How to read a peptide certificate of analysis

Why the certificate matters

A lyophilised peptide in a vial tells you nothing by looking at it. Powder of 99% purity, powder of 70% purity and a different peptide altogether look identical. The certificate of analysis (CoA) is the only document that connects the vial on your bench to measurements made on the batch it came from, and reading it takes about five minutes once you know what each section is for.

This guide walks through a typical certificate for a synthetic research peptide in the order the sections usually appear. The worked example uses figures of the kind you would see for a lipidated incretin peptide such as tirzepatide. The methods themselves are explained in the companion guide, HPLC and LC-MS explained; this page concentrates on the document. It is written for material supplied for in-vitro research, not pharmaceutical release, although the pharmacopoeial and ICH documents cited at the end are where the underlying expectations come from.

The header

The top of the certificate identifies what was tested, by whom, and when. It is the least technical part and the most often skimmed, which is a mistake, because most of the serious problems with a certificate show up here.

Laboratory name and address. The report should name the laboratory that did the work, with a physical address. If it is independent, that laboratory's own name should appear, not just the supplier's.

Report reference. A unique job number assigned by the laboratory, which you quote if you want to confirm the report with them directly.

Sample identification. The sample name as received, which should be the compound name rather than a brand or product code alone, and usually the laboratory's own internal ID.

Batch or lot number. The manufacturing batch the sample came from, which should match the vial label. A certificate without a batch number cannot be tied to the vial in your hand.

Dates. The date the sample was received and the date the report was issued. The gap is usually days to a couple of weeks. A certificate dated before the batch was made is an impossibility, and it happens more often than you might expect with reused templates.

Requested by. Who submitted the sample. On an independent report this is normally the supplier.

Identity

The identity section answers one question: is this the peptide it is supposed to be? The standard evidence is a molecular mass measured by mass spectrometry, almost always liquid chromatography coupled to electrospray mass spectrometry (LC-MS), occasionally MALDI-TOF.

Expected mass. The theoretical mass calculated from the molecular formula. The certificate should state whether this is the average or the monoisotopic mass; for a peptide of 4,000 to 5,000 Da the two differ by about 3 Da, so a mismatch in convention can look like an error when it is not. For tirzepatide the average molecular weight is 4813.45 g/mol.

Observed mass. What the instrument measured. Electrospray produces a series of multiply charged ions, so the raw result is a set of mass-to-charge (m/z) values, each corresponding to the same molecule carrying a different number of protons. A certificate may report the deconvoluted mass, where software has combined the charge states into a single neutral mass that can be compared directly with the expected value; one or more charge-state m/z values, such as the 4+ ion at m/z 1204.4, which you convert yourself (the arithmetic is in the companion guide); or both.

Acceptable tolerance. On a routine instrument, a deconvoluted mass within about 0.5 to 1 Da of the theoretical average mass is normal for a peptide of this size; high-resolution instruments may quote a monoisotopic mass agreeing to within 0.05 Da. A difference of 16 Da suggests oxidation, 1 Da deamidation, and tens or hundreds of daltons a missing or extra residue, in which case the sample is not what it claims to be regardless of the purity figure. What a mass does not tell you is sequence: two peptides with the same residues in a different order weigh the same, and sequence confirmation by tandem MS is rarely performed for research chemicals.

Purity

Purity is the figure quoted on product pages, and the section that deserves the most careful reading.

Method. Reversed-phase HPLC (RP-HPLC) is the standard. The certificate should state the column (usually a C18, with dimensions and particle size), the mobile phases (typically water and acetonitrile, each with 0.1% trifluoroacetic acid), the gradient, the flow rate and the detection wavelength (214 or 220 nm, because the peptide bond absorbs there). A certificate that gives none of this is reporting a number without a method.

The chromatogram. A good certificate includes the chromatogram, a plot of detector response against time, with the main peptide as one dominant peak. Note the retention time of the main peak and whether it matches earlier batches on the same method. Look at the baseline either side of it; small peaks close to the main peak, particularly on the leading edge, are where closely related impurities sit. Check the integration table, which lists each peak with its retention time, area and area percentage; purity is the main-peak area divided by the total integrated area. And look at the scale: a plot in which the main peak fills the frame hides small impurities, so a second plot with the y-axis expanded is good practice.

Purity figure. For a research peptide, 98% or above by area is typical of good synthesis and purification, and 99% or above is common. The number should be given to two decimal places with the method it came from. "Purity: 99%" with no chromatogram and no method is a claim, not a result.

Common impurities. The main families, as described in the peptide impurity literature and the FDA's guidance on synthetic peptides, are:

• Deletion and truncation sequences, missing one or more residues because a coupling step in solid-phase synthesis did not go to completion. They often elute close to the parent and are the impurities most likely to hide under the main peak.
• Oxidation. Methionine oxidises to the sulfoxide; tryptophan and cysteine are also susceptible. Oxidised species elute slightly earlier and add 16 Da in the mass spectrum.
• Deamidation. Asparagine and glutamine lose ammonia to form aspartate or glutamate (or the isoaspartate isomer), adding about 1 Da. This is also a storage degradation route, so it can grow over time.
• Diastereomers. Racemisation of one residue during synthesis gives a peptide with the same mass and sequence but one inverted stereocentre. These are invisible to MS and may or may not separate on HPLC.
• Protecting-group remnants, such as residual t-butyl groups, which add a fixed mass and elute later.

Counter-ion and water. Trifluoroacetic acid from purification remains in the lyophilised product as the trifluoroacetate salt, and the powder also holds bound water. Neither interferes at 214 nm, so neither reduces the area-percentage purity; they do reduce how much of the powder is actually peptide, which is the next section. Some certificates report TFA (by ion chromatography) and water (by Karl Fischer titration) separately.

