Short answer: A post-translational modification (PTM) is a covalent change to a protein or peptide after the polypeptide chain has been made. Directed PTMs are why the set of covalent protein forms in a cell greatly exceeds what the DNA sequence alone predicts [2]. UniProt's controlled vocabulary now lists hundreds of modification types, and laboratory training materials based on that list put the catalogue above 600 [3][15]. For research peptides and recombinant proteins, the practical question is usually analytical: which covalent forms are present, at which residues, and at what relative abundance?
This article is educational analytical chemistry for laboratory readers. It does not describe effects in people or animals and is not guidance for human or animal use.
What counts as a post-translational modification
In chemical terms, PTMs are the breaking or formation of covalent bonds on the polypeptide backbone or on amino-acid side chains after translation [2]. Walsh and colleagues classified the major eukaryotic types — phosphorylation, acetylation, glycosylation, methylation and ubiquitylation — by the side chain modified, the enzyme class involved, and whether the change is reversible [2]. They also noted the scale of the enzyme inventory behind those reactions: on the order of 500 human protein kinases, about 150 protein phosphatases and about 500 proteases [2].
Not every mass shift on a research peptide is an enzymatic cellular PTM. Storage chemistry and sample prep can add the same or similar mass changes (for example asparagine deamidation or methionine oxidation), so analytical reports usually separate intended biosynthetic modifications from process- or method-induced variants [1][4]. Our guides to deamidation and methionine oxidation cover those chemical routes in detail.
Why PTMs matter for identity and characterization
ICH Q6B, the harmonised guideline on specifications for biotechnological and biological products, states that an inherent degree of structural heterogeneity occurs in proteins because of the biosynthetic processes used to make them, so the desired product can be a mixture of anticipated post-translationally modified forms such as glycoforms [1]. The manufacturer is expected to define that heterogeneity pattern and show consistency with lots used in development [1]. Heterogeneity can also arise during manufacture or storage, and because that profile defines product quality it should be characterized for lot-to-lot consistency [1].
Q6B places peptide mapping, terminal amino-acid analysis and disulfide assignment among the tools used for structural confirmation [1]. In that framing, PTM analysis is part of establishing what the molecule is covalently — not a side topic.
Common PTM classes laboratories encounter
Phosphorylation
Phosphorylation adds a phosphate group, most often to serine, threonine or tyrosine (histidine and aspartate are rarer targets) [15]. In UniProt's PTM vocabulary the monoisotopic mass difference for phosphoserine, phosphothreonine and phosphotyrosine is +79.966331 Da [3]. Kinases install the modification and phosphatases remove it, which is why Walsh treated phosphorylation as a major reversible class [2].
Glycosylation
Glycosylation attaches one or more carbohydrate units to specific residues [15]. N-linked glycosylation is present across domains of life and is built around highly conserved biosynthetic steps even though mature glycan structures vary widely among species [6]. Classic acceptor sequons are Asn-Xaa-Ser/Thr, where Xaa is any residue except proline [5]. Not every sequon is occupied: some Asn-Xaa-Ser sites (for example with Xaa = Trp, Asp, Glu or Leu) are poor oligosaccharide acceptors, whereas the corresponding Asn-Xaa-Thr sequons are glycosylated more efficiently in the same experimental system [5]. O-linked sugars commonly attach to serine or threonine [15]. Because glycans are large and heterogeneous, intact-mass and glycopeptide LC-MS workflows are usually required to describe site occupancy and antennary pattern together with the peptide sequence [1][4].
Acetylation and methylation
Acetylation most often adds an acetyl group to lysine side chains; N-terminal acetylation is also common [15]. UniProt lists a monoisotopic mass difference of +42.010565 Da for N6-acetyllysine and for N-acetylalanine [3]. Methylation adds one or more methyl groups to lysine or arginine [15]. Both modifications change side-chain bulk and charge without adding a large prosthetic group, so high-resolution peptide MS/MS is the usual way to place them [4].
