- Email: [email protected]
Reduced glyoxalase 1 activity in carotid artery plaques of nondiabetic patients with increased hemoglobin A1c level
Reduced glyoxalase 1 activity in carotid artery plaques of nondiabetic patients with increased hemoglobin A1c level Andreas S. Peters, MD,a,b Marieluise Lercher,a Thomas H. Fleming, PhD,c,d Peter P. Nawroth, MD,c,d,e Moritz S. Bischoff, MD,a Susanne Dihlmann, PhD,a Dittmar Böckler, MD,a and Maani Hakimi, MD,a,b Heidelberg and Munich, Germany Objective: Glyoxalase 1 (GLO1) is ubiquitously expressed in the cytosol of the cell and is the major opponent against the reactive metabolite methylglyoxal, which is involved in the development of atherosclerosis. Nondiabetic individuals with an increased hemoglobin A1c (HbA1c) level are at higher risk for development of cardiovascular diseases. As such, this study investigated whether there was an association between reduced GLO1 activity in atherosclerotic lesions of nondiabetic patients with an increased HbA1c level. Methods: HbA1c level was determined in venous blood of patients with carotid artery disease. Protein level of GLO1 was measured in endarterectomy-derived carotid artery plaques by Western blotting. Activity was measured by spectrophotometric assay in the plaques as well as in the erythrocytes; GLO1 activity in erythrocytes was compared with that in a cohort of healthy individuals (n [ 15; 33% men; average age, 60 years). Results: There were 36 patients with carotid artery disease (69% men; average age, 69 years) included in this study and divided into two equal groups: group I, HbA1c < 5.7% (<39 mmol/mol); and group II, 5.7% # HbA1c < 6.5% (39 mmol/mol # HbA1c < 48 mmol/mol). GLO1 activity in carotid plaques was reduced by 29% in group II compared with group I (P [ .048), whereas protein expression was unchanged (P [ .25). Analysis of GLO1 activity in erythrocytes revealed no difference between the groups (P [ .36) or in comparison to healthy controls (P [ .15). Examination of clinical parameters showed an increased amount of patients with concomitant peripheral arterial disease in group II (44% vs 10%; P [ .020). Conclusions: Reduction of GLO1 activity in atherosclerotic lesions of nondiabetic patients with increased HbA1c is associated with a functional involvement of this protective enzyme in atherogenesis. Systemic GLO1 activity seems to be independent of both HbA1c and localized atherosclerosis as it was unchanged between group I and group II as well as compared with healthy controls, respectively. (J Vasc Surg 2016;-:1-5.)
The pathogenesis of atherosclerosis still remains unclear. According to the Horus study, lifestyle-associated factors are only one component of this multifactorial disease.1 Among age- and gender-dependent mechanisms, inflammation has been shown to play a major role in atherogenesis.2 Advanced glycation end products (AGEs) From the Department of Vascular and Endovascular Surgery,a Vaskuläre Biomaterialbank Heidelberg (Vascular Biobank Heidelberg),b and Department of Medicine I and Clinical Chemistry,c University of Heidelberg, Heidelberg; and the German Center for Diabetes Research (DZD),d and the Joint Heidelberg-IDC Translational Diabetes Program, Helmholtz-Zentrum,e Munich. This study was supported by grants from the Doktor Robert PflegerStiftung (to A.S.P.), Deutsche Forschungsgemeinschaft (SFB1118; to P.P.N.), Dietmar Hopp Foundation (to P.P.N.), and German Center for Diabetes Research (to P.P.N.). Author conflict of interest: none. Correspondence: Andreas S. Peters, MD, Department of Vascular and Endovascular Surgery, University of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany (e-mail: andreas1.peters@med. uni-heidelberg.de). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2016 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2016.04.025
have been associated with the initiation and progression of this inflammation response, irrespective of the status of diabetes.3 Moreover, in an inflammatory condition, AGEs have been shown to be a contributing factor to the development and progression of late diabetes-related complications.4 AGEs are produced by glycation of macromolecules, in particular proteins, through reactive metabolites, of which methylglyoxal (MG) is the most potent. MG is mainly detoxified by the glyoxalase (GLO) system, which consists of two enzymes, GLO1 and GLO2, and a catalytic amount of reduced glutathione. GLO1 is ubiquitously expressed in the cytosol of all mammalian cells. In 1960, it was reported that GLO1 activity is significantly reduced in atherosclerotic vessel walls compared with normal full arterial wall sections of the same blood vessels and that activity decreased with age.5 Since then, studies investigating the role of GLO1 in cardiovascular disease have been rare. Recently, we showed that patients with carotid artery disease (CarAD) had reduced GLO1 activity with respect to the overall atherosclerotic burden and that GLO1 reduction was more predominant in men.6 Interestingly, large prospective studies have shown an association of increased risk for cardiovascular morbidity and all-cause mortality in nondiabetic adults with an elevated hemoglobin A1c (HbA1c) level.7,8 Moreover, a positive association between HbA1c level and common 1
JOURNAL OF VASCULAR SURGERY --- 2016
2 Peters et al
Table. Comparison of clinical parameters between the groups of carotid artery disease (CarAD) patients divided by hemoglobin A1c (HbA1c) level
Total No. 40 (36) No. Age, years, mean 6 SD Gender Male Female HbA1c, %, mean 6 SD PAD Yes No CHD Yes No AH Yes No No. of AH drugs, median Statin Yes No Nicotine abuse Yes No BMI, mean 6 SD
Group I HbA1c < 5.7% 20 67.8 6 9.4 11 9 5.3 6 0.35
Group II 5.7% # HbA1c < 6.5% 16 70.8 6 8.3 12 4 5.9 6 0.14
P value .31a .21
2 18
7 9
.020
4 16
8 8
.058
17 3 2
15 1 3
.41
13 7
13 3
.28
9 11 25.8 6 4.0
7 9 27.7 6 6.0
.10b
.94 .28a
AH, Arterial hypertension; BMI, body mass index; CHD, coronary heart disease; DM, diabetes mellitus; PAD, peripheral arterial disease; SD, standard deviation. All statistical analysis was carried out using c2 test except as noted. A P value < .5 was considered to be significant. a Student t-test. b Mann-Whitney U test.
