Why are dietary recommendations for methionine consumption combined with cysteine?

Why are dietary recommendations for methionine consumption combined with cysteine?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

I want to understand the amino acids missing in certain vegetables. I looked up the US recommendations for amino acids (source: wikipedia).

I don't understand why they pair

  • Methionine + Cysteine: 25 mg/g protein
  • Phenylalanine + Tyrosine: 47 mg/g protein

I searched for a justification but I cannot find one. While I acknowledge the chemical and structural similarities between them, they are separate amino acids. If these pairs could simply be transformed from one to another, then why bother considering methionine and phenylalanine to be essential? If you could easily convert cysteine to methionine, then methionine could be manufactured by the body. Similarly, so could phenylalanine. But it is commonly assumed this does not occur. Thus, an answer based on the idea that they can be "easily transformed into one another", if true, needs to address these issues.

Here are references for methionine:

The intimate relation between amino acids and protein and nitrogen requirements is well recognized. Nutrition research has focused on the capacity of food to meet the need for nitrogen and indispensable amino acids (IAA) and led to the conclusion that the quality, not just the quantity, of protein is critical. This is especially relevant in regard to the sulfur amino acids (SAA) methionine and cysteine because of the increased understanding of their relation to chronic diseases (e.g., cardiovascular disease, dementia, cirrhosis), immunomodulation, DNA transcription, and RNA translation. Considerable effort has been expended to determine whether and to what extent cysteine can spare the requirement for the IAA methionine. In vivo studies in humans generally concur that the dietary requirement of the SAA ranges between 13 and 16 mg/kg/d, but how much can be met by cysteine relative to methionine remains controversial.

Using the indicator amino acid oxidation (IAAO) approach to further examine this issue, Di Buono et al. (9-11) confirmed the mean requirement for methionine as 12.6 mg/kg/d in the absence of exogenous cysteine but noted that a safe level of intake of total SAA for the population was substantially higher at 21 mg/kg/d.

Naomi K. Fukagawa; Sparing of Methionine Requirements: Evaluation of Human Data Takes Sulfur Amino Acids Beyond Protein, The Journal of Nutrition, Volume 136, Issue 6, 1 June 2006, Pages 1676S-1681S

… and phenylalanine:

Conclusion: On the basis of the 24-h IAAO and 24-h IAAB methods, a mean phenylalanine requirement of 38 mg/kg/d is proposed for healthy well-nourished Indian adults in the absence of tyrosine intake. This finding is similar to that in Western adults.

Anura V Kurpad, Meredith M Regan, Tony DS Raj, Vidya N Rao, Justin Gnanou, Vernon R Young; The daily phenylalanine requirement of healthy Indian adults, The American Journal of Clinical Nutrition, Volume 83, Issue 6, 1 June 2006, Pages 1331-1336


  • Why would amino acid requirements be grouped together?
  • If methionine is considered an essential amino acid, wouldn't grouping it with cysteine allow a food high in cysteine to appear as though it is satisfying your methionine requirement?
  • Suppose I consume 12.6 mg/kg/d of methionine, would this affect my ability to consume cysteine?
  • If the recommended level of methionine and cysteine is 19 mg/kg/d and this article proposes 12.6 mg/kg/d, then is it logical to consume approximately 6.4mg/kg/d of cysteine?
  • If true, would the same logic apply to phenylalanine and tyrosine?

In humans, cysteine can be synthesized from methionine and tyrosine from phenylalanine (note that the reverse pathways do not occur). Because their synthesis requires essential amino acids and the biosynthetic capacity of the organism does not always meet its need, they have been labelled conditionally essential. Under normal circumstances, an adult human can live with no dietary source of cysteine so long as dietary methionine is adequately supplied since the dietary deficiency can be compensated for by biosynthesis. Because of this, though, less dietary methionine is required when the diet is also supplemented with cysteine since less methionine is required to synthesize cysteine. This effect is known as cysteine sparing and a similar effect is observed with tyrosine/phenylalanine.

The data you cite are the reference values for PDCAAS, which is an indicative tool to evaluate the essential amino acid composition of protein sources. This system was introduced in 1991 by the WHO, and in this report they give the following rationale for their combined scoring of methionine and cysteine as sulfur amino acids:

The total of methionine and cystine [is] used for scoring purposes… Cystine is not an essential amino acid but can be synthesized from methionine. Cystine in a diet can thus "spare" methionine, and the total of the two has been found more satisfactory for scoring purposes than methionine alone.

