Like other amino acids, sulfur amino acids affect protein metabolism. They are the components of tissue proteins and, when provided at insufficient levels, they lead to reduced protein synthesis. Methionine is, for example, the first limiting factor in classical diets used for growing chickens because of the high requirement of sulfur amino acids for the synthesis of feathers, whereas poultry diets based on maize and soyabean meal without supplementation are deficient in sulfur amino acids(Reference Hocquette, Tesseraud and Cassar-Malek1). A positive effect of methionine supplementation on muscle growth has been reported in this species, since the addition of methionine to a methionine-deficient diet that is otherwise balanced in terms of other amino acids increases the accretion and synthesis of protein in skeletal muscles. Like methionine, cysteine is used for protein synthesis. Cysteine can be produced through the metabolism of methionine (Fig. 1) and can thus be considered as a non-essential amino acid. It can become conditionally indispensable in particular situations such as stress conditions or inflammatory states. In such situations, cysteine may be used for acute-phase proteins but above all for the synthesis of glutathione, i.e. the most important intracellular antioxidant of the body. This results in an increased cysteine requirement, with a demand for cysteine thus exceeding the body's capacity for cysteine production(Reference Obled, Papet, Breuille and Cynober2–Reference Soeters, van de Poll and van Gemert4). Although insufficient intake of sulfur amino acids is clearly detrimental, negative effects of excess methionine or cysteine have also been reported(Reference van de Poll, Dejong and Soeters5–Reference Garlick7). It is therefore important to obtain a deeper understanding of the interaction between sulfur amino acid metabolism and nutritional demand, especially in vivo.
Fig. 1 Methionine–cysteine metabolic pathways. (1), Transmethylation; (2), trans-sulfuration; (3), folate-dependent remethylation; (4), folate-independent remethylation; B6, vitamin B6; B12, vitamin B12; BHMT, betaine homocysteine methyltransferase; 5-CH3THF, 5-methyltetrahydrofolate; DMG, dimethylglycine; MS, methionine synthase; MT, methyl transferase; 5,10-MTHF, 5,10-methyl tetrahydrofolate; MTHFR, methylenetetrahydrofolate reductase; SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine; SHMT, serine hydroxymethyltransferase; THF, tetrahydrofolate.
Optimising sulfur amino acid nutrition involves considering the different roles of these amino acids, which are recognised to exert several significant influences. They are precursors of essential molecules and act as mediators affecting metabolism and cell functions. For example, cysteine is required for the synthesis of glutathione and taurine, which play a crucial role in oxidative stress conditions since they have the capacity to affect cellular redox status. Some mechanisms through which sulfur amino acids control oxidative status, and potentially amino acid signalling, have been detailed in a recent review(Reference Métayer, Seiliez and Collin8) and they will be not presented here. In the present review, we first address the implications of the roles related to oxidative stress and cell function for nutrition recommendations. Since sulfur amino acids may also be involved in the regulation of gene transcription and participate in methyl group metabolism, we then provide an overview of the roles of sulfur amino acids in controlling gene expression and methylation processes, focusing particularly on the consequences for amino acid nutrition.
Revisiting recommendations for methionine and cysteine in the light of their effects on oxidative stress and cell function
Sulfur amino acids participate in the control of oxidative status since they are involved in the synthesis of intracellular antioxidants, particularly glutathione(Reference Soeters, van de Poll and van Gemert4, Reference Métayer, Seiliez and Collin8, Reference Wu, Fang and Yang9), and in the methionine sulfoxide reductase antioxidant system(Reference Métayer, Seiliez and Collin8, Reference Moskovitz, Rahman and Strassman10, Reference Levine, Mosoni and Berlett11). Using cysteine derivatives (i.e. N-acetylcysteine) or cysteine-rich dietary proteins has been shown to increase blood and/or intracellular glutathione, which may affect the thiol redox status(Reference Stipanuk, Dominy and Lee6, Reference Mariotti, Simbelie and Makarios-Lahham12–Reference Dröge14). Interestingly, dietary cysteine may also improve glucose control and alleviate sucrose-induced insulin resistance(Reference Blouet, Mariotti and Azzout-Marniche15–Reference Mariotti, Hermier and Sarrat17). Moreover, several studies have reported a beneficial effect of cysteine supplementation at nutritional doses in the prevention of post-exercise oxidative stress in human subjects(Reference Tsakiris, Parthimos and Parthimos18–Reference Lands, Grey and Smountas21). There is also evidence that the cysteine/cystine redox couple acts as an extracellular thiol redox buffer (i.e. it represents the major plasma redox pool) and a modulator of cell function. In vitro studies have indicated that the extracellular cysteine/cystine redox potential affects proliferation and the p44/p42 mitogen-activated protein kinase pathway via a transforming growth factor α-dependent mechanism in enterocytes(Reference Jonas, Ziegler and Gu22, Reference Nkabyo, Go and Ziegler23), the production of reactive oxygen species and expression of cell–cell adhesion molecules in endothelial cells(Reference Go and Jones24), and modulates reactive oxygen species-induced apoptosis in retinal pigment epithelial cells(Reference Jiang, Moriarty-Craige and Orr25) and Burkitt's lymphoma cells(Reference Banjac, Perisic and Sato26). In these studies, the effects of the cysteine/cystine redox potential as a thiol/disulfide redox couple were independent of intracellular glutathione status, suggesting a specific role of the extracellular redox environment in cell function(Reference Go and Jones27).
