Effectiveness of the serum ratio of l-cysteine to glutathione for the prediction of cardiovascular events

Abstract

Aminothiols play an important role in the antioxidant defense system. Their serum profile may be a predictor of prognosis and cardiovascular events. This study followed-up 262 patients (202 men; age, 65 ± 13 years) who had been treated for cardiovascular disease. The patients were divided into two groups by the median total l-cysteine to total glutathione (tCySH/tGSH) ratio in serum at enrollment. There were 32 (11%) all-cause deaths, 20 (8%) cardiovascular deaths, and 32 (12%) major cardiovascular events in 5.5 ± 3.2 years. Twenty-nine (11%) patients were hospitalized for heart failure. The high tCySH/tGSH ratio group (≥80.70) had a higher incidence of all-cause death than the low tCySH/tGSH ratio group (<80.70; log-rank test, p = 0.025). Multivariate Cox regression analysis adjusted for age, sex, body mass index, current smoking, renal function, and log₁₀-transformed brain natriuretic peptide showed that the tCySH/tGSH ratio had predictive value for all-cause death, cardiovascular death, and heart failure. The adjusted hazard ratio for heart failure for the high versus low tCySH/tGSH ratio groups was 3.071 (95% confidence interval: 1.186–7.952; p = 0.021). The tCySH/tGSH ratio may be an useful biomarker to assess prognosis, cardiovascular events, and heart failure.

Introduction

In vivo processes such as metabolism and inflammation generate superoxide through various mechanisms, including nicotinamide adenine dinucleotide phosphate oxidase, xanthine oxidase, nitric oxide synthase, and the mitochondrial electron transport system.^(1,2) Amino acids with thiol group such as glutathione (GSH), l-cysteine (CySH) forms dimer at the disulfide bond and reduce their surroundings upon oxidation. Theses amino acids play an important role in the antioxidant defense system.

The aminothiol profiles have been investigated to estimate oxidative stress in individuals.⁽³⁾ Patel et al.⁽⁴⁾ found that the ratio of l-cystine (CySSCy) and GSH in plasma predicts mortality and cardiovascular events better than CySSCy or GSH alone. Interestingly, the combination of these two aminothiols has a better predictive value than the ratio of the oxidation/reduction states of single amino acids such as CySSCy to CySH or glutathione disulfide (GSSG) to GSH. This is because CySSCy and GSH exist predominantly in the extracellular region than CySH and GSSG.

However, since GSH in plasma is rapidly oxidized to GSSG after blood collection, caution has been required in sample handling and interpretation of results.^(5,6) Quantification of the reduced form requires rapid acidification, protein precipitation, and derivatization of the sample with specific reagents for the -SH group. This has led to the belief that it is more reasonable to measure the ratio of total CySH (CySH + CySSCy) (tCySH) to total GSH (GSH + GSSG) (tGSH), which does not distinguish between redox states. Several studies have used tCySH and tGSH levels for risk assessment.^(7–10) Based on the above, the tCySH/tGSH ratio, like the CySSCy/GSH ratio, is expected to be a potential predictor of cardiovascular events. In the present study, we conducted a cohort study to determine the tCySH/tGSH ratio in patients undergoing coronary angiography and to evaluate the 10-year prognosis.

Materials and Methods

Study population

We recruited patients who underwent elective cardiac catheterization for suspected or already known coronary artery disease at the National Defense Medical College (Tokorozawa, Japan) from November 2012 to November 2013 and performed serum amino acid measurements. The exclusion criteria were age <20 years, ongoing treatment for malignancy, and acute emergent status including acute coronary syndrome and acute heart failure. We obtained written informed consent from each patient. This study was approved by the Ethics Committee of the National Defense Medical College (No. 1084) and was conducted in accordance with the Declaration of Helsinki. The study protocol was registered on the website of University Hospital Medical Information Network (UMIN) Center in Japan (UMIN000009635).

Follow-up

We performed follow-up of the patients until July 2021. At least two of the authors assess the condition of each patient from the hospital records. The primary endpoint was all-cause death and cardiovascular death. The secondary endpoint was major cardiovascular events including myocardial infarction and cerebral infarction. Furthermore, we assessed hospitalization for heart failure. We excluded coronary revascularization therapy and cardiovascular surgery, which had already been planned and performed within 6 months after study entry, from the endpoints.