They also recognize and somewhat address the weakness in this methodology that inspired your question:

While it is known that cystine can spare part of the requirement for methionine, FAO/WHO/UNU 1985 does not give any indication of the proportion of total sulphur amino acids which can be met by cystine. For the rat, chick and pig, the proportion is about 50%. Most animal proteins are low in cystine; in contrast, many vegetable proteins, especially the legumes, contain substantially more cystine than methionine. Thus, for animal protein diets or mixed diets containing animal protein, cystine is unlikely to contribute more than 50% of the total sulphur amino acids and scores calculated using cystine plus methionine will be appropriate. However, in certain all vegetable combinations, e.g. wheat and legumes, part of the cystine value may not be realized. Because of insufficient data on human requirements, however, the total of the two sulphur amino acids should, for the present, remain the recommended approach for computing amino acid scores.

A further complication arises from our lack of knowledge of the proportion of the total sulphur amino acid requirement which can be met by cystine. Without that knowledge, expression of protein values in terms of the sum total of methionine and cystine has both theoretical and practical limitations.

It's because of their structural similarity. Here are some quick pics of the molecules:


Compare with tyrosine:

You can see that the only difference is the methyl group in para position on the benzyl group in the tyrosine.

Similarly, take a look at cysteine and methionine:


Compare with methionine:

These two look somewhat less similar; you have an extra methyl group on the methionine, a butyl chain instead of a propyl chain, and the alpha carbon bearing the amino group is in R position instead of S. Nonetheless, they are the only amino acids bearing sulfur Also, if you go back to the original paper, they say,

While RDAs are provided for each age group for the nine indispensable amino acids, histidine, isoleucine, leucine, lysine, sulfur amino acids (methionine + cysteine), aromatic amino acids (phenylalanine + tyrosine), threonine, tryptophan, and valine, the requirements for these amino acids are used to develop the FNB/IOM Protein Scoring Pattern.

Because of their similarities, some of them can be fairly easily transformed one from the one to the other in the body. For instance, cysteine can be produced from L-methionine. This results in it being easier to classify them together, since if they can be produced one from another, they become virtually indistinguishable between the intake and usage pathways.


  • Trumbo, Paula, Sandra Schlicker, Allison A. Yates, and Mary Poos. "Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein and amino acids. (Commentary)." Journal of the American Dietetic Association 102, no. 11 (2002): 1621+.

  • Institute of Medicine of the National Academies. 2003. Dietary Reference Intakes. p. 138.

  • U.S. National Library of Medicine. Toxicology Data Network: HSDB: (L)-Tryptophan.

  • Bin P, Huang R, Zhou X. 2017. Oxidation resistance of the sulfur amino acids: Methionine and cysteine. BioMed Research International 2017 : 9584932-6.

Altering Diet Enhances Response to Cancer Treatments in Mice

A diet low in methionine, a nutrient cells need to repair DNA damage, may make therapies like chemotherapy more effective.

People must eat to survive. And the cells that make up the body eat too. Or more accurately, cells break down and rebuild food into the individual molecules they need to stay alive and grow. This complex network of processes is called cellular metabolism.

Cancer cells can alter their metabolism to survive, so targeting cancer cell metabolism has become of great interest to researchers. Questions being asked include: Is it possible to attack a tumor’s nutritional needs as part of cancer treatment? And could this be done by tweaking a cancer patient’s diet?

A new NCI-supported study suggests that the latter may be possible. In the study, researchers showed that feeding mice a diet very low in the nutrient methionine improved the ability of chemotherapy and radiation therapy to shrink tumors.

When the researchers tested a low methionine diet in six healthy adults, methionine levels in their bodies fell and they experienced metabolic changes similar to those seen in the mouse studies. However, the study was not designed to test the effect of methionine restriction on cancer treatment in humans.

The concept of using specific dietary changes to enhance cancer treatment “is really at the very early stages,” said Jason Locasale, Ph.D., of Duke University, who led the new study. “And there’s not going to be one be-all, end-all diet for [treating] cancer. But these aspects of diet seem to have all kinds of really interesting effects on cancer outcomes, and we have to take them seriously.”