Ubiquitination and SUMOylation
Ubiquitin and SUMO (small ubiquitin-like modifier) are themselves small proteins conjugated to substrate lysines [7][15]. SUMOylation uses an enzymatic cascade analogous to, but distinct from, ubiquitination: E1 activation, E2 conjugation and E3-assisted transfer onto the target [7]. From an analytical standpoint these are large mass additions and branched isopeptide linkages, so detection usually means identifying the remnant tag on a lysine-containing peptide after digestion rather than treating the intact conjugate as a simple +X Da side-chain edit [4][7].
Proteolytic processing
Limited proteolysis is a PTM in the broad sense used by Walsh when counting dedicated proteases among modification enzymes [2]. Signal-peptide removal, propeptide cleavage and conversion of precursor chains into mature termini change both sequence length and the set of peptides a map will show [1][15]. On a research lot this shows up as truncated forms, ragged N- or C-termini, or unexpected peptides at the ends of a peptide map.
Disulfide bond formation
Disulfide bridges covalently link cysteine residues and are a major maturation step for extracellular and ER-folded proteins [12]. Protein disulfide isomerase (PDI) catalyzes disulfide formation and isomerization while also acting as a chaperone [12]. ICH Q6B asks that free sulfhydryls and disulfide bridges be determined where cysteine residues are expected, and notes that peptide mapping under reducing and non-reducing conditions can help locate them [1]. Our disulfide bond guide covers the linkage chemistry itself.
Lipidation
Protein lipidation includes cysteine prenylation, N-terminal glycine myristoylation, cysteine palmitoylation, and serine or lysine fatty acylation [8]. UniProt records N-myristoyl glycine at a monoisotopic mass difference of +210.198365 Da and S-palmitoyl cysteine at +238.229666 Da [3]. These hydrophobic adducts shift chromatographic behavior as well as mass, which matters when a lipidated peptide is being tracked through HPLC and LC-MS.
Hydroxylation
Prolyl 4-hydroxylase converts selected prolines to (2S,4R)-4-hydroxyproline in an irreversible reaction that Gorres and Raines describe as the most common posttranslational modification in humans [9]. UniProt lists 4-hydroxyproline with a monoisotopic mass difference of +15.994915 Da [3]. The same +16 Da signature appears for methionine sulfoxide, so residue localization by MS/MS (or orthogonal chemistry) is needed before a +16 Da peptide is assigned as hydroxyproline versus oxidation [3][4].
Pyroglutamate (pyroGlu)
N-terminal glutamine (and sometimes glutamate) can cyclize to pyroglutamate. Spontaneous cyclization of N-terminal glutamine is slow under physiological conditions [17]; glutaminyl cyclase accelerates conversion of glutaminyl peptides to pyroglutamyl peptides, and in one crude-extract comparison the enzymatic reaction was about 70 times faster than phosphate-enhanced nonenzymatic cyclization [10]. UniProt's mass difference for pyrrolidone carboxylic acid formed from glutamine is −17.026549 Da [3]. Cyclization of N-terminal glutamine to pyroglutamate is a common modification of recombinant monoclonal antibodies and is routinely identified by LC-MS [11]. Because pyroGlu lacks a free α-amino group, it blocks Edman sequencing until it is removed — for example with pyroglutamate aminopeptidase before N-terminal analysis [13]. That is one reason modern labs lean on MS for termini that classical Edman chemistry cannot start. See also our Edman degradation and N-terminus vs C-terminus guides.
Chemical variants that look like PTMs
Some covalent changes share mass signatures with enzymatic PTMs but often arise non-enzymatically during production, purification or digestion:
- Deamidation of asparagine or glutamine adds about +0.984 Da (UniProt monoisotopic difference 0.984016 Da) as the side-chain amide becomes a carboxylate [3].
- Methionine oxidation to the sulfoxide is also +15.994915 Da in UniProt — the same nominal oxygen addition as many hydroxylations [3].
- Digestion artifacts (long, warm, alkaline tryptic digests) can create deamidation and N-terminal glutamine cyclization; the amount of those artifacts scales with incubation time, and faster digests reduce them [16]. Related reading: deamidation, methionine oxidation and peptide mapping.
