Net peptide content vs HPLC purity: how amino acid analysis and mass balance measure the peptide in the vial

Three numbers describe how much peptide you have, and they are routinely confused. The gross weight is what the balance reads. The HPLC purity is what the chromatogram reports. The net peptide content is how much of that weight is actually the peptide molecule. For a lyophilized peptide salt, the third number can be 20–30 points below the second, and every concentration calculated from the gross weight inherits the error.

This guide explains how net peptide content is measured, from amino acid analysis to nitrogen analysis, qNMR and the mass balance approach USP uses for its reference standards, with a worked example of how a 99% purity result can still mean 70–80% peptide by weight. It goes deeper than the short net-content section in our peptide purity testing guide.

Gross weight, HPLC purity and net peptide content: three different numbers

Term	What it measures	Typical method	What it ignores
Gross weight	Total mass of powder weighed	Balance	Everything about composition
HPLC purity	Main peak area as % of total detected peak area	RP-HPLC with UV at 210–220 nm	Water, counter-ions, solvents, salts; anything not detected at that wavelength
Net peptide content	Mass of peptide as % of total mass	Amino acid analysis, nitrogen or CHN analysis, qNMR, mass balance	Nothing, if done properly; it is the mass fraction you weigh

Purity is a ratio within the peptide-related material. Content is a ratio of peptide to everything. A related-substances method that gives 99.0% purity says that 1% of the UV-detected peptide material is impurities. It says nothing about the 20% of the vial that may be counter-ion and water.

What else is in a peptide powder?

USP’s mass balance scheme for peptide reference standards lists the components that have to be measured besides the peptide: peptide-related impurities by HPLC, counter-ion (for example acetic acid), residual reagent such as trifluoroacetic acid, residual solvents, inorganic residue (residue on ignition or thermogravimetric analysis) and water by Karl Fischer titration [1].

• Counter-ions. Peptides purified with TFA “are obtained as trifluoroacetate salts”, while “most of the peptide drugs and peptide-based therapeutics are provided as acetate salts” [2]. Ion chromatography “is a technique of the first choice for counter-ion determination” [2].
• Water. EMA notes that “peptides are often very hygroscopic powders” [3] and expects water content to be “part of the stability protocols” for hygroscopic powders [3].
• Residual solvents and inorganics. Measured by headspace GC and by residue on ignition or ICP-MS, depending on the laboratory [1], [4].

Counter-ions are not just inert weight in biological work. In fetal rat osteoblast cultures, TFA at 10⁻⁸ to 10⁻⁷ M reduced cell numbers, and TFA salts of amylin and calcitonin consistently gave less proliferation than hydrochloride salts, “resulting in failure to detect a proliferative effect or wrongly attributing an antiproliferative effect” [5]. That was an in vitro finding, but it is a reason to know which salt you have as well as how much.

How mass balance works

Mass balance assigns content by measuring everything that is not the peptide and taking it away. USP’s paper on peptide reference standards gives the equation and a theoretical worked example [1]:

Purity (mg/mg) = [(100 − Σ HPLC impurities, % area) ÷ 100] × [(100 − Σ non-peptide components, % w/w) ÷ 100]

In USP’s example, HPLC impurities summed to 0.92% of total detected area, acetic acid was 5.58% w/w, trifluoroacetic acid 0.003%, residual solvents 0.003% and residue on ignition 0.00%. The result was 0.99 × 0.94 = 0.93 mg of peptide (free base) per mg of material on the anhydrous basis [1]. Adjusting for each laboratory’s measured water content of 1.5% and 2% gave water-corrected values of 0.92 and 0.91 mg/mg [1]. USP measures water on three independent weighings at the time of the HPLC assay, because humidity during testing changes it [1].

Two details in that equation are easy to miss. HPLC impurities are a percentage of chromatographic area, while the other components are weight percentages, so the two factors are multiplied rather than added [1]. And the result is peptide free base, the molecule without its counter-ion.