Methionine Restriction and Longevity

The Confusion

Some people assume that methionine is something that needs to be reduced in the diet in order to be optimally healthy.

But like almost everything else in biology, methionine is just not good or bad. We know that it is an essential amino acid &ndash we need to get certain amounts of it from food to be in good health.

On the other hand, people come across some scary dangers of this amino acid searching the internet. From brain damage to heart disease risk, methionine seems to be everything but healthy.

To start with, methionine is considered safe in the amounts people take in with food. It&rsquos also safe when used appropriately in medicinal amounts. Serious dangers occur only with using extremely high doses (orally or intravenously) [1, 2].

This post is meant to clarify the health effects of methionine and whether there are any benefits to higher or lower levels.

Restriction Science

Animal studies suggested that restricting methionine consumption can increase lifespan, but this has never been confirmed in humans [3].

A 2005 study showed methionine restriction without calorie restriction extends mouse lifespan [4].

Several studies found that methionine restriction also inhibits certain aging-related disease processes in mice. But no proper human studies have investigated the effects of methionine on aging-related pathways and diseases in humans [5, 6, 7].

In rats, dietary methionine increases mitochondrial ROS production and DNA oxidative damage the liver. Researchers suspect that this is a plausible mechanism for its liver toxicity in excess, but human data are lacking to confirm this [8].

Methionine and Genetics

There are a few genes that might affect the amount of dietary methionine, but their impact on methionine levels in humans is poorly understood.

The MTR gene coded for the MTR enzyme, which converts homocysteine to methionine (see related SNPs). The MTHFR gene indirectly affects the conversion of homocysteine to methionine, by producing the active form of folate [9].

People that have a poorly functioning gene usually require more folate. Not getting adequate folate may raise homocysteine and lower methionine [10].

Lynch syndrome is a type of inherited cancer syndrome associated with a genetic predisposition to different cancer types. In people with Lynch syndrome, low methionine intake was associated with an increased risk of Colorectal Tumor in MTHFR 677 (AA) individuals compared to people with low intake and the normal genotype [11].

However, no studies have replicated these findings. We also don&rsquot know how they relate to people without Lynch syndrome. Lastly, this study only identified potential associations. It does not provide information about causes [11].

Materials and methods

Animal experiments


The approval for all experiments was obtained from the institutional animal care and use committee (IACUC) of Pennington Biomedical Research Center. Male Sprague Dawley (SD) rats between 250 and 270 g were obtained from Envigo RMS, Inc. (Indianapolis, IN). The rats were weight-matched and divided into 4 groups (n = 7) and fed Control (Con), high Methionine (Met), high Cholesterol (Cho), or Methionine+Cholesterol (MetCho) -enriched diets. Purina rodent diet (#5001) with 0.5% cholic acid and 2% maltose dextrin served as the Con diet. The high Met diet was made by enriching the Con diet with 1.5% methionine, the high Cho diet by enriching the Con diet with 2% cholesterol, and the high Met+Cho diet by enriching the Con diet with 1.5% methionine + 2% cholesterol. The energy content of Con, Met, Cho and MetCho diets were 12.71 kJ/g, 12.77 kJ/g, 12.46 kJ/g and 12.52 kJ/g, respectively. All rats were provided their respective diets ad-libitum, with free access to water. The experiment lasted for 35 d and the rats were maintained in a light-controlled room (12:12 h day/night cycle) under a constant temperature (22 °C). Rats were housed in cages with standard bedding.


Weight-matched adult male SD rats were divided into 4 groups (n = 7) and fed the Con diet, the Met diet, the Cho diet, or the MetCho diet for a period of 35 days. From day 10 through day 35, all rats were administered an aqueous solution of sitagliptin (100 mg/kg body weight/day) by oral gavage. In this experiment we had an additional Con group and rats in this group were gavaged with vehicle (water) vehicle Con group. Fasting blood glucose was measured at weekly intervals using a glucometer and tail vein sampling. As was done in the first experiment, an initial blood sample was collected before starting the dietary regimen and a final blood sample was obtained at the end of the experiment.


