Introduction

Keloid scars are found only in humans and presently there is no cure for this condition because the underlying biochemical mechanisms of pathogenesis remain unknown (Haverstock, 2001). Keloids are characterized by the excessive deposition of extracellular collagens by abnormal fibroblasts in response to cutaneous injury. It is known that Type I and III collagens and their respective mRNAs are upregulated in these highly proliferative cells (Naitoh et al., 2001), which suggest an overactive system. As most of the energy in the cell is produced in the mitochondrion, this organelle is implicated in keloid pathophysiology. Interestingly, ATP levels remain high in the keloid scar tissue 10 years after the initial injury (Ueda et al., 2004). In this study, we examined the mechanism by which keloid fibroblasts (KFs) sustain a high ATP level.

Histological examination of keloid demonstrates a normal skin periphery, an inflammatory border, and a stable portion, which is hypo-vascular and hypoxic (Berry et al., 1985). Human hypertrophic burn scars were shown to be sites of relative hypoxia (Sloan et al., 1978) and the oxygen tension in a wound is low days after the injury (Ninikoski and Hunt, 1972; Haroon et al., 2000). The presence of hypoxia could possibly explain the induction of the expression of hypoxia-inducible factor 1 (HIF-1) mRNA and of HIF-1 protein in wound cells hours and days later (Albina et al., 2001). HIF-1 is a heterodimer consisting of two subunits: 1 and 1. The regulation by oxygen is through its 1 subunit, which is rapidly destroyed by proteasomal degradation under normal oxygenation (Beck et al., 1993; Maxwell et al., 1993; Wang and Semenza, 1993). Hypoxia results in decreased ubiquitination and accumulation of HIF-1 (Sutter et al., 2000). Keloid has been shown to express constitutively a higher level of HIF-1 protein compared with the adjacent skin (Zhang et al., 2003; Wu et al., 2004). The similarity in the hypoxic microenvironment in solid tumors and keloids led us to examine the response of KF to hypoxia and hypoxia-mimetic agents, such as desferrioxamine and cobalt chloride.

Results

Titration of digitonin permeabilization of fibroblasts

Both KFs and normal fibroblasts (NFs) were optimally permeabilized with 65 g digitonin per 10⁶ cells as shown by the maximal increase in the fluorescence of the J-aggregates following the entry of glutamate/malate and succinate. The energization of the mitochondrial membrane potential by these respiratory substrates (as previously shown in our publication, Ng et al., 2006) was apparent and the increase in mitochondrial membrane potential was inhibited by rotenone and malonate, respectively (data not shown).

ATP biosynthesis and intracellular ATP

From the five batches of KF and NF, the former were shown to biosynthesize ATP from exogenous ADP (1 mM) more rapidly than NF; their respective specific activities, expressed in nmol ATP per minute per 10⁶ cells were 11.93.7 and 3.20.8, giving a ratio of 3.7. Likewise, intracellular ATP was higher in KF with respective values, expressed in nmol ATP per 10⁶ cells of 8.91.2 and 4.60.9. Both sets of data showed significant difference between NF and KF with P<0.01.

Action of respiratory substrates, inhibitors, uncoupler, or oligomycin on ATP generation

The biosynthesis of ATP in KF was inhibited to a higher degree by 3-bromopyruvate, iodoacetate and oxamate, inhibitors of hexokinase, glyceraldehyde-3-phosphate dehydrogenase, and lactate dehydrogenase (LDH), respectively, compared with NF, which implied that their ATP production relied more on the glycolytic pathway (Table 1). On the other hand, the degree of inhibition by thenoyltrifluoroacetate, an inhibitor of respiratory complex II; carbonyl cyanide p-trifluoro-methoxyphenylhydrazone, an uncoupler; and oligomycin, an inhibitor of FoF₁ ATPase was more pronounced in NF, suggesting that they depend more on oxidative phosphorylation (OXPHOS) for their ATP production. In view of this finding, OXPHOS was studied by measuring the amount of ATP produced from the oxidation of respiratory substrates such as pyruvate/malate, glutamate/malate, and succinate in digitonin-permeabilized cells. Consistent with the data on ATP biosynthesis above, KF again showed inherently higher capacity of about 3.6 times compared with NF measured in the absence of exogenous substrates (Table 2). Addition of respiratory substrates increased significantly the rate of ATP production in both NF and KF to a similar maximum value. The increase represents the amount of ATP derived from their oxidation via OXPHOS. The magnitude of increase was greater in NF compared with KF supporting the earlier conclusion that contribution by OXPHOS was greater in NF when thenoyltrifluoroacetate, carbonyl cyanide p-trifluoro-methoxyphenylhydrazone, or oligomycin was added (Table 1).