carotid artery intima-media thickness in nondiabetic participants of a large-scale population-based cohort study was recently reported.9 Formation of HbA1c occurs by reducing sugars like glucose, reacting with the lysine and valine residues in hemoglobin. In addition, reactive metabolites like MG glycate hemoglobin to form an MG-H1 adduct.10,11 However, the physiologic relevance of this type of modification on hemoglobin remains unclear. In this study, the association between HbA1c level and GLO1 activity in nondiabetic CarAD patients was investigated with respect to both atherosclerotic plaques and red blood cells (RBCs; used as a surrogate to display the systemic level). Moreover, GLO1 activity in erythrocytes of healthy individuals served as a control. METHODS Study population. Between May and October 2014, 36 patients with CarAD and an indication for open repair by North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria and an HbA1c level of <6.5% (48 mmol/mol) were included in the study consecutively
to prevent selection bias. Patients with HbA1c level $6.5% ($48 mmol/mol), diabetes, or restenosis were excluded; in the case of bilateral disease, patients were included only once. Tissue and blood collection and further processing as well as examination of clinical data were conducted according to ethical guidelines and approved by the Medical Ethics Committee of the University of Heidelberg (S-495/2012 and S-301/2013). All patients gave their written informed consent. Patient characteristics. Patient characteristics were recorded prospectively, including baseline medication (antiplatelet therapy, statin, antihypertensive drugs), and are described in the Table. It was not distinguished between asymptomatic and symptomatic CarAD. Peripheral arterial disease (PAD) was considered relevant if either anklebrachial index was <0.9 or medical history revealed prior vascular open surgery or intervention indicating at least a prior Fontaine stage IIb disease. Nicotine abuse was indicated positive if patients were actively smoking. Existence of coronary heart disease was abstracted from the patient’s medical records. Characteristics of healthy controls. Controls were chosen to be healthy, especially in terms of vascular disease, ie, no history of vascular disease (CarAD, PAD, myocardial infraction, stroke), no previous intervention, no claudication, no arterial hypertension (no treatment with antihypertensive drugs), and no diabetes. Moreover, minimum age was defined to be 50 years (the same as for the CarAD patients). Analysis of blood samples. Venous blood was drawn from CarAD patients on the day of hospitalization, and HbA1c level was analyzed by standard procedures in the central hematology laboratory of the University Hospital Heidelberg. Erythrocytes were isolated from blood samples within 4 hours of donation using Ficoll-Paque PLUS (GE Healthcare, Freiburg, Germany) according to the manufacturer’s instructions and lysed in ice-cold buffer (100 mM sodium phosphate buffer, 0.1% Triton X-100, pH 7.4) supplemented with protease inhibitor (Serva, Heidelberg, Germany). Hemoglobin concentration was determined using a Drabkin assay (Sigma-Aldrich, St. Louis, Mo) with human lyophilized hemoglobin (Sigma-Aldrich) as a standard.12,13 Tissue processing. Carotid artery plaques were resected, collected, and processed within 30 minutes of removal according to the standard operating procedures of the Vascular Biobank Heidelberg. After washing with 0.9% sodium chloride solution to remove remaining blood, samples were flash frozen in liquid nitrogen and stored at 80 C until further use. Protein isolation. Frozen carotid artery plaques were processed as described before in liquid nitrogen to extract proteins.6 Protein concentration was determined by the Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, Ill) using bovine serum albumin as a standard. The resulting lysates were retained on 80 C until further analysis. Western blotting. Protein extracts (20 mg) were separated under denaturing conditions on precast 4-20%
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Peters et al 3
Mini-PROTEAN TGX gels (Bio-Rad, Munich, Germany), transferred to a nitrocellulose membrane, and blocked with 5% nonfat milk in TRIS-buffered saline. Membranes were incubated with either polyclonal rabbit anti-GLO1 (ab96032; Abcam, Cambridge, UK) or rabbit antiglyceraldehyde 3-phosphate dehydrogenase monoclonal antibody (D16H11; Cell Signaling Technology, Danvers, Mass) at 1:1000 dilution. Membranes were subsequently incubated with the appropriate horseradish peroxidaseconjugated immunoglobulin G secondary antibody (1:2000; Cell Signaling Technologies) and visualized on x-ray films using enhanced chemoluminescence detection reagents according to the manufacturer’s instructions. Intensity of bands corresponding to the GLO1 expression was quantified by densitometry using ImageJ (http://rsb. info.nih.gov/ij) and normalized to the intensity of glyceraldehyde 3-phosphate dehydrogenase signals. Assay of GLO1 activity. GLO1 activity was determined, as previously described, using a spectrophotometric method that monitors the initial rate of change in absorbance at 240 nm caused by the formation of S-Dlactoylglutathione.6 Statistical analysis. Statistical analysis was carried out using Student t-test, analysis of variance, Mann-Whitney U test, or c2 test. Results are presented as mean 6 standard deviation or as median. A P value < .05 was considered to be significant. RESULTS To test for an association between GLO1 and HbA1c levels, a Spearman test was used. However, no correlation could be observed (R ¼ 0.088; P ¼ .44). This could be explained by the low sample size of the study cohort. Another