USP also notes that “mass balance has been proposed as a benchmark for purity assessment of peptides”, and that in its own programme the mass balance approach gave “the least inter-laboratory variability” [1].

How a peptide reference standard gets its number

Net peptide content matters most for the reference standard itself, because every HPLC assay result is calculated against it. USP’s programme, described by USP scientists with co-authors from the UK’s Medicines and Healthcare products Regulatory Agency for six peptides (bivalirudin, desmopressin acetate, exenatide, gonadorelin, leuprolide and oxytocin), uses a two-step value assignment [1]:

1. Bulk purity by mass balance, as above, with each test run in several laboratories.
2. Content of each lyophilized vial by HPLC assay against the characterised bulk material, corrected for the bulk purity.

The physical data for the lyophilized lots show what a well-controlled fill looks like: residual moisture of 1.11–2.79% w/w and headspace oxygen of 0.57–1.58% across the six peptides, with fill-weight coefficients of variation of 0.20–0.26% [1]. USP measures headspace oxygen because it “is important to measure when working with peptides that contain oxidizable amino acids such as methionine” [1].

Data handling is as deliberate as the chemistry. For the desmopressin example, outliers were screened by residual analysis, one laboratory’s results were removed partly because it “did not perform water content testing”, and laboratories with higher within-laboratory variability were down-weighted by inverse-variance weighting [1]. A content value that skipped the water measurement would not have survived that review.

How low can net peptide content be? Published examples

Two metrology studies show real numbers for well-characterised peptides:

Material	Main non-peptide contributor	Assigned peptide content	Method	Ref
Angiotensin II candidate certified reference material	TFA counter-ion, “nearly 25% by mass”	691 ± 9 mg/g (k = 2), about 69%	qNMR and LC-MS/MS amino acid analysis combined with mass balance; TFA by ¹⁹F-qNMR	[6]
Synthetic oxytocin	Water, counter-ions, inorganics, solvents measured separately	796.5 mg/g, expanded uncertainty 6.5 mg/g (k = 2), about 80%	Mass balance: LC-hrMS for related impurities, coulometric KF, ion chromatography, ICP-MS, headspace GC-MS	[4]
USP theoretical acetate example	Acetic acid 5.58% w/w; water 1.5–2%	0.91–0.92 mg/mg as is	Mass balance	[1]

In the oxytocin material, the 21 structurally related impurities ranged from 0.05 to 15.65 mg/g each [4]. So both the peptide-related and the non-peptide fractions mattered.

Worked example: how 99% purity becomes 70–80% peptide

The figures below are illustrative, chosen to show the arithmetic, and are not results for any product. They use USP’s mass balance formula [1].

	Peptide as TFA salt	Same peptide as acetate salt
HPLC purity	99.0%	99.0%
Counter-ion	18.0% w/w TFA	6.0% w/w acetate
Water (Karl Fischer)	6.0% w/w	3.0% w/w
Residual solvents	0.5% w/w	0.5% w/w
Non-peptide total	24.5%	9.5%
Net peptide content	0.990 × 0.755 = 74.7%	0.990 × 0.905 = 89.6%

The TFA case is not extreme: the published angiotensin II reference material carried nearly 25% TFA on its own [6]. The point of the table is that purity barely moves while content swings by 15 points with the salt form and the water content.

Amino acid analysis: hydrolysis

Amino acid analysis (AAA) answers two questions at once: which amino acids are present in what ratio (composition, an identity check) and how many moles of them are in a weighed sample (content). It “is used to determine the amino acid content of amino acid-, peptide- and protein-containing samples”: the peptide bonds are hydrolysed, and the liberated amino acids “are then separated, detected, and quantified” [7]. The method goes back to Moore, Stein and co-workers in the 1950s, and “the basic methodology remains relatively unchanged” [7].