For experiment 3, weight-matched adult male SD rats were randomly assigned to 3 groups (n = 16 per dietary group) and fed the Con diet or diets enriched with Cho or with MetCho. From each group 50% of rats (n = 8) were gavaged with an aqueous solution of sitagliptin (100 mg/kg body weight/day) while the remaining 50% were gavaged with vehicle (water) orally starting from day 10 through day 35. Fasting blood glucose was measured at weekly intervals using a glucometer and tail vein sampling.

Measurement of body composition

Body weight and body composition of all rats were measured at weekly intervals for the duration of the experiments. Body composition was measured using time domain-NMR spectroscopy (Bruker Minispec, Billerica, MA). The instrument was calibrated using the appropriate standards for fat, lean mass and water as per the protocol of the manufacturer.

Sample collection

The final blood sample was collected at the end of each experiment by cardiac puncture (after CO2 inhalation just before euthanasia). Serum was separated by centrifugation and stored at −80°C for the analysis of biochemical parameters. All the lobes of the liver were carefully dissected and a small segment from the largest lobe of the liver was processed for fixation, paraffin embedding, and sectioning for histological analysis. The remaining tissue was snap frozen in liquid nitrogen and stored at −80°C for further analysis.

RNA isolation and real time PCR

RNA was isolated from liver using TRIzol (MRC, Inc., Cincinnati, OH) and homogenized by a hand-held homogenizer. After incubation for 5 min at room temperature, 1-bromo-3-chloropropane (Sigma-Aldrich, St. Louis, MO) was added and vortexed. After centrifugation at 12,000 rpm for 15 min at 4 o C, the supernatant was transferred to a fresh tube for the addition of 70% ethanol (1:1). Total RNA was isolated using RNeasy mini kit (Qiagen, Germantown, MD) according to the manufacturer’s protocol and RNA samples were quantified on a NanoDrop spectrophotometer (Thermo Fischer Scientific, Waltham, MA). 2.0 μg of total RNA was reverse-transcribed using Oligo-(dT)20 primers and M-MLV reverse transcriptase using the kit from Promega (Madison, WI). 10 ng of cDNA was used for quantitative real-time PCR on a Step One Plus System (Applied Biosystems, Foster City, CA). The sequences of primers used in this study are provided in Table 1. Target gene expression in each sample was normalized to the endogenous control gene cyclophilin in each sample.

Histological evaluation by H&E staining

The liver samples were fixed in 10% Neutral Buffer Formalin and processed on a TissueTek VIP 6 Vacuum Infiltration Processor. They were embedded in paraffin and 5 μm sections were obtained for staining with hematoxylin and eosin (H&E). The H&E staining was performed using a Leica St 5020 Autostainer (Buffalo Grove, IL) and the slides were used for microscopy and histopathological examination. Also, the sections were scanned at 20X using a Hamamatsu Nanozoomer Digital Pathology system (Hamamatsu City, Japan).


Immunohistochemistry was performed on paraffin embedded liver sections. Briefly, the slides were first deparaffinized using xylene and dehydrated using ethanol. Theses slides were then pressure heated at 100 °C for 20 min in Na-citrate buffer. To inactivate the endogenous peroxide activity slides were kept for 12 min at room temperature in 3% H2O2 in TBS. Further, the slides were incubated with blocking buffer for 30 min to block non-specific binding sites followed by overnight incubation in anti 4-HNE primary antibody (Abcam). Detection was performed using Leica Bond Polymer Refine kit and slides were counterstained with hematoxylin. The stained slides were scanned using Hamamatsu Nanozoomer Digital Pathology system (Hamamatsu City, Japan) and the digital information were stored for analysis.

Measurement of liver triglycerides

The triglyceride content in the liver tissue was assayed using a commercially available kit from Cayman chemicals (Ann Arbor, MI). Briefly, 40–50 mg of liver tissue was homogenized in the diluted NP40 buffer containing protease inhibitor cocktail. The homogenate was centrifuged at 10,000 g for 10 min at 4°C and the supernatant was collected and diluted 10 times before assaying for triglycerides. The procedure followed was according to the manufacturers protocol.