Table 1 - Action of inhibitors of glycolysis and OXPHOS on ATP biosynthesis in NF and KF¹.

Full table

Table 2 - Biosynthesis of ATP in digitonin-permeabilized fibroblasts from respiratory substrates¹.

Full table

Glucose consumption and lactate production

The uptake of glucose from the culture medium was not different in KF and NF for the first 2 days of culture when the cells were <50% confluent. For the subsequent 3 days, KF consumed more glucose as reflected by a greater decrease of glucose in the medium (Figure 1); this was accompanied by a higher accumulation of lactate compared with NF. The cell number and protein contents per well were essentially similar in NF compared with KF during the 5 days of culture.

Figure 1.

Glucose uptake and lactate accumulation. Glucose and lactate in the culture medium were measured by the glucose oxidase and LDH assays, respectively. From day 3 to 5 when the cells became confluent, KF consumed significantly more glucose compared with NF, concomitant with a greater production of lactate by KF. ^*P<0.01 for glucose values; ^#P<0.05, ^##P<0.005 for lactate values.

Full figure and legend (14K)

Activity of glycolytic enzymes

The specific activities of hexokinase, glyceraldehyde-3-phosphate dehydrogenase, and LDH, expressed as nmol reduced nicotinamide adenine dinucleotide or reduced nicotinamide adenine dinucleotide phosphate per minute per mg protein were 1.8.0.51, 343.128.3, and 554.312.1, respectively, for five samples of NF. The corresponding values for KF, which were significantly different (with P<0.05) were 130, 63, and 35% higher.

Glucose utilization

When cells metabolize D-[2-³H]glucose, the tritium label is lost as ³H₂O at the phosphoglucose isomerase reaction (Rose and O'Connell, 1961). The rate of glucose utilization measured by the amount of tritiated water formed from D-[2-³H]glucose was linear for the first 3 hours and was significantly higher in the KF compared with NF (Figure 2).

Figure 2.

Rate of glucose utilization in keloid and normal fibroblasts. Tritiated water formed from radiolabeled glucose was separated by Dowex-1-borate column chromatography and measured by liquid scintillation counting. The rate of glucose utilization was significantly higher in KF compared with NF as shown by two representative plots.

Full figure and legend (10K)

Effects of hypoxia, desferrioxamine, and cobalt chloride on ATP biosynthesis in KF

The rate of production of ATP was significantly higher in KF exposed to hypoxia, desferrioxamine, or cobalt chloride compared with values obtained under normoxic condition (Table 3). Under these conditions, the cell viability of the fibroblasts measured by the 3-(4,5,dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide assay was not compromised (data not shown). The increase was significantly reduced by inhibitors of glycolysis, namely 3-bromopyruvate, iodoacetate and oxamate under both normoxic and hypoxic conditions.

Table 3 - Effect of inhibitors of glycolysis on ATP biosynthesis in KFs under normoxic and hypoxic conditions¹.

Full table

Production of reactive oxygen species and hydrogen peroxide

Total reactive oxygen species (ROS) production measured by the 2',7'-dichlorofluorescein-diacetate assay was significantly higher in three batches of NF compared with KF (Figure 3a). Known antioxidants, ascorbate, and Trolox (a vitamin E analog) were able to reduce the ROS production. The generation of H₂O₂ was higher in NF and the antioxidant effect of Trolox was also apparent (Figure 3b).

Figure 3.

Production of ROS and H₂O₂ in NF and KF. The rates of production of (a) ROS and (b) H₂O₂ were significantly higher in NF compared with KF. The antioxidants, Trolox and/or ascorbate, at final concentration of 50 M reduced the production of ROS and H₂O₂. Values are meansSD for n=3; ^**P<0.01 when compared with basal ROS generation and ^#P<0.005 for comparison of basal H₂O₂ production between NF and KF. RFU, relative fluorescence unit.