reason could be that the relationship between GLO1 and HbA1c is rather displayed by a threshold. Therefore, dividing the patients into two groups seems reasonable. According to the HbA1c value, patients were assigned into two different groups: group I, HbA1c < 5.7% (<39 mmol/mol; n ¼ 20); and group II, 5.7% # HbA1c < 6.5% (39 mmol/mol # HbA1c < 48 mmol/mol; n ¼ 16). The applied thresholds for the HbA1c of 6.5% and 5.7% were chosen for the following reasons. An HbA1c value $6.5% indicates the diagnosis of diabetes by definition; an HbA1c value between 5.7% and 6.5% indicates patients being at risk for development of or even having diabetes. Mean age was 67.8 and 70.8 years, respectively, whereas mean age for all CarAD patients together amounted to 69 years. Mean age of the healthy control group was 60 years. Distribution of demographic and clinical data of the CarAD patients can be seen in the Table. All CarAD patients from both groups were treated with acetylsalicylic acid. Use of a statin was reported in 65% of patients in group I compared with 81% in group II. The median number of antihypertensive drugs was two and three in group I and group II, respectively (Table). Of the 36 CarAD patients, 25 were asymptomatic and 11 were symptomatic (group I, 15 vs 5; group II, 10 vs 6).
Fig 1. A, Glyoxalase 1 (GLO1) activity in carotid artery lesions is reduced in patients with increased hemoglobin A1c (HbA1c) level (n ¼ 20 vs 16; single-tailed Student t-test, P ¼ .048). B and C, On protein level, GLO1 was unchanged between the groups (n ¼ 20 vs 16; P ¼ .25), exemplified by a Western blot of three relevant patients for each group (group I: Biobank No. 14-69, 14-119, and 14-120; group II: Biobank No. 14-91, 14-115, and 14-116). Signals for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as controls.
GLO1 activity and expression in carotid artery plaques. GLO1 was found to be regulated differently on activity and protein level. GLO1 activity in carotid artery plaques was significantly reduced by 29% in group II compared with group I (0.61 6 0.11 vs 0.86 6 0.54 U/mg; P ¼ .048; Fig 1, A). In contrast, GLO1 protein levels did not differ between the groups (P ¼ .25; Fig 1, B and C). GLO1 activity in erythrocytes of CarAD patients and healthy controls. To determine the systemic activity of GLO1 and to reveal a possible compensatory effect,
JOURNAL OF VASCULAR SURGERY --- 2016
4 Peters et al
compared with group I; however, this difference was not statistically significant (P ¼ .10; Table). DISCUSSION
Fig 2. A and B, Analysis of red blood cell (RBC) glyoxalase 1 (GLO1) activity did not reveal a difference between the groups divided by hemoglobin A1c (HbA1c) level (n ¼ 20 vs 16; P ¼ .36) or between an apparently healthy control group (n ¼ 15) and the patients with carotid artery disease (CarAD ; n ¼ 36; P ¼ .15).
the activity of GLO1 in RBCs was analyzed from the two groups as well as from a cohort of healthy individuals (n ¼ 15) serving as a control group. No significant differences were found between groups I and II (0.012 6 0.004 vs 0.013 6 0.004 U/mg; P ¼ .358; Fig 2, A). Furthermore, no difference in GLO1 activity of RBCs could be observed between the healthy controls and the group of all CarAD patients together, irrespective of HbA1c level (0.011 6 0.003 vs 0.012 6 0.004; P ¼ .15, Fig 2, B). Even if the healthy control group was compared to groups I and II, there was no difference in GLO1 activity (P ¼ .28). Analysis of clinical parameters of the CarAD patients. The distribution of clinical parameters between group I and group II revealed an increased number of patients suffering from concomitant PAD in group II (44% vs 10%; P ¼ .020; Table). All patients from both groups were found to have arterial hypertension as a risk factor. Patients in group II were treated with more antihypertensive drugs
In this study, GLO1 activity was found to be significantly reduced in atherosclerotic carotid artery lesions of nondiabetic patients with increased HbA1c. These data are consistent with large prospective studies, which have shown that there is an increased risk for cardiovascular morbidity and all-cause mortality in nondiabetic patients with an elevated HbA1c.7,8 This association with reduced GLO1 activity suggests that GLO1 can act as a protective enzyme in atherogenesis and could be a marker for an increased risk of cardiovascular disease. The reduction in GLO1 activity was not due to a reduction in expression of the GLO1 protein. This suggests that the decrease in activity is due to post-translational modification of the protein, particularly as the activity was measured under saturating concentrations of the hemiacetal substrate. Various post-translational modifications of GLO1 have been described, with only a few having an effect on activity. In one study, four sites of post-translational modification were identified where glutathionylation on Cys139 strongly inhibited GLO1 activity.14 It has also been described that nitric oxide affects GLO1 activity in a reduced glutathione-dependent manner.15 GLO1 was recently found to be a substrate of transglutaminase 2, resulting in either deamination or polyamine conjugation (if present), leading to activation of GLO1 alone or additional stabilization against denaturation, respectively.16 Further studies are required to determine whether there are differences in the kinetic parameters of GLO1 in the atherosclerotic carotid artery lesions from CarAD patients. Furthermore, proteomic studies will help establish the type and extent of post-translational modifications on GLO1. No differences in GLO1 activity were found in the RBCs between the groups studied as well as compared with the control group, suggesting that the effects