Fountoulakis and Lahm stress that “a properly performed hydrolysis is a prerequisite of a successful analysis” [8]. Typical conditions and their known weaknesses:

Step or residue	What happens	How it is handled	Ref
Standard acid hydrolysis	6 M HCl containing 1% phenol at 100 °C for 24 h, as used for ganirelix	Default for most residues	[9]
Asparagine, glutamine	Not distinguished from Asp and Glu after hydrolysis; reported as Asx and Glx	Interpret composition accordingly	[10]
Methionine, cysteine	Need their own sample treatment	Performic acid oxidation	[7]
Tryptophan	Needs its own hydrolysis conditions	Base hydrolysis, or 4 N methanesulfonic acid with 3-(2-aminoethyl)indole	[7], [11]
Tyrosine, cysteine, methionine	Partial decomposition	Thioglycolic acid as protectant (Met recovery still inconsistent)	[10]
Stereochemistry	Hydrolysis causes some epimerisation	Hydrolysis in deuterated HCl when chiral purity is measured	[3]

Simpson and colleagues’ methanesulfonic acid method (4 N, 115 °C, 22–72 h in vacuo) gave values “close to the expected theoretical values” for all amino acids including tryptophan and half-cystine from a single hydrolysate [11].

The Asx/Glx point shows up in real data. In USP’s amino acid analysis of bivalirudin (sequence FPRPGGGGNGDFEEIPEEYL, one Asn and one Asp), the theoretical ratio listed for Asp is 2, and the measured value was 2.15 [1]. The same table shows why AAA is still useful for identity: bivalirudin contains one leucine and one isoleucine, and the measured ratios of 0.94 and 0.97 confirmed both [1]. USP notes that AAA has been “gradually replaced” as an identity test by HPLC and MS, but “is still a useful tool when discrimination between Leucine and Isoleucine is required” [1].

Derivatisation and detection

Most amino acids lack a strong chromophore, so they are derivatised before or after separation.

• Phenylisothiocyanate (PITC). Heinrikson and Meredith’s pre-column method forms phenylthiocarbamyl derivatives “in 5 to 10 min at room temperature”, detected at 254 nm, with all amino acids “including proline” converted quantitatively and eluted in about 30 minutes [12].
• AQC (6-aminoquinolyl-N-hydroxysuccinimidyl carbamate). Cohen and Michaud’s reagent forms stable urea derivatives in one step, detected by fluorescence (excitation 250 nm, emission 395 nm), linear over 2.5–200 µM, with detection limits from 40 fmol for phenylalanine to 800 fmol for cystine [13]. Ummiti and Kumar used it to confirm the composition of ganirelix [9].
• o-Phthalaldehyde (OPA). Used for post-column labelling, as in Hashimoto and colleagues’ HPLC method [10].
• Isotope dilution LC-MS/MS. The conventional reference method for metrologically traceable peptide standards, with the caveat that, like qNMR, it “is susceptible to interference from related peptide impurities” unless those are measured and corrected for [6]. Burkitt and colleagues built an SI-traceable chain from amino acid standards of known purity to peptides to proteins this way [14].

How amino acid analysis gives peptide content

The arithmetic is simple. For each reliably recovered residue, the measured moles divided by the number of times that residue occurs in the sequence gives the moles of peptide. Averaging across stable residues, multiplying by the peptide’s molecular weight (free base) and dividing by the weighed sample mass gives the content. Rutherfurd and Dunn’s protocol unit describes these calculations [15].

Precision depends on the laboratory as much as the method. Vemuri compared four methods for thymalfasin and found that “the AAA method is highly variable in one laboratory while quite precise in another laboratory”, that HPLC results depended on the laboratory, and that CHN analysis “yielded highly precise (i.e. <2% coefficient of variation) values” and was recommended for that peptide [16].

Nitrogen, CHN and qNMR: the other content methods

EMA lists the accepted options for assay or content of a synthetic peptide as LC against a reference standard, “elemental analysis, amino acid analysis, nitrogen analysis by Kjeldahl, or qNMR” [3].

