In colorectal cancer cells, dichloroacetate causes apoptosis and cell-cycle arrest.

Background: Glycolysis is a crucial process for cancer cells. Our goal was to investigate the underlying mechanisms and evaluate if switching from glycolysis to mitochondrial respiration would inhibit growth preferentially in colorectal cancer cells as opposed to normal cells.

Methods: Dichloroacetate, a pyruvate dehydrogenase kinase inhibitor, was used to treat representative colorectal cancer and non-cancerous cell lines.

Results: Dichloroacetate (20 mM) significantly reduced cancer cell proliferation (P=0.009), which was linked to apoptosis and G2 phase cell-cycle arrest. Dichloroacetate (20 mM) had no effect on the growth of non-cancerous cells. Metastatic LoVo cells showed the greatest apoptotic effect; after 48 hours, DCA caused an up to ten-fold increase in the number of apoptotic cells. In well-differentiated HT29 cells, where DCA generated an eight-fold increase in cells in G2 phase after 48 hours, the most significant G2 arrest was visible. Dichloroacetate decreased the amount of lactate present in growth medium and caused the dephosphorylation of the E1 subunit of the pyruvate dehydrogenase complex in all cell lines, however it only affected cancer cells’ intrinsic mitochondrial membrane potential (P=0.04).

Conclusions: Pyruvate dehydrogenase kinase inhibition inhibits glycolysis and promotes mitochondrial oxidative phosphorylation, which prevents non-cancerous cells from proliferating but slows down colorectal cancer cells.

Keywords: pyruvate dehydrogenase, colorectal cancer, dichloroacetate, and pyruvate dehydrogenase kinase

The fourth-leading cause of cancer-related death worldwide and the third-most prevalent cancer worldwide is colorectal cancer (Shike et al, 1990). In the United Kingdom, 17.1 deaths per 100,000 people were caused by colorectal cancer in 2007. (UK Bowel Cancer Statistics, 2009). The prognosis for patients with advanced and metastatic colorectal cancer is still poor, despite recent improvements. The field of cancer therapy that targets tumor metabolism is one that is expanding quickly (Pan and Mak, 2007). Otto Warburg produced the first observations of the metabolic differences between cancer cells and normal cells, demonstrating that cancer cells are inherently dependent on glycolysis for the production of chemical energy (Warburg, 1956). There is growing evidence that this enhanced glycolysis is caused by a number of biological mechanisms, such as mitochondrial malfunction, oncogenic signaling, and adaptive responses to the hypoxic tumor microenvironment (Gatenby and Gillies, 2004; Gillies and Gatenby, 2007; Wu et al, 2007). By preventing apoptosis and promoting tumor metastasis and spread, the glycolytic phenotype benefits cancer cells’ ability to develop (Yeluri et al, 2009).

Pyruvate dehydrogenase is a crucial metabolic regulator in cells (PDH). Pyruvate dehydrogenase changes pyruvate, a byproduct of glycolysis, into acetyl-CoA, which is then oxidized in the mitochondrial tricarboxylic acid cycle. The inhibitory phosphorylation of pyruvate dehydrogenase kinase tightly controls pyruvate dehydrogenase activity. The PDH E1 subunit (PDHE1) is phosphorylated three times at Ser232, Ser293, and Ser300 (Rardin et al, 2009). Recently, it has been demonstrated that the drug dichloroacetate (DCA), which inhibits all four of the PDK(1-4) isoenzymes, inhibits the growth of breast, endometrial, and lung cancer cell lines (Bonnet et al, 2007; Wong et al, 2008; Sun et al, 2009). In order to promote mitochondrial oxidative phosphorylation and induce apoptosis through mitochondrial, NFAT-Kv 1.5, and p53 upregulated modulator of apoptosis (PUMA)-mediated pathways, it has been reported to primarily reduce inhibitory phosphorylation of PDH.

Increased glycolysis has been observed in colorectal cancer cells (Bi et al., 2006), and the tumor microenvironment has been found to be hypoxic and acidotic, primarily because of the underdeveloped blood supply(Dewhirst et al, 1989; Milosevic et al, 2004). Expression of key markers of hypoxia is increased in colorectal cancer, particularly at the invasive margin, as we have previously demonstrated (Thorn et al., 2009). This is particularly true for the more aggressive phenotype(Rajaganeshan et al, 2008, 2009). In order to evaluate PDK inhibition as a novel treatment option for colorectal cancer, the aim of this study was to look into how DCA affects the growth of colorectal cancer cells.

