Short-Term Vitamin B-6 Restriction Does Not Affect Plasma Concentrations of Hydrogen Sulfide Biomarkers Lanthionine and Homolanthionine in Healthy Men and Women^,,

Abstract

Background: Suboptimal vitamin B-6 status is associated with increased cardiovascular disease risk, although the mechanism is unknown. The synthesis of the vasodilator hydrogen sulfide occurs through side reactions of the transsulfuration enzymes cystathionine β-synthase and cystathionine γ-lyase, with pyridoxal 5′-phosphate as a coenzyme. Two proposed hydrogen sulfide biomarkers, lanthionine and homolanthionine, are produced concurrently.

Objective: To determine whether hydrogen sulfide production is reduced by vitamin B-6 deficiency, we examined the relations between plasma concentrations of lanthionine and homolanthionine, along with other components of the transsulfuration pathway (homocysteine, cystathionine, and Cys), in a secondary analysis of samples from 2 vitamin B-6 restriction studies in healthy men and women.

Methods: Metabolite concentrations were measured in plasma from 23 healthy adults (12 men and 11 women) before and after 28-d controlled dietary vitamin B-6 restriction (0.37 ± 0.04 mg/d). Vitamin B-6 restriction effects on lanthionine and homolanthionine concentrations were assessed. Associations between hydrogen sulfide biomarkers, transsulfuration metabolites, and functional indicators of vitamin B-6 deficiency were analyzed by linear regression.

Results: Preprandial plasma lanthionine and homolanthionine concentrations ranged from 89.0 to 372 nmol/L and 5.75 to 32.3 nmol/L, respectively, in healthy adults. Mean lanthionine and homolanthionine concentrations were not affected by vitamin B-6 restriction (P < 0.66), with marked heterogeneity of individual responses. After restriction, homolanthionine was positively associated with functional indicators of vitamin B-6 deficiency, which differed from hypothesized negative associations. Plasma lanthionine was positively correlated with the concentration of its precursor, Cys, before (R² = 0.36; P = 0.002) and after (R² = 0.37; P = 0.002) restriction. Likewise, homolanthionine concentration was positively correlated with its precursor homocysteine, but only in vitamin B-6 adequacy (R² = 0.41; P < 0.001).

Conclusions: The resiliency of plasma lanthionine and homolanthionine concentrations after short-term vitamin B-6 restriction suggests a minimal effect of moderate vitamin B-6 deficiency on hydrogen sulfide production. Additional research is needed to better understand the metabolism and disposal of these biomarkers in humans. This study was registered at clinicaltrials.gov as NCT00877812.

Introduction

Hydrogen sulfide is produced in mammalian cells by the pyridoxal 5′-phosphate (PLP)⁴–dependent transsulfuration enzymes cystathionine β-synthase (CBS) and cystathionine γ-lyase (CSE). As a regulatory molecule, hydrogen sulfide promotes vasodilation through the opening of K_ATP channels (1–5), which attenuates myocardial ischemia–reperfusion injury and hypertension (6, 7). Accurate measurement of hydrogen sulfide is difficult and prone to inaccuracy because of its volatility and susceptibility to oxidation (8). Lanthionine and homolanthionine, which previously have been described as markers of hydrogen sulfide synthesizing reactions, are formed concurrently with hydrogen sulfide in side reactions of CBS and CSE from Cys and homocysteine, respectively (9–12).

Previous research has shown that vitamin B-6 insufficiency causes a differential effect on the 2 enzymes of the trans-sulfuration pathway. Although PLP is well established as a coenzyme of CBS (13), vitamin B-6 deficiency in rats yields only modest losses of CBS activity (14). In contrast, CSE exhibits substantial losses of activity consistent with the extent of lowering of liver PLP in rats (14). The increase in plasma and urinary cystathionine in humans, along with elevations of liver cystathionine in rats, during moderate vitamin B-6 deficiency documents the existence of a metabolic bottleneck created by reduced cellular PLP supply and the resulting reduction of CSE activity (14–18). In spite of the sensitivity of CSE, vitamin B-6 restriction in humans does not reduce overall flux of the trans-sulfuration pathway (19), nor does it alter overall Cys flux (15). The elevation of cystathionine concentration is postulated to provide a driving force (i.e., increased ratio of cystathionine concentration to its K_m) that compensates for the reduced CSE activity (i.e., lower V_max), thus maintaining transsulfuration flux and Cys production at most degrees of vitamin B-6 insufficiency.