Coronary angiography and measurement of amino acids

We performed coronary angiography using a 4 Fr catheter system. We obtained angiograms from four standard projections for each of the right and left coronary arteries. We defined obstructive stenosis as a visual narrowing of the lumen of >75%; we defined coronary artery disease as coronary stenosis in at least one major coronary artery or its branches or a clinical history of myocardial infarction, percutaneous coronary intervention, or coronary artery bypass graft surgery. We collected blood samples in plain tubes from the guide sheath during coronary angiography prior to heparin administration. We cooled the blood samples in iced water immediately. We obtained serum by centrifugation at 3,000 ‍rpm for 10 ‍min at 4°C. We stored whole blood and serum at −80°C until analysis. We used the derivatives of reactive oxidative metabolites (d-ROMs) test to measure serum hydroperoxides, a reactive oxygen species as described previously.⁽¹¹⁾ We measured the serum levels of total l-homocysteine (tHcy), tCySH and tGSH by high-performance liquid chromatography using a Shimadzu RF-20A system (Shimadzu Corporation, Kyoto, Japan) with a Symmetry C18 column (3.9 × 150 ‍mm, particle size 5 ‍μm; Waters Corp., Milford, MA). The detection method was based on fluorescence derivatization using the AccQ-Fluor^TM reagent (Waters Corp.), as described previously.^(12,13)

Clinical data

We defined hypertension as blood pressure >140/90 mmHg or the use of antihypertensive medication. We diagnosed diabetes mellitus as fasting blood glucose >7.0 ‍mmol/L or the use of insulin or oral hypoglycemic agents. We defined hyperlipidemia as total cholesterol >5.7 ‍mmol/L, low-density lipoprotein cholesterol >3.6 ‍mmol/L, or the use of anti-hyperlipidemic medication. We calculated the estimated glomerular filtration rate using the Modification of Diet in Renal Disease equation modified for the Japanese population.⁽¹⁴⁾ The lower limit of detection of conventional C-reactive protein was 3,000 ‍μg/L. We obtained the left ventricular end-systolic diameter, end-diastolic diameter, and left atrial diameter by echocardiography. We measured ejection fraction by the Teichholz method or biplane-modified Simpson method as appropriate. We also obtained the Doppler-derived septal E/E' ratio, an indicator of cardiac diastolic function.

Statistical analysis

We evaluated the distribution of continuous clinical characteristics and measurements by examining a histogram and performing the Shapiro–Wilk test. Summary data are presented as the mean ± SD with a 95% confidential interval (CI) for a normal distribution or median (1st, 3rd quartile) for a non-normal distribution. We used the independent samples t test or Mann–Whitney U test for two-group comparisons, as appropriate. We performed cross-table analyses using the chi-squared test or Fisher’s exact test, as appropriate. We assessed the correlation between two variables using Pearson’s method or Spearman’s method.

We used Kaplan–Meier curves to show the unadjusted event-free rate among groups. We assessed differences by the log-rank test. We performed multivariate Cox regression analysis to exclude the influence of the variance of clinical features and laboratory parameters from the prediction of event-free survival. In the assessment, we adjusted the tCySH/tGSH ratio for age and sex (Model 1) or body mass index, hypertension, hyperlipidemia, diabetes mellitus, current smoking, estimated glomerular filtration rate, and log₁₀-transformed brain natriuretic peptide (pmol/L) (Log₁₀BNP) in addition to Model 1 (Model 2). In addition, we performed a subgroup analysis with forest plots. We performed most statistical analyses using JMP ver. 15.0 (SAS Institute, Inc., Cary, NC) and SPSS ver. 22.0 (SPSS Japan, Tokyo, Japan). We performed cubic spline analysis using R ver. 4.2.2 (R Core Team). For all analyses, we considered p<0.05 to be statistically significant.

Results

Patient characteristics and examinations

Of the 302 consecutive patients, 18 were excluded due to undergoing cancer treatment and 22 due to blood sampling at emergent coronary angiography. In total, 262 patients participated in the evaluation. The observation period was 5.5 ± 3.2 years, and the patients were divided equally into two groups by the median tCySH/tGSH ratio (high tCySH/tGSH ratio group ≥80.70; low tCySH/tGSH ratio group <80.70). The high tCySH/tGSH ratio group was older and had a higher prevalence of diabetes despite coronary artery disease was less common (Table 1). The serum concentration of gamma-glutamyl transferase (GGT) was higher in the high tCySH/tGSH ratio group.

