s) seem to play a role for to induce apoptosis when NADH/NAD+ ratios are low. The identification of these factors could be, next to LIP, important biomarkers for the toxicity of 2-DG.Taken together, we found that the transcriptional repressor LIP makes cells sensitive to 2-DG treatment by stimulating the use of cytoplasmic NADH. We describe that the consequently low NADH/NAD+ ratio is an important mediator of 2-DG induced cells death in triple negative breast cancer cells. Furthermore, we suggest a new model for 2-DG sensitivity in which a low NADH/NAD+ ratio (mediated by high LIP or MAS) has to be accompanied by the presence or absence of one or more additional factor(s) that respond to the low NADH/NAD+ ratio and induce apoptosis.
Acknowledgements
We thank Stefan Juranek and Floris Foijer for supervising the CRISPR/CAS9 facility where the BT20 CBko cells were generated and Hilde Jalving for sharing BT549, MDA-MB-231 and ZR-75-1 breast cancer cell lines with us. T.A. and G.K. were supported by the Dutch Cancer Society (KWF, #10080) through grants to C.F.C. and C.M. G.H. was supported by Deutsche Krebshilfe through a grant (DFH, 612100) to C.F.C.
Authorship contributions
T.A. designed and performed the research, and collected and analysed the data; H.R.Z., G.K and M.A.Z. performed research and collected data; G.H. and C.F.C. designed research and supervised the project; T.A. and C.F.C. wrote the manuscript.
Materials and methods
Cell culture
Hepa1-6 cells, HEK293T cells, BT20 cells, MDA-MB-231 and all immortalized MEF cell lines were culture in high glucose DMEM supplemented with 10% FBS, 10mM HEPES, 1mM Sodium Pyruvate and 100U/ml Penicillin Streptomycin. BT549, ZR-75-1, T47D and MCF7 breast cancer cells were maintained in RPMI1640 medium supplemented with 10% FBS, 25mM HEPES, 1mM Sodium Pyruvate and 100U/ml Penicillin/Streptomycin. C/EBPβ ko MEF were described before (Zidek et al., 2015).
132
DNA constructs
Plasmids containing rat C/EBPβ-LAP, rat C/EBPβ-LIP, human C/EBPβ-LIP and Flag-tagged rat C/EBPβ-LIP were described before 11. For CRISPR/Cas9 mediated knock out of both C/EBPβ isoforms, guide RNA sequences (LAP: 5’-GAGTGGCCAACTTCTACTACG-3’, LIP: 5’-GCGCTTACCTCGGCTACCAGG-3’) were cloned into pSpCas9(BB)-2A-puro (PX459) v2.0 (http://www.addgene.org/62988/).
Transfection
Immortalized MEFs were transfected with an empty, rat C/EBPβ-LIP or -LAP containing pcDNA3 or pSV2Zeo vector by using FugeneHD (Promega) according to the manufactures protocol. For stable overexpression, C/EBPβko MEFs were treated with 0.2 mg/ml Zeocin (Invitrogen). To maintain the expression cells were culture with 0.1 mg/ml Zeocin in the medium. T47D cells and MCF7 cells were transfected with empty or human LIP-containing pcDNA3.1 via Fugene HD (Promega) using the manufactures protocol. For stable expression, MCF7 cells were selected with 0.8 mg/ml, T47D with 0.4mg/ml G418. For CRISPR/Cas9 mediated knock out of both C/EBPβ isoforms, BT20 cells cells were transfected with Fugene HD (Promega) according to the manufactures protocol and selected with puromycine (1 µg/ml). After the selection, clones were grown out and C/EBPβ level were analysed by western blot.
Proliferation assays
To determine the proliferation and survival of cancer cells, relative cell numbers were measured using the CellTiter-Fluor™ Cell Viability Assay (Promega) after 3 days of treatment. Measurements were performed according to the suppliers manual.
Metabolic flux analysis
Metabolic flux analyses were performed using a Seahorse XF96 Extracellular Flux analyser (Agilent Bioscience). 3x104 EV and LIP overexpressing T47D cells were seeded 16 hours before the experiments. Assays were performed according to the manufactures protocol. Injected drugs were UK5099 (5 µM) for blockage of mitochondrial pyruvate transporter, BPTES (3 µM) for blockage of glutaminase, etomoxir (40 µM) for blockage of fatty acid transport into the mitochondrion, 2-DG (100 mM) inhibition of glycolysis, Rotenone (4 µM) blockage of complex 1 and oligomycin (2,5 µM) for blockage of ATP related respiration. To measure the activity
V
133
of the MAS, cells were permeabilized by injection of saponin (25 µg/ml) and the substrates of the MAS (Glutamate (1 mM) and Malate (2 mM)) were added separately by single injections.Luciferase based assays
NADH, NAD+, ATP and ADP level were distinguished using luciferase based assays. 24 hours before the assay, 7500 cells per well were seeded in a 96-well plate. Experiments were performed according to manufactures protocols (NADH/NAD+: Promega, G9071; ATP/ADP: Biovision, K255-200). For detection, a GloMax-Multi Detection System (Promega) was used.
