Molecular, catalytic and structural properties of glyoxalase pathway enzymes of many species are now known. Current research has focused on the regulation of activity and expression of Glo1 (glyoxalase I) and Glo2 (glyoxalase II) and their role in health and disease. Human GLO1 has MRE (metal-response element), IRE (insulin-response element), E2F4 (early gene 2 factor isoform 4), AP-2α (activating enhancer-binding protein 2α) and ARE (antioxidant response-element) regulatory elements and is a hotspot for copy number variation. The human Glo2 gene, HAGH (hydroxyacylglutathione hydrolase), has a regulatory p53-response element. Glo1 is linked to healthy aging, obesity, diabetes and diabetic complications, chronic renal disease, cardiovascular disease, other disorders and multidrug resistance in cancer chemotherapy. Mathematical modelling of the glyoxalase pathway predicts that pharmacological levels of increased Glo1 activity markedly decrease cellular methylglyoxal and related glycation, and pharmacological Glo1 inhibition markedly increases cellular methylglyoxal and related glycation. Glo1 inducers are in development to sustain healthy aging and for treatment of vascular complications of diabetes and other disorders, and cell-permeant Glo1 inhibitors are in development for treatment of multidrug-resistant tumours, malaria and potentially pathogenic bacteria and fungi.

Introduction

The enzymatic components of the glyoxalase system are Glo1 (glyoxalase I) and Glo2 (glyoxalase II) [1]. Much is known of the molecular properties, structure and catalytic mechanism of glyoxalase enzymes. Ongoing and developing areas of glyoxalase research are on the regulation of glyoxalases and their role in health and disease (Table 1).

Table 1
Characteristics of human Glo1 and Glo2

Data from [1,8,9,28] and references therein. AP-2α, activating enhancer-binding protein 2α; CNV, copy number variation; E2F4, early gene 2 factor isoform 4; IRE, insulin-response element; MRE, metal-response element.

CharacteristicGlo1Glo2
Molecular mass 46 kDa (gel filtration) or 42 kDa (sequence) 29 kDa (cytosolic), 34 kDa (mitochondrial) 
Number of subunits and structure Two. Monomers consists of two structurally equivalent domains with the active site in the dimer interface One. Two domains: β-lactamase-like four-layered β-sandwich; and α-helical domain 
pI 4.8–5.1 8.3 
Prosthetic groups Zn2+ Zn2+, Fe2+ 
Reaction catalysed and kinetics CH3COCH(OH)-SG→CH3CH(OH)CO-SG Km=192 μM, kcat=1.1×105 min−1 CH3CH(OH)CO-SG→GSH+CH3CH(OH)CO2+H+Km=146 μM, kcat=4.4×104 min−1 
Genetics and polymorphism GLO1. Locus: 6p21.2. Polymorphism: A/E111, GLO1/GLO2 (common); frameshift mutation (rare). Hotspot for CNV HAGH or GLX2. Locus: 16p13.3. Polymorphism: HAGH1/HAGH2 (rare) 
Transcriptional regulatory elements MRE, IRE, E2F4, AP-2α, ARE p53-RE 
Effect of mild or severe accumulation of substrate Mild: glycation of nucleotides and proteins with inhibition and increased proteolysis of target proteins. Severe: activation of apoptosis and anoikis Mild: S-D-lactoyl transfer to protein with protein dysfunction, particularly mitochondrial. Severe: inhibition of de novo pyrimidine synthesis and apoptosis; depletion of GSH 
Inducers Sulforaphane, allyl isothiocyanate, trans-resveratrol Unknown 
Therapeutic application of inducers Healthy aging and treatment of early-stage vascular complications of diabetes, renal failure and others  
Inhibitors S-p-bromobenzylglutathione (Ki=160 nM), bis[(S-N-4-chlorophenyl-N-hydroxycarbamoyl)glutathione-hexa(β-alanyl)] suberate diamide (Ki=0.96 nM) S-(p-nitrobenzoxycarbonyl)glutathione (Ki=1.2 μM) 
Therapeutic application of inhibitors/prodrugs Glo1-linked MDR tumours, malaria Plasmodium falciparum, pathogenic bacteria and fungi  
CharacteristicGlo1Glo2
Molecular mass 46 kDa (gel filtration) or 42 kDa (sequence) 29 kDa (cytosolic), 34 kDa (mitochondrial) 
Number of subunits and structure Two. Monomers consists of two structurally equivalent domains with the active site in the dimer interface One. Two domains: β-lactamase-like four-layered β-sandwich; and α-helical domain 
pI 4.8–5.1 8.3 
Prosthetic groups Zn2+ Zn2+, Fe2+ 
Reaction catalysed and kinetics CH3COCH(OH)-SG→CH3CH(OH)CO-SG Km=192 μM, kcat=1.1×105 min−1 CH3CH(OH)CO-SG→GSH+CH3CH(OH)CO2+H+Km=146 μM, kcat=4.4×104 min−1 
Genetics and polymorphism GLO1. Locus: 6p21.2. Polymorphism: A/E111, GLO1/GLO2 (common); frameshift mutation (rare). Hotspot for CNV HAGH or GLX2. Locus: 16p13.3. Polymorphism: HAGH1/HAGH2 (rare) 
Transcriptional regulatory elements MRE, IRE, E2F4, AP-2α, ARE p53-RE 
Effect of mild or severe accumulation of substrate Mild: glycation of nucleotides and proteins with inhibition and increased proteolysis of target proteins. Severe: activation of apoptosis and anoikis Mild: S-D-lactoyl transfer to protein with protein dysfunction, particularly mitochondrial. Severe: inhibition of de novo pyrimidine synthesis and apoptosis; depletion of GSH 
Inducers Sulforaphane, allyl isothiocyanate, trans-resveratrol Unknown 
Therapeutic application of inducers Healthy aging and treatment of early-stage vascular complications of diabetes, renal failure and others  
Inhibitors S-p-bromobenzylglutathione (Ki=160 nM), bis[(S-N-4-chlorophenyl-N-hydroxycarbamoyl)glutathione-hexa(β-alanyl)] suberate diamide (Ki=0.96 nM) S-(p-nitrobenzoxycarbonyl)glutathione (Ki=1.2 μM) 
Therapeutic application of inhibitors/prodrugs Glo1-linked MDR tumours, malaria Plasmodium falciparum, pathogenic bacteria and fungi  