The cysteine/cystine redox potential in plasma has recently been shown to vary during the day, with a pattern influenced by meals which may reflect acute changes in plasma thiol redox state with dietary intake of sulfur amino acids(Reference Blanco, Ziegler and Carlson28). Interestingly, it has been found in rats that the cysteine/cystine redox potential was affected by both decrease and increase in the sulfur amino acid (l-methionine+l-cystine) content of the diet, in contrast to glutathione redox potential(Reference Nkabyo, Gu and Jones29). Taken together, these studies provide new insights into the roles of cysteine and dietary sulfur amino acids in controlling cell function through modulation of the extracellular redox environment. These findings also highlight the complexity of the underlying mechanisms and do not support any systematic benefit from supplementing sulfur amino acids. In agreement with this assumption, decreasing methionine ingestion could be recommended in some situations. For example, due to the role of the methionine sulfoxide reductase antioxidant system, lowering dietary methionine levels appears to decrease the sensitivity of proteins to oxidative damage, oxidative stress and reactive oxygen species production(Reference Sanz, Caro and Ayala30, Reference Caro, Gómez and López-Torres31). Possible dietary implications for humans were discussed in a recent review(Reference López-Torres and Barja32). Since the intake of proteins, and therefore methionine, is higher than required in humans in developed countries, the authors suggest that decreasing excessive consumption could be an efficient way to reduce tissue oxidative stress and improve healthy life span.
Sulfur amino acids, particularly cysteine, may also act on cell function as precursors of hydrogen sulfide (H2S)(Reference Stipanuk33). First considered as a toxic gas, H2S is now recognised for its physiological function as the third (with NO and CO) gaseous transmitter in the body. In the brain, H2S may act as a neuromodulator and protect neurons from oxidative stress(Reference Abe and Kimura34–Reference Kimura and Kimura38). H2S has been shown to directly activate KATP channels in vascular smooth muscle cells and insulin-secreting cells(Reference Zhao, Zhang and Lu39–Reference Yang, Yang and Jia41), thus promoting vasorelaxation and inhibiting insulin secretion(Reference Zhao, Zhang and Lu39–Reference Kaneko, Kimura and Kimura45). Although little is known regarding the production of endogenous H2S in response to changes in cysteine concentrations, it is important to consider the possibility that sulfur amino acids might affect cell function through modulation of H2S and not to ignore the possible consequences in terms of nutrition. Similarly, it is necessary to consider other products of sulfur amino acid metabolism such as sulfates. Sulfate is the predominant endproduct of sulfur amino acid catabolism, and a relationship has been reported between sulfate excretion and sulfur amino acid intake(Reference Hamadeh and Hoffer46, Reference Hamadeh and Hoffer47). Sulfates contribute to glycosaminoglycan synthesis and detoxification processes through reactions depending on 3′-phosphoadenosine-5′-phosphosulfate, and these sulfation reactions account for 27 % of inorganic sulfate turnover in humans(Reference Hoffer, Hamadeh and Robitaille48). In rats, ingesting a sulfur-deficient diet decreases plasma sulfate and elimination of the analgesic acetaminophen, with concomitant increase in its toxication(Reference Gregus, Kim and Madhu49), highlighting the importance of adequate methionine and/or cysteine intake for detoxification processes. Nevertheless, sulfates have long been thought to increase urinary excretion of Ca(Reference Hu, Zhao and Parpia50, Reference Whiting and Draper51); a recent study suggests that sulfur amino acids counterbalance the positive association between protein intake and bone mineral density(Reference Thorpe, Mojtahedi and Chapman-Novakofski52). These positive and negative effects of sulfates should be taken into account when establishing recommendations for sulfur amino acid intake.