Oil-red-O staining

To examine lipid deposition in livers from experimental animals, 10 μm thick liver sections were subjected to Oil Red O staining using the NovaUltra (Woodstock, MD) Oil Red O staining kit. Briefly, frozen liver sections were air dried and fixed in formalin and washed under running water for 1–10 min. After rinsing with 60% isopropanol these sections were stained with freshly prepared Oil red O solution for 15 min. After rinsing again with 60% isopropanol and distilled water, the slides were mounted in mounting media and scanned using a Hamamatsu Nanozoomer Digital Pathology system (Hamamatsu City, Japan).

Picrosirius staining

In order to evaluate hepatic collagen deposition, liver sections were stained using the Picrosirius Red staining kit (Polyscience, Warrington, PA). Briefly, liver sections were deparaffinized and hydrated with distilled water. Then the samples were immersed into solution A (Phosphomolybdic acid) for 2 min and rinsed with distilled water. Subsequently, the slides were kept in solution B (Picrosirius red stain) for 60 min and then in solution C (0.1 N hydrochloric acid) for 2 min. Following this the slides were placed in 70% ethanol for 45 s, again dehydrated, and used for mounting. The stained slides were scanned using Hamamatsu Nanozoomer Digital Pathology system (Hamamatsu City, Japan) and the digital information were stored for analysis.


Both initial and final serum samples and liver samples from all 3 experiments were subjected to metabolomic analysis by Dr. Shawn Campagna, the Director of the Biological and Small Molecule Mass Spectrometry Core facility at the University of Tennessee (

Statistical analysis

For the analysis of variables measured at the end of each experiment, one-way (for diet only) or two-way analysis of variance for multiple comparisons was performed with diet and sitagliptin treatment as main effects followed by post-hoc analysis using Tukey correction for multiple comparisons. Data are presented as mean +/− SEM. P values of 0.05 or less were considered as statistically significant.


We gratefully acknowledge support from the National Institutes of Health (NIH) R01CA193256, R21CA201963 and P30CA014236 (J.W.L.), R35CA197616 (D.G.K.), T32CA93240 (D.E.C.) and the Canadian Institutes of Health Research (CIHR, 146818) (X.G.). We thank M. L. Kiel and T. Hartman for assistance in designing the diets, and S. Heim for help with food preparation in the human study. The human study was partially supported by the Clinical Research Center at Penn State University (NIH M01RR10732). We gratefully acknowledge members of the Locasale laboratory for discussions and apologize to those whose work we could not cite owing to space constraints.

Estimating the tolerable upper intake of sulfur amino acid

Pertinent to this discussion of the effect of sulfur amino acid intake on requirement estimates is the problem of determining a tolerable upper intake (i.e., probable safe intake) for sulfur amino acids. Very little data exist in this area in humans for most amino acids ( 40), mainly because toxicity studies in humans are not ethical. Requirement experiments, if they include diets with sufficiently high intake of amino acids, have the opportunity to contribute data to this current lack of safety information. Those conducting amino acid experiments in the future should acquire blood, urine, and tissue samples when possible and search for appropriate biomarkers of toxicity. These analyses would supply information about the potential safety of these amino acids to help guide further research.

An additional approach to potentially identifying tolerable upper intakes of amino acids is to use requirement experiments in a different way. There is a pattern of change in amino acid retention with increasing intake, which is typical of many amino acids. This generalized pattern is comprised of 3 phases: a phase with a positive slope, a plateau, and a second positive slope. The first phase results from increasing retention of the amino acid as a result of increasing utilization of the limiting amino acid for protein synthesis and other required metabolic functions. The second phase usually has no slope (i.e., a plateau) or some minimum slope, depending on the amino acid, and the intercept of phase 1 and 2 represents a measure of dietary requirement often called the breakpoint estimate. Phase 2 represents a range of intakes where additional increments of the test amino acid are primarily catabolized in proportion to the extra intake. This increasing catabolism in proportion to intake occurs because each additional increment in intake is in excess of the requirements for metabolism, and the amino acid is broken down and used for energy. The third phase is characterized by a positive slope, and the change in slope creates a second intercept or breakpoint. This change in slope results from increasing retention of the amino acid in body pools. This increase in retention is a result of dietary intake exceeding the metabolic capacity to catabolize the amino acid in direct proportion to intake. This third phase, and the second breakpoint, are usually also characterized by an increasing rate of excretion of the amino acid in urine. This second breakpoint may be regarded as one estimate of the tolerable upper intake because it represents the intake where the normal regulatory mechanisms are no longer sufficient to dispose of the excess. The second breakpoint does not necessarily mean that the amino acid or its metabolites are toxic at intakes above this level, nor that toxicity cannot occur at lower intakes—this will probably vary with each amino acid. However, the dietary intake represented by the second intercept could reasonably be described as an intake above which the risk of adverse events is increasing. As a result, we suggest that the intake at which this second intercept occurs would be a good starting point for assessing toxicity of certain amino acids.