Top of page

Materials and methods

Cell cultures

All cell lines were purchased from American Type Culture Collection (Manassas, VA, USA) or European Collection of Cell Cultures (Salisbury, Wiltshire, UK): HB2 (breast epithelial cells of non-cancer origin), 293 (epithelial cells from human embryo kidney), HT29 (well-differentiated primary colorectal adenocarcinoma), SW480 (poorly differentiated primary colorectal adenocarcinoma), and LoVo (metastatic left supraclavicular lymph node from colorectal adenocarcinoma). 293 and HB2 cells were maintained in DMEM medium, HT29 and SW480 in RPMI 1640 medium, and LoVo in F12 medium (all from Invitrogen, Carlsbad, CA, USA), supplemented with 10% fetal calf serum, in a 37°C, 5% CO2 humidified incubator. For experiments under hypoxic conditions, we incubated cells in a humidified, hypoxic incubator (1% O2, 5% CO2, 94% N2, 37°C). Sodium dichloroacetate (Specials Lab, Prudhoe, UK) was donated by the Pharmacy Department at St. James’s University Hospital, Leeds, UK.

MTT assays

Cells (1 × 104) per well were seeded in 96-well tissue culture plates. After overnight incubation, we replaced media with fresh media containing increasing doses of DCA (0, 10, 15, 20, 30, 50, and 100 mM). After 24 and 48 h of incubation, we performed MTT assay by replacing the media with 50 μl of 1 mg ml−1 MTT solution and the plates were incubated in the dark for 3 h. MTT solution was then removed and the dark blue formazan precipitates were dissolved in 100 μl of propan-1-ol. Optical density was measured using microplate reader (Opsys MR; Dynex Technologies Ltd, Worthing, West Sussex, UK) at 570 nm.

Annexin V and 7-AAD assays

Cells were seeded in 25 cm2 tissue culture flasks and incubated overnight in standard conditions. Media was replaced with fresh media containing a range of doses of DCA (0, 10, 20, and 50 mM). Flow cytometric analysis was performed after 24 and 48 h of incubation. Cells were washed twice with cold PBS and resuspended in 1 × binding buffer (BD Bioscience, Franklin Lakes, NJ, USA) at 5 × 106 cells per ml. 100 μl of solution (5 × 105 cells) was transferred to 5 ml culture tubes. These cells were stained with 5 μl annexin V-FITC and 10 μl 7-AAD (BD Bioscience), gently vortexed, and incubated at ambient temperature for 15 min in dark. Following this 400 μl 1 × binding buffer was added to each tube and analysed within an hour on LSR II flow cytometer (BD Bioscience).

Propidium iodide assays

Cells were propagated as mentioned for the apoptosis assay. Dichloroacetate (50 mM) was used and compared to vehicle control. After harvesting, we resuspended cells in 350 μl of PBS at a concentration of 0.5–1.0 × 106 cells per ml. 100 μl of 0.25 mg ml−1 propidium iodide (PI)/5% Triton (Sigma, St Louis, MO, USA) was added to the cell suspension. 50 μl of 1 mg ml−1 ribonuclease A (Sigma) was then added. Sample tubes were thoroughly vortexed and incubated for 10 min in the dark at room temperature. Flow cytometry was performed on LSR II flow cytometer (BD Bioscience) and data were analysed using FlowJo software (FlowJo, Ashland, OR, USA).

Lactate measurements

Lactate measurements in growth media were performed by the chemical pathology department at the General Infirmary, Leeds Teaching Hospitals NHS Trust. Cells were incubated in 25 cm2 flasks overnight in normoxia. Media was replaced next day with a range of doses of DCA (0, 10, 20, and 50 mM). After 48 h of incubation, we collected 2 ml of media in fluoride tubes and transferred immediately to the chemical pathology laboratory. The tubes were maintained on ice during the transfer. Lactate levels were measured using an automated analyser (Advia 1200 Chemistry system; Siemens Healthcare Diagnostics, Camberley, Surrey, UK).