Potential function of sulfur amino acids in controlling gene expression
Amino acids are recognised to act as modulators of signal transduction pathways that control gene transcription and translation(Reference Jefferson and Kimball53–Reference Jousse, Averous and Bruhat55). Some in vitro experiments performed in mammals(Reference Shigemitsu, Tsujishita and Miyake56, Reference Stubbs, Wheelhouse and Lomax57) and avian species(Reference Tesseraud, Bigot and Taouis58) have suggested that methionine may have such a signal function by inducing the activation of an intracellular kinase involved in the control of mRNA translation, i.e. the p70 S6 kinase (p70S6K, also called S6K1). In addition, amino acid availability regulates the mammalian general control non-depressible 2/eukaryotic initiation factor 2α (eIF2α) pathway that is thought to be affected by tRNA deacylation in the case of amino acid depletion(Reference Kimball and Jefferson59–Reference Tremblay, Lavigne and Jacques61). By influencing the phosphorylation of eIF2α, this pathway affects (1) the first step in the initiation of mRNA translation through the formation of eIF2-GTP, and (2) amino acid-controlled gene expression through the induction of activating transcription factor 4 (ATF4). Indeed, there is evidence that ATF4 can bind to the amino acid response element in the promoters of genes such as CCAAT/enhancer binding protein (C/EBP) homologous protein (CHOP) and asparagine synthetase, thereby up-regulating their expression(Reference Kilberg, Pan and Chen54, Reference Averous, Bruhat and Jousse62). In relation to the present review, CHOP and asparagine synthetase are over-expressed when cells are deprived of a particular amino acid, for instance leucine, lysine, phenylalanine and methionine(Reference Jousse, Bruhat and Ferrara63, Reference Bruhat, Jousse and Wang64). Different mechanisms are probably involved in such regulation since CHOP expression is strongly induced in response to methionine deprivation but only slightly affected by cysteine, asparagine or histidine deprivation, whereas asparagine synthetase expression is induced equally whatever the amino acid depleted(Reference Kilberg, Pan and Chen54, Reference Jousse, Bruhat and Ferrara63). Cysteine restriction may induce the expression of several genes via the mammalian general control non-depressible 2 protein kinase/ATF4-dependent integrated stress response pathway to cope with oxidative and chemical stresses(Reference Lee, Dominy and Sikalidis65). It has also been reported that homocysteine (a metabolite of methionine) increases the expression of vascular endothelial growth factor by a mechanism involving ATF4(Reference Roybal, Yang and Sun66). It is of note that CHOP can interfere with C/EBP α/β expression and function, thus regulating adipogenesis, which could provide a mechanism through which amino acids have an effect on the pathogenesis of insulin resistance, as proposed by Tremblay et al. (Reference Tremblay, Lavigne and Jacques61). Moreover, methionine and cysteine are potent inhibitors of insulin-stimulated glucose transport in muscle cells(Reference Tremblay and Marette67). These results have been reported in non-physiological conditions, and the importance of a p70 S6 kinase-dependent regulatory loop in the development and/or the maintenance of insulin resistance remains to be determined(Reference Hinault, Van Obberghen and Mothe-Satney68). More information is therefore needed on the potential adverse effects of sulfur amino acid supplementation before use for nutritional purposes.