Full figure and legend (53K)

Discussion

Keloid scars are characterized by excessive collagen deposition during the wound healing process. ATP is required for this uncontrolled production of extracellular matrix. Although cellular ATP is normally synthesized in the mitochondria by OXPHOS, the increased lactate in keloid tissue (Hoopes et al., 1971) suggests that glycolysis may provide the main energy source. To examine the separate contribution of mitochondrial and extra-mitochondrial compartments to the biosynthesis of ATP, it is necessary to examine OXPHOS activity in mitochondria to delineate any cytoplasmic contribution from glycolysis. Isolation of mitochondria from fibroblasts is a technical challenge because of the scarcity of this organelle, which represents only 2.4% of the volume of human skin fibroblasts compared with 28% of the cell volume in hepatocytes (Mourant et al., 1998). On the other hand, to measure ATP synthesis by mitochondria in whole cells, it is necessary to permeabilize the plasma membrane to facilitate the entry of respiratory substrates without compromising the integrity of the mitochondrial membranes. Digitonin permeabilization has previously been monitored by ATP synthesis (Ouhabi et al., 1998; Manfredi et al., 2002) or by oxygen uptake (Kunz et al., 1995; Villani et al., 1998). Our method of measuring the mitochondrial membrane potential with 5,5',6,6'-tetrachloro 1,1, 3,3'tetraethylbenzimidazolcarbocyanine iodide (JC-1) to monitor digitonin titration is rapid and sensitive and the assay also provides an assessment of the activity (or lack of activity) of the mitochondrial respiratory complexes. Using this procedure, complex I and II activities in both NF and KF were observed when their respective substrates were introduced (data not shown).

Under normal basal conditions, KF appeared to preferentially use the alternative pathway of glycolysis instead of OXPHOS for their ATP biosynthesis (see Figure 4). This conclusion was supported by the observation of increased accumulation of lactate in the culture medium. The higher activities of glycolytic enzymes (hexokinase, glyceraldehyde-3-phosphate dehydrogenase, and LDH) contributed to the greater rate of glucose utilization by KF compared with NF. Biochemical data on ATP biosynthesis using specific inhibitors of glycolysis and OXPHOS have alluded to glycolysis as the preferred pathway in KFs, because ATP generation was inhibited to a greater extent by 3-bromopyruvate, iodoacetate, and oxamate. In contrast, inhibition by thenoyltrifluoroacetate (a respiratory inhibitor), carbonyl cyanide p-trifluoro-methoxyphenylhydrazone (an uncoupler), and oligomycin (an inhibitor of F_oF₁ ATP synthase) was more pronounced in NF.

Figure 4.

An overview of glycolysis and OXPHOS showing sites of action of respiratory substrates and inhibitors (in parentheses) used in this study.

Full figure (19K)

Proliferative cells preferentially use aerobic glycolytic rather than oxidative metabolism to generate ATP (Brand and Hermfisse, 1997), a phenomenon proposed as the primary defect in cancer (Warburg, 1956). Clinically, keloids are benign tumors as they continue to grow beyond the periphery of the original wound margins. Although our data showed that KF appeared to rely mainly on glycolysis for their energy supply under basal conditions, these cells are also capable of generating ATP from added respiratory substrates via OXPHOS. It is tempting to propose that this dual capability with a greater reliance on glycolysis confers on KF the capacity to proliferate and survive in spite of (or because of) its hypoxic environment. This advantage was translated into a higher rate of biosynthesis of ATP with its accumulation in the cell. In addition, a switch to glycolysis may enable the KF to minimize their production of ROS, which are generated normally from mitochondrial oxidation. Although mitochondria isolated from human skin fibroblasts have been shown to produce ROS, including superoxide and nitric oxide (Chen et al., 1996; Cobbold, 2001), we found that the rate of ROS production was lower in KF compared with NF.

The aerobic glycolysis described above in KF is a common feature of many cancer cells and this Warburg effect has been linked to the HIF-1, which affects angiogenesis, glucose transporter, and glucose metabolism (Semenza, 1999). The HIF-1 subunit has been reported in healing wounds (Albina et al., 2001) and demonstrated to be expressed at higher levels in keloid tissues compared with the adjacent skins (Zhang et al., 2003; Wu et al., 2004). Our measurement of the HIF-1 protein in KF and NF (data not shown) corroborated these findings. The enhanced ATP biosynthesis in KF was further increased on exposure to hypoxia possibly the result of an upregulation of the HIF-1 protein (Zhang et al., 2006). Desferrioxamine and cobalt chloride, known to mimic intracellular hypoxia (Gleadle et al., 1995) exhibited similar effects. HIF-1 has been shown to activate the transcriptional regulation of genes encoding glycolytic enzymes (Semenza et al., 1994). The co-ordinated upregulation of all glycolytic enzymes on exposure to hypoxia has previously been demonstrated, whereas the activity of non-glycolytic enzymes remained unchanged (Robin et al., 1984). These observations were consistent with the conclusion that the rate of glycolysis was positively controlled by the glycolytic enzymes (Ainscow and Brand, 1999), a relationship that was also apparent in our study of a higher rate of glycolysis with higher activities of hexokinase, glyceraldehyde-3-phosphate dehydrogenase, and LDH in KF.