Table 1.

Patient characteristics

	tCys/tGSH ratio		p value
	<80.70	≥80.70
Number of subjects	130	132
Age (years)	63 ± 14	67 ± 11	0.017
Sex (male/female), n (%)	33 (25)	27 (20)	0.342
BMI (kg/m2)	24 ± 4	23 ± 4	0.909
Hypertension, n (%)	75 (58)	90 (68)	0.079
Hyperlipidemia, n (%)	64 (49)	72 (55)	0.389
Diabetes mellitus, n (%)	34 (26)	54 (41)	0.011
Current smoking, n (%)	42 (32)	33 (25)	0.191
Previous conditions
CAD, n (%)	76 (58)	58 (44)	0.019
OMI, n (%)	19 (15)	28 (21)	0.164
CABG, n (%)	5 (31)	11 (69)	0.129
PCI, n (%)	32 (50)	32 (50)	0.944
HCM, n (%)	13 (10)	7 (5)	0.152
DCM, n (%)	14 (11)	13 (10)	0.806
Atrial fibrillation, n (%)	13 (10)	16 (12)	0.584
Valvular heart disease, n (%)	7 (9)	12 (15)	0.282
Aortic aneurysm, n (%)	7 (5)	15 (11)	0.081
Medications
Beta-blocker, n (%)	47 (36)	52 (39)	0.589
ACE inhibitor, n (%)	9 (7)	15 (11)	0.213
ARB, n (%)	57 (44)	64 (48)	0.451
Calcium channel blocker, n (%)	43 (33)	51 (39)	0.348
Furosemide, n (%)	16 (12)	25 (19)	0.140
Spironolactone, n (%)	10 (8)	17 (13)	0.167
Statin, n (%)	58 (45)	68 (52)	0.264
Insulin, n (%)	7 (5)	14 (11)	0.120
Warfarin, n (%)	16 (12)	20 (15)	0.504
Hb (g/L)	140 ± 20	130 ± 20	0.015
AST (IU/L)	25 ± 8	28 ± 12	0.010
ALT (IU/L)	22 ± 12	28 ± 20	0.007
GGT (IU/L)	34 ± 50	58 ± 46	<0.001
LDL cholesterol (mmol/L)	2.8 ± 0.8	2.6 ± 0.8	0.006
TG (mmol/L)	1.6 ± 1.9	1.7 ± 1.1	0.566
HDL cholesterol (mmol/L)	1.4 ± 0.4	1.4 ± 0.4	0.548
UA (μmol/L)	350.9 ± 107.1	374.7 ± 113.0	0.097
HbA1c (%)	5.8 ± 1.1	6.0 ± 0.9	0.272
Cr (μmol/L)	73 (62, 85)	78 (67, 105)	0.002
CRP (μg/L)	(3,000, 3,000)	(3,000, 3,000)	0.283
eGFR (ml/min)	69.9 ± 20.7	62.1 ± 23.5	0.005
Log10BNP	1.2 ± 0.1	1.3 ± 0.2	0.038
d-ROMs (U.CARR)	330 ± 82	361 ± 91	0.005
Echocardiography
Dd (mm)*	49 ± 10	50 ± 7	0.326
Ds (mm)*	33 ± 11	34 ± 9	0.263
EF (%)†	62 ± 16	60 ± 15	0.260
E/E'‡	14.8 ± 9.5	14.1 ± 6.0	0.580
LAD (mm)§	38 ± 9	42 ± 9	0.003
Aminothols
tCySH/tGSH ratio	54 ± 16	120 ± 42	<0.001
tCySH (μmol/L)	174 ± 60	181 ± 47	0.256
tGSH (μmol/L)	3.58 ± 1.89	1.60 ± 0.49	<0.001
tHcy (μmol/L)	6.11 ± 3.58	6.50 ± 3.85	0.400