Many other examples demonstrate that sulfur amino acids are able to regulate the expression of numerous genes, thereby controlling nutrient metabolism and cell functions. Methionine supply has been shown (1) to affect growth hormone-induced insulin-like growth factor I gene expression in ovine hepatocytes(Reference Stubbs, Wheelhouse and Lomax57) and (2) to regulate the expression of an important gene controlling ubiquitin-proteasome-dependent proteolysis (i.e. E3 ubiquitin ligase atrogin-1) in avian fibroblasts(Reference Tesseraud, Métayer-Coustard and Boussaid69). Cysteine and cysteine derivatives (for example, N-acetylcysteine, or N-acetylcysteine amide) can modulate the activity of NF-κB(Reference Mihm, Ennen and Pessara70–Reference Lee, Kim and Park73), which induces the expression of many genes that are involved in cell survival and proliferation, and in the regulation of immune and inflammatory responses (for reviews, see Barnes & Karin(Reference Barnes and Karin74), Chen et al. (Reference Chen, Castranova and Shi75), Grimble(Reference Grimble76) and Wek et al. (Reference Wek, Jiang and Anthony60)). The activation pathways of this transcription factor involve the thiol redox status that responds to antioxidants including cysteine (see the section ‘Revisiting recommendations for methionine and cysteine in the light of their effects on oxidative stress and cell function’ above). Despite the clear theoretical importance of sulfur amino acids in immune function, studies directly linking methionine and cysteine intake and immune function are still needed(Reference Grimble76). Sulfur amino acids are also recognised to control genes involved in lipid metabolism, including cholesterol 7α-hydroxylase and apoA-I(Reference Oda77), which could be in favour of sulfur amino acid supplementation in cases of abnormal lipid metabolism. Nevertheless, high methionine intake can promote the development of atheromatous disorder(Reference Troen, Lutgens and Smith78), indicating that any such supplementation has to be undertaken with care to prevent atherosclerosis. The underlying mechanisms are still unclear. Methionine is the precursor of homocysteine (Fig. 1) and elevated levels of plasma homocysteine (hyperhomocysteinaemia) have been associated with occlusive vascular disease(Reference Gerhard and Duell79). However, Troen et al. (Reference Troen, Lutgens and Smith78) have shown in mice that the atherogenic effect of methionine exists even in the absence of hyperhomocysteinaemia. High plasma homocysteine levels appear not to be independently atherogenic as recently reported(Reference Zhou, Werstuck and Lhoták80, Reference Liu, Wang and Guo81). Vitamin B supplementation (its effect on methionine metabolism is discussed in the section ‘Role of sulfur amino acids in the methylation processes’ below) does not prevent CVD in humans despite lowering homocysteine(Reference Jamison, Hartigan and Kaufman82, Reference Albert, Cook and Gaziano83), similarly challenging the role of homocysteine as an independent cardiovascular risk factor.
Role of sulfur amino acids in the methylation processes
In addition to the above roles, sulfur amino acids have another major function since methionine is a source of the methyl groups needed for all biological methylation reactions, including methylation of DNA and histones, a process that influences chromatin structure and gene expression(Reference Egger, Liang and Aparicio84–Reference Waterland89). DNA methylation generally occurs at cytosines within cytosine-guanine dinucleotides (CpG). The CpG sites are dispersed throughout the genome, with specific CpG-rich regions called CpG islands. The methylation status of cytosine residues within CpG dinucleotides and in the context of CpG islands determines which genes are active or not, and plays an important role in maintaining gene silencing that is needed for tissue- and development-specific gene expression, epigenetic mechanisms involved in the establishment and maintenance of gene expression patterns, and genomic imprinting.
Being converted to S-adenosylmethionine, the major biological methylating agent, methionine provides the methyl groups that are needed for cellular methylation reactions. However, in order to adapt dietary methionine intake effectively, it is important to keep in mind the specific features of the methyl metabolism that is more generally involved in methylation regulation. S-adenosylmethionine is converted to S-adenosylhomocysteine following methyl donation, and to homocysteine (reactions of transmethylation; Fig. 1). Homocysteine can be remethylated into methionine through two separate pathways: (1) the folate-dependent remethylation pathway that involves 5,10-methyl tetrahydrofolate; (2) the folate-independent remethylation pathway in which the methyl group is provided by betaine. Disorders of methyl balance lead to several diseases (liver disease, CVD, closure of the neural tube, synthesis of creatine, cancer), illustrating the key role played by methylation reactions(Reference Brosnan, da Silva and Brosnan90, Reference Williams and Schalinske91). Whether DNA methylation is increased by high levels of methionine is still unclear, since, for example, excess methionine may impair DNA methylation by inhibiting remethylation of homocysteine(Reference Dunlevy, Burren and Chitty92). These findings need further investigation to quantify their physiological relevance, particularly using more physiological amounts of methionine. In addition, based on the methionine cycle, nutritional factors affecting the supply of S-adenosylmethionine and/or removal of homocysteine (for example, dietary choline, betaine, folic acid and vitamin B12) potentially affect DNA methylation and must therefore be taken into account when assessing the supply of methionine. Mathematical models have been developed to consider the complexity of methionine metabolic pathways, and they will be useful to complement and help guide laboratory studies(Reference Reed, Nijhout and Sparks93, Reference Reed, Nijhout and Neuhouser94).