Caspase3/7 activity was measured 3 days after the treatment with a commercial available kit (Caspase-Glo 3/7 Assay, Promega).
Immunoblot analysis
Cells and tissues were lysed using RIPA buffer. Equal amounts of protein were separated via SDS-PAGE and transferred to a PVDF membrane using Trans-Blot Turbo System (Bio-rad).
The following antibodies were used for detection: C/EBPβ (E299) from Abcam, α-Tubulin (GT114) from GeneTex and β-actin (clone C4) (#691001) from MP Biomedicals. For detection, HRP-conjugated secondary antibodies (Amersham Life Technologies) were used. The signals were visualized by chemiluminescence (ECL, Amersham Life Technologies) using ImageQuant LAS 4000 mini imaging machine (GE Healthcare Bioscience AB) and the supplied software was used for the quantification of the bands.
References
1 Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646-674, doi:10.1016/j.cell.2011.02.013 (2011).
2 Koppenol, W. H., Bounds, P. L. & Dang, C. V. Otto Warburg's contributions to current concepts of cancer metabolism. Nat Rev Cancer 11, 325-337, doi:10.1038/nrc3038 (2011).
3 DeBerardinis, R. J. & Chandel, N. S. Fundamentals of cancer metabolism. Sci Adv 2, e1600200, doi:10.1126/sciadv.1600200 (2016).
4 Vander Heiden, M. G. & DeBerardinis, R. J. Understanding the Intersections between Metabolism and Cancer Biology. Cell 168, 657-669, doi:10.1016/j.cell.2016.12.039 (2017).
134
5 Pavlova, N. N. & Thompson, C. B. The Emerging Hallmarks of Cancer Metabolism. Cell Metab 23, 27-47, doi:10.1016/j.cmet.2015.12.006 (2016).
6 DeBerardinis, R. J., Lum, J. J., Hatzivassiliou, G. & Thompson, C. B. The biology of cancer: metabolic reprogramming fuels cell growth and proliferation. Cell Metab 7, 11-20, doi:10.1016/j.cmet.2007.10.002 (2008).
7 Gui, D. Y. et al. Environment Dictates Dependence on Mitochondrial Complex I for NAD+ and Aspartate Production and Determines Cancer Cell Sensitivity to Metformin. Cell Metab 24, 716-727, doi:10.1016/j.cmet.2016.09.006 (2016).
8 Birsoy, K. et al. An Essential Role of the Mitochondrial Electron Transport Chain in Cell Proliferation Is to Enable Aspartate Synthesis. Cell 162, 540-551, doi:10.1016/j.cell.2015.07.016 (2015).
9 Greenhouse, W. V. & Lehninger, A. L. Occurrence of the malate-aspartate shuttle in various tumor types. Cancer Res 36, 1392-1396 (1976).
10 Hanse, E. A. et al. Cytosolic malate dehydrogenase activity helps support glycolysis in actively proliferating cells and cancer. Oncogene 36, 3915-3924, doi:10.1038/onc.2017.36 (2017).
11 Zidek, L. M. et al. Deficiency in mTORC1-controlled C/EBPbeta-mRNA translation improves metabolic health in mice. EMBO Rep 16, 1022-1036, doi:10.15252/embr.201439837 (2015).
12 Millward, C. A. et al. Mice with a deletion in the gene for CCAAT/enhancer-binding protein beta are protected against diet-induced obesity. Diabetes 56, 161-167, doi:10.2337/db06-0310 (2007).
13 Calkhoven, C. F., Muller, C. & Leutz, A. Translational control of C/EBPalpha and C/EBPbeta isoform expression. Genes Dev 14, 1920-1932 (2000).
14 Descombes, P. & Schibler, U. A liver-enriched transcriptional activator protein, LAP, and a transcriptional inhibitory protein, LIP, are translated from the same mRNA. Cell 67, 569-579 (1991).
15 Calkhoven, C. F., Bouwman, P. R., Snippe, L. & Ab, G. Translation start site multiplicity of the CCAAT/enhancer binding protein alpha mRNA is dictated by a small 5' open reading frame. Nucleic Acids Res 22, 5540-5547 (1994).
16 Begay, V. et al. Deregulation of the endogenous C/EBPbeta LIP isoform predisposes to tumorigenesis. J Mol Med (Berl) 93, 39-49, doi:10.1007/s00109-014-1215-5 (2015).
