What a Tag Adds: Mass, pI and Charge, Tag by Tag
7 min read · Updated November 11, 2026
Every tag is a peptide, so it has a mass, and adding it to a protein moves the band on a gel. That much is expected. What surprises people is how little mass some tags add and how much they move the isoelectric point, and how differently the small and large tags have to be handled.
This page runs the common tags through a real protein so the shifts are numbers rather than adjectives.
Small tags: compute the mass from the residues
A peptide tag is six to twenty-four residues, so its mass is the sum of residue masses plus one water — the same arithmetic as for any protein. On a 250-residue model protein of 26.92 kDa, the measured additions are:
- His6, HHHHHH, 6 aa: +822.85 Da, taking 26.92 to 27.74 kDa.
- FLAG, DYKDDDDK, 8 aa: +994.97 Da, to 27.92 kDa.
- Strep-tag II, WSHPQFEK, 8 aa: +1,040.15 Da, to 27.96 kDa.
- HA, YPYDVPDYA, 9 aa: +1,084.15 Da, to 28.01 kDa.
- c-Myc, EQKLISEEDL, 10 aa: +1,185.30 Da, to 28.11 kDa.
- V5, GKPIPNPLLGLDST, 14 aa: +1,403.64 Da, to 28.33 kDa.
- 3×FLAG, DYKDDDDKDYKDDDDKDYKDDDDK, 24 aa: +2,984.90 Da, to 29.91 kDa.
Why a catalogue says 0.84 kDa and the gel says 0.82
Both figures for His6 are in circulation and neither is a typo. A tag quoted on its own is a free peptide, so it carries its own water: six histidines plus one H₂O is 840.86 Da, the 0.84 kDa a datasheet prints. A tag fused to a protein loses that water in forming the peptide bond, so what the fusion gains is 822.85 Da, and 822.85 is what a mass computed from the fused sequence will report.
The difference is 18 Da, which is invisible on any gel and matters only if you are reconciling a datasheet against a calculator. It is worth knowing which convention you are looking at, because the same 18 Da sits behind every tag in the list above and a construct mass that is short by exactly one water per junction is a convention mismatch, not a missing tag.
The pI moves far more than the mass does
His6 adds 0.82 kDa, three percent of the protein, and moves the pI from 5.48 to 7.10 — 1.62 pH units, because six histidines are six titratable imidazoles with pKa near 6. The same protein with V5, which is 1.4 kDa of mostly neutral residues, moves from 5.48 to 5.49, a hundredth of a unit.
This is the number that decides whether a purification works. An ion-exchange step run at pH 6 that was chosen for the untagged protein will behave differently for the tagged one, and a protein whose pI crosses the running pH reverses its charge and binds the wrong resin entirely. FLAG and 3×FLAG move the pI the other way — 5.48 to 4.89 and 4.52 — because their aspartates are acidic.
How much of the shift you get depends on the starting protein. Replacing every lysine in the same model with glutamate drops its pI to 2.26, and His6 then moves it only to 3.15 — 0.90 units rather than 1.62, because six histidines are a smaller fraction of an already strongly negative molecule's charge.
An epitope tag can change the extinction coefficient too
ε₂₈₀ is computed from Trp, Tyr and cystine counts, so a tag carrying any of those changes it. On the model protein the untagged value is 42,650 M⁻¹cm⁻¹ (42,400 reduced, which assumes the cystines are not formed). Strep-tag II contains a tryptophan and adds exactly the 5,500 one Trp contributes; HA adds 4,470 for its three tyrosines; FLAG adds 1,490 for its single tyrosine too, and 3×FLAG adds three times that, 4,470, because the tag is three copies of DYKDDDDK and so three tyrosines.
His6, c-Myc and V5 contain none of the three, so they leave the coefficient alone — an important list to get right, because c-Myc contains no Trp, Tyr or Cys despite being ten residues long. That matters because the coefficient is what turns an A280 reading into a concentration: using the untagged value on a Strep-tagged protein overstates the concentration by about 13%.
Large tags: the mass comes from the table, not the sequence
GST is 218 residues and about 26 kDa, MBP 396 residues and 43.4 kDa (the precursor, signal peptide included), SUMO 98 residues and 11 kDa, HaloTag 297 and 33, SNAP-tag 182 and 20, NusA 495 and 55, thioredoxin 109 and 11.7. On the same 26.92 kDa model, the fusion sizes are 52.92, 70.32, 37.92, 59.92, 46.92, 81.92 and 38.62 kDa.
There is a practical difference in how those numbers are obtained. For a six-to-twenty-two residue tag the full sequence is carried and the mass is computed exactly. For the large tags the reference table carries length and mass but not a full residue sequence, so the addition is the table's mass rather than a computed one — which is the right way round, because a 396-residue MBP is a folded domain whose mass is a published property of that protein, not something to re-derive from a partial sequence.
That also means a large-tag fusion's pI should not be estimated by appending residues to your protein. A folded domain's surface charge is not the sum of its residues' pKas, and the table does not claim otherwise.
Cleavage changes the arithmetic again
A cleavable tag leaves residues behind, and the scar has a mass. TEV protease cuts ENLYFQ↓G, so a target with an N-terminal TEV site keeps the glycine: the tag leaves and one residue stays. Enterokinase cuts after DDDDK in FLAG, which is designed to leave a native N-terminus when the tag is N-terminal. SUMO protease cuts after the C-terminal di-glycine of the SUMO fold and likewise leaves the target's own N-terminus.
So there are three masses worth computing for a construct, not one: tagged, cleaved, and the scar. A gel that resolves a 1 kDa difference — a high-percentage gel, or mass spectrometry — will show which of the three you actually have, and that is often the fastest evidence that cleavage worked.
FAQ
How much does a His6 tag add to molecular weight?
822.85 Da, or 0.82 kDa — six histidines plus one water. It is small enough that it barely moves a band on a standard gel, but it can shift the pI by more than a unit because histidine titrates near pH 6.
How big is a GST tag?
218 amino acids, about 26 kDa. Fused to a 26.92 kDa protein the construct is 52.92 kDa. MBP is 396 residues and 43.4 kDa, and SUMO is 98 residues and 11 kDa.
Does adding a tag change the extinction coefficient?
Only if the tag contains Trp, Tyr or cystine. Strep-tag II has a tryptophan and adds 5,500 M⁻¹cm⁻¹; HA adds 4,470 for its three tyrosines and FLAG 1,490 for its one, so 3×FLAG adds 4,470. His6, c-Myc and V5 contain none of the three and leave it unchanged.
Why does my tagged protein run higher than the calculated mass?
Usually because the calculation is of the sequence and the gel is of the protein. Glycosylation, phosphorylation, bound detergent and incomplete denaturation all add apparent mass, and a histidine-rich or proline-rich region can retard migration directly. Check the cleaved sample against the tagged one before assuming the tag is present twice.
Related references
Common affinity, epitope and solubility protein tags with amino acid sequences, mass, purification/detection and protease cleavage.
One-letter and three-letter amino acid codes with key properties.
Side-chain and terminal pKa values and how they set the isoelectric point.
Related tools
Molecular weight, isoelectric point, extinction coefficient and composition.
In silico digestion with trypsin, Lys-C, chymotrypsin and others, with peptide masses.
Back-translation of proteins using most-frequent or degenerate IUPAC codons.
Domains, GO terms, transmembrane helices and signal peptides via EBI InterProScan.