TMRM assays

Cells were treated with DCA as described for the apoptosis assay. After 24 and 48 h of incubation, we washed cells in PBS, and suspended 1 × 106 cells per ml in Hank’s buffered salt solution with 50 nM tetramethylrhodamine methyl ester (TMRM) (Invitrogen). 100 μl of the cell suspension (1 × 105 cells per well) was transferred to opaque 96-well plates, incubated for 30 min, and fluorescence was measured at 530/620 nm at 37°C using a plate reader (Mithras LB 40; Berthold Technologies, Bad, Wildbad, Germany).

Western blotting

Cells were treated with DCA as described above. After 8 h of treatment, we extracted proteins from cells in Laemmli buffer (2% SDS, 10% glycerol, 0.7% 2-mercaptoethanol, 0.05% bromophenol blue, and 0.5 M Tris-HCl). Lysates were resolved by electrophoresis on NuPAGE Novex 12% Bis-Tris gels (Invitrogen) in MOPS-SDS running buffer (Invitrogen). Proteins were transferred to a polyvinylidene fluoride membrane (GE Healthcare, Chalford St Giles, Bucks, UK). The membrane was blocked for 1 h at ambient temperature in 5% skimmed milk in TBS-T (Tris-buffered saline with 0.1% Tween). The membrane was then probed with primary antibodies in 1% skimmed milk in TBS-T for 90 min, washed in TBS-T, and then probed with the appropriate horseradish peroxidase (HRP)-conjugated secondary antibody for 60 min. Primary antibodies rabbit polyclonal phosphodetect anti-PDH-E1α (pSer293), 1 : 500 (AP1062; EMD Chemicals, Darmstadt, Germany), and mouse monoclonal anti-PDHE1α, 1 : 500 (459400; Invitrogen). Secondary antibodies anti-rabbit or anti-mouse HRP conjugates, 1 : 1000 (Dako, Glostrup, Denmark). Proteins were visualised with Supersignal West Pico or Femto chemiluminescent substrate (Pierce Biotechnology, Rockford, IL, USA) and the Chemidoc XRS system (Bio-Rad, Hercules, CA, USA). β-Actin was used as a loading control.

Statistical analyses

Flow cytometry data were acquired using specific software, BD FACSDiva 6.0 and FlowJo software. Statistical analyses were performed using SPSS for Windows (SPSS version 15.0, Chicago, IL, USA). Differences between DCA-treated and vehicle control groups were assessed using the Mann–Whitney U-test and the 95% confidence intervals of the difference in means between the two groups. A P-value of less than 0.05 was considered to be statistically significant. Data are represented as mean from at least three independent experiments and error bars represent standard deviation of mean.

Top of page


DCA inhibits the proliferation of cancer cells, and the impact is the same in both normoxia and hypoxia.

First, we wanted to know whether DCA therapy decreased cellular proliferation and whether malignant and non-cancerous cells would react differently in normoxic and hypoxic settings. In terms of hypoxia, our theory was that DCA’s effects would be especially strong when there is not enough oxygen present to enable further oxidative phosphorylation. A variety of dosages of DCA were administered to all cell lines (HB2, 293, HT29, SW480, and LoVo) for 24-48 h in normoxic and hypoxic settings. Using MTT assays, relative cell counts were evaluated.


DCA treatment inhibited cellular proliferation in a dose-dependent manner, with increasing dosages (Figure 1A-D). In contrast to what we expected, both hypoxia and normoxia had identical profiles of impaired cell proliferation. Up to 20 mM DCA had no impact on the HB2 and 293 non-cancerous cell culture development at 24 and 48 hours. However, all three colorectal cancer cell line cultures’ growth was greatly slowed by 20 mM DCA (P0.009). Compared to the well-differentiated HT29 cells, the metastatic LoVo cells and the poorly differentiated SW480 cells were more affected by DCA. Compared to cell cultures treated with vehicle control, the growth of LoVo cell cultures treated with 20 mM DCA was inhibited by up to 40%. Further research was restricted to normoxia because there wasn’t much of a difference between the DCA-treated cultures’ growth reduction in hypoxic and normoxic environments.

Figure 1.

Dichloroacetate (20 mM) considerably slowed down the growth of all colorectal cancer cell cultures but did not significantly slow down cultures of 293 and HB2 non-cancerous cells (*P0.009). The relative number of viable cells was determined at 24 hours (A and B) and 48 hours (C and D) using the MTT assay after cells were treated with various doses of DCA or vehicle control under normoxia (A and C) or hypoxia (B and D). Data are presented as a percentage of the control (0 mM dose); an asterisk (*) denotes a substantial departure from the control (0 mM).