Assessment of the tolerable upper intake must also consider the adaptation that occurs with excess amino acid intake. The catabolic mechanisms for most amino acids are up-regulated with chronic intake in excess of requirement ( 41). In addition there are overflow pathways for several amino aids that do not become evident until an excess intake is consumed for a period of time ( 41). The dietary intake at which the second intercept occurs may therefore shift to the right (i.e., greater intake) following adaptation. The up-regulated normal pathways or the new mechanisms that are recruited could either reduce or increase the toxic effects of the amino acid. Therefore, a solid understanding of the metabolism of the test amino acid is required to interpret these results.


In determining the requirement for protein, the subcommittee first considered requirements for the essential amino acids. The required amounts of the nine essential amino acids must be provided in the diet, but because cystine can replace approximately 30% of the requirement for methionine, and tyrosine about 50% of the requirement for phenylalanine, these amino acids must also be considered. The essential amino acid requirements of infants, children, men, and women were studied extensively from 1950 to 1970. Except for infants, where the criterion was growth and nitrogen accretion, the requirement was accepted to be the amount of intake needed to achieve nitrogen equilibrium in short-term studies of adults or positive balance in children (see review by FAO/WHO, 1973 NRC, 1974 WHO, 1985). Estimates of amino acid requirements for various age groups are listed in Table 6-1.


Estimates of Amino Acid Requirements.

In a novel approach to examining these requirements, the need for four amino acids was examined in children whose diets were strictly controlled because of inborn errors of metabolism and who were developing normally (Kindt and Halvorsen, 1980). Requirements determined in this way during the first 3 years of life are in good agreement with the values for isoleucine, leucine, phenylalanine plus tyrosine, and valine given in Table 6-1 for infants and 2-year-old children.

The requirement for histidine has not been quantified beyond infancy. Requirement values are difficult to establish because deficiency symptoms occur only after long periods of low intake. Kopple and Swendseid (1981) demonstrated that nitrogen balance diminished when histidine intake was less than 2 mg/kg per day, and increased when intake was increased to 4 mg/kg per day. WHO (1985) estimated the probable adult histidine requirement to be between 8 and 12 mg/kg per day by extrapolation from the infant requirement this estimate is likely to be high, but safe.

The relatively low requirements estimated for adults have been confirmed by Inoue et al. (1988) using the nitrogen balance method. Studies of whole body lysine, leucine, valine, and threonine oxidation rates suggest that adult requirements for these essential amino acids have been underestimated. Approximations of average requirements according to the 13 C tracer studies are leucine, 40 mg/kg (Meguid et al., 1986a) lysine, 35 mg/kg (Meredith et al., 1986) threonine, 15 mg/kg (Zhao et al., 1986) and valine, 16 mg/kg (Meguid et al., 1986b). These new estimates have been challenged on methodologic and theoretical grounds (Millward and Rivers, 1986) and require further confirmation.

Studies on requirements for individual essential amino acids in the elderly are inconsistent. Some suggest that requirements are increased in the elderly others indicate that they are decreased (Munro, 1983). In the one study in which the same methodology and design were applied to the elderly as in a study of young men, no differences in requirements between age groups were found (Watts et al., 1964). The pattern of requirement for essential amino acids in the elderly is accepted to be the same as for younger adults.

There is no information on amino acid requirements of pregnant and lactating women.

The data demonstrate the unsatisfactory state of knowledge concerning amino acid requirements. The values in Table 6-1 are the best available and serve as the basis for calculation of amino acid requirement patterns at various ages and for procedures for the amino acid scoring of diets (see below).