Glyoxalase I

Glo1 catalyses the isomerization of the hemithioacetal formed spontaneously from MG (methylglyoxal) and glutathione (GSH), CH3COCHO+GSH⇋CH3COCH(OH)-SG, to the pathway intermediate S-D-lactoylglutathione: CH3COCH(OH)-SGCH3CH(OH)CO-SG. The dynamics of the pre-equilibrium forming the hemithioacetal are rapid with respect to the Glo1-catalysed step in situ such that Glo1 activity is directly proportional to cellular GSH concentration [2]. Glo1 is present in the cytosol of nearly all organisms. Where GSH is not the major non-protein thiol, the cofactor differs (for example, trypanothione in some protozoans) [3]. Human Glo1 is a dimer, expressed at a diallelic genetic locus GLO1 which encodes expression products differing only in amino acid residue at position 111: Ala111 and Glu111. All alloenzymes have a molecular mass of 46 kDa (gel filtration) or 42 kDa (sequence), pI of 4.8–5.1 and distinctive charge densities/shapes. Each subunit contains one Zn2+ ion. The translation product of 184 amino acids has the N-terminal methionine residue removed in post-translational processing and is acetylated. Cys19 and Cys20 form a disulfide bridge; Cys139 and Cys61 may also form a disulfide bridge or Cys139 forms a mixed disulfide with GSH which inhibits Glo1 activity in vitro. Glo1 may be S-nitrosylated on Cys139. Glo1 is a substrate for Ca2+/calmodulin-dependent protein kinase II and is phosphorylated at Thr107. The structure of human Glo1 has two domains per subunit and the active site situated in the dimer interface. The catalytic mechanism involves base-catalysed shielded-proton transfer from C-1 to C-2 of the hemithioacetal, bound in the active site, to form an enediol intermediate and rapid ketonization to the thioester product (reviewed in [1]).

Human GLO1 is at genetic locus 6p21.2 [4]. GLO1 is a hotspot for functional CNV (copy number variation) in human and mouse genomes, giving rise to a 2–4-fold increase in Glo1 expression [5,6]. Deletion of GLO1 is embryonically lethal in mice and humans [7]. Human GLO1 contains regulatory elements: MRE (metal-response element), IRE (insulin-response element), E2F4 (early gene 2 factor isoform 4) and AP-2α (activating enhancer-binding protein 2α) elements (reviewed in [1]). It also contains a functional ARE (antioxidant-response element) in exon 1, indicating that Glo1 expression is regulated by Nrf2 (nuclear factor erythroid 2-related factor 2) [8]. Nrf2 activators with Glo1 inducer activity are sulforaphane, allyl isothiocyanate and resveratrol, whereas bardoxolone methyl was only a weak inducer, indicating that not all Nrf2 activators are Glo1 inducers [8,9]. Glo1 expression may be decreased by activation of RAGE (receptor for advanced glycation end-products), although the mechanism remains unclear [10,11].

Glyoxalase I in health and disease

Aging

A link of Glo1 to aging was found in nematode Caenorhabditis elegans model of aging where overexpression of Glo1 increased mean lifespan by 29% and maximum lifespan by 32%, and Glo1 silencing decreased mean lifespan by 52% and maximum lifespan by 36% (reviewed in [1]). Overexpression of Glo1 and Glo2 together also increased the lifespan of the fungus Podospora anserina [12].

Obesity

There is a genetic link of Glo1 to obesity: a meta-analysis of 34 mouse cross-breeding experiments linked Glo1 to body weight [13], and, in humans, Glo1 was linked to upper-arm circumference and supra-iliac skinfold thickness [14]. The leptin-deficient ob/ob mouse model of experimental obesity had 3-fold decreased Glo1 protein in the liver. Studies of quantitative trait loci for high carbohydrate intake in inbred strains of mice linked increased Glo1 mRNA and activity of Glo1 to a preference for a high carbohydrate diet (reviewed in [15]).

Diabetes and diabetic complications

Following the discovery of increased MG formation by cells incubated with a high glucose concentration in vitro and increased MG concentration in patients with diabetes [16,17], MG emerged as a likely factor linked to the damaging effects of hyperglycaemia, particularly chronic microvascular complications (nephropathy, retinopathy and neuropathy) (reviewed in [15]). Dysfunction induced by high glucose concentration in endothelial cells, endothelial progenitor cells, renal mesangial cells, tubular epithelial cells, retinal pericytes and bone marrow cells linked to vascular complications of diabetes was corrected by overexpression of Glo1 [1822]. Where Glo1 was knocked down, dysfunction of cells in hyperglycaemia was exacerbated [2325]. Glo1 protein delivered into tissues as a cell-permeant tat (transactivator of transcription)-peptide fusion construct prevented streptozotocin-induced diabetes [26]. These studies suggest that increased Glo1 expression may be protective against development of both diabetes and diabetic complications, and small-molecule Glo1 inducers may provide improved prophylaxis and therapy.