How labs detect PTMs
Mass spectrometry and peptide mapping
Mann and Jensen summarized the core strategy still used in most labs: enrich or otherwise focus a modified subpopulation when needed, then use mass spectrometry to find modified peptides and map the modified sites in molecular detail [4]. High-resolution precursor masses flag candidate mass shifts; tandem MS/MS places the modification on a residue; isotope-labeling strategies can add quantitative kinetic information [4].
In product characterization, LC-MS peptide mapping is the workhorse for site-specific PTMs because the digest converts one large molecule into sequence-readable peptides whose masses encode both sequence and modification state [1][4]. Intact-mass analysis is complementary when the question is global heterogeneity (for example glycoform envelopes) rather than exact site assignment [1].
Edman sequencing caveats
Edman degradation still confirms N-terminal sequence when a free α-amine is present, and combining Edman data with mass fingerprinting has been used to assign PTMs such as N-terminal methionine loss, acetylation, methylation and hydroxylation on ribosomal proteins [14]. Limits matter:
- N-terminal pyroglutamate blocks the Edman cycle until the residue is enzymatically removed [13].
- Many side-chain PTMs do not yield a standard PTH-amino acid, so an "X" cycle plus a mass difference is often how a modification is inferred [13][14].
- Internal modifications are invisible to N-terminal Edman unless the protein is first cleaved and the modified peptide is isolated [4][14].
Orthogonal clues
Charge-based separations (ion exchange or capillary isoelectric focusing) respond whenever a modification changes net charge — for example deamidation, sialylated glycans, C-terminal lysine variants or N-terminal pyroGlu. They do not by themselves locate the residue — MS still does that — but they are useful orthogonal profiles when a lab is monitoring lot consistency of anticipated PTM patterns, which ICH Q6B expects manufacturers to define [1].
Putting it together for research peptides and proteins
A practical characterization checklist for a research protein or large peptide usually asks:
- What covalent forms does the gene sequence plus expected maturation predict (disulfides, glycans, processed termini, pyroGlu)? [1][10][12]
- Which of those forms are observed by intact mass and by peptide map LC-MS/MS? [1][4]
- Which additional mass shifts match known chemical routes (deamidation, Met oxidation) rather than enzymatic installation? [3][4]
- Did sample preparation itself create pyroGlu or deamidation artifacts? [11][16]
FAQ
What is a post-translational modification?
A covalent chemical change to a polypeptide after translation — on a side chain or the backbone — that creates a distinct molecular form [2].
How many kinds are there?
UniProt maintains a controlled vocabulary of PTMs used in its knowledgebase; training materials based on that list cite more than 600 types [3][15].
Which PTMs show up most often in recombinant protein analytics?
Glycosylation, disulfide pairing, N-terminal pyroglutamate, C-terminal heterogeneity, deamidation and methionine oxidation are among the forms ICH-aligned characterization programs routinely document [1][11].
Can mass spectrometry tell enzymatic PTMs from chemical damage?
MS reports mass and site. Biology versus damage is an interpretation that uses process knowledge, time-course or stress studies, and sometimes enzyme-specific biology — the spectrum alone gives the covalent structure [4].
Why does Edman sequencing miss some PTMs?
Blocked N-termini such as pyroglutamate prevent the first coupling cycle, and many side-chain modifications are not recovered as standard PTH derivatives [13][14].
Is PEGylation a PTM?
Cellular PTMs are enzyme-guided edits inside biological systems [2]. PEGylation is typically a deliberate chemical conjugation used in research and product chemistry; it is related as a covalent mass tag but not as a cellular PTM [2].
Bottom line
Post-translational modifications are covalent edits that multiply the number of distinct protein forms beyond the coding sequence [2][3]. Phosphorylation, glycosylation, acylation/methylation, ubiquitin-like conjugation, proteolysis, disulfides, lipids, hydroxylation and pyroglutamate cover most of what analytical labs are asked to identify [2][6][7][8][9][10][12]. ICH Q6B treats anticipated PTM heterogeneity as part of product definition [1], and LC-MS peptide mapping — with Edman used carefully at free N-termini — is how those forms are usually confirmed residue by residue [1][4][13][14].
All products are for research use only — not for human or animal consumption.