observed in the atherosclerotic carotid artery lesions are specific to the microenvironment of the damaged tissue, consistent with previous studies.5,6 Analysis of the risk factors between the groups showed that there was a significantly higher number of PAD patients with elevated HbA1c. Previously, we have shown that GLO1 activity is decreased in carotid artery plaques of patients with an increased systemic burden of atherosclerosis in terms of concomitant PAD.6 In this study, arterial hypertension was distributed evenly between the two groups, although patients in group II were found to be treated, nonsignificantly, with more antihypertensive drugs. Numerous studies have investigated the influence of MG on arterial hypertension. However, the effect of GLO1 on blood pressure remains unknown. In an in vivo model, it has been shown that subchronic oxidative stress, in terms of increased reactive oxygen species, resulted in a reduction of GLO1 protein in multiple regions of the brain, leading to a nuclear factor kB-mediated elevation of proinflammatory factors and angiotensin 1
JOURNAL OF VASCULAR SURGERY Volume -, Number -
receptor levels, indicating a role for GLO1 in antihypertensive control.17 Additional studies are required to determine whether a similar proinflammatory signaling pathway is activated in the atherosclerotic carotid artery lesions. CONCLUSIONS GLO1 activity is reduced in atherosclerotic lesions of nondiabetic patients with an increased HbA1c level, indicating a functional role of this detoxifying enzyme in the development of atherosclerosis. The fact that systemic GLO1 activity is regulated irrespective of the HbA1c level emphasizes the importance of the change in the local microenvironment influencing activity of GLO1 and hence induction or progress of atherosclerosis. Further studies, especially focusing on post-translational modifications, are needed to reveal the underlying mechanisms. GLO1 and the control of reactive metabolites and AGEs could be a potential therapeutic target, particularly in regard to nondiabetic patients with an elevated HbA1c level. Moreover, nondiabetic CarAD patients with elevated HbA1c more often suffer from additional PAD and are treated with additional antihypertensive drugs. We thank Anja Spieler for excellent technical assistance and Dr Markus Wortmann for his support. AUTHOR CONTRIBUTIONS Conception and design: AP, TF, PN, SD, DB, MH Analysis and interpretation: AP, ML, SD Data collection: AP, ML, TF Writing the article: AP, MB Critical revision of the article: AP, TF, PN, MB, SD, MH Final approval of the article: AP, ML, TF, PN, MB, SD, DB, MH Statistical analysis: AP Obtained funding: AP, PN Overall responsibility: MH REFERENCES 1. Thompson RC, Allam AH, Lombardi GP, Wann LS, Sutherland ML, Sutherland JD, et al. Atherosclerosis across 4000 years of human history: the Horus study of four ancient populations. Lancet 2013;381: 1211-22.
Peters et al 5
2. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med 2005;352:1685-95. 3. Peppa M, Uribarri J, Vlassara H. The role of advanced glycation end products in the development of atherosclerosis. Curr Diab Rep 2004;4: 31-6. 4. Fleming T, Nawroth PP. Reactive metabolites as a cause of late diabetic complications. Biochem Soc Trans 2014;42:439-42. 5. Kirk JE. The glyoxalase I activity of arterial tissue in individuals of various ages. J Gerontol 1960;15:139-41. 6. Peters AS, Hakimi M, Vittas S, Fleming TH, Nawroth PP, Bockler D, et al. Gender difference in glyoxalase 1 activity of atherosclerotic carotid artery lesions. J Vasc Surg 2015;62:471-6. 7. Selvin E, Steffes MW, Zhu H, Matsushita K, Wagenknecht L, Pankow J, et al. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med 2010;362:800-11. 8. Skriver MV, Borch-Johnsen K, Lauritzen T, Sandbaek A. HbA1c as predictor of all-cause mortality in individuals at high risk of diabetes with normal glucose tolerance, identified by screening: a follow-up study of the Anglo-Danish-Dutch Study of Intensive Treatment in People with Screen-Detected Diabetes in Primary Care (ADDITION), Denmark. Diabetologia 2010;53:2328-33. 9. Haring R, Baumeister SE, Lieb W, von Sarnowski B, Volzke H, Felix SB, et al. Glycated hemoglobin as a marker of subclinical atherosclerosis and cardiac remodeling among non-diabetic adults from the general population. Diabetes Res Clin Pract 2014;105:416-23. 10. Rahbar S. The discovery of glycated hemoglobin: a major event in the study of nonenzymatic chemistry in biological systems. Ann N Y Acad Sci 2005;1043:9-19. 11. Gao Y, Wang Y. Site-selective modifications of arginine residues in human hemoglobin induced by methylglyoxal. Biochemistry 2006;45: 15654-60. 12. Fleming T, Cuny J, Nawroth G, Djuric Z, Humpert PM, Zeier M, et al. Is diabetes an acquired disorder of reactive glucose metabolites and their intermediates? Diabetologia 2012;55:1151-5. 13. Kender Z, Fleming T, Kopf S, Torzsa P, Grolmusz V, Herzig S, et al. Effect of metformin on methylglyoxal metabolism in patients with type 2 diabetes. Exp Clin Endocrinol Diabetes 2014;122:316-9. 14. Birkenmeier G, Stegemann C, Hoffmann R, Gunther R, Huse K, Birkemeyer C. Posttranslational modification of human glyoxalase 1 indicates redox-dependent regulation. PLoS One 2010;5:e10399. 15. Mitsumoto A, Kim KR, Oshima G, Kunimoto M, Okawa K, Iwamatsu A, et al. Nitric oxide inactivates glyoxalase I in cooperation with glutathione. J Biochem 2000;128:647-54. 16. Lee DY, Chang GD. Methylglyoxal in cells elicits a negative feedback loop entailing transglutaminase 2 and glyoxalase 1. Redox Biol 2014;2: 196-205. 17. Wortmann M, Peters AS, Hakimi M, Bockler D, Dihlmann S, Glyoxalase I. (Glo1) and its metabolites in vascular disease. Biochem Soc Trans 2014;42:528-33.