In conclusion, KF under basal conditions are characterized by a higher rate of ATP biosynthesis, with glycolysis as their primary energy source. This enhanced glycolytic flux provides a mechanistic basis to treat keloid scar formation with glycolytic inhibitors of which 3-bromopyruvate has been noted to be highly efficacious against liver cancers in experimental animal models (Ko et al., 2004; Vali et al., 2007). The KF appear to retain their keloid characteristics during culture by their ability to synthesize collagens in vitro (Diegelmann et al., 1979) and to respond to hypoxia and hypoxia mimetics with further increase in ATP biosynthesis (this study). It is tempting to relate the latter changes to the presence of a functional HIF-signaling pathway. Thus, KF in culture could provide a human cell model to study the effects of drug candidates on their bioenergetics including their response to hypoxia, conceivably via the activation of the HIF pathway. Enhanced glycolysis and attenuated OXPHOS as observed in KF seem to be a salient feature of many cancer cells (Moreno-Sanchez et al., 2007) and the dependency on glycolysis appeared to more pronounced with the progression of tumorigenesis (Ramanathan et al., 2005).

Materials and Methods

Materials – 5,5',6,6'-tetrachloro 1,1, 3,3'tetraethylbenzimidazolcarbocyanine iodide and the Amplex^® Red hydrogen peroxide/peroxidase assay kit were from Molecular Probes Inc. (Eugene, OR). P1,P5-Di(adenosine-5') pentaphosphate pentalithium salt and glucose-6-phosphate dehydrogenase (204 U ml^-1) were obtained from Fluka Chemie (Buchs, Switzerland). The following were purchased from Sigma Aldrich Co. Ltd (St Louis, MO): ADP, ATP, 3-bromopyruvate, 2',7'-dichlorofluorescein diacetate; carbonyl cyanide p-trifluoro-methoxyphenylhydrazone; glucose oxidase (151 kU mg^-1), L-glutamic acid, homovanillic acid, horseradish peroxidase, (298 U mg^-1), iodoacetate, LDH (5.3 kU ml^-1), L-malic acid, NAD⁺, NADP⁺, oligomycin (containing 59, 18, and 14% of oligomycin A, B, and C), oxamate, pyruvate, rotenone, succinate, thenoyltrifluoroacetate, 6-hydroxy-2,5,7,8-tetramethyl chromane-2-carboxylic acid (Trolox), FL-ASC Bioluminescent cell assay kit. D-[2-³H]glucose (16 Ci mmol^-1) was purchased from GE Healthcare UK Limited (Buckinghamshire, UK).

Collection and processing of skin specimens

Excised normal skin and keloid skin specimens were collected from plastic surgical procedures (Table 4) as reported in Mukhopadhyay et al. (2006) under the guidelines approved by the National University Institutional Review Board. Signed informed consent had been obtained before operative excision. For babies, informed consent was obtained from the parents. All clinical investigations were conducted according to the Declaration of Helsinki Principles. A portion of the excised specimen was sent for histological confirmation of keloid identity and other portions were used for primary cell culture to isolate fibroblasts.

Table 4 - Profiles of Normal Skin and Keloid Scar Fibroblasts used in the study.

Full table

Primary fibroblast cell culture

Remnant dermis was minced and incubated in a solution of collagenase type-I (0.5 mg ml^-1) and trypsin (0.2 mg ml^-1) for 6 hours at 37°C. Cells were cultured in DMEM medium supplemented with 10% fetal bovine serum and 100 U ml^-1 of penicillin-streptomycin mixture at 37°C in 5% CO₂.

Permeabilization of fibroblasts with digitonin

Digitonin permeabilization was carried out on skin fibroblasts from passages 3 to 7 by the method described in our publication (Ng et al., 2006). The titration of digitonin was monitored by measuring the mitochondrial membrane potential using 5,5',6,6'-tetrachloro 1,1, 3,3'tetraethylbenzimidazolcarbocyanine iodide. The rate of formation of the J-aggregates was measured upon sequential addition of 5 mM each of glutamate/malate and 5 mM succinate. Specific inhibitors of respiratory complexes I and II, namely rotenone and malonate, were used to validate that the oxidation of these substrates occurred via the mitochondrial electron transport system. This assay also provides a means of assessing the activity (or lack of activity) of these respiratory complexes.