Net peptide content

The purity figure says what fraction of the UV-absorbing material is the right peptide. It does not say what fraction of the powder by mass is peptide. A vial labelled 10 mg contains 10 mg of lyophilised solid, and that solid is peptide plus counter-ions plus water; trifluoroacetate alone can account for 10% to 20% of the mass, and water a further 3% to 8%.

Net peptide content (sometimes "peptide content" or "assay") expresses the actual peptide fraction, typically 70% to 90%. It is measured by amino acid analysis, in which the peptide is hydrolysed to its constituent amino acids and these are quantified against standards, or by UV absorbance at 280 nm where the sequence contains tryptophan or tyrosine. A researcher preparing a 1 mg/mL solution from a vial with 80% net peptide content who ignores that figure actually makes 0.8 mg/mL. If net peptide content is not reported, treat the gross mass as an upper bound on the peptide mass.

Endotoxin

Endotoxin is lipopolysaccharide from the outer membrane of Gram-negative bacteria. It is not removed by HPLC purification, survives lyophilisation, and is biologically active at very low concentrations, so it can confound cell-based assays even when the peptide itself is pure. The standard test is the limulus amoebocyte lysate (LAL) assay, described in USP general chapter 85 and European Pharmacopoeia chapter 2.6.14, with recombinant Factor C now accepted as an alternative in both. Results are reported in endotoxin units per milligram of peptide (EU/mg), or as less than a detection limit, for example "<1.0 EU/mg". Endotoxin testing is not performed on every research batch and its absence is not in itself a red flag; where your assay is sensitive to it, look for a batch that has the result or test the material yourself.

Sterility

Research peptides are not sterile products, and a certificate for one will normally not include a sterility test. The manufacturing process is not conducted under the aseptic conditions needed to support a sterility claim, and a sterility test under USP general chapter 71 or Ph. Eur. 2.6.1 is rarely applied to research chemicals. A certificate that claims "sterile" without naming the method and standard should be read with suspicion rather than relief. Where sterility is needed for a cell-culture experiment, it is achieved by filtration at the point of use, not inherited from the vial.

Who signed it, and what kind of certificate it is

The foot of the certificate carries the name and title of the analyst and usually a reviewer, with a signature and date. A report with no named person behind it is weaker evidence than one with.

Two kinds of certificate are in circulation. A manufacturer's certificate of analysis is produced by the facility that made the peptide from tests on its own product; it is the primary document for the batch, often the most complete, but self-reported. An independent third-party report is produced by a separate laboratory that received a sample of the finished product, usually commissioned by the supplier, and tested it with its own methods. It is a check from someone with no stake in the result, though it tests only what was sent. The strongest position is to have both and see that they agree. Every batch in our certificate register carries an independent report, and the product pages quote the figures from it.

A worked example

The table below shows the kind of results you would expect on a sound certificate for a batch of research-grade tirzepatide. The figures are illustrative.

Notice that nothing in this table is a round number apart from the ones that should be.

Red flags

The following should make you stop and ask questions, or put the certificate down altogether.

• No batch number, or one that does not match the vial. The certificate cannot be linked to your material.
• No laboratory name, or only the supplier's logo with no indication of who did the work.
• Purity stated without a chromatogram or a method. A number with nothing behind it.
• A template certificate. The same document appearing for different batches or compounds with only the name changed; look for identical retention times, impurity profiles and dates across batches that should differ.
• Results that are too round. "Purity: 99%" and "Mass: 4813" rather than 99.41% and 4813.6. Instruments do not produce round numbers; people filling in forms do.
• An expected mass that is wrong, or an observed mass off by tens of daltons. The first means nobody checked; the second means a different peptide.
• Claims of sterility or pharmaceutical grade on a research chemical with no supporting method.
• A reported date before the batch date, or dates implausibly close together for the tests claimed.
• A chromatogram with no scale, time axis or integration table. A picture of a peak is not a result.

None of these on its own proves anything is wrong with the peptide. Together they tell you how much weight the certificate can bear, which is the point of reading it.

Finally, a certificate describes a batch at the time of testing, not a vial after a week in a warm courier van or six months at the back of a refrigerator; the storage and stability guide covers how to keep material in the state the certificate describes. And a certificate for a research chemical says nothing about suitability for any use other than laboratory research; the material is supplied for in-vitro work only and is not for human or veterinary administration.

References

1. International Council for Harmonisation. ICH Q6B: Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products. Step 4, March 1999. EMA CPMP/ICH/365/96.
2. International Council for Harmonisation. ICH Q2(R2): Validation of Analytical Procedures. Step 4, November 2023. EMA/CHMP/ICH/82072/2006.
3. U.S. Food and Drug Administration. ANDAs for Certain Highly Purified Synthetic Peptide Drug Products That Refer to Listed Drugs of rDNA Origin: Guidance for Industry. May 2021. fda.gov/media/107622/download.
4. D'Hondt M, Bracke N, Taevernier L, et al. Related impurities in peptide medicines. Journal of Pharmaceutical and Biomedical Analysis. 2014;101:2–30. doi:10.1016/j.jpba.2014.06.012.
5. United States Pharmacopeia. General chapter <85> Bacterial Endotoxins Test. USP–NF. Rockville, MD: United States Pharmacopeial Convention.
6. European Pharmacopoeia. Chapter 2.6.14 Bacterial endotoxins. Strasbourg: European Directorate for the Quality of Medicines and HealthCare (EDQM).

This monograph summarises the published literature for laboratory researchers. It is not medical advice, and nothing on this site is for human or veterinary use.