Chronic renal disease

Aging-related decline in renal function and interstitial thickening was prevented in transgenic rats overexpressing Glo1. Glo1 overexpression prevented and Glo1 knockdown enhanced senescence of primary proximal tubular epithelial cells in vitro. Mice deficient for Glo1 developed nephropathy similar to diabetic nephropathy [25]. Decreased Glo1 activity in rare GLO1 frameshift mutation heterozygote human subjects was associated with a decreased glomerular filtration rate [27]. This suggests a link of Glo1 expression to chronic renal disease in the healthy population, where small-molecule Glo1 inducers may provide improved prophylaxis and therapy.

CVD (cardiovascular disease)

Glo1 deficiency was linked to high frequency of recurrent clinical CVD events despite correction of conventional modifiable risk factors. In haemodialysis patients, the GLO1 419A>C polymorphism was associated with increased risk of CVD complications. Further studies showed a high mortality rate in patients with homozygous GLO1 419CC mutation (reviewed in [15]).

Cancer and cancer chemotherapy

Cytotoxic cancer chemotherapy may be achieved by inhibition of Glo1 with a cell-permeant inhibitor. The prototype Glo1 inhibitor was S-p-bromobenzylglutathione which had poor cell permeability and extracellular stability through cleavage by cell-surface γ-glutamyltransferase. Diesterification conferred cell permeability and stability to extracellular degradation; when delivered to the tumour, ester groups were removed by non-specific esterases and the Glo1 inhibitor was revealed. BBGD (S-p-bromobenzylglutathione diester) was the first specific cell-permeant Glo1 inhibitor with potent anti-tumour activity in vitro and in vivo. Inbred strains of laboratory mice had markedly higher plasma esterase activity than humans and so esterase-deficient DBA/2 C57BL/6 mice were identified as appropriate for in vivo evaluation. BBGD was particularly effective in human tumour cancer cell lines and mouse bearing MAC 15A colon, DMS114 lung and DU-145 prostate tumours which had high expression of Glo1. High Glo1 expression conferred MDR (multidrug resistance) to conventional anticancer agents, but sensitivity to Glo1 inhibitors and siRNA silencing of Glo1 (reviewed in [28]).

It is likely that high Glo1 expression and high endogenous flux of MG formation are required for a cytotoxic response to Glo1 inhibitors or silencing. A cause of increased Glo1 expression in human tumours is increased GLO1 copy number. A systematic screen of 750 tumour cell lines and 225 primary tumours identified increased gene copy number at 6p21.2, including GLO1. In a survey of 520 human tumours, increased GLO1 copy number was found at a prevalence of 8%. The highest prevalence was in breast cancer (19%), small-cell lung cancer (16%) and non-small-cell lung cancer (11%) [29]. Glo1 overexpression is also acquired by oncogene-linked malignant transformation [30] and chronic treatment with anti-tumour agents [31]. Glo1 expression in tumours may also be increased by increased Nrf2 signalling. Hyperactivation of Nrf2 has been identified in human tumours and contributes to Nrf2-associated MDR [32,33]. It is likely that there is no threshold effect for Glo1-linked MDR, but rather a continuum of increasing severity of MDR linked positively to increased expression of Glo1. The severity of MDR is also likely to depend on the anti-tumour agent and the contribution of MG accumulation to its cytotoxicity. In HEK (human embryonic kidney)-293T cells transfected to overexpress Glo1 with activity increase to 6-fold in vitro, for example, median growth inhibition concentration GC50 values of anticancer drugs were increased up to 34-fold; the highest GC50 increase was found for mitomycin C and mechlorethamine (S.J. Larkin, N. Rabbani and P.J. Thornalley, unpublished work). Glo1 overexpression linked to MDR suggests that MG-induced cytotoxicity mediates, in part, the anti-tumour activity of current clinical anti-tumour agents. MG-induced cytotoxicity may involve apoptosis, necroptosis and anoikis [28].

Malaria

BBGD also had potent anti-malarial activity against the red blood cell stage of Plasmodium falciparum. This stage of the malarial parasite growth cycle has only anaerobic glycolysis with an associated high flux of MG formation. It is therefore particularly sensitive to pharmacological inhibition of Glo1 [21]. Glo1 inhibitor development is required to improve treatment of refractory tumours, particularly of the breast and lung, with likely spin-offs for anti-microbial therapy for inhibitors optimized for bacterial, protozoan and fungal enzymes.

Arthritis

Glo1 regulation may also be distinctive in arthritis. Glo1 expression decreased in fibroblast-like synoviocytes of patients with osteoarthritis compared with those with rheumatoid arthritis [34].

Glyoxalase II

Glo2 catalyses the conversion of S-D-lactoylglutathione into D-lactate and GSH, thereby reforming the GSH consumed in the Glo1-catalysed step of the glyoxalase pathway. It is a highly efficient catalyst of this process. Human Glo2 contains a Fe(II)Zn(II) centre, where the Fe(II) ion has little effect on catalytic activity. Its structure consists of a metallo-β-lactamase-like and a α-helical domain. The active site contains the Fe(II)Zn(II) centre and a substrate-binding site in the two-domain interface. An OH ion co-ordinated to the Zn(II) ion is a base catalyst.