Measurement of ATP production

1. Assay conditions. ATP generation from the oxidation of respiratory substrates such as glutamate/malate and succinate has been measured in isolated mitochondria (Vincent et al., 2004; Zhang et al., 2004) and was extended in this study to digitonin-permeabilized skin fibroblasts. The chemiluminescence generated by the luciferin-luciferase reaction was read in a luminometer (PerkinElmer Life and Analytical Sciences, Turku, Finland; Victor³) as described in our report (Ng et al., 2006).
2. In the presence of inhibitors of glycolysis or OXPHOS. In addition to 500 M P1,P5-Di(adenosine-5') pentaphosphate, which was routinely introduced to inhibit adenylate kinase (Kurebayashi et al., 1980), each of the following chemicals was pre-incubated with fibroblasts for 5 minutes before the mixture was added to the assay incubate described above for ATP biosynthesis: 10 mM each of 3-bromopyruvate, iodoacetate or oxamate, or 250 M thenoyltrifluoroacetate, 10 M carbonyl cyanide p-trifluoro-methoxyphenylhydrazone or 10 g oligomycin per ml.
3. Under normoxic and hypoxic conditions. KF were incubated under normoxic condition or in a hypoxic chamber (Billups-Rothenberg Inc, Del Mar, CA) containing 0.5% oxygen, 94.5% N₂ and 5% CO₂ for 16 hours followed by measurement of ATP biosynthesis. In another set of experiments, KFs were exposed to 130 M desferrioxamine or 70 M cobalt chloride for 24 hours.

Intracellular ATP

The amount of ATP present intracellularly was determined in five different batches of confluent KF and NF, following the assay protocol reported in Jeong et al. (2004).

Glycolysis and glycolytic enzymes

1. Glycolysis – The rate of glucose utilization was determined by the production of ³H₂O from D-[2-³H] glucose as described in McKay et al. (1983). NF and KF were plated in six-well plates at 10⁵ cells per well in glucose-free DMEM supplemented with 10% fetal bovine serum. The cells were used the next day when they were approximately 60–70% confluent as recommended. The reaction was started by adding 0.25 mM unlabeled glucose and 93.75 nM D-[2-³H]glucose (containing 2.25 Ci). At hourly intervals up to 4 hours, 100 l aliquots were removed to assay for ³H₂O.
   The separation of ³H₂O from the radiolabeled glucose was carried out by column chromatography using Dowex-1-borate and eluting with 0.9 ml deionized water (Hammerstedt, 1973). The radioactivity in the eluate was measured by liquid scintillation counting. At the end of the experiment, the cells were lysed with a buffer containing 7 M urea, 2 M thiourea, 4% 3-[3-cholamidopropyl) dimethylammonio]-1-propanesulfonic acid (CHAPS), and 10 mM Tris. The protein content in each well was determined (Bradford, 1976).
2. The activities of hexokinase, glyceraldehyde 3-p dehydrogenase, and LDH were measured in extracts (containing 0.03 mg protein) of NF and KF by the methods described in Krebs (1955), Clark and Lai (1989) and Jenkins and Thompson (1989). The fluorescence of NADH and NADPH was monitored at Ex/Em of 352/464 nm in a luminescence spectrophotometer (Perkin-Elmer LS55) and the activities were expressed as an increase in fluorescence of NADH or NADPH per minute per mg protein.

Determination of glucose and lactate in cell culture medium

A total of 40,000 fibroblasts were plated in each well of a six-well plate containing 3 ml culture medium of DMEM supplemented with 5 mM glucose, 10% fetal bovine serum, and 100 U ml^-1 of penicillin-streptomycin mixture. The concentrations of glucose and lactate in the culture medium were measured as follows:

1. Glucose oxidase assay – 20 l of the culture medium in triplicate wells were added to 2 ml of buffer (adjusted to pH 7.4) containing 145 mM KCl, 30 mM HEPES, 5 mM KH₂PO₄, 3 mM MgCl₂ 0.1 mM EGTA, 0.2 mM homovanillic acid, 1 U ml^-1 horseradish peroxidase, and 20 mU ml^-1 glucose oxidase. Hydrogen peroxide produced from the oxidation of glucose was coupled to the reaction between horseradish peroxidase and homovanillic acid to produce a fluorescent dimer (Barja, 2002), which was measured at Ex/Em 312/429 nm. The concentration of glucose was extrapolated from a standard of 0.1–5 mM glucose oxidized by glucose oxidase under identical conditions.
2. LDH assay – Likewise, 20 l aliquots of the culture medium from triplicate wells were added to 2 ml of buffer containing 0.4 M hydrazine and 0.5 M glycine adjusted to pH 9.0, 0.15 mM NAD⁺ and 26 U LDH (Clark and Lai, 1989).