In spite of this resilience and adaptability of the canonical reactions of transsulfuration pathway, the effects of vitamin B-6 deficiency on hydrogen sulfide production in humans or a mammalian animal model are unclear. A recent cell culture study showed that the adequacy of vitamin B-6 supply to cells influences their capacity to form hydrogen sulfide and governs the rate and extent of lanthionine and homolanthionine synthesis (9), suggesting that hydrogen sulfide production may be reduced by vitamin B-6 insufficiency in vivo. We hypothesized that the condensation of 2 molecules of Cys by CBS to form lanthionine would be affected minimally by vitamin B-6 restriction in humans. Homolanthionine formation by CSE-catalyzed condensation of 2 homocysteine molecules is postulated to be lowered after vitamin B-6 restriction because of the reduction in CSE activity and the potential competitive inhibition by the alternate substrate, cystathionine. However, the in vivo nutritional regulation of hydrogen sulfide production by transsulfuration enzymes, including the influence of vitamin B-6 status, has not been fully determined.

We report here the quantitative analysis of the surrogate markers of hydrogen sulfide production, lanthionine and homolanthionine, in healthy men and women before and after vitamin B-6 restriction in 2 previously reported dietary protocols (16, 19). This secondary analysis from the same plasma samples allows assessment of lanthionine and homolanthionine, within the context of previous kinetic, metabolite profiling, and metabolomic analyses (20, 21), to provide initial information regarding the influence of short-term vitamin B-6 insufficiency on hydrogen sulfide production.

Methods

Human vitamin B-6 restriction protocols.

Plasma samples analyzed in this study were obtained from 23 healthy participants from 2 identical dietary vitamin B-6 restriction studies that were previously reported (19, 22). In these studies, health was determined through routine tests of hepatic, renal, thyroid, and hematologic function, as well as a physical examination (19, 22). Further exclusions were no history of gastrointestinal surgeries, chronic disease, chronic smoking, or alcohol or drug use, and a BMI (in kg/m²) <28. Additional exclusion criteria included consumption of supplements containing vitamins, amino acids, or proteins. Preprandial assessment determined nutritional adequacy of serum folate (>7 nmol/L), serum vitamin B-12 (>200 pmol/L), plasma PLP (>30 nmol/L), and plasma total homocysteine (<12 μmol/L). The baseline blood sample was taken after a 2-d standardized diet that was nutritionally adequate (total vitamin B-6 = 1.02 ± 0.11 mg/d). After completion of the 28-d restriction diet (total vitamin B-6 = 0.37 ± 0.04 mg/d), a second preprandial plasma sample was collected. Samples were collected in EDTA-coated tubes and separated by centrifugation of whole blood at 1650 × g for 15 min at 4°C. All samples were stored at 80°C with minimal freeze-thaw cycles. These protocols and subsequent metabolite analyses were approved by the University of Florida Institutional Review Board and the University of Florida Clinical Research Center Scientific Advisory Committee (NCT00877812).

Analytical methods.

Plasma PLP concentrations were determined by reverse-phase HPLC (Dionex) with fluorescence detection (9, 23). Aminothiol concentrations (i.e., total homocysteine, Cys, glutathione, and cysteinylglycine) were quantified as 7-fluorescence-2-oxa-1,3-diazole-4-sulfonate derivatives by reverse-phase HPLC with fluorescence detection (24). Concentrations of Met and the transsulfuration metabolites lanthionine, homolanthionine, and cystathionine were determined by gas chromatography–mass spectroscopy (Thermo DSQII) as N-propyl ester heptafluorobutyramide derivatives in negative chemical ionization mode with selected-ion monitoring (25, 26). Lanthionine and homolanthionine were measured relative to a norleucine internal standard, whereas cystathionine and Met were quantified relative to [¹³C₅]-Met and [D₄]-cystathionine internal standards (Cambridge Isotopes). As described previously (12), recombinant CSE (provided by Ruma Banerjee, University of Michigan) was used for enzymatic synthesis of a homolanthionine reference standard. Homolanthionine was quantified relative to the cystathionine response curve. Previously reported concentrations of homocysteine, Cys, 3-hydroxykynurenine and xanthurenic acid were used for vitamin B-6 restriction study data comparison (20, 21, 27).