• Nitrogen (Kjeldahl) and CHN elemental analysis measure the nitrogen or carbon, hydrogen and nitrogen fraction of the sample and convert it to peptide using the molecular formula. They were among the four methods in Vemuri’s comparison, with CHN the most precise [16]. The conversion assumes the measured element belongs to the peptide, so nitrogen-containing counter-ions or solvents bias the result.
• Quantitative NMR (qNMR) measures the intact molecule against an internal standard. Pauli and colleagues describe it as “nearly universal” in detection and “orthogonal” to chromatography and elemental analysis, and note that absolute qNMR “captures analytes that frequently escape detection (water, sorbents)” [17]. Melanson’s group used ¹H-qNMR for angiotensin II and ¹⁹F-qNMR for its TFA counter-ion [6].

Water by Karl Fischer titration

Karl Fischer titration “is based on the oxidation of sulphur dioxide by iodine with consumption of water” [18]. For lyophilized peptides the sample is small and hygroscopic, so laboratories use coulometric KF and protect the sample from air: in the USP/MHRA reference standard work, residual moisture in vials was determined by coulometric KF with vials opened “within a dry nitrogen environment to minimize moisture ingress” [1]. Bulk water was measured per USP general chapter <921> [1].

Water is the component that changes after release. A powder that met its specification can pick up moisture every time a cold container is opened in humid air, which lowers net peptide content without any chemical change.

Why net peptide content matters for assay stock concentration

If you weigh a peptide powder and treat the weight as pure peptide, every solution you make is weaker than you think by the factor of the net peptide content. With the TFA-salt example above (74.7% content), 1.00 mg of powder dissolved in 1.00 mL of buffer contains 0.747 mg/mL of peptide, not 1.00 mg/mL. Every IC50, EC50 or binding constant derived from that stock is shifted by the same 25%.

The correction for an in vitro assay stock is:

• Mass concentration: weighed mass × net peptide content (as a fraction) ÷ volume.
• Molar concentration: that mass concentration ÷ the molecular weight of the peptide itself (free base), because net peptide content already excludes the counter-ion. Use the average molecular weight here, not the monoisotopic mass; see peptide mass spectrometry for the difference.

When content is unknown, concentration can be measured directly by UV absorbance. At 280 nm, Pace and colleagues’ predictive equation is ε(280) = 5,500 × (number of Trp) + 1,490 × (number of Tyr) + 125 × (number of cystines) M⁻¹ cm⁻¹, reliable for proteins containing Trp, but “the best approach is to measure rather than predict” [19]. Gill and von Hippel’s composition-based calculation was accurate to ±5% in most cases for the proteins they studied [20]. For peptides without Trp or Tyr, absorbance near 214 nm can be used: the peptide bond’s molar extinction coefficient there is 923 M⁻¹ cm⁻¹, with tryptophan about 30 times higher and Phe, Tyr and His about six times higher [21].

These are laboratory calculations for preparing analytical and in vitro stocks. Our guide to reconstituting lyophilized peptides in the lab covers solvent choice and handling.

What to look for on a certificate of analysis

1. Is net peptide content reported, and by which method? AAA, CHN, nitrogen, qNMR or mass balance.
2. On which basis? “As is”, anhydrous, or counter-ion free. EMA notes that LC assay limits are expressed on “the counter-ion free, anhydrous substance, unless otherwise justified” [3].
3. Which counter-ion, and how much? TFA, acetate or chloride, by ion chromatography or NMR [2].
4. Water content by Karl Fischer, with the date of measurement.
5. Does the arithmetic close? HPLC purity × (100 − counter-ion − water − solvents − residue) ÷ 100 should be close to the reported content [1].

Our guide on how to read a peptide COA covers the rest of the certificate.

How Winhycare Pharma supplies research peptides

Winhycare Pharma supplies research peptides from Hyderabad, India, as research-use-only material, at purity of ≥99% as stated in our product brochure. Semaglutide, tirzepatide, retatrutide, cagrilintide and BPC-157 are also supplied as API to licensed manufacturers; see peptide APIs. If your work needs a specific salt form or a net peptide content result, say so when you ask for a quote, or use our custom peptide synthesis service: request research peptides.

For laboratory research use only. Not for human or veterinary use.