The gene for Glo1 is conventionally designated by HAGH (hydroxyacylglutathione hydrolase) and recently GLX2. Human HAGH is at locus 16p13.3 and genetic polymorphism is extremely rare. In yeast and higher plants, discrete genes encode the cytosolic and mitochondrial forms of Glo2. In mammals, there is a single gene that gives rise to two distinct mRNA species transcribed from nine and ten exons respectively. The nine-exon-derived transcript encodes both mitochondrial matrix-targeted and cytosolic Glo2, whereas the ten-exon transcript encodes only cytosolic Glo2. The molecular mass of the human cytosolic form is 29 kDa and that of the mitochondrial form is 34 kDa. Human Glo2 is up-regulated at the transcriptional level by p63 and p73, transcription factors of the p53 family involved in development. Glo2 was also decreased in p53-knockout mice. A p53-response element, RRRC(A/T)(A/T)GYYY, was found in intron 1 of HAGH, which was activated when bound by p63 and p73. Only cytosolic Glo2 was increased. Cells deficient in Glo2 were hypersensitive to MG-induced apoptosis and DNA-damage-induced apoptosis (reviewed in [1]).

Role of mitochondrial Glo2

The role of Glo2 in mitochondria is uncertain as there is no mitochondrial targeting of Glo1. Mitochondrial Glo2 may be involved in delivery of GSH into mitochondria via S-D-lactoylglutathione [35], but since GSH and S-D-lactoylglutathione had similar mitochondrial uptake kinetics and the cytosolic concentration of S-D-lactoylglutathione is usually <1% of GSH, this appears unlikely. Glo2 hydrolyses other acyl-GSH derivatives such as S-acetyl-GSH and S-succinyl-GSH [36]. Recent research suggests that there is significant non-enzymatic acetyl and succinyl transfer from acetyl-CoA and succinyl-CoA in mitochondria [37], and a likely acceptor is mitochondrial GSH. Glo2 may therefore maintain GSH by repairing endogenous acylations where high concentrations of acetyl-CoA and succinyl-CoA in mitochondria require targeting of Glo2 to this compartment.

Toxicity of Glo2 deficiency

The obvious explanation for toxicity of Glo2 deficiency is accumulation of S-D-lactoylglutathione and thereby depletion of GSH and in situ activity of Glo1, resulting in MG toxicity. Cellular S-D-lactoylglutathione concentration, however, is normally maintained at low levels. With S-D-lactoylglutathione accumulation, there may be significant lactoyl transfer to cysteine forming S-D-lactoylcysteine which rearranges to N-D-lactoylcysteine. The latter is an inhibitor of dihydro-orotase and de novo pyrimidine synthesis which is toxic and also potentiates toxicity of DNA damage [38].

Other putative glyoxalase enzymes

A protein which converted MG into D-lactate without GSH cofactor and without formation of intermediate S-D-lactoylglutathione, an MG oxidoreductase, was purified from Escherichia coli and called ‘glyoxalase III’ [39]. This was confirmed by independent researchers, determining the specificity constant kcat/Km of 1.1×105 M1·min1 [40], approximately 7000-fold lower than Glo1 of E. coli (7.4×108 M1·min1 [41]). Such low specific activity casts doubt as to whether this protein has significant MG metabolizing activity in vivo. Moreover, MG oxidoreductase activity was measured under conditions where MG is chemically unstable (alkaline pH). DJ-1 isoenzymes were also considered as possible MG oxidoreductases, suggested by sequence analogy with glyoxalase III [42]. Conditions used for activity assay were non-physiological (45°C and 0.96 M NaOH in sample processing). Again, the kcat/Km for human DJ-1 was very low, at approximately 10000-fold lower than that of human Glo1 (0.5–1.5×105 M1·min1 [42] compared with 1.1×109 M1·min1 [43]). It is unlikely that DJ-1 contributes significantly to cellular metabolism of MG in mammals. DJ-1 is secreted into plasma and could possibly provide a basis for extracellular metabolism of MG. However, at DJ-1 concentrations found in health and disease (1–2 nM [44]), DJ-1 would not compete effectively with non-enzymatic glycation of plasma protein as a fate for MG. Similar doubt applies to related putative novel glyoxalase DJ-1 of Arabidopsis thaliana and Candida albicans which similarly have low specific activity (kcat/Km=1.7×104 M1·min1 and 8.5×104 M1·min1 respectively) [45,46]. It is therefore currently doubtful that ‘glyoxalase III’ and the DJ-1 superfamily are indeed glyoxalases and contribute significantly to MG metabolism in vivo. An improved activity assay of MG oxidoreductase and attention to catalytic efficiency is required in future studies.

Metabolism of methylglyoxal by aldoketo reductases

MG and glyoxal is metabolized mainly by Glo1 and the glyoxalase system, with normally minor metabolism by AKRs (aldoketo reductases) and aldehyde dehydrogenases. When the glyoxalase system is impaired, AKR isoenzymes 1A4, 1B1 (aldose reductase) and 1B3 may metabolize MG to mainly hydroxyacetone and AKR isoenzymes 1B1, 1B3 and 1B8 may metabolize glyoxal to glycolaldehyde. Metabolism by AKR 1B1 may be a major fate of MG and glyoxal in the human renal medulla where high expression of AKR 1B1 outcompetes Glo1. AKRs are ARE-linked genes with expression regulated by transcription factor Nrf2 (reviewed in [47]).