Legend and full figure (149K)


DCA induces apoptosis in cancer cells while protecting healthy cells

The next thing we wanted to do was look at whether DCA treatment caused cultures to grow less quickly or if it caused apoptosis to be induced. DCA (0, 10, 20, and 50 mM) was administered to cells for 24 and 48 hours, and the percentage of cells passing through apoptosis was determined by identifying membrane phosphatidylserine with annexin V-FITC. Using flow cytometry, cells were labeled with annexin V-FITC and the essential dye 7-AAD. After 24 and 48 hours of treatment, the cancer cell lines underwent a dose-dependent induction of apoptosis, but the non-cancerous cells underwent minimal to no apoptosis (Figure 2A and B). The proportion of apoptotic cells increased by 10 times in the metastatic LoVo cells after 48 hours of exposure to 50 mM DCA, as opposed to seven and five times in HT29 and SW480 cells, respectively. With 50 mM DCA, the mean percentage of all apoptotic cells increased by 2.8 (95% CI: 2-3), 3.5 (95% CI: 2-5), and 21 (95% CI: 8-34) in HT29, SW480, and LoVo cells, respectively. Even with 50 mM DCA, only 0.2 (95% CI: 0.2 to 0.6) of the 293 cells underwent modest apoptosis. After treatment with 50 mM DCA, there was a nonsignificant drop in the proportion of apoptotic cells in HB2 cells, 0.9 (95% CI: 2.2 to 0.4).

Figure 2.

With little apoptosis in non-cancerous cells, dichloroacetate increased the percentage of the apoptotic population in cancer cells in a dose-dependent manner. DCA was administered to the cells for 24 or 48 hours, respectively. The cells were then stained with annexin V-FITC and 7-AAD and subjected to flow cytometric analysis. The data points represent the average (plus or minus standard deviation) of three separate experiments using 0 and 50 mM DCA.

Full figure and legend (71K)

In colorectal cancer cells, DCA produces G2 phase arrest, whereas it has no impact on the cell-cycle profile of non-cancerous 293 cells.

We also wanted to look into whether the DCA treatment-induced slowdown in culture growth was linked to the induction of growth arrest. Using flow cytometric analysis of DNA content following PI labeling, cell-cycle patterns were analyzed after cells were treated with 50 mM DCA for 24 or 48 hours. The non-cancerous cells were unaffected by the dichloroacetate treatment, while all cancer cells had modifications in their cell-cycle patterns. Beyond 24 hours of therapy, it was possible to see changes in the cell-cycle profile, and they persisted after 48 hours (Figure 3A and B).

Figure 3.

Dichloroacetate caused G2 phase arrest in colorectal cancer cells but had no impact on 293 or HB2 cells’ cell-cycle characteristics. Cells were exposed to 50 mM DCA or a vehicle control after 24 or 48 hours, respectively, before being stained with PI and subjected to flow cytometric analysis. For statistical significance assessments, we compared the average percentage of cells in each cell cycle phase (G1, S, and G2) in DCA-treated cells to the average percentage of cells in same phases in untreated cells (* indicates a significant difference compared to the control).

Full figure and legend (87K)

In HT29 and SW480 cells, there was an eight-fold increase in G2 phase cells after 48 hours of treatment with 50 mM DCA, and a three-fold rise in LoVo cells. For HT29, 19 (95% CI: 13-24) for SW480, and 14 (95% CI: 10-21) for LoVo cells, the mean percentage of all cancer cells in the G2 phase increased; however, there was no difference for the 293 cells, 1 (95% CI: 4 to 7), and HB2 cells, 0.3 (95% CI: 9 to 9). In all cancer cell lines, there was a similar decline in the number of cells in the G0/G1 phase. It’s interesting to note that while the percentage of cells thought to be in the S phase increased significantly in SW480 and LoVo cells, it decreased only little in HT29 cells (see Discussion section). After being treated with DCA, 293 and HB2 cells’ cell-cycle profiles barely changed.


Extracellular lactate levels in growth medium are decreased by DCA.