The patients were grouped according to the median tCySH/tGSH ratio. ACE, angiotensin-converting enzyme; ALT, alanine aminotransferase; ARB, angiotensin receptor blocker; AST, aspartate aminotransferase; BMI, body mass index; BNP, brain natriuretic peptide; CABG, coronary artery bypass graft; CAD, coronary artery disease; Cr, creatinine; CRP, C-reactive protein; DCM, dilated cardiomyopathy; Dd, left ventricular end-diastolic diameter; d-ROMs, derivatives of reactive oxidative metabolites; Ds, left ventricular end-systolic diameter; E/E', tissue Doppler-derived septal E/E' ratio; EF, ejection fraction; eGFR, estimated glomerular filtration rate; GGT, gamma-glutamyl transferase; Hb, hemoglobin; HbA1c, hemoglobin A1c; HCM, hypertrophic cardiomyopathy; HDL, high-density lipoprotein; LAD, left atrial diameter; LDL, low-density lipoprotein; OMI, old myocardial infarction; PCI, percutaneous coronary intervention; tCySH, total l-cysteine; TG, triglyceride; tGSH, total glutathione; tHcy, total l-homocysteine; UA, uric acid. n = *122 vs 121, ^†122 vs 122, ^‡105 vs 101, ^§122 vs 119.

Relationship between the tCySH/tGSH ratio and other cardiovascular disease risk factors

The tCySH/tGSH ratio increased with age (r = 0.205, p = 0.001). The tCySH/tGSH ratio correlated with d-ROMs (r = 0.267, p<0.001), hemoglobin (r = −0.271, p<0.001), and Log₁₀BNP (r = 0.222, p<0.001), suggesting a relationship with heart failure. The tCySH/tGSH ratio correlated with GGT (r = 0.297, p<0.001) stronger than tGSH alone (r = −0.222, p<0.001). The tCySH was not associated with GGT. The other correlations between the tCySH/tGSH ratio and clinical parameters are shown in Supplemental Table 1*.

Kaplan–Meier curves for clinical endpoints

The primary endpoint of all-cause death was reached in 31 patients (12%) (Table 2). The Kaplan–Meier Curve for each endpoint is shown in Fig. 1, with significantly higher rates of all-cause death, MACEs, and hospitalization for heart failure in the high tCySH/tGSH ratio group than in the low tCySH/tGSH ratio group.

Table 2.

Cardiovascular events that occurred in the observation period

tCySH/tGSH ratio		<80.70	≥80.70
Number of subjects	262	130	132
All-cause death, n (%)	31 (12)	10 (8)	21 (16)
Cardiovascular death, n (%)	20 (8)	6 (5)	14 (11)
Heart failure, n (%)	5 (2)	1 (1)	4 (3)
AMI, n (%)	1 (0.4)	0	1 (1)
Stroke, n (%)	4 (2)	0	4 (3)
Aortic dissection, n (%)	2 (1)	1 (1)	1 (1)
Sudden death, n (%)	6 (2)	3 (2)	3 (2)
Infective endocarditis, n (%)	2 (1)	1 (1)	1 (1)
MACEs, n (%)	32 (12)	10 (8)	22 (17)
Non-fatal AMI, n (%)	3 (1)	2 (2)	1 (1)
Non-fatal stroke, n (%)	8 (3)	2 (2)	6 (5)
Survivor of VF, n (%)	1 (0.4)	0	1 (1)
Heart failure, n (%)*	29 (11)	6 (5)	23 (17)

Successful resuscitation from fatal arrythmia was classified into MACEs. AMI, acute myocardial infarction; MACEs, major adverse cardiovascular events; tCySH, total l-cysteine; tGSH, total glutathione; VF, ventricular fibrillation. *Including patients who died from heart failure.

Fig. 1.

Kaplan–Meier curves were generated for tCySH/tGSH and (A) all-cause death, (B) cardiovascular death, (C) major adverse cardiovascular events (MACEs), and (D) heart failure. The median of tCySH/tGSH was 80.70. MACEs, major adverse cardiovascular events; tCySH, total l-cysteine; tGSH, total glutathione.

Cox regression analysis

Adjusted Cox regression analysis showed that the tCySH/tGSH ratio was associated with an increased risk of cardiovascular events. The adjusted hazard ratio (HR) for all-cause death was 1.008 (95% CI: 1.001–1.015; p = 0.027) (Table 3, Model 2). For hospitalization for heart failure, the adjusted HR for the tCySH/tGSH ratio was 1.009 (95% CI: 1.003–1.016; p = 0.004) (Table 3, Model 2). The serum GSH levels were not associated with any clinical endpoint. The serum level of tCySH was an important predictor of heart failure. But the predictive value was further improved by dividing with tGSH. The serum GGT was also a predictor for death. However, it could not be used to estimate the risk of MACEs and hospitalization for heart failure.

