A dipeptide is two amino acids joined by a single peptide bond. It is the smallest possible peptide. For a long time dipeptides were studied far less than amino acids or proteins. Only a few, above all the sweetener aspartame and the infusion component L-alanyl-L-glutamine (Ala-Gln), found commercial use, partly because there was no cheap way to make most of them[1]. That has begun to change with fermentation methods that produce dipeptides directly, and with growing evidence that some dipeptides do things their amino acids alone cannot.
The interesting question about dipeptides is therefore how two amino acids joined together differ from the same two amino acids on their own. The answer runs through this page. A dipeptide can dissolve better or survive heat better, it is absorbed from the gut by a different route, and a few dipeptides have effects of their own. Getting them cheaply enough to use is the last part of the story.
| Structure | Two amino acid residues and one peptide bond; unless modified, a free amino end and a free carboxyl end |
|---|---|
| Possible combinations | 400 from the 20 standard amino acids, because the order matters |
| First made in the lab | Glycylglycine, by Emil Fischer and Ernest Fourneau in 1901 |
| Best-known examples | Aspartame, carnosine, anserine, Ala-Gln |
| Uptake from the gut | Whole, by the proton-coupled transporter PEPT1 |
Structure and naming
A dipeptide forms when the carboxyl group of one amino acid reacts with the amino group of another and a molecule of water is released. Unless one of its ends is later modified, as in aspartame, the product keeps a free amino group at one end, the N-terminus, and a free carboxyl group at the other, the C-terminus. By convention the sequence is written from the N-terminus. Ala-Gly (alanylglycine) and Gly-Ala (glycylalanine) therefore contain the same two amino acids but are different compounds. With 20 standard amino acids there are 20 × 20 = 400 possible dipeptides[2], and many more once unusual amino acids such as β-alanine are included.
The first dipeptide made in the laboratory was glycylglycine. In 1901 Emil Fischer and Ernest Fourneau obtained it by opening glycine anhydride, a ring of two glycine units, in hot hydrochloric acid[3]. The starting ring was itself a cyclic dipeptide.
To calculate the mass, formula or charge of any dipeptide, use the peptide calculator.
Well-known dipeptides
Most dipeptides simply deliver their two amino acids, but the best-known ones are known precisely because they do something more: they taste sweet, they are stable in a drip, or they act on the body in their own right.
| Dipeptide | Sequence | Known for |
|---|---|---|
| Aspartame | Asp-Phe methyl ester | Low-calorie sweetener, about 180 times as sweet as sugar[1] |
| Alanylglutamine | Ala-Gln | Source of glutamine in intravenous nutrition[1] |
| Glycyltyrosine | Gly-Tyr | Soluble source of tyrosine in intravenous nutrition[1] |
| Carnosine | β-Ala-His | Abundant in muscle of mammals, birds and fish[1] |
| Anserine | β-Ala-(N-methyl)His | Methylated relative of carnosine[1] |
| Kyotorphin | Tyr-Arg | Analgesic peptide first isolated from bovine brain[4] |
| Valyltyrosine | Val-Tyr | Inhibits angiotensin-converting enzyme; used in a blood-pressure food in Japan[1] |
β-Ala is β-alanine, an amino acid that is not used to build proteins.
Aspartame is by far the most important commercially: more than 19,000 tonnes a year were used worldwide by 2008[1]. It is an exception in taste as well. When synthetic dipeptides were tasted systematically, most turned out to be bitter[1].
Several dipeptides found in hydrolysed fish, seaweed and mushrooms, among them Val-Tyr, Ile-Tyr, Ile-Trp and Lys-Trp, lower blood pressure in animal studies by inhibiting angiotensin-converting enzyme, the same target as the ACE-inhibitor drugs. In Japan, extracts containing them are approved as foods for specified health uses[1].
Kyotorphin relieves pain when it is injected into the brain but has little effect when given by other routes, because it hardly crosses the blood–brain barrier. It has since been found in the brains of other mammals and in human cerebrospinal fluid[4].
Dipeptides in clinical nutrition
In the body, most dipeptides are quickly split into their amino acids, so they act as a delivery form for those amino acids. Their advantage lies in their chemistry. Glutamine breaks down when an infusion is heat-sterilised, while Ala-Gln survives it. Tyrosine hardly dissolves in water, while Ala-Tyr dissolves up to 14 g/L. Some dipeptides even dissolve far better than either of their parts: alanine dissolves to 89 g/L and glutamine to 36 g/L, but Ala-Gln to 586 g/L[1].
This is why Ala-Gln and Gly-Tyr are used in intravenous feeding solutions. Glutamine is in demand after surgery and injury, when muscle stores of it fall sharply, but on its own it is too unstable for ready-made infusions[2]. Synthetic Ala-Gln was first given intravenously to healthy volunteers in December 1985. It was cleared from the blood quickly, and alanine and glutamine rose in equal amounts as it was cleaved. In patients after operations, 20 g of Ala-Gln a day, which supplies about 13 g of glutamine, improved nitrogen balance compared with feeding without it[2].
How the gut absorbs dipeptides
It was long believed that only free amino acids, the end products of digestion, are absorbed from the gut. In fact the small intestine also takes up di- and tripeptides intact, through a dedicated carrier called PEPT1, although how much of all dietary amino acids enters this way is still not known[5]. It ignores free amino acids and tetrapeptides, but among di- and tripeptides it is very broad: it is thought to carry essentially every one released during digestion[5]. The driving force is the proton gradient across the gut lining, which lets PEPT1 pull peptides into the cell even against a concentration gradient[5]. Inside the cell most of them are quickly split into amino acids, which then pass on into the blood[4]. A related carrier, PEPT2, binds the same substrates more tightly but moves them more slowly. It is found in the kidney tubules, the brain, the lungs and the mammary gland[4].