Mathematical modelling of the glyoxalase pathway

With the prospect of developing Glo1 inducers for diabetic complications and other metabolic disorders and Glo1 inhibitors for MDR tumours, there is an increasing requirement to model mathematically the glyoxalase pathway and its metabolites. In Figure 1, a one-compartment metabolic model for related enzymatic and non-enzymatic pathways is given and outcomes of 90% inhibition or 3-fold induction of Glo1 simulated. It shows that the cellular concentration of MG and flux of formation of related glycation adducts may be increased 6.4-fold or decreased by 59% respectively. This agrees approximately with the changes found with DNA glycation adduct increase in cells treated with BBDG [48] and MG and protein glycation adduct decrease by induction of Glo1 with sulforaphane [8] respectively. Further modelling refinements and experimental validations are required. The current model indicates that approximately 90% of MG measured in assays is reversibly bound to protein thiols in situ and the concentration of S-D-lactoylglutathione is ≪1% GSH.

Systems biology of glyoxalase pathway: mathematical modelling of disturbance in disease and targeted therapeutics in a one-compartment cell model

Figure 1
Systems biology of glyoxalase pathway: mathematical modelling of disturbance in disease and targeted therapeutics in a one-compartment cell model

(A) Kinetic model. (B) Simulation of the effect of 90% inhibition and 3-fold induction of Glo1. Model characteristics (from experimental estimates of variables of human endothelial HMEC-1 cells in vitro): [GSH], 750 μM; protein-SH, 20 mM; [arginine residues], 40 mM; [lysine residues], 80 mM; flux of MG formation, 0.15 μmol/l per min; Glo1, Km, 100 μM and Vmax, 6500 μmol/l per min; Glo2, Km, 146 μM and Vmax, 14000 μmol/l per min; cellular MG (MGfree+HA+HA-protein ≈ 1.3 μM; HA is hemithioacetal). Modelling in COPASI program [49].

Figure 1
Systems biology of glyoxalase pathway: mathematical modelling of disturbance in disease and targeted therapeutics in a one-compartment cell model

(A) Kinetic model. (B) Simulation of the effect of 90% inhibition and 3-fold induction of Glo1. Model characteristics (from experimental estimates of variables of human endothelial HMEC-1 cells in vitro): [GSH], 750 μM; protein-SH, 20 mM; [arginine residues], 40 mM; [lysine residues], 80 mM; flux of MG formation, 0.15 μmol/l per min; Glo1, Km, 100 μM and Vmax, 6500 μmol/l per min; Glo2, Km, 146 μM and Vmax, 14000 μmol/l per min; cellular MG (MGfree+HA+HA-protein ≈ 1.3 μM; HA is hemithioacetal). Modelling in COPASI program [49].

Glyoxalase Centennial: 100 Years of Glyoxalase Research and Emergence of Dicarbonyl Stress: A Biochemical Society Focused Meeting held at the University of Warwick, U.K., 27–29 November 2013. Organized and Edited by Naila Rabbani and Paul Thornalley (University of Warwick, U.K.).

Abbreviations

     
  • AKR

    aldoketo reductase

  •  
  • ARE

    antioxidant-response element

  •  
  • BBGD

    S-p-bromobenzylglutathione diester

  •  
  • CVD

    cardiovascular disease

  •  
  • Glo

    glyoxalase

  •  
  • HAGH

    hydroxyacylglutathione hydrolase

  •  
  • MDR

    multidrug resistance

  •  
  • MG

    methylglyoxal

  •  
  • Nrf2

    nuclear factor erythroid 2-related factor 2

We thank the sponsors of the Glyoxalase Centennial conference, University of Warwick, 27–29 November 2013 and past and present members of our research team for their contributions to glyoxalase research.

Funding

We thank the Biotechnology and Biological Sciences Research Council, Medical Research Council, Wellcome Trust, British Heart Foundation, Cancer Research UK, Juvenile Diabetes Research Fund and British Council for support for our glyoxalase-related research.

References

References
1
Xue
 
M.
Rabbani
 
N.
Thornalley
 
P.J.
 
Glyoxalase in ageing
Semin. Cell Dev. Biol.
2011
, vol. 
22
 (pg. 
293
-
301
)
2
Vander Jagt
 
D.L.
Daub
 
F.
Krohn
 
J.A.
Han
 
L.P. B.
 
Effects of pH and thiols on the kinetics of yeast glyoxalase I: an evaluation of the random pathway mechanism
Biochemistry
1975
, vol. 
14
 (pg. 
3669
-
3675
)
3
Greig
 
N.
Wyllie
 
S.
Vickers
 
T.J.
Fairlamb
 
A.H.
 
Trypanothione-dependent glyoxalase I in Trypanosoma cruzi
Biochem. J.
2006
, vol. 
400
 (pg. 
217
-
223
)
4
Thornalley
 
P.J.
 
Population genetics of human glyoxalases
Heredity
1991
, vol. 
67
 (pg. 
139
-
142
)
5
Cahan
 
P.
Li
 
Y.
Izumi
 
M.
Graubert
 
T.A.
 