Table 3.

Unadjusted and adjusted risk for cardiovascular events with increasing tCySH/tGSH ratio and serum aminothiols

	For all-cause death		For cardiovascular death		For MACEs		For heart failure
Events, n (%)	31 (12)		20 (8)		32 (13)		29 (11)
<Scale>
tCySH/tGSH ratio
Unadjusted HR (95% CI), p	1.009 (1.004–1.013)	<0.001	1.010 (1.005–1.015)	<0.001	1.008 (1.004–1.013)	<0.001	1.011 (1.006–1.016)	<0.001
Adjusted HR (95% CI), p, Model 1	1.009 (1.004–1.015)	0.001	1.010 (1.004–1.016)	0.002	1.008 (1.002–1.013)	0.004	1.010 (1.005–1.016)	<0.001
Adjusted HR (95% CI), p, Model 2	1.008 (1.001–1.015)	0.027	1.010 (1.002–1.019)	0.019	1.005 (0.998–1.011)	0.157	1.009 (1.003–1.016)	0.004
tCySH
Unadjusted HR (95% CI), p	1.003 (0.997–1.009)	0.275	1.003 (0.996–1.010)	0.382	1.002 (0.996–1.008)	0.561	1.007 (1.002–1.013)	0.008
Adjusted HR (95% CI), p, Model 1	1.001 (0.995–1.007)	0.788	1.001 (0.993–1.008)	0.869	1.000 (0.994–1.006)	0.898	1.006 (1.000–1.011)	0.042
Adjusted HR (95% CI), p, Model 2	0.996 (0.990–1.002)	0.205	0.994 (0.987–1.002)	0.142	0.995 (0.988–1.002)	0.156	1.003 (0.997–1.009)	0.389
tGSH
Unadjusted HR (95% CI), p	0.937 (0.749–1.173)	0.571	0.859 (0.624–1.183)	0.351	0.874 (0.684–1.116)	0.279	0.882 (0.681–1.142)	0.340
Adjusted HR (95% CI), p, Model 1	0.927 (0.753–1.140)	0.472	0.868 (0.650–1.161)	0.340	0.884 (0.704–1.109)	0.285	0.896 (0.698–1.151)	0.391
Adjusted HR (95% CI), p, Model 2	0.933 (0.763–1.141)	0.499	0.850 (0.646–1.120)	0.248	0.938 (0.749–1.175)	0.579	0.939 (0.721–1.222)	0.638
tHcy
Unadjusted HR (95% CI), p	1.032 (0.955–1.116)	0.426	1.047 (0.958–1.143)	0.311	1.082 (1.010–1.159)	0.025	1.039 (0.962–1.122)	0.332
Adjusted HR (95% CI), p, Model 1	1.027 (0.947–1.114)	0.520	1.039 (0.950–1.136)	0.407	1.068 (0.998–1.142)	0.056	1.033 (0.957–1.115)	0.411
Adjusted HR (95% CI), p, Model 2	0.912 (0.800–1.038)	0.164	0.921 (0.788–1.076)	0.301	1.025 (0.929–1.130)	0.627	0.956 (0.858–1.065)	0.411
GGT
Unadjusted HR (95% CI), p	1.004 (1.000–1.008)	0.035	1.003 (0.998–1.009)	0.193	1.001 (0.996–1.007)	0.614	1.003 (0.999–1.008)	0.184
Adjusted HR (95% CI), p, Model 1	1.005 (1.001–1.009)	0.022	1.005 (0.999–1.010)	0.092	1.002 (0.997–1.008)	0.431	1.004 (0.999–1.009)	0.091
Adjusted HR (95% CI), p, Model 2	1.008 (1.002–1.013)	0.004	1.007 (1.000–1.014)	0.054	1.001 (0.995–1.008)	0.671	1.005 (0.999–1.011)	0.128
<Category>
tCySH/tGSH ratio High vs Low
Unadjusted HR (95% CI), p	2.324 (1.090–4.955)	0.029	2.505 (0.961–6.525)	0.060	2.370 (1.121–5.008)	0.024	4.413 (1.790–10.879)	0.001
Adjusted HR (95% CI), p, Model 1	2.019 (0.944–4.318)	0.070	2.197 (0.840–5.747)	0.109	2.129 (1.004–4.518)	0.049	4.092 (1.648–10.158)	0.002
Adjusted HR (95% CI), p, Model 2	1.409 (0.636–3.124)	0.398	1.568 (0.553–4.448)	0.398	1.368 (0.617–3.032)	0.441	3.071 (1.186–7.952)	0.021