We thank Erica Goodoff, ELS(D), from Editing Services, Research Medical Library at The University of Texas MD Anderson Cancer Center, for editorial support. This work was supported by National Key Research and Development Program of China [2020YFA0112300 and 2018YFC2000400 to C.C.], the National Natural Science Foundation of China [82072622, 81860488, and 81560432 to Y.R. 81830087 and 31771516 to C.C. 81772847 to L.R. 81672639 to Z.Z. and 81872414 and 81802671 to J.D.], Yunnan Leading Medical Talents Program [L-201610] to Y.R., Yunnan Fundamental Research Projects [2019FB112] and Yunnan excellent young scientist foundation (2020) to J.D., and Project of Innovative Research Team of Yunnan Province [2018HC004 and 2019HC005]. S.-C.J.Y. was a member of an expert panel for Celgene, Inc. S.-C.J.Y. had funding support from Bristol-Myer Squibb, Inc. and DepoMed, Inc. for investigator-initiated clinical studies.

Cysteine plasma pools and bioavailability

In plasma, cysteine is the major thiol that contributes to glutathione levels and protein synthesis. 172 Under normal conditions, protein synthesis prevails over the other cysteine-dependent pathways. Although the degradation of glutathione contributes to the cysteine pool, the resulting levels are not sufficient for ‘normal’ metabolism upon cystine scarcity. 173

In healthy volunteers, the total cysteine availability in plasma is 200–300 μM, distributed across three pools—free reduced, free oxidised and protein bound. Up to 65% of cysteine is bound to proteins (protein s-cysteinylation Fig. 4) 174,175 and this pool increases with age. 176 The remaining cysteine circulates mostly as cystine (25–30%, 40–50 μM) and the low abundance pool constituted by reduced cysteine. The concentration of cystine in blood is higher in women than in men and also increases with age. 177 Cystine bioavailability across various tissues is ensured by different strategies, including drug-transporter-dependent mechanisms. The ability of NRF2 to regulate xCT coupled with the decline of NRF2 with age 178 might account for the increased levels of plasma cystine seen with increasing age. Plasma from xCT-knockout mice contains a higher proportion of oxidised cysteine 179 and xCT expression is increased in many tumours, 180 pointing out its relevance in the context of cancer and eventually contributing for cancer metabolic rewiring. As many cysteine-containing proteins (transporters, receptors and enzymes) at extracellular surfaces or in extracellular fluids are prone to oxidation, their activity might be influenced by the thiol/disulphide redox microenvironment. 181

The oxidation of a cysteine residue within a protein can result in the formation of a cysteinyl radical. l -Cystine is reduced to l -cysteine under the action of l -cystine reductase. Reaction between protein cysteinyl residues and low molecular weight thiols such as free cysteine can yield s -cysteinylated proteins.


Plasma cysteine is strongly associated with body fat mass in human cohorts and diets low in cysteine prevents fat accumulation in mice. It is unclear if plasma cysteine affects fat development or if fat accumulation raises plasma cysteine. To determine if cysteine affects adipogenesis, we differentiated 3T3-L1 preadipocytes in medium with reduced cysteine. Cells incubated in media with 10–20 μM cysteine exhibited reduced capacity to differentiate into triacylglycerol-storing mature adipocytes compared with cells incubated with 50 μM cysteine. Low cysteine severely reduced expression of peroxisome proliferator-activated receptor gamma2 (Pparγ2) and its target genes perlipin1 (Plin1) and fatty acid binding protein-4 (Fabp4). Expression of stearoyl-CoA desaturase-1 (Scd1), known to be repressed with cysteine depletion, was also reduced with low cysteine. Medium depletion of the essential amino acids leucine, valine, and isoleucine had only a modest effect on adipocyte specific gene expression and differentiation. Stimulation with the PPARγ agonist BRL-49653 or addition of a hydrogen sulfide donor enhanced differentiation of 3T3-L1 cells cultured in low cysteine. This demonstrates that the ability to induce PPARγ expression is preserved when cells are cultured in low cysteine. It therefore appears that cysteine depletion inhibits adipogenesis by specifically affecting molecular pathways required for induction of PPARγ expression, rather than through a general reduction of global protein synthesis. In conclusion, we show that low extracellular cysteine reduces adipocyte differentiation by interfering with PPARγ2 and PPARγ target gene expression. Our results provide further evidence for the hypothesis that plasma cysteine is a casual determinant for body fat mass.