Submitted Feb 4, 2016; accepted Apr 18, 2016.
The pathogenesis of atherosclerosis still remains unclear. According to the Horus study, lifestyle-associated factors are only one component of this multifactorial disease.1 Among age- and gender-dependent mechanisms, inflammation has been shown to play a major role in atherogenesis.2 Advanced glycation end products (AGEs) From the Department of Vascular and Endovascular Surgery,a Vaskuläre Biomaterialbank Heidelberg (Vascular Biobank Heidelberg),b and Department of Medicine I and Clinical Chemistry,c University of Heidelberg, Heidelberg; and the German Center for Diabetes Research (DZD),d and the Joint Heidelberg-IDC Translational Diabetes Program, Helmholtz-Zentrum,e Munich. This study was supported by grants from the Doktor Robert PflegerStiftung (to A.S.P.), Deutsche Forschungsgemeinschaft (SFB1118; to P.P.N.), Dietmar Hopp Foundation (to P.P.N.), and German Center for Diabetes Research (to P.P.N.). Author conflict of interest: none. Correspondence: Andreas S. Peters, MD, Department of Vascular and Endovascular Surgery, University of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany (e-mail: andreas1.peters@med. uni-heidelberg.de). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2016 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2016.04.025
have been associated with the initiation and progression of this inflammation response, irrespective of the status of diabetes.3 Moreover, in an inflammatory condition, AGEs have been shown to be a contributing factor to the development and progression of late diabetes-related complications.4 AGEs are produced by glycation of macromolecules, in particular proteins, through reactive metabolites, of which methylglyoxal (MG) is the most potent. MG is mainly detoxified by the glyoxalase (GLO) system, which consists of two enzymes, GLO1 and GLO2, and a catalytic amount of reduced glutathione. GLO1 is ubiquitously expressed in the cytosol of all mammalian cells. In 1960, it was reported that GLO1 activity is significantly reduced in atherosclerotic vessel walls compared with normal full arterial wall sections of the same blood vessels and that activity decreased with age.5 Since then, studies investigating the role of GLO1 in cardiovascular disease have been rare. Recently, we showed that patients with carotid artery disease (CarAD) had reduced GLO1 activity with respect to the overall atherosclerotic burden and that GLO1 reduction was more predominant in men.6 Interestingly, large prospective studies have shown an association of increased risk for cardiovascular morbidity and all-cause mortality in nondiabetic adults with an elevated hemoglobin A1c (HbA1c) level.7,8 Moreover, a positive association between HbA1c level and common 1
JOURNAL OF VASCULAR SURGERY --- 2016
2 Peters et al
Table. Comparison of clinical parameters between the groups of carotid artery disease (CarAD) patients divided by hemoglobin A1c (HbA1c) level
Total No. 40 (36) No. Age, years, mean 6 SD Gender Male Female HbA1c, %, mean 6 SD PAD Yes No CHD Yes No AH Yes No No. of AH drugs, median Statin Yes No Nicotine abuse Yes No BMI, mean 6 SD
Group I HbA1c < 5.7% 20 67.8 6 9.4 11 9 5.3 6 0.35
Group II 5.7% # HbA1c < 6.5% 16 70.8 6 8.3 12 4 5.9 6 0.14
P value .31a .21
2 18
7 9
.020
4 16
8 8
.058
17 3 2
15 1 3
.41
13 7
13 3
.28
9 11 25.8 6 4.0
7 9 27.7 6 6.0
.10b
.94 .28a
AH, Arterial hypertension; BMI, body mass index; CHD, coronary heart disease; DM, diabetes mellitus; PAD, peripheral arterial disease; SD, standard deviation. All statistical analysis was carried out using c2 test except as noted. A P value < .5 was considered to be significant. a Student t-test. b Mann-Whitney U test.