HR was calculated per increase of the following: 1 ‍μmol/L for total l-homocysteine and total glutathione; 10 ‍μmol/L for total l-cysteine. For categorical analysis, patients were divided by the median tCySH/tGSH ratio (80.70). Adjusted factors were age and sex (Model 1). Furthermore, body mass index, hypertension, hyperlipidemia, diabetes mellitus, current smoking, estimated glomerular filtration rate, and Log₁₀BNP were added to those of Model 1 (Model 2). BNP, brain natriuretic peptide; CI, confidence interval; GGT, gamma-glutamyl transferase; HR, hazard ratio; tCySH, total l-cysteine; tGSH, total glutathione; tHcy, total l-homocysteine.

Cubic spline analysis for the primary and secondary endpoints

The results of cubic spline analysis for the primary and secondary endpoints in the univariate analysis of the tCySH/tGSH ratio are shown in Fig. 2. The HRs for all-cause death, cardiovascular death, MACEs, and heart failure increased with increasing tCySH/tGSH ratio. The graphs show that the threshold of tCySH/tGSH 80.70 was optimal to distinguish the outcome patients for every endpoint.

Fig. 2.

The spline curves of hazard ratios for all-cause death, cardiovascular death, MACEs, and heart failure for unadjusted (A–D) and adjusted models (Model 2) of the tCySH/tGSH ratio (E–H) are described. Adjusted factors are age, sex, body mass index, hypertension, hyperlipidemia, diabetes mellitus, current smoking, eGFR, and Log₁₀BNP. BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; MACEs, major adverse cardiovascular events; tCySH, total l-cysteine; tGSH, total glutathione.

Subgroup analysis

We described the forest plots to clarify the association between pathological backgrounds of patients and the predictive values of tCySH/tGSH for the endpoints (Supplemental Fig. 1*). The subgroup analysis suggested that tCySH/tGSH ratio was more effective in predicting cardiovascular death especially in patients with coronary artery disease.

Discussion

The serum tCySH/tGSH ratio was shown to be useful for secondary risk assessment of cardiovascular events. The endpoints of all-cause death, cardiovascular death, and MACEs were reached more frequently in the group with a high tCySH/tGSH ratio. Furthermore, hospitalization for heart failure was increased in the high tCySH/tGSH ratio group. The tCySH/tGSH ratio correlated negatively with hemoglobin concentration and positively with Log₁₀BNP, suggesting an association with heart failure. We also evaluated serum tGSH, tHcy, and tCySH alone, which constitute the major extracellular aminothiol pool in serum. Their association with aging, mortality, cardiovascular events, and vascular function has been investigated extensively,^(4,15,16) but conflicting results also exist.⁽¹⁷⁾ In the present study, none of these serum aminothiols could predict any of the endpoints, except for the association of tHcy with MACEs in univariate analysis.

In the aminothiol antioxidant system, peroxiredoxins, glutaredoxins, and glutathione peroxidases have active site thiolates or selenolates.⁽¹⁸⁾ These enzymes react with oxidants or an oxidized target protein and become oxidized itself by direct or trans-S-glutathionylation. GSH serves as a reductant for the enzymes and forms GSSG, restoring the enzyme’s function. Glutathione reductase then returns the oxidized form of GSSG to the reduced form of GSH with the consumption of nicotinamide adenine dinucleotide phosphate and restores the cellular redox equilibrium. GSH conjugation reaction may occur non-enzymatic (i.e., chemical), but it is significantly accelerated by GSH S-transferase.⁽¹⁹⁾ The GSH conjugate can be released outside from the cell via a transporter, multidrug resistance-associated protein.

Once excreted, GSH cannot return to the cell in their native state and must be degraded into its constituent amino acids (CySH, glycine, and l-glutamate) for recycling. GGT is anchored on the outside of the cell membrane and hydrolyzes GSH into l-glutamate and l-cysteinylglycine. Dipeptidase further hydrolyzes l-cysteinylglycine into CySH, and glycine. Each amino acid is then taken up into the cell and GSH is resynthesized intracellularly. Thus, GSH may also serve as a cysteine source for other tissues.