ROS production

ROS production was measured by the procedure based on the oxidation of 2',7'-dichlorofluorescein (Wang and Joseph, 1999) using 30,000 cells per well of a 96-well plate and grown overnight in 5% CO₂. The action of 50 M each of Trolox and ascorbate on the production of ROS was also examined. The rate of 2',7'-dichlorofluorescein oxidation was measured for 30 minutes at Ex/Em 485/526 nm in a fluorescence plate-reader (Gemini XS from Molecular Devices Corporation, Sunnyvale, CA). Cells were lysed at the end of the experiment with 20% Triton-X and the protein content in each well was measured (Bradford, 1976). The generation of H₂O₂ was determined by the Amplex Red hydrogen peroxide/peroxidase assay following the protocol provided by the supplier (Molecular Probes).

Conflict of Interest

The authors state no conflict of interest.

References

1. Ainscow EK, Brand MD (1999) Top-down control analysis of ATP turnover, glycolysis and oxidative phosphorylation in rat hepatocytes. Eur J Biochem 263:671–685 | Article | PubMed | ISI | ChemPort |
2. Albina JE, Mastrofrancesco B, Vessella JA, Louis CA, Henry WL Jr, Reichner JS (2001) HIF-1 expression in healing wounds: HIF-1alpha induction in primary inflammatory cells by TNF-alpha. Am J Physiol Cell Physiol 281:C1971–C1977 | PubMed | ISI | ChemPort |
3. Barja G (2002) The quantitative measurement of H₂O₂ generation in isolated mitochondria. J Bioenerg Biomembr 34:227–233 | Article | PubMed | ISI | ChemPort |
4. Beck I, Weinmann R, Caro J (1993) Characterization of hypoxia-responsive enhancer in the human erythropoietin gene shows presence of hypoxia-inducible 120-Kd nuclear DNA-binding protein in erythropoietin-producing and nonproducing cells. Blood 82:704–711 | PubMed | ISI | ChemPort |
5. Berry RB, Tan OT, Cooke ED, Gaylarde PM, Bowcock SA, Lamberty BG et al. (1985) Transcutaneous oxygen tension as an index of maturity in hypertrophic scars treated by compression. Br J Plast Surg 38:163–173 | Article | PubMed | ISI | ChemPort |
6. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 | Article | PubMed | ISI | ChemPort |
7. Brand KA, Hermfisse U (1997) Aerobic glycolysis by proliferating cells: a protective strategy against reactive oxygen species. FASEB J 11:388–395 | PubMed | ISI | ChemPort |
8. Chen X, Guo G, Li S (1996) A study of human keloid with electron spin resonance. Zhonghua Zheng Xing Shao Shang Wai Ke Za Zhi 12:9–11 | PubMed | ChemPort |
9. Clark JB, Lai CK (1989) Glycolysis, tricarboxylic acid and related enzymes in brain. In: Neuromethods (Boulton A, Baker G, Butterworth RF eds) vol. 11. New Jersey: Humana Press, 233–281
10. Cobbold CA (2001) The role of nitric acid in the formation of keloid and hypertrophic lesions. Med Hypotheses 57:497–502 | Article | PubMed | ISI | ChemPort |
11. Diegelmann RF, Cohen IK, McCoy BJ (1979) Growth kinetics and collagen synthesis of normal skin, normal scar and keloid fibroblasts in vitro. J Cell Physiol 98:341–346 | Article | PubMed | ISI | ChemPort |
12. Gleadle JM, Ebert BL, Firth JD, Ratcliffe PJ (1995) Regulation of angiogenic growth factor expression by hypoxia, transition metals, and chelating agents. Am J Physiol 268:C1362–C1368 | PubMed | ISI | ChemPort |
13. Hammerstedt RH (1973) The use of Dowex-1-borate to separate ³HOH from 2-³H-glucose. Anal Biochem 56:292–293 | Article | PubMed | ISI | ChemPort |
14. Haroon ZA, Raleigh JA, Greenberg CS, Dewhirst MW (2000) Early wound healing exhibits cytokine surge without evidence of hypoxia. Ann Surg 231:137–147 | Article | PubMed | ISI | ChemPort |
15. Haverstock BD (2001) Hypertrophic scars and keloids. Clin Podiatr Med Surg 18:147–159 | PubMed | ChemPort |
