06 Identity & adulteration

How we prove an alkaline protease lot has the declared activity

Technical document 06 · Reviewed 2026-11-19 · Alkaline Protease

On this page
  1. The four layers of identity
  2. Units per milligram of protein
  3. What a diffraction scan cannot tell you
  4. The ways a lot can be other than it claims
  5. Writing the specification so it can be checked
  6. Five questions to put to a supplier

Proving an alkaline protease lot has the declared activity takes four layers: what the material is, how the activity was measured, which protein carries it, and which documents tie the numbers to the drum. A U/g figure is one layer, and on its own it cannot separate a genuine 200,000 U/g concentrate from a diluted lot still described at the original activity, or from a preparation assayed against a different substrate. The lot code is what joins the four layers to the pack in the warehouse.

In one line

Identity is the organism, the assay and the protein figure read together, and the lot code is what ties all three to the bag.

The four layers of identity

The first layer is the preparation: a powder from Bacillus licheniformis, CAS 9014-01-1, standardised to a declared activity, with a named carrier. The second is the activity assay, with its substrate, pH, temperature, incubation time and detection principle written down. The third is the protein the activity sits on, because a rate per gram becomes comparable only when it is set against the protein that produces it. The fourth is the paper trail: certificate, lot code on the pack, retained sample and strain statement, all pointing at the same material.

Each layer answers a different question about the same gram of powder, and a weakness in one leaves a gap the other three cannot close. A certificate with a method but no strain statement describes an activity, not a source.

Units per milligram of protein

Protein is measured colorimetrically against a bovine serum albumin standard, and the activity divided by the protein gives specific activity in U per mg. That ratio separates a concentrate from a dilution: dilution lowers the units per gram and the protein per gram together, so the ratio holds while both absolute figures fall. A lot at 200,000 U/g with 2% protein and a lot at 200,000 U/g with 6% protein are not the same material even when the activity line matches.

The pair of numbers also sorts the failure modes. High protein with low activity suggests an enzyme that has been partly inactivated, or non-enzyme protein added as a filler. Low protein with normal activity suggests a more purified preparation, which is a different product and should be described as one. Neither conclusion can be drawn from the activity alone, which is why the COA page asks for both lines on the same certificate.

The carrier matters to the reading. A maltodextrin or salt carrier contributes mass but almost no protein, so a heavily extended lot shows a low protein figure and a low specific activity. That is not fraud by itself — a declared dilution is a legitimate product — but it must be declared, because the buyer is paying for units and receiving carrier.

What a diffraction scan cannot tell you

An enzyme powder is amorphous. An X-ray diffraction scan of it shows a broad halo and no crystalline peaks, so the method can confirm the absence of a crystalline carrier phase but cannot identify the enzyme or measure its activity. A laboratory offering XRD as the identity test for a protease is selling a pattern that will look much the same for any spray-dried or freeze-dried preparation.

Infrared spectroscopy is more informative and still not a release test: the amide I and amide II bands show that the material is protein, and the shape of the amide I band reflects secondary structure, but the spectrum does not name the protease or separate an alkaline from a neutral one.

The pattern method that does work is gel electrophoresis. Denaturing polyacrylamide gels separate the proteins by mass, and a lot can be compared with a reference lot of the same product: the band pattern, the position of the dominant protease band and the absence of unexpected bands are evidence of consistency. It is a comparison, not an absolute identity, and it is strongest when run beside the activity and the strain statement.

The ways a lot can be other than it claims

Failure modes and the test that exposes each
ModeWhat it looks likeWhat catches it
Dilution with carrierNormal powder, lower U/g than the labelActivity and protein both low; specific activity held
Neutral protease substitutedActivity plausible in a generic assaypH–activity profile; activity collapses at pH 10; EDTA inhibition
Another alkaline proteaseSame unit name, different kineticsProtein pattern; strain statement; assay at two pH values
Heat-inactivated blendCorrect colour and flow, low activityActivity at two incubation times; specific activity
Activity boosted by free amino acidsAssay reads high, protein normal or lowBlank correction; dialysis or precipitation before assay
Undeclared carrier changeSpecification sheet unchangedCarrier identity; change-control notice

The last two rows are the ones a buyer rarely expects. The Folin-Ciocalteu reagent reacts with tyrosine and tryptophan, and with reducing substances, so a powder spiked with free amino acids or a reducing agent reads high without carrying more enzyme. A blank and a clean-up step on the sample expose it. The carrier change is quieter: the activity can stay inside the specification while the blend’s flow, dust and label declaration all move.

Writing the specification so it can be checked

A testable specification for this material carries the activity as “not less than” with the unit and method reference, the protein or specific-activity range, loss on drying, particle size, the carrier and its amount, the production organism and strain, the activity basis, and the storage statement.

Two habits make the difference. Define the unit once, in a method appendix, and keep that appendix under change control, so that a method change cannot pass as a specification change. Then state which identity check is applied to every new lot — the protein pattern against a reference lot, the strain statement, or both — so identity is a routine release step rather than a reaction to an odd result. The assay glossary entry covers the vocabulary, and the specification entry covers how the limits are written.

Five questions to put to a supplier

First, which strain produced this lot, and where is it held? Second, what exactly does one U mean in your assay — substrate, pH, temperature, incubation time, detection and tyrosine basis? Third, what are the activity, the protein and the specific activity for this lot? Fourth, what test would show that this powder is not a dilution or another protease, and will you run it? Fifth, which document carries the lot code that matches the pack?

A supplier who answers all five in writing has given a buyer something to audit. One who answers only the third has given a number, and a number is the layer that is easiest to produce and hardest to verify. The catalog figures are on the alkaline protease product page, the comparison with the neutral enzyme is on the alkaline versus neutral page, the grade options are on the grades page, and the release process behind the answers is described on the quality page.

Send the specification you intend to use — grade, quantity and destination market — and we will reply with the batch COA, the technical data sheet and a quotation.

Request a quotation