The impact of copy number variation on local gene expression in mouse hematopoietic stem and progenitor cells
Nat. Genet.
2009
, vol. 
41
 (pg. 
430
-
437
)
6
Redon
 
R.
Ishikawa
 
S.
Fitch
 
K.R.
Feuk
 
L.
Perry
 
G.H.
Andrews
 
T.D.
Fiegler
 
H.
Shapero
 
M.H.
Carson
 
A.R.
Chen
 
W.
, et al 
Global variation in copy number in the human genome
Nature
2006
, vol. 
444
 (pg. 
444
-
454
)
7
Arai
 
M.
Yuzawa
 
H.
Nohara
 
I.
Ohnishi
 
T.
Obata
 
N.
Iwayama
 
Y.
Haga
 
S.
Toyota
 
T.
Ujike
 
H.
Arai
 
M.
, et al 
Enhanced carbonyl stress in a subpopulation of schizophrenia
Arch. Gen. Psychiatry
2010
, vol. 
67
 (pg. 
589
-
597
)
8
Xue
 
M.
Rabbani
 
N.
Momiji
 
H.
Imbasi
 
P.
Anwar
 
M.M.
Kitteringham
 
N.R.
Park
 
B.K.
Souma
 
T.
Moriguchi
 
T.
Yamamoto
 
M.
Thornalley
 
P.J.
 
Transcriptional control of glyoxalase 1 by Nrf2 provides a stress responsive defence against dicarbonyl glycation
Biochem. J.
2012
, vol. 
443
 (pg. 
213
-
222
)
9
Cheng
 
A.S.
Cheng
 
Y.H.
Chiou
 
C.H.
Chang
 
T.L.
 
Resveratrol upregulates Nrf2 expression to attenuate methylglyoxal-induced insulin resistance in Hep G2 cells
J. Agric. Food Chem.
2012
, vol. 
60
 (pg. 
9180
-
9187
)
10
Zeng
 
S.
Zhang
 
Q.Y.
Huang
 
J.Z.
Vedantham
 
S.
Rosario
 
R.
Ananthakrishnan
 
R.
Yan
 
S.F.
Ramasamy
 
R.
DeMatteo
 
R.P.
Emond
 
J.C.
, et al 
Opposing roles of RAGE and Myd88 signaling in extensive liver resection
FASEB J.
2012
, vol. 
26
 (pg. 
882
-
893
)
11
Reiniger
 
N.
Lau
 
K.
McCalla
 
D.
Eby
 
B.
Cheng
 
B.
Lu
 
Y.
Qu
 
W.
Quadri
 
N.
Ananthakrishnan
 
R.
Furmansky
 
M.
, et al 
Deletion of the receptor for advanced glycation end products reduces glomerulosclerosis and preserves renal function in the diabetic OVE26 mouse
Diabetes
2010
, vol. 
59
 (pg. 
2043
-
2054
)
12
Scheckhuber
 
C.Q.
Mack
 
S.J.
Strobel
 
I.
Ricciardi
 
F.
Gispert
 
S.
Osiewacz
 
H.D.
 
Modulation of the glyoxalase system in the aging model Podospora anserina: effects on growth and lifespan
Aging
2010
, vol. 
2
 (pg. 
969
-
980
)
13
Wuschke
 
S.
Dahm
 
S.
Schmidt
 
C.
Joost
 
H.G.
Al Hasani
 
H.
 
A meta-analysis of quantitative trait loci associated with body weight and adiposity in mice
Int. J. Obes.
2006
, vol. 
31
 (pg. 
829
-
841
)
14
Wilson
 
A.F.
Elston
 
R.C.
Tran
 
L.D.
Siervogel
 
R.M.
 
Use of the robust sib-pair method to screen for single-locus, multiple-locus, and pleiotropic effects: application to traits related to hypertension
Am. J. Hum. Genet.
1991
, vol. 
48
 (pg. 
862
-
872
)
15
Rabbani
 
N.
Thornalley
 
P.J.
 
Glyoxalase in diabetes, obesity and related disorders
Semin. Cell Dev. Biol.
2011
, vol. 
22
 (pg. 
309
-
317
)
16
Thornalley
 
P.J.
 
Modification of the glyoxalase system in human red blood cells by glucose in vitro
Biochem. J.
1988
, vol. 
254
 (pg. 
751
-
755
)
17
McLellan
 
A.C.
Thornalley
 
P.J.
Benn
 
J.
Sonksen
 
P.H.
 
The glyoxalase system in clinical diabetes mellitus and correlation with diabetic complications
Clin. Sci.
1994
, vol. 
87
 (pg. 
21
-
29
)
18
Shinohara
 
M.
Thornalley
 
P.J.
Giardino
 
I.
Beisswenger
 
P.J.
Thorpe
 
S.R.
Onorato
 
J.
Brownlee
 
M.
 
Overexpression of glyoxalase I in bovine endothelial cells inhibits intracellular advanced glycation endproduct formation and prevents hyperglycaemia-induced increases in macromolecular endocytosis
J. Clin. Invest.
1998
, vol. 
101
 (pg. 
1142
-
1147
)
19
Ceradini
 
D.J.
Yao
 
D.
Grogan
 
R.H.
Callaghan
 
M.J.
Edelstein
 
D.
Brownlee
 
M.
Gurtner
 
G.C.
 
Decreasing intracellular superoxide corrects defective ischemia-induced new vessel formation in diabetic mice
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
10930
-
10938
)
20
Kim
 
K.M.
Kim
 
Y.S.
Jung
 
D.H.
Lee
 
J.
Kim
 
J.S.
 