GSH and GSSG are in equilibrium in the cell at a ratio of 400–500:1.⁽²⁰⁾ On the other hand, GSH and GSSG are also present in very low concentrations extracellularly, and their equilibrium state is dominated by the reduced form of GSH in a ratio of 40:1.⁽²¹⁾ CySH and its dimer, CySSCy, are the most abundant extracellular aminothiols, and their equilibrium state is dominated by the oxidized form, CySSCy, in a ratio of 1:8.⁽²¹⁾ Therefore, the tCySH/tGSH can be considered similar to the CySSCy/GSH in serum.

It is of great interest that the ratio of different aminothiols in serum can be validated as an independent predictor of cardiovascular events. To clarify why a composite assessment would improve the predictive value of aminothiols, we need to look at the properties of each aminothiol in serum. De Chiara et al.⁽²²⁾ investigated the relationship between aminothiols and a number of cardiovascular risk factors comprising smoking, hypertension, hypercholesterolemia, hyperhomocysteinemia, diabetes mellitus, and obesity. They showed that plasma tCySH levels increased with the number of risk factors, but CySH, the reduced form, did not change between the groups classified by the number of risk factors. This suggests that the difference between the groups could be attributed to the concentration of CySSCy, the oxidized form. On the other hand, plasma GSH levels are known to be associated with aging, chronic kidney disease, coronary artery disease, and total cholesterols levels.⁽⁴⁾ We found that only tGSH levels correlated with d-ROMs, an oxidative stress test in the serum aminothiols. These concentration of amino acids in serum can also be affected by metabolisms and renal excretion as well as oxidative and reductive responses. Therefore, the higher predictive values of tCySH/tGSH for cardiovascular events may be due to the interaction effects of the features of individual amino acids.

Our results of correlation analysis suggested that the prognostic effect of GGT activity may contribute to the predictive value of the tCySH/tGSH ratio.^(23,24) The level of GGT activity in the blood has a very strong association with cardiovascular diseases such as myocardial infarction and heart failure and is an independent risk factor, although the causal relationship is unclear.⁽²⁵⁾ GGT is considered an antioxidant enzyme and overexpressed on the surface of oxidative-stressed plaques as an adaptive response.^(26,27) However, the active thiol of l-cysteinylglycine, a degradation product of GGT, reduces oxygen one-electron via metal ions such as Fe³⁺ and Cu²⁺ to generate reactive oxygen species, leading to lipid peroxidation and arterial atheroma formation.^(26–30) Thus, once GGT becomes unregulated, it is thought to form a negative chain of oxidative stress due to its pro-oxidant effect while playing a role in the antioxidant system.⁽³¹⁾ Serum GGT reflects various aspects of hepatocellular damage including alcoholic fatty liver, and serum concentration of GGT was significantly associated with prognosis.^(24,32,33) However, in this study, the serum tCySH/tGSH ratio was a better marker than GGT, especially for cardiovascular risk assessment.

The present study has several limitations that should be considered. First, the sample size was small relative to the incidence of events. Second, although no significant correlation was found between the tCySH/tGSH ratio at baseline and specific cardiac disease, we did not consider the specificity of cardiac disease. We retrospectively performed a subgroup analysis in the study, but the number of patients of each cardiac disease was not sufficient. Further study is needed in this regard. Third, the measurements were performed only once at enrollment, and time-course changes were not considered. Fourth, the serum concentrations of aminothiols can be influenced by diet and supplements,^(34–36) and total protein intake is related to tCySH concentration.⁽³⁴⁾ However, we performed blood sampling in the fasting condition.

In conclusion, the ratio of tCySH to tGSH in serum strengthens the association of these aminothiols with cardiovascular disease and reveals the hidden risks of further events. The tCySH/tGSH ratio is suggested to be of use for risk stratification and the management of patients with cardiovascular disease.

Author Contributions

All authors contributed to the conception and design of the study. YIshinoda and NM carried out the collection, analysis, and interpretation of data. YIshinoda wrote the first draft of the manuscript. NM was the supervisor of the present study and participated in drafting, revising the manuscript, and approving the final version. All authors read and approved the final version of the manuscript and declared that the contents have not been published elsewhere.

Funding

This research was supported in part by a grant (to TA) from the Ministry of Defense of Japan. NM is supported by a Grant-in-aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan (20K08503).

Conflict of Interest

No potential conflicts of interest were disclosed.