We evaluated the lactate levels in growth media to determine whether the adjustments in growth and apoptosis brought about by DCA coincided with decreased glycolysis. The last byproduct of glycolysis is lactic acid. Lactate levels in the growth media would fall if DCA were to induce mitochondrial oxidative phosphorylation because pyruvate would be decarboxylated to acetyl-CoA rather than reduced to lactate. After 48 hours of treatment with various dosages of DCA, lactate levels in growth media of all cell lines were assessed (Figure 4). The tests are based on a colorimetric reaction catalyzed by lactate oxidase, and lactate levels were measured using an auto-analyser, which is frequently used for biochemical assessment of lactate levels. In both malignant and non-cancerous cell lines, DCA treatment decreased extracellular lactate levels in growth media in a dose-dependent way.

Figure 4.

Both malignant and non-cancerous cells responded dose-dependently to dichloroacetate by lowering the levels of lactate in growth media. Extracellular lactate levels in the growth media were assessed using an auto-analyser after cells were treated with a range of DCA dosages for 48 hours. Results are given as a comparison to the control.

Full figure and legend (65K)

In colorectal cancer cells, DCA depolarizes the intrinsic mitochondrial membrane, but not in non-cancerous cells.

We assessed the intrinsic mitochondrial membrane potential (m) to see if the induction of apoptosis in cancer cells following treatment with DCA was linked to the enhancement of mitochondrial oxidative phosphorylation. Increasing mitochondrial respiration would deactivate the cancer cell’s hyperpolarized m and reactivate the electron transport chain. Cells were given dosages of DCA for 24 and 48 hours before being stained with the fluorescent measurement-enabling dye TMRM.

Similar to other tests, the impact of DCA was seen after 24 hours and persisted for 48 hours (Figure 5A and B). Treatment with dichloroacetate dose-dependently decreased the hyperpolarized m in all cancer cells. Contrary to expectations, dichloroacetate enhanced the non-cancerous 293 cells’ m in a dose-dependent way while having no effect on the non-cancerous HB2 cells’ m. In all cancer cells, 50 mM DCA significantly decreased m at 24 hours after treatment; however, in LoVo cells, a significant drop was still seen at 20 mM DCA (Figure 5A, P=0.02). On DCA treatment, there was a tendency for an increase in m in the non-cancerous 293 cells, albeit this was not statistically significant (P=0.08). After 48 hours of treatment, there was a noticeable increase in the 293 cells and a decrease in all cancer cells’ m at 20–50 mM DCA (Figure 5B, P0.04).

Figure 5.

The intrinsic mitochondrial membrane potential (m) was decreased in all cancer cells after dichloroacetate treatment, increased in non-cancerous 293 cells, and remained unchanged in non-cancerous HB2 cells. Cells were given dosages of DCA for 24 or 48 hours (A and B), respectively. Fluorescence was then measured at 530/620 nm at 37°C (* – significant difference from control).

Full figure and legend (112K)

The PDHE1 sub-unit is dephosphorylated as a result of DCA therapy because DCA is assumed to inhibit all four isoenzymes of PDK. This reduces the phosphorylation of the PDHE1 sub-unit, which in turn activates the PDH complex. We utilized western blot analysis on the lysates of DCA-treated and untreated cells to confirm whether the dephosphorylation of PDHE1 was happening with DCA treatment in the cell lines studied. Treatment with 20 mM DCA for 8 hours resulted in a noticeably reduced signal for phosphorylation at the pSer293 location in all cell lines, but no change in the levels of total PDHE1 was found (Figure 6). The other two phosphorylation sites, Ser232 and Ser300, do not yet have commercially accessible phospho-specific antibodies.

Figure 6.

Treatment with dichloroacetate decreased PDHE1 phosphorylation at the pSer293 location but had no impact on total PDHE1 levels across all tested cell lines. Western blot studies were carried out on whole-cell lysates that had been made from both treated and untreated cells that had been exposed to 20 mM DCA for 8 hours.

Full figure and legend (70K)

Top of page


Differential effects of DCA on the development of cancer cells versus normal cells

We have demonstrated that DCA inhibits the proliferation of colorectal cancer cells and non-cancerous cells in in vitro cultures in a dose-dependent manner. With a dose of 20 mM, DCA significantly slowed the growth of cancer cells, but had no effect on non-cancerous cells because cancer cells were more susceptible to it. As we have demonstrated, the elements of this differential effect include the powerful activation of apoptosis and cell-cycle arrest in cancer cells but not in non-cancerous cells.