carotid artery intima-media thickness in nondiabetic participants of a large-scale population-based cohort study was recently reported.9 Formation of HbA1c occurs by reducing sugars like glucose, reacting with the lysine and valine residues in hemoglobin. In addition, reactive metabolites like MG glycate hemoglobin to form an MG-H1 adduct.10,11 However, the physiologic relevance of this type of modification on hemoglobin remains unclear. In this study, the association between HbA1c level and GLO1 activity in nondiabetic CarAD patients was investigated with respect to both atherosclerotic plaques and red blood cells (RBCs; used as a surrogate to display the systemic level). Moreover, GLO1 activity in erythrocytes of healthy individuals served as a control. METHODS Study population. Between May and October 2014, 36 patients with CarAD and an indication for open repair by North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria and an HbA1c level of <6.5% (48 mmol/mol) were included in the study consecutively
to prevent selection bias. Patients with HbA1c level $6.5% ($48 mmol/mol), diabetes, or restenosis were excluded; in the case of bilateral disease, patients were included only once. Tissue and blood collection and further processing as well as examination of clinical data were conducted according to ethical guidelines and approved by the Medical Ethics Committee of the University of Heidelberg (S-495/2012 and S-301/2013). All patients gave their written informed consent. Patient characteristics. Patient characteristics were recorded prospectively, including baseline medication (antiplatelet therapy, statin, antihypertensive drugs), and are described in the Table. It was not distinguished between asymptomatic and symptomatic CarAD. Peripheral arterial disease (PAD) was considered relevant if either anklebrachial index was <0.9 or medical history revealed prior vascular open surgery or intervention indicating at least a prior Fontaine stage IIb disease. Nicotine abuse was indicated positive if patients were actively smoking. Existence of coronary heart disease was abstracted from the patient’s medical records. Characteristics of healthy controls. Controls were chosen to be healthy, especially in terms of vascular disease, ie, no history of vascular disease (CarAD, PAD, myocardial infraction, stroke), no previous intervention, no claudication, no arterial hypertension (no treatment with antihypertensive drugs), and no diabetes. Moreover, minimum age was defined to be 50 years (the same as for the CarAD patients). Analysis of blood samples. Venous blood was drawn from CarAD patients on the day of hospitalization, and HbA1c level was analyzed by standard procedures in the central hematology laboratory of the University Hospital Heidelberg. Erythrocytes were isolated from blood samples within 4 hours of donation using Ficoll-Paque PLUS (GE Healthcare, Freiburg, Germany) according to the manufacturer’s instructions and lysed in ice-cold buffer (100 mM sodium phosphate buffer, 0.1% Triton X-100, pH 7.4) supplemented with protease inhibitor (Serva, Heidelberg, Germany). Hemoglobin concentration was determined using a Drabkin assay (Sigma-Aldrich, St. Louis, Mo) with human lyophilized hemoglobin (Sigma-Aldrich) as a standard.12,13 Tissue processing. Carotid artery plaques were resected, collected, and processed within 30 minutes of removal according to the standard operating procedures of the Vascular Biobank Heidelberg. After washing with 0.9% sodium chloride solution to remove remaining blood, samples were flash frozen in liquid nitrogen and stored at 80 C until further use. Protein isolation. Frozen carotid artery plaques were processed as described before in liquid nitrogen to extract proteins.6 Protein concentration was determined by the Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, Ill) using bovine serum albumin as a standard. The resulting lysates were retained on 80 C until further analysis. Western blotting. Protein extracts (20 mg) were separated under denaturing conditions on precast 4-20%
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Peters et al 3
Mini-PROTEAN TGX gels (Bio-Rad, Munich, Germany), transferred to a nitrocellulose membrane, and blocked with 5% nonfat milk in TRIS-buffered saline. Membranes were incubated with either polyclonal rabbit anti-GLO1 (ab96032; Abcam, Cambridge, UK) or rabbit antiglyceraldehyde 3-phosphate dehydrogenase monoclonal antibody (D16H11; Cell Signaling Technology, Danvers, Mass) at 1:1000 dilution. Membranes were subsequently incubated with the appropriate horseradish peroxidaseconjugated immunoglobulin G secondary antibody (1:2000; Cell Signaling Technologies) and visualized on x-ray films using enhanced chemoluminescence detection reagents according to the manufacturer’s instructions. Intensity of bands corresponding to the GLO1 expression was quantified by densitometry using ImageJ (http://rsb. info.nih.gov/ij) and normalized to the intensity of glyceraldehyde 3-phosphate dehydrogenase signals. Assay of GLO1 activity. GLO1 activity was determined, as previously described, using a spectrophotometric method that monitors the initial rate of change in absorbance at 240 nm caused by the formation of S-Dlactoylglutathione.6 Statistical analysis. Statistical analysis was carried out using Student t-test, analysis of variance, Mann-Whitney U test, or c2 test. Results are presented as mean 6 standard deviation or as median. A P value < .05 was considered to be significant. RESULTS To test for an association between GLO1 and HbA1c levels, a Spearman test was used. However, no correlation could be observed (R ¼ 0.088; P ¼ .44). This could be explained by the low sample size of the study cohort. Another reason could be that the relationship between GLO1 and HbA1c is rather displayed by a threshold. Therefore, dividing the patients into two groups seems reasonable. According to the HbA1c value, patients were assigned into two different groups: group I, HbA1c < 5.7% (<39 mmol/mol; n ¼ 20); and group II, 5.7% # HbA1c < 6.5% (39 mmol/mol # HbA1c < 48 mmol/mol; n ¼ 16). The applied thresholds for the HbA1c of 6.5% and 5.7% were chosen for the following reasons. An HbA1c value $6.5% indicates the diagnosis of diabetes by definition; an HbA1c value between 5.7% and 6.5% indicates patients being at risk for development of or even having diabetes. Mean age was 67.8 and 70.8 years, respectively, whereas mean age for all CarAD patients together amounted to 69 years. Mean age of the healthy control group was 60 years. Distribution of demographic and clinical data of the CarAD patients can be seen in the Table. All CarAD patients from both groups were treated with acetylsalicylic acid. Use of a statin was reported in 65% of patients in group I compared with 81% in group II. The median number of antihypertensive drugs was two and three in group I and group II, respectively (Table). Of the 36 CarAD patients, 25 were asymptomatic and 11 were symptomatic (group I, 15 vs 5; group II, 10 vs 6).