Increased glyoxalase I levels inhibit accumulation of oxidative stress and an advanced glycation end product in mouse mesangial cells cultured in high glucose
Exp. Cell Res.
2012
, vol. 
318
 (pg. 
152
-
159
)
21
Miller
 
A.G.
Smith
 
D.G.
Bhat
 
M.
Nagaraj
 
R.H.
 
Glyoxalase I is critical for human retinal capillary pericyte survival under hyperglycemic conditions
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
11864
-
11871
)
22
Vulesevic
 
B.
McNeill
 
B.
Geoffrion
 
M.
Kuraitis
 
D.
McBane
 
J.E.
Lochhead
 
M.
Vanderhyden
 
B.C.
Korbutt
 
G.S.
Milne
 
R.W.
Suuronen
 
E.J.
 
Glyoxalase-1 overexpression in bone marrow cells reverses defective neovascularization in STZ-induced diabetic mice
Cardiovasc. Res.
2013
, vol. 
101
 (pg. 
306
-
316
)
23
Engelbrecht
 
B.
Stratmann
 
B.
Hess
 
C.
Tschoepe
 
D.
Gawlowski
 
T.
 
Impact of GLO1 knock down on GLUT4 trafficking and glucose uptake in L6 myoblasts
PLoS ONE
2013
, vol. 
8
 pg. 
e65195
 
24
Queisser
 
M.A.
Yao
 
D.
Geisler
 
S.
Hammes
 
H.P.
Lochnit
 
G.
Schleicher
 
E.D.
Brownlee
 
M.
Preissner
 
K.T.
 
Hyperglycemia impairs proteasome function by methylglyoxal
Diabetes
2010
, vol. 
59
 (pg. 
670
-
678
)
25
Giacco
 
F.
Du
 
X.
D’Agati
 
V.D.
Milne
 
R.
Sui
 
G.
Geoffrion
 
M.
Brownlee
 
M.
 
Knockdown of Glo1 mimics diabetic nephropathy in non-diabetic mice
Diabetes
2014
, vol. 
63
 (pg. 
291
-
299
)
26
Kim
 
M.J.
Kim
 
D.W.
Lee
 
B.R.
Shin
 
M.J.
Kim
 
Y.N.
Eom
 
S.A.
Park
 
B.J.
Cho
 
Y.S.
Han
 
K.H.
Park
 
J.
, et al 
Transduced Tat-glyoxalase protein attenuates streptozotocin-induced diabetes in a mouse model
Biochem. Biophys. Res. Commun.
2013
, vol. 
430
 (pg. 
294
-
300
)
27
Ikeda
 
Y.
Inagi
 
R.
Miyata
 
T.
Nagai
 
R.
Arai
 
M.
Miyashita
 
M.
Itokawa
 
M.
Fujita
 
T.
Nangaku
 
M.
 
Glyoxalase I retards renal senescence
Am. J. Pathol.
2011
, vol. 
179
 (pg. 
2810
-
2821
)
28
Thornalley
 
P.J.
Rabbani
 
N.
 
Glyoxalase in tumourigenesis and multidrug resistance
Semin. Cell Dev. Biol.
2011
, vol. 
22
 (pg. 
318
-
325
)
29
Santarius
 
T.
Bignell
 
G.R.
Greenan
 
C.D.
Widaa
 
S.
Chen
 
L.
Mahoney
 
C.L.
Butler
 
A.
Edkins
 
S.
Waris
 
S.
Thornalley
 
P.J.
, et al 
GLO1: a novel amplified gene in human cancer
Genes Chromosomes Cancer
2010
, vol. 
49
 (pg. 
711
-
725
)
30
Young
 
T.W.
Mei
 
F.C.
Yang
 
G.
Thompson-Lanza
 
J.A.
Liu
 
J.
Cheng
 
X.
 
Activation of antioxidant pathways in Ras-mediated oncogenic transformation of human surface ovarian epithelial cells revealed by functional proteomics and mass spectrometry
Cancer Res.
2004
, vol. 
64
 (pg. 
4577
-
4584
)
31
Yang
 
Y.X.
Chen
 
Z.C.
Zhang
 
G.Y.
Yi
 
H.
Xiao
 
Z.Q.
 
A subcelluar proteomic investigation into vincristine-resistant gastric cancer cell line
J. Cell. Biochem.
2008
, vol. 
104
 (pg. 
1010
-
1021
)
32
Hayes
 
J.D.
McMahon
 
M.
 
NRF2 and KEAP1 mutations: permanent activation of an adaptive response in cancer
Trends Biochem. Sci.
2009
, vol. 
34
 (pg. 
176
-
188
)
33
Wang
 
X.J.
Sun
 
Z.
Villeneuve
 
N.F.
Zhang
 
S.
Zhao
 
F.
Li
 
Y.J.
Chen
 
W.M.
Yi
 
X.F.
Zheng
 
W.X.
, et al 
Nrf2 enhances resistance of cancer cells to chemotherapeutic drugs, the dark side of Nrf2
Carcinogenesis
2008
, vol. 
29
 (pg. 
1235
-
1243
)
34
Devauchelle
 
V.
Marion
 
S.
Cagnard
 
N.
Mistou
 
S.
Falgarone
 
G.
Breban
 
M.
Letourneur
 
F.
Pitaval
 
A.
Alibert
 
O.
Lucchesi
 
C.
, et al 
DNA microarray allows molecular profiling of rheumatoid arthritis and identification of pathophysiological targets
Genes Immun.
2004
, vol. 
5
 (pg. 
597
-
608
)
35
Armeni
 
T.
Cianfruglia
 
L.
Piva
 
F.
Urbanelli
 
L.
Caniglia
 
M.L.
Pugnaloni
 
A.
Principato
 
G.
 