17 Müller, M. et al. Reduced expression of C/EBPβ-LIP extends health- and lifespan in mice. bioRxiv, doi:https://doi.org/10.1101/250100 (2018).
135
18 Zahnow, C. A., Younes, P., Laucirica, R. & Rosen, J. M. Overexpression of C/EBPbeta-LIP, a naturally occurring, dominant-negative transcription factor, in human breast cancer. J Natl Cancer Inst 89, 1887-1891 (1997).19 Jundt, F. et al. A rapamycin derivative (everolimus) controls proliferation through down-regulation of truncated CCAAT enhancer binding protein {beta} and NF-{kappa}B activity in Hodgkin and anaplastic large cell lymphomas. Blood 106, 1801-1807, doi:2004-11-4513 [pii]10.1182/blood-2004-11-4513 (2005).
20 Zahnow, C. A., Cardiff, R. D., Laucirica, R., Medina, D. & Rosen, J. M. A role for CCAAT/enhancer binding protein beta-liver-enriched inhibitory protein in mammary epithelial cell proliferation. Cancer Res 61, 261-269 (2001).
21 Sundfeldt, K. et al. The expression of CCAAT/enhancer binding protein (C/EBP) in the human ovary in vivo: specific increase in C/EBPbeta during epithelial tumour progression. Br J Cancer 79, 1240-1248, doi:10.1038/sj.bjc.6690199 (1999). 22 Rask, K. et al. Increased expression of the transcription factors CCAAT-enhancer
binding protein-beta (C/EBBeta) and C/EBzeta (CHOP) correlate with invasiveness of human colorectal cancer. Int J Cancer 86, 337-343 (2000).
23 Quintanilla-Martinez, L. et al. NPM-ALK-dependent expression of the transcription factor CCAAT/enhancer binding protein beta in ALK-positive anaplastic large cell lymphoma. Blood 108, 2029-2036, doi:10.1182/blood-2005-10-014258 (2006).
24 Vander Heiden, M. G., Cantley, L. C. & Thompson, C. B. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029-1033, doi:10.1126/science.1160809 (2009).
25 Thornburg, J. M. et al. Targeting aspartate aminotransferase in breast cancer.
Breast Cancer Res 10, R84, doi:10.1186/bcr2154 (2008).
26 Wang, Y. P. et al. Arginine Methylation of MDH1 by CARM1 Inhibits Glutamine Metabolism and Suppresses Pancreatic Cancer. Mol Cell 64, 673-687, doi:10.1016/j.molcel.2016.09.028 (2016).
27 Canto, C., Menzies, K. J. & Auwerx, J. NAD(+) Metabolism and the Control of Energy Homeostasis: A Balancing Act between Mitochondria and the Nucleus.
Cell Metab 22, 31-53, doi:10.1016/j.cmet.2015.05.023 (2015).
28 Canto, C., Sauve, A. A. & Bai, P. Crosstalk between poly(ADP-ribose) polymerase and sirtuin enzymes. Mol Aspects Med 34, 1168-1201, doi:10.1016/j.mam.2013.01.004 (2013).
29 Zhao, Y. et al. SoNar, a Highly Responsive NAD+/NADH Sensor, Allows High-Throughput Metabolic Screening of Anti-tumor Agents. Cell Metab 21, 777-789, doi:10.1016/j.cmet.2015.04.009 (2015).
136
30 Brosey, C. A. et al. Defining NADH-Driven Allostery Regulating Apoptosis-Inducing Factor. Structure 24, 2067-2079, doi:10.1016/j.str.2016.09.012 (2016). 31 Sevrioukova, I. F. Apoptosis-inducing factor: structure, function, and redox
regulation. Antioxid Redox Signal 14, 2545-2579, doi:10.1089/ars.2010.3445 (2011). 32 Kowenz-Leutz, E., Pless, O., Dittmar, G., Knoblich, M. & Leutz, A. Crosstalk
between C/EBPbeta phosphorylation, arginine methylation, and SWI/SNF/Mediator implies an indexing transcription factor code. EMBO J 29, 1105-1115, doi:10.1038/emboj.2010.3 (2010).
33 Zhang, D. et al. 2-Deoxy-D-glucose targeting of glucose metabolism in cancer cells as a potential therapy. Cancer Lett 355, 176-183, doi:10.1016/j.canlet.2014.09.003 (2014).