Recent in vivo studies have demonstrated opposite diet-related changes in DNA methylation according to the tissue studied: long-term administration of a folate/methyl-deficient diet in rats results in progressive hypomethylation of DNA in the liver(Reference Pogribny, Ross and Wise95), but an increase in DNA methylation in the brain(Reference Pogribny, Karpf and James96). Although the underlying mechanisms and significance of these findings are not understood, insufficient supply of the nutrients involved in methyl metabolism clearly affects methylation reactions and concomitantly alters patterns of gene expression(Reference Pogribny, Karpf and James96). This function of methyl group donors represents another mechanism though which sulfur amino acids can regulate gene expression (see section ‘Potential function of sulfur amino acids in controlling gene expression’ above) with potentially both negative and positive consequences, for example, either unsuitable dysregulation of gene expression or conversely restoration of appropriate locus-specific DNA methylation using therapeutic pro-methylation diets(Reference Waterland89). Epigenetic regulation thus plays an important role in the potential development (or in contrast prevention) of cancer, CVD, type 2 diabetes and obesity. For example, the availability of methyl group donors regulates the expression of genes involved in the development of the digestive tract through a mechanism dependent on DNA methylation, at least for the homeobox gene CDX1, which could explain the protective role attributed to folates in colon cancer(Reference Lu, Freund and Muller97).
An increasing number of reviews have highlighted the importance of perinatal nutrition (i.e. nutrition in early life), since aberrant methyl metabolism in utero is linked with certain disorders (for reviews, see Ulrey et al. (Reference Ulrey, Liu and Andrews85), Waterland(Reference Waterland89), Rees et al. (Reference Rees, Wilson and Maloney98), Waterland & Michels(Reference Waterland and Michels87) and Nafee et al. (Reference Nafee, Farrell and Carroll88)). For example, a murine metastable epiallele (axin fused, AxinFu) exhibits epigenetic plasticity to the maternal diet, since supplementation with methyl group donors before and during pregnancy increases offspring DNA methylation at AxinFu(Reference Waterland, Dolinoy and Lin99). As discussed by some authors(Reference Waterland89, Reference Rees, Wilson and Maloney98, Reference Rees100), manipulating the sulfur amino acid content of the early diet may induce chronic changes in cell functions that have implications for long-term health. More recent findings have provided evidence that changes in the supply of methionine and specific B vitamins such as B12 during the periconception period can lead to widespread epigenetic changes in DNA methylation in offspring, and modify adult health-related phenotypes(Reference Sinclair, Allegrucci and Singh101). For instance, adult offspring exhibit changes in body composition (fatness), immune function and insulin response. Interestingly, these effects were obtained with modest early dietary intervention (changes within physiological ranges), indicating the importance of adequate dietary methyl group supply for metabolic programming and the need for further studies to define the level of intake. Such studies will certainly provide a rationale for nutritional advice in the future.
Conclusion
Sulfur amino acids have a very significant place among amino acids due to their numerous roles. They are precursors of major components such as glutathione, taurine, H2S and sulfates, and thus act on oxidative status and various signalling pathways. In addition, methionine plays a unique role in epigenetic regulation by affecting DNA methylation. Sulfur amino acids consequently affect metabolism, gene expression and cell functions, as illustrated in Fig. 2, and a change in their dietary levels may have either beneficial or deleterious consequences. An increasing number of studies have been devoted to exploring the key roles of sulfur amino acids, but further experiments are still necessary to obtain direct evidence regarding some of their effects on nutrient metabolism, cell functions and more generally physiological responses in animals and humans. Understanding of the mechanisms of amino acid actions is essential to optimise dietary amino acid requirements whatever the conditions, whether physiological (pregnancy, early nutrition) or physiopathological (disease, ageing). In particular, the effects of sulfur amino acids should be analysed under circumstances affecting oxidative status and gene expression. Recommendations for methionine and cysteine intake have to be revisited in the light of their roles and the intricacies of sulfur amino acid metabolism. The availability of other dietary nutrients (for example, cysteine, vitamins) has of course to be taken into account when adjusting the supply of methionine, for instance. However, a major challenge for future experiments is to improve the integration of the various roles of sulfur amino acids at the whole-body level for rational use in nutrition.
Fig. 2 Regulation of metabolism and cell functions, and possible roles of sulfur amino acids: (1) precursors of protein synthesis; (2) antioxidant function; (3) nutrient signal; (4) methyl group donor. The figure represents a summary of the findings discussed in the present review. MSR, methionine sulfoxide reductase.
Acknowledgements
The authors thank S. Kaushik (INRA, St Pée-sur-Nivelle, France) for helpful comments and discussions, and D. Raine (Surrey, UK) for revision of the English language. Studies on sulfur amino acids in the laboratories of the authors were funded by INRA and Adisseo France SA. The contribution of each author to the paper was as follows: S. T., analysis of the literature, writing of the manuscript and supervision of the review; S. M. C. and A. C., writing of the manuscript, production of the figures and referencing the literature cited; I. S., writing of the manuscript, and discussion to define the aims and structure of the review.
S. T., S. M. C., A. C. and I. S. declare no conflicts of interest.