16. Hoopes JE, Su CT, Im MJ (1971) Enzyme activities in hypertrophic scars and keloids. Plast Reconstr Surg 47:132–137 | Article | PubMed | ISI | ChemPort |
17. Jenkins WT, Thompson CC (1989) Two kinetic methods to study the regulation of mammalian hexokinases. Anal Biochem 177:396–401 | Article | PubMed | ISI | ChemPort |
18. Jeong DW, Kim TS, Cho IT, Kim IY (2004) Modification of glycolysis affects cell sensitivity to apoptosis induced by oxidative stress and mediated by mitochondria. Biochem Biophys Res Commun 313:984–991 | Article | PubMed | ISI | ChemPort |
19. Ko YH, Smith BL, Wang Y, Pomper MG, Rini DA, Torbenson MS et al. (2004) Advanced cancers: eradication in all cases using 3-bromopyruvate therapy to deplete ATP. Biochem Biophys Res Commun 324:269–275 | Article | PubMed | ISI | ChemPort |
20. Krebs E (1955) Glyceraldehyde-3-phosphate dehydrogenase from yeast: GAP+DPN+P1,3-Diphosphoglycerate+DPNH+H⁺. Methods Enzymol 1:407–411 | ISI | ChemPort |
21. Kunz D, Luley C, Fritz S, Bohnensack R, Winkler K, Kunz WS et al. (1995) Oxygraphic evaluation of mitochondrial function in digitonin-permeabilized mononuclear cells and cultured skin fibroblasts of patients with chronic progressive external ophthalmoplegia. Biochem Mol Med 54:105–111 | Article | PubMed | ISI | ChemPort |
22. Kurebayashi N, Kodama T, Ogawa Y (1980) P1,P5-Di(adenosine-5')pentaphosphate(Ap5A) as an inhibitor of adenylate kinase in studies of fragmented sarcoplasmic reticulum from bullfrog skeletal muscle. J Biochem (Tokyo) 88:871–876 | PubMed | ChemPort |
23. Manfredi G, Yang L, Gajewski CD, Mattiazzi M (2002) Measurements of ATP in mammalian cells. Methods 26:317–326 | Article | PubMed | ISI | ChemPort |
24. Maxwell PH, Pugh CW, Ratcliffe PJ (1993) Inducible operation of the erythropoietin 3' enhancer in multiple cell lines: evidence for a widespread oxygen-sensing mechanism. Proc Natl Acad Sci USA 90:2423–2427 | Article | PubMed | ChemPort |
25. McKay ND, Robinson B, Brodie R, Rooke-Allen N (1983) Glucose transport and metabolism in cultured human skin fibroblasts. Biochim Biophys Acta 762:198–204 | Article | PubMed | ISI | ChemPort |
26. Moreno-Sanchez R, Rodriguez-Enriquez S, Marin-Hernandez A, Saavedra E (2007) Energy metabolism in tumor cells. FEBS J 274:1393–1418 | Article | PubMed | ISI | ChemPort |
27. Mourant JR, Freyer JP, Hielscher AH, Eick AA, Shen D, Johnson TM (1998) Mechanisms of light scattering from biological cells relevant to noninvasive optical-tissue diagnostics. Applied Optics 37:3586–3593 | ISI |
28. Mukhopadhyay A, Chan SY, Lim I, Phillips DJ, Phan TT (2006) The role of activin system in keloid pathogenesis. Am J Physiol Cell Physiol 292:1331–1338 | Article | ChemPort |
29. Naitoh M, Hosokawa N, Kubota H, Tanaka T, Shirane H, Sawada M et al. (2001) Upregulation of HSP47 and collagen type III in the dermal fibrotic disease, keloid. Biochem Biophys Res Commun 280:1316–1322 | Article | PubMed | ISI | ChemPort |
30. Ng LE, Vincent AS, Halliwell B, Wong KP (2006) Action of diclofenac on kidney mitochondria and cells. Biochem Biophys Res Commun 348:494–500 | Article | PubMed | ISI | ChemPort |
31. Ninikoski J, Hunt TK (1972) Measurement of wound oxygen with implanted Silastic tube. Surgery 71:22–26 | PubMed | ChemPort |
32. Ouhabi R, Boue-Grabot M, Mazat JP (1998) Mitochondrial ATP synthesis in permeabilized cells: assessment of the ATP/O values in situ. Anal Biochem 263:169–175 | Article | PubMed | ISI | ChemPort |
33. Ramanathan A, Wang C, Schreiber S (2005) Perturbational profiling of a cell-line model of tumorigenesis by using metabolic measurements. Proc Natl Acad Sci USA 102:5992–5997 | Article | PubMed | ChemPort |
34. Robin ED, Murphy BJ, Theodore J (1984) Coordinate regulation of glycolysis by hypoxia in mammalian cells. J Cell Physiol 118:287–290 | Article | PubMed | ISI | ChemPort |