When a segment of the small intestine was perfused in human volunteers, dipeptides were usually absorbed faster than a mixture of the same amino acids in free form, probably because PEPT1 has a very high capacity[5]. People with cystinuria show this clearly. Their gut absorbs free lysine poorly, but lysine given as a lysyl dipeptide is absorbed as well as in healthy people[5]. PEPT1 also explains why some drugs work well by mouth. Aminocephalosporin antibiotics and some ACE inhibitors resemble dipeptides closely enough to be carried by it, and reach oral availability of up to 90%[5].
Histidine dipeptides
The histidine dipeptides are the clearest case of dipeptides with a job of their own, rather than a convenient way of delivering amino acids.
Carnosine was discovered in meat extract by Vladimir Gulevich (Gulewitsch) and S. Amiradžibi in 1900. Together with anserine, it occurs in the muscle of mammals, birds and fish[1]. Both are thought to help keep muscle pH stable during intense work and to act as antioxidants[1]. The muscle of skipjack tuna, a fast-swimming fish, is particularly rich in them, which is one reason they are sold in sports nutrition. Zinc carnosine is used as an anti-ulcer drug[1]. N-acetylcarnosine has been proposed as eye drops against cataracts[1].
Cyclic dipeptides
A dipeptide can also close on itself. When its free amino end attacks its own carboxyl end, it forms a six-membered ring, a 2,5-diketopiperazine, the smallest of the cyclic peptides. They are among the most common peptide derivatives in nature[6]. Many form without any enzyme during fermentation, food processing and the storage of peptides and proteins, which is why they turn up in food and in protein hydrolysates[6]. Bacteria also make them on purpose, and some species use them as signals to sense their own population density[7].
The ring makes cyclic dipeptides resistant to protein-digesting enzymes and lets some of them cross the gut wall and the blood–brain barrier[7]. Several have effects in mammals, but few are known to be made in the mammalian body. The best studied is cyclo(His-Pro), found throughout the brain and in blood, cerebrospinal fluid and urine; cyclo(Pro-Gly) has also been found in rat brain[7]. In the brain at least part of it forms from the hormone TRH (pGlu-His-Pro-NH2): an enzyme removes the first residue, and the remaining His-Pro closes into a ring on its own[6].
How dipeptides are made
Cost was the obstacle mentioned at the start: for most dipeptides there was no cheap way to make them. There are three broad routes, chemical, chemoenzymatic and enzymatic synthesis, and they differ mainly in how many steps and how much protection they need[1].
Chemical synthesis follows four steps. All functional groups except those that will form the peptide bond are protected. The free carboxyl group of one amino acid is activated. The activated amino acid reacts with the other one. Finally, all protecting groups are removed. Any dipeptide can be made this way, usually in high yield. The drawbacks are cost, because of the many steps and reagents, the risk of racemisation and sometimes the need for hazardous reagents, so the method is used mostly in research laboratories[1]. One industrial route to Ala-Gln goes through an alanine N-carboxyanhydride made with phosgene[1]. For longer chains the same cycle of coupling and deprotection is repeated on a resin, in solid-phase peptide synthesis.
Chemoenzymatic synthesis runs a protease in reverse, so that it joins amino acids instead of cutting them apart, with at least one partner protected[1]. Aspartame is made this way with thermolysin, an enzyme from Bacillus thermoproteolyticus. Thanks to the selectivity of the enzyme, the process avoids the bitter β-linked by-product that chemical synthesis gives and can start from a cheap mixture of D- and L-phenylalanine[1].
Enzymatic synthesis uses enzymes that join unprotected amino acids. By 2008 the most practical was L-amino acid α-ligase, found in 2005 by searching the genome of Bacillus subtilis. It makes only dipeptides, never longer chains, and fixes their order: it forms Ala-Gln but not Gln-Ala. It does not accept acidic or basic amino acids, and 44 different dipeptides were confirmed as its products. An Escherichia coli strain engineered to carry this enzyme makes Ala-Gln directly from glucose and ammonia, without any amino acids being added[1]. Such fermentation needs neither protecting groups nor purified amino acids, which removes the main costs of the other two routes, and it may make many more dipeptides available for study and use[1].
References
- Yagasaki M, Hashimoto S (2008). Synthesis and application of dipeptides; current status and perspectives. Applied Microbiology and Biotechnology 81:13–22. DOI: 10.1007/s00253-008-1590-3
- Fürst P (1993). Glutamine containing dipeptides: an overview. Clinical Nutrition 12 (Suppl 1):62–65. DOI: 10.1016/0261-5614(93)90156-X
- Fischer E, Fourneau E (1901). Ueber einige Derivate des Glykocolls. Berichte der Deutschen Chemischen Gesellschaft 34:2868–2877. DOI: 10.1002/cber.190103402249
- Santos S, Torcato I, Castanho MARB (2012). Biomedical applications of dipeptides and tripeptides. Biopolymers (Peptide Science) 98:288–293. DOI: 10.1002/bip.22067
- Daniel H (2004). Molecular and integrative physiology of intestinal peptide transport. Annual Review of Physiology 66:361–384. DOI: 10.1146/annurev.physiol.66.032102.144149
- Prasad C (1995). Bioactive cyclic dipeptides. Peptides 16:151–164. DOI: 10.1016/0196-9781(94)00017-Z
- Bellezza I, Peirce MJ, Minelli A (2014). Cyclic dipeptides: from bugs to brain. Trends in Molecular Medicine 20:551–558. DOI: 10.1016/j.molmed.2014.08.003