Statistical analyses.

All data are presented as means ± SDs. The changes in lanthionine and homolanthionine concentrations, prerestriction compared with postrestriction, were analyzed by Student’s paired t test after log transformation. Linear regression analysis was performed to compare functional indicators of vitamin B-6 deficiency (i.e., cystathionine, 3-hydroxykynurenine, and 3-hydroxykynurenine:xanthurenic acid) and transsulfuration amino acids to hydrogen sulfide biomarkers before and after dietary restriction with the use of SigmaPlot 12.5. A Shapiro-Wilk normality test and a constant variance test were performed before modeling. Statistical significance was determined at the 0.05 level for all procedures used.

Results

Effect of vitamin B-6 restriction protocol.

As reported previously, all participants had serum folate, vitamin B-12, and total homocysteine concentrations in the normal range at the beginning and end of the study (16, 19). The 28-d dietary vitamin B-6 restriction protocol lowered plasma PLP concentrations from 52.6 ± 2.93 nmol/L to 21.5 ± 0.95 nmol/L (P < 0.001). After vitamin B-6 restriction, 15 of the participants were considered to be in the marginally deficient range (plasma PLP 20–30 nmol/L) and 8 were considered to be vitamin B-6 deficient (plasma PLP <20 nmol/L) after a 28-d dietary vitamin B-6 restriction. A 2-sample t test after restriction revealed no significant difference between male and female participants (P = 0.10; data not shown). Descriptive data of the 23 participants in this protocol have been reported previously (21). In addition, elevations of cystathionine and 3-hydroxykynurenine and the ratio of 3-hydroxykynurenine:xanthurenic acid constituted functional evidence of vitamin B-6 insufficiency (P < 0.001; Table 1), as reported previously. At this extent of deficiency, the concentrations of Cys and homocysteine, which are precursors of lanthionine, homolanthionine, and hydrogen sulfide, were not altered (Table 1).

TABLE 1.

Concentrations of amino acid and vitamin B-6 functional biomarkers in preprandial human plasma before and after 28-d dietary vitamin B-6 restriction¹

Metabolite	Prerestriction	Postrestriction	P
Lanthionine, nmol/L	188 ± 18	193 ± 16	0.659
Homolanthionine, nmol/L	14.9 ± 1.7	13.5 ± 1.9	0.460
Total cysteine,2 μmol/L	256 ± 39	253 ± 36	0.577
Total homocysteine,2 μmol/L	6.97 ± 1.26	6.97 ± 1.33	0.925
Cystathionine, nmol/L	145 ± 60.4	232 ± 79.3	<0.001
3-hydroxykynurenine,2 nmol/L	24.5 ± 9.46	32.4 ± 11.0	<0.001
3-hydroxykynurenine:xanthurenic acid	3.36 ± 1.92	5.20 ± 3.10	<0.001
Plasma PLP,2 nmol/L	52.6 ± 2.93	21.5 ± 0.95	<0.001

¹Values are means ± SDs; n = 23. PLP, pyridoxal 5′-phosphate.

²Previously reported concentrations (21).

Hydrogen sulfide biomarkers in healthy human subjects.

Lanthionine and homolanthionine concentrations ranged from 89.0 to 372 nmol/L and 5.75 to 32.3 nmol/L, respectively, in plasma during normal vitamin B-6 status. There was no significant effect of the vitamin B-6 restriction on mean lanthionine and homolanthionine concentrations (Table 1). Plasma lanthionine exceeded homolanthionine by a factor of ∼10 both before and after vitamin B-6 restriction. Lanthionine and homolanthionine concentrations before and after dietary restriction did not differ between male and female participants (P = 0.35). The sum of lanthionine and homolanthionine concentrations also did not change after dietary restriction (P = 0.75).