These findings back with a straightforward theory of differential susceptibility to DCA. Some facts, however, call for more discussion. First, cultures of non-cancerous 293 and HB2 cells grew less quickly when 50 mM DCA was added, but no change in these cells’ cell-cycle profiles or an increase in apoptotic cells was seen. These results might be explained by the fact that this amount of DCA caused these non-cancerous cells to move through the cell cycle more slowly overall, without altering the relative proportions within each stage. Furthermore, according to our findings, DCA caused G2 arrest in colorectal cancer cells. In contrast, prior investigations found that treatment with DCA resulted in G1 arrest or no change to the cell-cycle profile (Cao et al, 2008; Wong et al, 2008). Wong et al (2008) All of the endometrial cancer cell lines that responded to DCA by apoptosizing displayed elevated expression of PUMA, and Wong et al. (2008) came to the conclusion that this p53 activation caused G1 arrest. But in our investigation, colorectal cancer cells treated with DCA arrested in the G2 phase, and we did not detect any activation of p53 in our colorectal cancer cell lines (data not shown). Though DCA alone had no effect on the cell-cycle profile, Cao et al. (2008) discovered that the combination of DCA and radiation effectively stopped prostate cancer cells in the G2 phase. Third, DCA treatment increased the percentage of SW480 and LoVo cells that were thought to be in the S phase. This implies both a proliferative increase and an apoptotic induction. Wong et al. (2008) reported a similar discovery in one of the numerous endometrial cancer cells examined. As previously documented in lymphoma cells, a part of the cells that were observed to be in the “S phase” after DCA treatment of the cancer cell lines may actually be apoptotic cells in the “sub-G2” region (Klucar and Al-Rubeai, 1997).


Cellular metabolic changes after treatment with DCA

Both malignant and non-cancerous cells showed a reduction in lactic acid generation from pyruvate when treated with DCA. Furthermore, DCA treatment resulted in dephosphorylation of PDHE1 and subsequent activation of PDH in all of the cell lines examined. Therefore, the difference between DCA’s effects on malignant and non-cancerous cells may be due to how it affects mitochondrial function. Treatment with DCA decreased all cancer cells’ high m, but not non-cancerous cells’. This shows that DCA induces apoptosis through the proximal mitochondrial route by inhibiting PDK and so activating PDH. Promoting mitochondrial respiration results in depolarization of the intrinsic mitochondrial membrane (Bonnet et al, 2007; Cao et al, 2008; Wong et al, 2008). Compared to the less invasive HT29 and SW480 cells, the highly invasive and metastatic LoVo cells showed the most dramatic induction of apoptosis and alterations in mitochondrial function. Given that highly invasive metastatic malignancies are typically the most resistant to conventional chemotherapy and may also be the most responsive to PDK inhibition, this may have clinical implications for the treatment of metastatic colorectal cancer. This is supported by a recent study that found that 5-fluorouracil-resistant colorectal tumors are more likely to have increased glycolysis, making them more receptive to treatment that targets cancer metabolism (Shin et al, 2009). In this respect, our results differ with those of Wong et al. (2008), who discovered that DCA therapy was more effective against highly invasive endometrial cancer cells.


