What a Certificate of Analysis Tells You

1. What a CoA is — and what it isn't

A Certificate of Analysis is a written record of analytical tests performed on a specific lot of material against a pre-defined specification. It documents what was tested, by what method, and what result was obtained. It is the primary evidence that a given vial of peptide meets the manufacturer's claimed quality attributes.

What a CoA is not:

• Not a regulatory approval. A CoA does not establish that a compound is approved by any regulatory authority for human or veterinary use.
• Not a guarantee of biological activity. Purity and identity can both be acceptable while a peptide is biologically inactive due to misfolding, aggregation, or improper post-translational modification.
• Not a substitute for in-house qualification. Critical assays may warrant orthogonal verification of identity or activity, particularly for new lots or new suppliers.
• Not transferable across lots. Every lot has its own analytical record. Using one lot's CoA to characterise a different lot is invalid.

Read every CoA against the written specification it claims to meet. A CoA without a clear specification — or with internally inconsistent results — is a quality-system failure, not a benign formatting issue.

2. Appearance, solubility, and physical attributes

Most CoAs begin with a description of the lyophilised material's appearance: 'white to off-white powder' or 'white lyophilised cake' are typical entries. Discoloration (yellow, pink, brown) is a flag for oxidative degradation, residual reagent contamination, or, in extreme cases, identity error. Off-spec appearance should be reported to the supplier before use.

Solubility is sometimes stated as a written claim (e.g., 'soluble in water at ≥10 mg/mL'). Solubility is sequence-dependent and can change with the salt form and counter-ion fraction; the laboratory's experience reconstituting a specific peptide may diverge from the supplier's stated value.

3. HPLC purity

High-performance liquid chromatography (HPLC) is the principal method for quantifying peptide purity. A reversed-phase HPLC trace separates the target peptide from synthesis-related impurities (incomplete couplings, deletion sequences, aggregates) and reports each peak as an area-percent of the total.

What to look for

• Purity value (% area at the detection wavelength). Research-grade peptides are commonly specified at ≥98.0% area. Lower thresholds (95–97%) may be acceptable for certain applications but should be matched to the experimental requirement. (see USP General Chapters and current peptide-CMC industry guidance)
• Detection wavelength. 220 nm (peptide-bond absorbance) is standard. 280 nm (aromatic residues) gives different relative responses and is appropriate only for sequences containing tryptophan or tyrosine.
• Method conditions. Column chemistry, gradient, and run time should be documented or method-coded. Different gradients can resolve or hide different impurities.
• Single dominant peak. The chromatogram should show a clearly resolved main peak. Tailing, shoulders, or unresolved doublets indicate either an isomeric impurity or a non-optimised method.

4. Mass spectrometry — identity and confirmation

Mass spectrometry (MS) confirms that the molecule isolated by HPLC has the expected molecular mass. For peptides, electrospray ionisation (ESI) and matrix-assisted laser desorption ionisation (MALDI) are the two routine techniques.

Reading the MS line

• Theoretical mass. Calculated from the sequence, using monoisotopic or average atomic weights. The CoA should state which.
• Observed mass. The mass actually detected. For ESI on multi-charge ions, this is the deconvoluted mass; for MALDI, typically [M+H]⁺.
• Tolerance. Observed should match theoretical within an acceptable window — typically ±0.5 Da for standard-resolution MS, smaller for high-resolution instruments.
• Adducts and modifications. Sodium and potassium adducts (+22, +38) and oxidation peaks (+16, methionine sulfoxide) are common but should not dominate.

Identity confirmation is non-substitutable. A peptide that is 99% pure by HPLC but the wrong sequence is 100% useless.

5. Counter-ion and salt-form determination

Synthetic peptides are isolated as salts. The most common counter-ions are trifluoroacetate (TFA), acetate, and hydrochloride. The counter-ion is part of the dry mass of the peptide and meaningfully affects the mass-to-active-content ratio.

• TFA salt. Standard byproduct of Fmoc SPPS purification with TFA-containing mobile phases. Can be 5–15% of dry mass for highly basic sequences. TFA can be cytotoxic in cell-culture applications at high concentrations and may interfere with some assays; switching to acetate by anion-exchange is a common downstream step.
• Acetate salt. Generally preferred for biological work because of better assay compatibility. Often produced by lyophilisation from dilute acetic acid.
• HCl salt. Less common for synthetic peptides; more typical of recombinant material.

The CoA should state both the salt form and either the counter-ion content (% by mass) or the peptide content (% peptide free base). Without this, mg-to-mole conversions are systematically biased.

6. Water content (Karl Fischer)

Lyophilised peptide retains a small fraction of bound water. Karl Fischer titration measures this directly and is reported as a percentage of dry mass. Loss-on-drying is an alternative method but less specific for water vs other volatiles.