S-D-Lactoylglutathione can be an alternative supply of mitochondrial glutathione
Free Radic. Biol. Med.
2014
, vol. 
67
 (pg. 
451
-
459
)
36
Talesa
 
V.
Uotila
 
L.
Koivusalo
 
M.
Principato
 
G.
Giovann
 
E.
 
Isolation of glyoxalase II from two different compartments of rat liver mitochondria: kinetic and immunochemical characterization of the enzymes
Biochim. Biophys. Acta
1989
, vol. 
993
 (pg. 
7
-
11
)
37
Wagner
 
G.R.
Payne
 
R.M.
 
Widespread and enzyme-independent Nε-acetylation and Nε-succinylation of proteins in the chemical conditions of the mitochondrial matrix
J. Biolog. Chem.
2013
, vol. 
288
 (pg. 
29036
-
29045
)
38
Edwards
 
L.G.
Adesida
 
A.
Thornalley
 
P.J.
 
Inhibition of human leukaemia 60 cell growth by S-D-lactoylglutathione in vitro: mediation by metabolism to N-D-lactoylcysteine and induction of apoptosis
Leuk. Res.
1996
, vol. 
20
 (pg. 
17
-
26
)
39
Misra
 
K.
Banerjee
 
A.B.
Ray
 
S.
Ray
 
M.
 
Glyoxalase III from Escherichia coli: a single novel enzyme for the conversion of methylglyoxal into D-lactate without reduced glutathione
Biochem. J.
1995
, vol. 
305
 (pg. 
999
-
1003
)
40
Subedi
 
K.P.
Choi
 
D.
Kim
 
I.
Min
 
B.
Park
 
C.
 
Hsp31 of Escherichia coli K-12 is glyoxalase III
Mol. Microbiol.
2011
, vol. 
81
 (pg. 
926
-
936
)
41
Clugston
 
S.L.
Barnard
 
J.F.J.
Kinach
 
R.
Miedema
 
D.
Ruman
 
R.
Daub
 
E.
Honek
 
J.F.
 
Overproduction and characterization of a dimeric non-zinc glyoxalase I from Escherichia coli: evidence for optimal activation by nickel ions
Biochemistry
1998
, vol. 
37
 (pg. 
8754
-
8763
)
42
Lee
 
J.-y.
Song
 
J.
Kwon
 
K.
Jang
 
S.
Kim
 
C.
Baek
 
K.
Kim
 
J.
Park
 
C.
 
Human DJ-1 and its homologs are novel glyoxalases
Hum. Mol. Genet.
2012
, vol. 
21
 (pg. 
3215
-
3225
)
43
Allen
 
R.E.
Lo
 
T.W. C.
Thornalley
 
P.J.
 
A simplified method for the purification of human red blood cell glyoxalase I: characteristics, immunoblotting and inhibitor studies
J. Protein Chem.
1993
, vol. 
12
 (pg. 
111
-
119
)
44
Maita
 
C.
Tsuji
 
S.
Yabe
 
I.
Hamada
 
S.
Ogata
 
A.
Maita
 
H.
Iguchi-Ariga
 
S.M.M.
Sasaki
 
H.
Ariga
 
H.
 
Secretion of DJ-1 into the serum of patients with Parkinson's disease
Neurosci. Lett.
2008
, vol. 
431
 (pg. 
86
-
89
)
45
Kwon
 
K.
Choi
 
D.
Hyun
 
J.K.
Jung
 
H.S.
Baek
 
K.
Park
 
C.
 
Novel glyoxalases from Arabidopsis thaliana
FEBS J.
2013
, vol. 
280
 (pg. 
3328
-
3339
)
46
Hasim
 
S.
Hussin
 
N.A.
Alomar
 
F.
Bidasee
 
K.R.
Nickerson
 
K.W.
Wilson
 
M.A.
 
A glutathione-independent glyoxalase of the DJ-1 superfamily plays an important role in managing metabolically generated methylglyoxal in Candida albicans
J. Biol. Chem
2013
, vol. 
289
 (pg. 
1662
-
1674
)
47
Rabbani
 
N.
Thornalley
 
P.J.
 
Niwa
 
T.
 
Dicarbonyls (glyoxal, methylglyoxal, and 3-deoxyglucosone)
Uremic Toxins
2012
Chichester
John Wiley & Sons
(pg. 
177
-
192
)
48
Thornalley
 
P.J.
Waris
 
S.
Fleming
 
T.
Santarius
 
T.
Larkin
 
S.J.
Winklhofer-Roob
 
B.M.
Stratton
 
M.R.
Rabbani
 
N.
 
Imidazopurinones are markers of physiological genomic damage linked to DNA instability and glyoxalase 1-associated tumour multidrug resistance
Nucleic Acids Res.
2010
, vol. 
138
 (pg. 
5432
-
5442
)
49
Hoops
 
S.
Sahle
 
S.
Gauges
 
R.
Lee
 
C.
Pahle
 
J.
Simus
 
N.
Singhal
 
M.
Xu
 
L.
Mendes
 
P.
Kummer
 
U.
 
COPASI: COmplex PAthway SImulator
Bioinformatics
2006
, vol. 
22
 (pg. 
3067
-
3074
)