PDK inhibition as a colorectal cancer treatment

We discovered that cancer cells responded differently to dosages of 20–50 mM DCA than did non–cancerous cells. Therefore, DCA therapeutic dosages could range from 20 to 50 mM. The IC50 of DCA for breast cancer cells, according to a recent study, is also between 20 and 30 mM. (Ko and Allalunis-Turner, 2009). This is in contrast to earlier research that claimed DCA might cause apoptosis and inhibit proliferation in cancer cells at levels as low as 0.5–10 mM. (Bonnet et al, 2007; Wong et al, 2008; Sun et al, 2009). Dichloroacetate has been proven to be quite safe for use in treating lactic acidosis in people (Stacpoole et al, 2003). Up to 100 mg kg1 DCA has the most detrimental effects on the neurological system and liver, resulting in mild sedation or sleepiness, reversible peripheral neuropathy, and a mildly asymptomatic increase of serum transaminases that indicates hepatic damage (Stacpoole et al, 1998). Additionally, recent investigations revealed that DCA successfully slowed tumor growth in both in vitro and in vivo settings at levels that were clinically feasible (Bonnet et al, 2007; Sun et al, 2009). It was proposed that DCA might be quickly added to clinical studies for early-stage cancer (Michelakis et al, 2008). However, it is unclear that the clinically-achievable dose of DCA needed to limit the proliferation of colorectal cancer cells in our investigation could be attained without having severe adverse effects. The amount of DCA needed to produce the same plasma concentrations in vivo would be five to 10 times greater than what was utilized in the lactic acidosis clinical studies. It seems that compared to lung, endometrial, and breast cancer cells, colorectal cancer cells in our study are more resistant to DCA. Intriguingly, Sun et al. (2009) discovered that DCA decreased cancer cell proliferation but did not cause apoptosis or cell death in their investigation on breast cancer cells. These outcomes differed significantly from the effects of DCA on colorectal, endometrial, and lung cancer cells in our investigation (Bonnet et al., 2007, Wong et al., 2008). Therefore, despite the fact that DCA slows the proliferation of numerous cancer cells, the effect and the underlying processes appear to depend on the kind of cancer cell. The varied ways that the PDK isoenzymes are expressed in the cancer cells under investigation may be a plausible explanation for these disparate effects. According to Whitehouse and Randle (1973), dichloroacetate is a non-specific inhibitor of PDK with a distinct Ki for each of the four PDK isoenzymes (Bowker-Kinley et al, 1998). The four PDK isoenzymes are also known to express differently in distinct tissues. Therefore, it is necessary to create PDK isoenzyme-specific inhibitors that should enable metabolic modification unique to different cancer cell types.

Top of page


Shen HM, Ong CN, Cheah PY, Eu KW, Hew CL, Bi X, Lin Q, Foo TW, Joshi S, You T, and (2006) A colorectal cancer proteomic analysis revealed changes in metabolic pathways as the carcinogenesis process. Mol Cell Proteomics 5: 1119–1130 | Article | PubMed | ChemPort | Bonnet S, Archer SL, Allalunis-Turner J, Haromy A, Beaulieu C, Thompson R, Lee CT, Lopaschuk GD, Puttagunta L, Bonnet S, Harry G, Hashimoto K, Porter CJ, Andrade MA, Thebaud B, Michelakis ED (2007) Cancer suppresses a mitochondria-K+ channel axis, and normalizing it encourages apoptosis and slows the spread of the disease. Cancer Cell 11: 37-51 | Article | PubMed | ChemPort | Davis WI, Wu P, Harris RA, and Popov KM Bowker-Kinley (1998) evidence that the mammalian pyruvate dehydrogenase complex is regulated in certain tissues. Cao W, Yacoub S, Shiverick KT, Namiki K, Sakai Y, Porvasnik S, Urbanek C, Rosser CJ. Biochem J 329(Part 1): 191–196 | PubMed | ChemPort (2008) Both wild-type and overexpressed Bcl-2 prostate cancer cells are radiation sensitive in vitro when exposed to dichloroacetate (DCA). Article | PubMed | ChemPort | Dewhirst MW, Tso CY, Oliver R, Gustafson CS, Secomb TW, Gross JF; Prostate 68: 1223–1231 (1989) Comparing the morphology and hemodynamics of the microvasculature in a tumor and normal tissue after healing Gatenby RA, Gillies RJ, Int J Radiat Oncol Biol Phys 17: 91–99 | PubMed | ISI | ChemPort (2004) Why is aerobic glycolysis more prevalent in cancers? Gillies RJ, Gatenby RA, Nat Rev Cancer 4: 891-899 | Article | PubMed | ISI | ChemPort (2007) Why do tumors have high levels of glycolysis? Adaptive landscapes and emergent phenotypes Klucar J, Al-Rubeai M, J Bioenerg Biomembr 39: 251-257 | Article | PubMed | ChemPort (1997) The anticancer drug oracin causes G2 cell cycle arrest and death in Burkitt’s lymphoma cells. Ko L. and Allalunis-Turner J. (2009) Investigation into the mechanism of dichloroacetate (DCA) induced apoptosis in breast cancer. FEBS Lett 400: 127–130 | Article | PubMed | ISI | ChemPort. e14637 in J Clin Oncol 27(Suppl 15).