Typical water content for well-lyophilised peptides is in the 1–8% range. Higher water content reduces the effective per-mg active fraction and accelerates several degradation pathways during long-term storage.

7. Endotoxin and microbial attributes

Endotoxin (lipopolysaccharide) is a product of gram-negative bacteria and is the most relevant microbial attribute for peptides intended for cell-culture or animal research. Even sterile-filtered material can carry endotoxin if upstream water or buffers were contaminated.

• LAL test. Limulus amebocyte lysate testing per USP <85> is the standard quantitative method. Results are typically reported in EU/mg.
• Acceptable limits. Limits depend on the planned application. For sensitive cell-culture work, levels below 1 EU/mg are typically targeted; for in vivo rodent work, route- and dose-dependent thresholds apply.
• Bioburden. Total microbial count is sometimes reported separately. A 'pass' on bioburden does not establish endotoxin status; both should be specified for biological-research material.

8. Residual solvents

Synthesis and purification leave trace solvents that can affect downstream assays and, in some cases, biological compatibility. Residual-solvent testing — typically by gas chromatography — is graded against ICH Q3C limits, which classify solvents by toxicity (Class 1: avoid; Class 2: limited; Class 3: low toxic potential).

For research-grade peptides, common residual-solvent entries include acetonitrile, methanol, ethanol, and dichloromethane traces. The CoA should report each solvent against its ICH limit. (see ICH Q3C(R8) Guideline for Residual Solvents (ICH.org))

9. Red flags and what to ask the supplier

• No lot number on the CoA. Without a lot number, the document is not lot-specific and is not a real CoA.
• Generic or template-style language. 'Meets specification' without numerical results is not a quality record. Demand the underlying data.
• Missing detection wavelength on HPLC. Purity is wavelength-dependent; 220 nm should be the default and should be stated.
• No mass-spectrometry identity confirmation. HPLC purity without MS identity is incomplete; you do not know that the major peak is the intended sequence.
• Counter-ion not specified. Salt form affects active-content calculations and assay compatibility.
• Date of analysis older than the stated stability window. An expired CoA is a procurement question, not a quality finding, but should be flagged.
• Inconsistent units or rounding. Quality records should be internally consistent. Disagreement between sections is usually a transcription error but warrants a fresh document.

Reputable suppliers will provide raw chromatograms and mass spectra on request. The willingness to share underlying data is itself a quality signal.

Comparing CoAs across suppliers

Apparent purity values from different suppliers are often not directly comparable because of differences in HPLC method, column chemistry, and gradient profile. A 99.0% peak in a fast 15-minute gradient may resolve into a 97.5% peak in a longer gradient that pulls a previously co-eluting impurity into a separate peak. When evaluating two suppliers' CoAs side-by-side, the most informative attributes are not the headline purity numbers but the method documentation, the orthogonal-method confirmation (where present), and the granularity of the reported impurity profile. A supplier that reports each major impurity at >0.1% area, with retention-time-relative identification, is providing substantially more usable data than one that reports a single 'purity 99%' value.

Mass-spectrometry sections are similarly variable in informational density. A high-quality MS section reports the theoretical monoisotopic mass, the observed mass, the deconvoluted charge envelope (for ESI), and the absolute mass error. A weaker section reports only an observed mass without theoretical reference, leaving the reader to perform identity confirmation themselves.

10. Tying the CoA to laboratory practice

The CoA is the entry point to several practical decisions. Three are particularly common.

• Per-mg dose calculations. Use the peptide content (free-base fraction) from the CoA, not the labelled vial mass. For a vial labelled 10 mg with a CoA peptide content of 87%, the effective active mass is 8.7 mg. Failure to apply this correction introduces a systematic 13% overestimation of stock concentration. See the worked examples in the handling pillar guide.
• Salt-form switching. If the CoA reports a TFA salt and the assay requires acetate, plan an anion-exchange step or order from a supplier that ships the acetate form natively. TFA traces above approximately 50 µg/mL can affect sensitive cell-culture readouts. (see PubMed; in vitro reference pending verification)
• Endotoxin gating. Decide before reconstitution whether the planned use is purely chemical (HPLC, mass spec) or biological (cells, animals). For biological use, gate by the CoA endotoxin number; for in vivo work, additionally apply route- and dose-specific limits.

None of these decisions require interpreting the CoA on the day of use; they are setup decisions that should be made when the lot is received. A CoA filed against the wrong lot, or filed without being read, is a paperwork artefact rather than a quality control.

References

This guide cites primary peer-reviewed and regulatory sources where the underlying claim has been verified. A small number of supporting items remain in active verification and are listed below for transparency.

Still seeking primary citation

• Research-grade HPLC purity threshold (≥98% area at 220 nm) — USP / industry reference
• ICH Q3C(R8) Guideline for Residual Solvents (ICH.org)
• TFA cytotoxicity threshold in cell culture — primary in vitro reference