Despite the lack of significant effect of vitamin B-6 restriction on mean concentrations of plasma lanthionine and homolanthionine, we observed substantial variability in individuals in their response of both lanthionine and homolanthionine to vitamin B-6 restriction (Figure 1A and B). Of the 23 total subjects, 12 (52% of total) had higher lanthionine concentrations after the 28-d vitamin B-6–restriction diet. More women had higher lanthionine concentrations after restriction, whereas the majority of men exhibited lower lanthionine concentrations after restriction. The mean increase was 22% in plasma lanthionine for all subjects who had higher concentrations after restriction, whereas the mean decrease was 19% in plasma lanthionine for all subjects who had lower concentrations after restriction. The heterogeneity in lanthionine concentrations was not related to the concurrent changes in Cys concentration. The majority of subjects (61%) had lower homolanthionine concentrations after restriction. Fifty-five percent of women and 67% of men had lower homolanthionine concentrations after the 28-d vitamin B-6 restriction. The mean changes in homolanthionine concentration for the subjects who increased compared with those who decreased were 43% and 44%, respectively. Increases or decreases in homolanthionine concentration after restriction were not related to changes in homocysteine concentrations.

FIGURE 1.

Human plasma concentrations of homolanthionine (A) and lanthionine (B) before and after 28-d dietary vitamin B-6 restriction. Dashed lines are female subjects and solid lines are male subjects; n = 23.

Relations between lanthionine, homolanthionine, total homocysteine, total Cys, and functional indicators of vitamin B-6 deficiency.

Previously reported plasma metabolite concentrations from the 23 subjects were compared with newly determined concentrations of lanthionine and homolanthionine. Linear regression analysis indicated no significant relation between the concentration of lanthionine and the functional indicators of vitamin B-6 deficiency (i.e., cystathionine, 3-hydroxykynurenine, and 3-hydroxykynurenine:xanthurenic acid) before dietary restriction. Homolanthionine was not associated with 3-hydroxykynurenine or 3-hydroxykynurenine:xanthurenic acid before dietary restriction but was positively associated with cystathionine (R² = 0.38, P = 0.002; Supplemental Figure 1). After the 28-d restriction, homolanthionine was positively correlated with 3-hydroxykynurenine (R² = 0.22, P = 0.025) and the ratio of 3-hydroxykynurenine:xanthurenic acid (R² = 0.39, P = 0.033; Supplemental Figure 1). Lanthionine concentrations were significantly correlated with concentrations of Cys (the precursor of lanthionine) in both adequate (R² = 0.36, P = 0.002) and vitamin B-6–restricted (R² = 0.37, P = 0.002) states (Figure 2A and B). Before dietary restriction of vitamin B-6, homolanthionine was positively correlated with its precursor, homocysteine (R² = 0.41, P < 0.001), even though all participants exhibited plasma total homocysteine concentrations within the normal range (<12 μmol/L). The positive correlation between homocysteine and homolanthionine concentrations disappeared after vitamin B-6 restriction (Figure 2C and D). Homolanthionine and lanthionine were positively correlated when vitamin B-6 status was adequate (R² = 0.36, P = 0.002); vitamin B-6 restriction caused the relation to disappear (R² = 0.062, P = 0.25) (Figure 3). Cystathionine and homocysteine concentrations were positively correlated in vitamin B-6–adequate (R² = 0.23, P = 0.021) and –restricted (R² = 0.18, P = 0.047) states, suggesting that the precursor concentration determines product concentration (data not shown). However, only in vitamin B-6 adequacy was cystathionine positively correlated with Cys production (R² = 0.17, P = 0.049). Homocysteine was positively correlated to Cys concentration before (R² = 0.76, P < 0.001) and after (R² = 0.51, P < 0.001) dietary restriction (data not shown).

FIGURE 2.

Relations between lanthionine and its precursor Cys (A and B) and between homolanthionine and its precursor homocysteine (C and D) in human plasma before and after 28-d dietary vitamin B-6 restriction; n = 23.

FIGURE 3.

Relation between lanthionine and homolanthionine in human plasma before (A) and after (B) 28-d dietary vitamin B-6 restriction; n = 23.