Dichloroacetate (DCA) as a potential metabolic-targeting therapy for cancer, Michelakis ED, Webster L, and Mackey JR 2008. Milosevic, Fyles, Hedley, and Hill (2004) The human tumor microenvironment: invasive (needle) monitoring of oxygen and interstitial fluid pressure. Br J Cancer 99: 989-994 | Article | PubMed | ChemPort. Article | PubMed | Semin Radiat Oncol 14: 249–258

Metabolic targeting as an anticancer strategy: the start of a new era? Pan JG, Mak TW 2007. e14 in Sci STKE 2007

  1. Rajaganeshan, R. Prasad, P. J. Guillou, G. Poston, N. Scott, and D. G. Jayne (2008) the part hypoxia plays in recurrence after Dukes’ B colorectal cancer resection. Rajaganeshan R, Prasad R, Guillou PJ, Scott N, Poston G, Jayne DG, Int J Colorectal Dis 23: 1049–1055 | Article | PubMed | ChemPort (2009) patterns of hypoxia marker expression at colorectal cancer and liver metastases’ invasive margins. Monitoring phosphorylation of the pyruvate dehydrogenase complex in Eur J Surg Oncol 35(12): 1286–1294 | PubMed | ChemPort | Rardin MJ, Wiley SE, Naviaux RK, Murphy AN, Dixon JE (2009) 157–164 Anal Biochem 389 Shike M, Winawer SJ, Greenwald PH, Bloch A, Hill MJ, and Swaroop SV (1990) Primary prevention of colorectal cancer. | Article | PubMed | ChemPort. The WHO Collaborating Centre for Colorectal Cancer Prevention (2009) Upregulation of glycolytic enzymes in proteins produced by human colon cancer cells with 5-fluorouracil resistance. Bull World Health Organ 68: 377-385 | PubMed | ChemPort | Shin YK, Yoo BC, Hong YS, Chang HJ, Jung KH, Jeong SY, Park JG. The pharmacology of dichloroacetate was covered in the article Electrophoresis 30: 2182-2192 | Article | PubMed | ChemPort | Stacpoole PW (1989). Stacpoole PW, Henderson GN, Yan Z, and James MO (1998) Clinical pharmacology and toxicity of dichloroacetate. Metabolism 38: 1124–1144. Article | PubMed | ChemPort. Stacpoole PW, Nagaraja NV, Hutson AD, Environ Health Perspect 106(Suppl 4): 989-994 | Article | PubMed | ChemPort (2003) effectiveness of the lactate-lowering medication dichloroacetate. Sun RC, Fadia M, Dahlstrom JE, Parish CR, Board PG, Blackburn AC. J Clin Pharmacol 43: 683-691 | PubMed | ChemPort (2009) Dichloroacetate suppresses the development of metastatic breast cancer cells both in vitro and in vivo by reversing the glycolytic phenotype. Thorn CC, Freeman TC, Scott N, Guillou PJ, and Jayne DG (2009) Laser microdissection expression profiling of marginal margins of colorectal tumors reveals indications of enhanced lactate metabolism in the aggressive phenotype. Breast Cancer Res Treat 120(1): 253-260 | Article | PubMed | ChemPort. Bowel Cancer Statistics in the UK (2009) Gut 58: 404-412 | Article | PubMed | ChemPort |

On the origin of cancer cells, Warburg O. (1956). Whitehouse S. and Randle PJ (1973) “Activation of pyruvate dehydrogenase in perfused rat heart by dichloroacetate” Science 123: 309-314 | Article | PubMed | ISI | ChemPort. PubMed, ChemPort, Wong JY, Huggins GS, Debidda M, Munshi NC, and De Vivo I. Biochem J 134: 651-653 (2008) Endometrial cancer cells undergo apoptosis when exposed to dichloroacetate. Gynecol Oncol 109: 394–402 | Article | PubMed | ChemPort | Wu, Neilson, Swift, Moran, Tamagnine, Parslow, Armistead, Lemire, Orrell, Teich, Chomicz, and Ferrick DA (2007) In human tumor cells, increased glycolysis reliance and reduced mitochondrial bioenergetic performance are closely related, according to multiparameter metabolic study. AJPCP 292: C125-C136 | Article | PubMed | ChemPort | Yeluri S, Madhok B, Prasad KR, Quirke P, Jayne DG (2009) Clinical chance to exploit cancer’s sugar addiction. 135: 867-877 | Article | PubMed | ChemPort | J Cancer Res Clin Oncol