Fig 1. A, Glyoxalase 1 (GLO1) activity in carotid artery lesions is reduced in patients with increased hemoglobin A1c (HbA1c) level (n ¼ 20 vs 16; single-tailed Student t-test, P ¼ .048). B and C, On protein level, GLO1 was unchanged between the groups (n ¼ 20 vs 16; P ¼ .25), exemplified by a Western blot of three relevant patients for each group (group I: Biobank No. 14-69, 14-119, and 14-120; group II: Biobank No. 14-91, 14-115, and 14-116). Signals for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as controls.
GLO1 activity and expression in carotid artery plaques. GLO1 was found to be regulated differently on activity and protein level. GLO1 activity in carotid artery plaques was significantly reduced by 29% in group II compared with group I (0.61 6 0.11 vs 0.86 6 0.54 U/mg; P ¼ .048; Fig 1, A). In contrast, GLO1 protein levels did not differ between the groups (P ¼ .25; Fig 1, B and C). GLO1 activity in erythrocytes of CarAD patients and healthy controls. To determine the systemic activity of GLO1 and to reveal a possible compensatory effect,
JOURNAL OF VASCULAR SURGERY --- 2016
4 Peters et al
compared with group I; however, this difference was not statistically significant (P ¼ .10; Table). DISCUSSION
Fig 2. A and B, Analysis of red blood cell (RBC) glyoxalase 1 (GLO1) activity did not reveal a difference between the groups divided by hemoglobin A1c (HbA1c) level (n ¼ 20 vs 16; P ¼ .36) or between an apparently healthy control group (n ¼ 15) and the patients with carotid artery disease (CarAD ; n ¼ 36; P ¼ .15).
the activity of GLO1 in RBCs was analyzed from the two groups as well as from a cohort of healthy individuals (n ¼ 15) serving as a control group. No significant differences were found between groups I and II (0.012 6 0.004 vs 0.013 6 0.004 U/mg; P ¼ .358; Fig 2, A). Furthermore, no difference in GLO1 activity of RBCs could be observed between the healthy controls and the group of all CarAD patients together, irrespective of HbA1c level (0.011 6 0.003 vs 0.012 6 0.004; P ¼ .15, Fig 2, B). Even if the healthy control group was compared to groups I and II, there was no difference in GLO1 activity (P ¼ .28). Analysis of clinical parameters of the CarAD patients. The distribution of clinical parameters between group I and group II revealed an increased number of patients suffering from concomitant PAD in group II (44% vs 10%; P ¼ .020; Table). All patients from both groups were found to have arterial hypertension as a risk factor. Patients in group II were treated with more antihypertensive drugs
In this study, GLO1 activity was found to be significantly reduced in atherosclerotic carotid artery lesions of nondiabetic patients with increased HbA1c. These data are consistent with large prospective studies, which have shown that there is an increased risk for cardiovascular morbidity and all-cause mortality in nondiabetic patients with an elevated HbA1c.7,8 This association with reduced GLO1 activity suggests that GLO1 can act as a protective enzyme in atherogenesis and could be a marker for an increased risk of cardiovascular disease. The reduction in GLO1 activity was not due to a reduction in expression of the GLO1 protein. This suggests that the decrease in activity is due to post-translational modification of the protein, particularly as the activity was measured under saturating concentrations of the hemiacetal substrate. Various post-translational modifications of GLO1 have been described, with only a few having an effect on activity. In one study, four sites of post-translational modification were identified where glutathionylation on Cys139 strongly inhibited GLO1 activity.14 It has also been described that nitric oxide affects GLO1 activity in a reduced glutathione-dependent manner.15 GLO1 was recently found to be a substrate of transglutaminase 2, resulting in either deamination or polyamine conjugation (if present), leading to activation of GLO1 alone or additional stabilization against denaturation, respectively.16 Further studies are required to determine whether there are differences in the kinetic parameters of GLO1 in the atherosclerotic carotid artery lesions from CarAD patients. Furthermore, proteomic studies will help establish the type and extent of post-translational modifications on GLO1. No differences in GLO1 activity were found in the RBCs between the groups studied as well as compared with the control group, suggesting that the effects observed in the atherosclerotic carotid artery lesions are specific to the microenvironment of the damaged tissue, consistent with previous studies.5,6 Analysis of the risk factors between the groups showed that there was a significantly higher number of PAD patients with elevated HbA1c. Previously, we have shown that GLO1 activity is decreased in carotid artery plaques of patients with an increased systemic burden of atherosclerosis in terms of concomitant PAD.6 In this study, arterial hypertension was distributed evenly between the two groups, although patients in group II were found to be treated, nonsignificantly, with more antihypertensive drugs. Numerous studies have investigated the influence of MG on arterial hypertension. However, the effect of GLO1 on blood pressure remains unknown. In an in vivo model, it has been shown that subchronic oxidative stress, in terms of increased reactive oxygen species, resulted in a reduction of GLO1 protein in multiple regions of the brain, leading to a nuclear factor kB-mediated elevation of proinflammatory factors and angiotensin 1