Discussion

This study quantified lanthionine and homolanthionine plasma concentrations in healthy adults before and after a 28-d dietary vitamin B-6 restriction. With respect to these vitamin B-6 restriction studies, we previously reported an elevation in cystathionine, disruptions in the tryptophan pathway, and other consequences (16, 19–21). In the present study, we used several functional biomarkers of vitamin B-6 insufficiency that reflect metabolic effects, including cystathionine, 3-hydroxykynurenine, and the 3-hydoxykynurenine:xanthurenic acid ratio (28), all of which are increased during deficiency. In this manner, we could evaluate the relation of lanthionine and homolanthionine to functional indicators of the deficiency state.

Although vitamin B-6 restriction did not alter the mean concentrations of lanthionine and homolanthionine or their precursors, linear regression analysis showed significant associations between the transsulfuration metabolites. The precursor–product relation between Cys and lanthionine existed before and after vitamin B-6 restriction, which suggests that lanthionine production by CBS is not affected by short-term vitamin B-6 insufficiency. There was, however, an apparent relation between homolanthionine and vitamin B-6 status. The precursor–product relation of homocysteine and homolanthionine was diminished after the vitamin B-6 restriction, likely because of the reduction in CSE activity caused by vitamin B-6 insufficiency. The positive correlation between homolanthionine and the functional biomarkers 3-hydroxykynurenine and 3-hydroxykynurenine:xanthurenic acid after restriction was unexpected, because we predicted lower homolanthionine concentrations with vitamin B-6 deficiency. These findings suggest that homolanthionine production is associated with vitamin B-6 status in humans, although a moderate short-term deficiency did not change plasma homolanthionine concentrations significantly.

Kinetic simulations by Singh et al. (29) predicted that, at normal cellular homocysteine concentrations, CBS would account for 70% of transsulfuration-derived hydrogen sulfide production, with the relative contribution from CBS declining in conditions of elevated homocysteine. From these simulations, the CBS-catalyzed β-replacement reaction of homocysteine and Cys to form cystathionine produced the majority of hydrogen sulfide, with significantly less hydrogen sulfide from the condensation of 2 molecules of Cys to form lanthionine (29, 30). In our study, mean plasma total Cys was 37 times higher than plasma total homocysteine, irrespective of vitamin B-6 status, which corresponded to lanthionine concentrations that were 10 times greater than homolanthionine concentrations. Because essentially all lanthionine is produced by CBS and homolanthionine by CSE (9), our results provide in vivo evidence that CBS-catalyzed hydrogen sulfide–generating reactions greatly exceeded those of CSE and were dependent on the concentration of the respective substrates, Cys and homocysteine.

Interestingly, the concentration of plasma lanthionine was equivalent to that of cystathionine in this study. The canonical and predominant reaction producing cystathionine is the CBS-catalyzed condensation of serine and homocysteine, which does not result in hydrogen sulfide formation. Because only a small portion of cystathionine is derived from the condensation of homocysteine with Cys (30), which produces 1 molecule of hydrogen sulfide concurrently, this suggests that more hydrogen sulfide is produced by the CBS-catalyzed formation of lanthionine than previously expected (29). The limitation of our current findings is that we report only plasma concentrations, which are governed by rates of appearance of Cys, homocysteine, cystathionine, lanthionine, and homolanthionine from cells, in addition to their disposal rates by metabolism or excretion. Absolute quantitative inferences regarding the functions of CBS and CSE in hydrogen sulfide production will require additional information about in vivo rates of synthesis and disposal of the biomarkers lanthionine and homolanthionine.

Homolanthionine has been found in the urine of patients with homocystinuria, thereby defining the relation between precursor and product (31). Furthermore, Chiku et al. (12) suggested that homolanthionine could serve as a biomarker of hydrogen sulfide production in elevated homocysteine conditions. Our results show that homolanthionine also is correlated with homocysteine under normal conditions, extending its application as a hydrogen sulfide biomarker in normal physiologic conditions.

In summary, the major implication of this study was that moderate, short-term vitamin B-6 insufficiency may not substantially reduce the production of hydrogen sulfide. In cell culture studies, severe vitamin B-6 deficiency yielded reduced cellular concentrations of lanthionine and homolanthionine and the hydrogen sulfide production capacity (9). This suggests that a more severe deficiency in humans, potentially accentuated by inflammatory conditions or vitamin B-6 antagonists, could have more extensive effects on hydrogen sulfide production by trans-sulfuration enzymes.