35. Rose IA, O'Connell EL (1961) Intramolecular hydrogen transfer in the phosphoglucose isomerase reaction. J Biol Chem 236:3086–3092 | PubMed | ISI | ChemPort |
36. Semenza GL (1999) Regulation of mammalian O₂ homeostasis by hypoxia-inducible factor 1. Annu Rev Cell Dev Biol 15:551–578 | Article | PubMed | ISI | ChemPort |
37. Semenza GL, Roth PH, Fang HM, Wang GL (1994) Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 269:23757–23763 | PubMed | ISI | ChemPort |
38. Sloan DF, Brown RD, Wells CH, Hilton JG (1978) Tissue gases in human hypertrophic burn scars. Plast Reconstr Surg 61:431–436 | Article | PubMed | ISI | ChemPort |
39. Sutter CH, Laughner E, Semenza GL (2000) Hypoxia-inducible factor 1 alpha protein expression is controlled by oxygen-regulated ubiquitination that is disrupted by deletions and missense mutations. Proc Natl Acad Sci USA 97:4748–4753 | Article | PubMed | ChemPort |
40. Ueda K, Yasuda Y, Furuya E, Oba S (2004) Inadequate blood supply persists in keloids. Scand J Plast Reconstr Surg Hand Surg 38:267–271 | Article | PubMed | ISI |
41. Vali M, Liapi E, Kowalski J, Hong K, Khwaja A, Torbenson MS et al. (2007) Intraarterial therapy with a new potent inhibitor of tumor metabolism (3-bromopyruvate): identification of therapeutic dose and method of injection in an animal model of liver cancer. Interv Radiol 18:95–101 | Article |
42. Villani G, Greco M, Papa S, Attardi G (1998) Low reserve of cytochrome c oxidase capacity in vivo in the respiratory chain of a variety of human cell types. J Biol Chem 273:31829–31836 | Article | PubMed | ISI | ChemPort |
43. Vincent AS, Lim BG, Tan J, Whiteman M, Cheung NS, Halliwell B et al. (2004) Sulfite-mediated oxidative stress in kidney cells. Kidney Int 65:393–402 | Article | PubMed | ISI | ChemPort |
44. Wang GL, Semenza GL (1993) General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci USA 90:4304–4308 | Article | PubMed | ChemPort |
45. Wang H, Joseph JA (1999) Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radic Biol Med 27:612–616 | Article | PubMed | ISI | ChemPort |
46. Warburg O (1956) On the origin of cancer cells. Science 123:309–314 | Article | PubMed | ISI | ChemPort |
47. Wu Y, Zhang Q, Ann DK, Akhondzadeh A, Duong HS, Messadi DV et al. (2004) Increased vascular endothelial growth factor may account for elevated level of plasminogen activator inhibitor-1 via activating ERK1/2 in keloid fibroblasts. Am J Physiol Cell Physiol 286:C905–C912 | Article | PubMed | ISI | ChemPort |
48. Zhang Q, Oh CK, Messadi DV, Duong HS, Kelly AP, Soo C et al. (2006) Hypoxia-induced HIF-1 alpha accumulation is augmented in a co-culture of keloid fibroblasts and human mast cells: involvement of ERK1/2 and PI-3K/Akt. Exp Cell Res 312:145–155 | Article | PubMed | ISI | ChemPort |
49. Zhang Q, Wu Y, Ann DK, Messadi DV, Tuan TL, Kelly AP et al. (2003) Mechanisms of hypoxic regulation of plasminogen activator inhibitor-1 gene expression in keloid fibroblasts. J Invest Dermatol 121:1005–1012 | Article | PubMed | ISI | ChemPort |
50. Zhang X, Vincent AS, Halliwell B, Wong KP (2004) A mechanism of sulfite neurotoxicity: direct inhibition of glutamate dehydrogenase. J Biol Chem 279:43035–43045 | Article | PubMed | ISI | ChemPort |

Acknowledgments

This study was supported by research grants from the National Medical Research Council (NMRC 0655/2002; R-183-000-089-213) and the Academic Research Fund (R-183-000-154-112) from the National University of Singapore awarded to KPW and from the Biomedical Research Council, Singapore (03/1/21/19/251; 04/1/21/19/338) awarded to TTP. We thank Ms Foo Ling Yann for her skilful technical support.