JOURNAL OF VASCULAR SURGERY Volume -, Number -
receptor levels, indicating a role for GLO1 in antihypertensive control.17 Additional studies are required to determine whether a similar proinflammatory signaling pathway is activated in the atherosclerotic carotid artery lesions. CONCLUSIONS GLO1 activity is reduced in atherosclerotic lesions of nondiabetic patients with an increased HbA1c level, indicating a functional role of this detoxifying enzyme in the development of atherosclerosis. The fact that systemic GLO1 activity is regulated irrespective of the HbA1c level emphasizes the importance of the change in the local microenvironment influencing activity of GLO1 and hence induction or progress of atherosclerosis. Further studies, especially focusing on post-translational modifications, are needed to reveal the underlying mechanisms. GLO1 and the control of reactive metabolites and AGEs could be a potential therapeutic target, particularly in regard to nondiabetic patients with an elevated HbA1c level. Moreover, nondiabetic CarAD patients with elevated HbA1c more often suffer from additional PAD and are treated with additional antihypertensive drugs. We thank Anja Spieler for excellent technical assistance and Dr Markus Wortmann for his support. AUTHOR CONTRIBUTIONS Conception and design: AP, TF, PN, SD, DB, MH Analysis and interpretation: AP, ML, SD Data collection: AP, ML, TF Writing the article: AP, MB Critical revision of the article: AP, TF, PN, MB, SD, MH Final approval of the article: AP, ML, TF, PN, MB, SD, DB, MH Statistical analysis: AP Obtained funding: AP, PN Overall responsibility: MH REFERENCES 1. Thompson RC, Allam AH, Lombardi GP, Wann LS, Sutherland ML, Sutherland JD, et al. Atherosclerosis across 4000 years of human history: the Horus study of four ancient populations. Lancet 2013;381: 1211-22.
Peters et al 5
2. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med 2005;352:1685-95. 3. Peppa M, Uribarri J, Vlassara H. The role of advanced glycation end products in the development of atherosclerosis. Curr Diab Rep 2004;4: 31-6. 4. Fleming T, Nawroth PP. Reactive metabolites as a cause of late diabetic complications. Biochem Soc Trans 2014;42:439-42. 5. Kirk JE. The glyoxalase I activity of arterial tissue in individuals of various ages. J Gerontol 1960;15:139-41. 6. Peters AS, Hakimi M, Vittas S, Fleming TH, Nawroth PP, Bockler D, et al. Gender difference in glyoxalase 1 activity of atherosclerotic carotid artery lesions. J Vasc Surg 2015;62:471-6. 7. Selvin E, Steffes MW, Zhu H, Matsushita K, Wagenknecht L, Pankow J, et al. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med 2010;362:800-11. 8. Skriver MV, Borch-Johnsen K, Lauritzen T, Sandbaek A. HbA1c as predictor of all-cause mortality in individuals at high risk of diabetes with normal glucose tolerance, identified by screening: a follow-up study of the Anglo-Danish-Dutch Study of Intensive Treatment in People with Screen-Detected Diabetes in Primary Care (ADDITION), Denmark. Diabetologia 2010;53:2328-33. 9. Haring R, Baumeister SE, Lieb W, von Sarnowski B, Volzke H, Felix SB, et al. Glycated hemoglobin as a marker of subclinical atherosclerosis and cardiac remodeling among non-diabetic adults from the general population. Diabetes Res Clin Pract 2014;105:416-23. 10. Rahbar S. The discovery of glycated hemoglobin: a major event in the study of nonenzymatic chemistry in biological systems. Ann N Y Acad Sci 2005;1043:9-19. 11. Gao Y, Wang Y. Site-selective modifications of arginine residues in human hemoglobin induced by methylglyoxal. Biochemistry 2006;45: 15654-60. 12. Fleming T, Cuny J, Nawroth G, Djuric Z, Humpert PM, Zeier M, et al. Is diabetes an acquired disorder of reactive glucose metabolites and their intermediates? Diabetologia 2012;55:1151-5. 13. Kender Z, Fleming T, Kopf S, Torzsa P, Grolmusz V, Herzig S, et al. Effect of metformin on methylglyoxal metabolism in patients with type 2 diabetes. Exp Clin Endocrinol Diabetes 2014;122:316-9. 14. Birkenmeier G, Stegemann C, Hoffmann R, Gunther R, Huse K, Birkemeyer C. Posttranslational modification of human glyoxalase 1 indicates redox-dependent regulation. PLoS One 2010;5:e10399. 15. Mitsumoto A, Kim KR, Oshima G, Kunimoto M, Okawa K, Iwamatsu A, et al. Nitric oxide inactivates glyoxalase I in cooperation with glutathione. J Biochem 2000;128:647-54. 16. Lee DY, Chang GD. Methylglyoxal in cells elicits a negative feedback loop entailing transglutaminase 2 and glyoxalase 1. Redox Biol 2014;2: 196-205. 17. Wortmann M, Peters AS, Hakimi M, Bockler D, Dihlmann S, Glyoxalase I. (Glo1) and its metabolites in vascular disease. Biochem Soc Trans 2014;42:528-33.
Submitted Feb 4, 2016; accepted Apr 18, 2016.