137
Supplementary figures
Figure S1. LIP overexpression addicts fibroblasts to Glucose consumption
A Schematic representation of malate-aspartate-shuttle and glycerol-3-phosphate-shuttle B Extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) of of C/EBPβ
ko fibroblasts and C/EBPβ ko fibroblasts with ectopic LIP and LAP expression (n=6)
0 10 20 30 40 0 50 100 150 200 Time (minutes) p mo l/ mi n EV LAP LIP DNP Oligomycin Aspartate Oxaloacetate Malate NAD+ NADH Oxaloacetate Malate NAD+ NADH α-Keto glutarate Glutamate α-Keto glutarate Glutamate Dihydroxyacetone-phosphate phosphate Dihydroxyacetone-phosphate Glycerol-3-phosphate Glycerol-3-phosphate NAD+ NADH FAD FADH2 malate-aspartate-shuttle Glycerol-3-phosphate shuttle ETC complex I NAD+ ETC complex II FAD A B 0 10 20 30 40 0 25 50 75 100 125 Time (minutes) m p H /m in 2-DG Oligomycin ECAR OCR
V
138
Figure S2. high LIP/LAP ratios make breast cancer cells sensitive to 2-deoxyglucose A Schematic representation of CEBPB locus and the CRISPR/Cas9 based knockout
strategy (arrows mark the guide RNAs)
B Western blot of cell extracts from EV and LIP overexpressing MCF7 breast cancer cells C Dose-response-curve of EV and LIP overexpressing MCF7 breast cancer cells after 3
days of 2-DG treatment (n=5)
D Relative Caspase 3/7 activity of EV control and LIP overexpressing T47D cells after 3 days of 2-DG treatment (n=5)
Figure S3. LIP overexpression induces the usage of electrons from cytoplasmic NADH in the mitochondrion
Extracellular acidification rate of EV control and LIP overexpressing T47D breast cancer cells upon control conditions and 2-DG treatment (n=6)
B -5 -4 -3 -2 -1 0 0.0 0.5 1.0 1.5 2-DG sensitivity log c [M] MCF7 LIP LAP EV LIP ββ-actin C A
ATG2 - LAP ATG3 - LIP ATG1
ATGugRNA1 gRNA2
Cebpb locus re la ti v e c e ll n um be r EV LIP R² 91.2 mM IC50 39.4 mM 0.98 0.979 T47D 0.0 0.5 1.0 1.5 relative ECAR m p H /m in *** *** 50 mM 2-DG ctr. ctr. 50 mM 2-DG LIP EV
V
β139
Figure S4. Blockage of NADH consuming processes in the cytoplasm reduces toxicity of 2-DG
A Relative NADH/NAD+ ratios in BT549 breast cancer cells after 1 day of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=4)
B Relative caspase3/7 activity of BT549 breast cancer cells after 3 days of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=4)
C Relative ATP/ADP ratios in BT549 breast cancer cells after 1 day of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=4)
D Relative NADH/NAD+ ratios in MDA-MB-231 breast cancer cells after 1 day of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=3)
E Relative ATP/ADP ratios in MDA-MB-231 breast cancer cells after 1 day of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=3)
control 5 mM DG 0.0 0.5 1.0 1.5 NADH/NAD+ ratio * p=0.055 A B C control 2 mM AOA MDA-MB-231 control 0.25 mM AOA BT549 control 5 mM DG 0.0 0.5 1.0 1.5 2.0 2.5 caspase3/7 activity * * *** control 5 mM DG 0.0 0.5 1.0 1.5 NADH/NAD+ ratio *** *** control 5 mM DG 0.0 0.5 1.0 1.5 ATP/ADP ratio *** *** G T47D 0.0 0.5 1.0 1.5 NADH/NAD+ ratio *** *** *** ns -47% -71% +15% LIP EV 2-DG ctr. AOA 2-DG + AOA 2-DG + AOA ctr. 2-DG D E F control 5 mM DG 0.0 0.5 1.0 1.5 ATP/ADP ratio *** control 5 mM DG 0 2 4 6 caspase 3/7 activity * *** ***
V
140
F Relative caspase3/7 activity in MDA-MB-231 breast cancer cells after 3 days of treatment with solvent, 2-DG, AOA, or 2-DG and AOA (n=4)
G Relative NADH/NAD+ ratios in EV and LIP overexpressing T47D breast cancer cells after 1 day of treatment with solvent, 2-DG, AOA or 2-DG and AOA (n=4)
Figure S5. Pharmacological reduction of NADH level makes cells sensitive to DG A Relative caspase3/7 activity in MCF7 breast cancer cells after 3 days of treatment with
solvent, 2-DG, duroquinone, or 2-DG and duroquinone (n=4)
B Relative caspase3/7 activity in BT20 breast cancer cells after 3 days of treatment with solvent, 2-DG, duroquinone, or 2-DG and duroquinone (n=4)
A B MCF7 BT20 control 5 mM DG 0 1 2 3 4 5 caspase 3/7 activity control 100 µM DQ *** control 5 mM DG 0.0 0.5 1.0 1.5 caspase 3/7 activity * ***
