What the COA for lactase actually contains
A lactase certificate is short: seven release parameters, one activity figure and a lot code that ties the page to a pack. It is a report of measurements, not a list of promises, so the ALU row has to name the substrate, the pH, the temperature and the time, or nobody can check the number. The values below are the catalog specification; a batch COA reports what one lot actually measured.
One ALU is defined by the assay that measures it, so the certificate has to say which assay produced the figure in ALU/g.
Seven rows, and what each one decides
Six rows are release checks that a lot passes or fails; the seventh, appearance, describes handling rather than identity.
| Parameter | Method | Release limit |
|---|---|---|
| Appearance | Visual | White to light brown, free-flowing powder |
| Lactase activity | Chromogenic assay on ONPG, ALU/g | ≥100,000 ALU/g |
| pH (1% solution) | Potentiometry | 3.0–8.0 |
| Loss on drying | Gravimetric | ≤8.0% |
| Particle size | Sieve analysis | ≥95% through 40 mesh |
| Heavy metals | ICP-MS | Pb ≤3 · As ≤2 · Cd ≤1 · Hg ≤0.1 mg/kg |
| Microbiology | Plate count and absence tests | APC ≤10,000 cfu/g; yeast and mould ≤100 cfu/g; E. coli and Salmonella absent |
Two rows deserve a second look. The pH window of 3.0–8.0 is broad enough that most powders, carrier included, fall inside it, so it catches gross error rather than describing behaviour. The microbiological row pairs a count with two absence tests, and an absence result is only as good as the enrichment behind it.
Where the ALU/g number comes from
Activity is the only row that needs a definition, because the unit exists only inside a procedure. The assay dilutes a weighed sample into a range where the reaction is linear, adds o-nitrophenyl-β-D-galactopyranoside buffered at pH 4.5, holds it at 37 °C, stops the reaction and reads the yellow o-nitrophenol near 420 nm. Absorbance goes through a standard curve to micromoles, and the result is divided by incubation minutes and by the grams of powder in the tube.
| Element | Typical entry | Consequence if it is missing |
|---|---|---|
| Unit | ALU per gram of preparation as sold | A figure per gram of protein would mean something else |
| Substrate | ONPG at a stated molarity | A lactose-based assay cannot be compared with it |
| pH | 4.5 | Activity at pH 6 is a different number |
| Temperature | 37 °C | Rate changes with temperature, and sharply near unfolding |
| Time | 15 minutes | Only part of the reaction course is linear |
| Standard | o-Nitrophenol, micromolar | The curve sets the scale of the whole result |
| Basis | As-is, or dry-solids corrected | Moisture shifts the per-gram figure |
The number is not a purity: a gram of powder holds enzyme, fermentation residues, carrier and water, and the assay counts only the enzyme. Activity per milligram of protein is the way to see past the carrier, as the identity page sets out.
A floor is not a result
The release limit is written as ≥100,000 ALU/g, which means the assay has one boundary: below it the lot fails, above it the lot is released. A powder at 102,000 ALU/g and one at 165,000 ALU/g would both pass the same specification, which cannot tell a formulator which of the two is in the drum.
That is why the measured figure matters. It shows how much activity the lot carries beyond the floor, which is the margin available for a process that will consume some of it. A result column that repeats ≥100,000, or a single word "conforms", leaves the buyer with a release decision and no data.
Three lines belong beside it: the release date, the storage condition assumed, and whether the figure is guaranteed at release or at end of shelf life. Activity declines slowly in a sealed pack, so a lot shipped late in its life arrives below the number on the page. Add the assay's repeatability: where the relative standard deviation is wide, a lot a few percent above the floor may fail another laboratory's retest.
Why two lots carry different figures
Fermentation is a biological process, and the amount of β-galactosidase secreted by Aspergillus oryzae moves with the strain, the medium, the induction and the point of harvest. Downstream, recovery and ultrafiltration decide how much of that protein reaches the concentrate, and blending with carrier sets the activity per gram. When a supplier promises ≥100,000 ALU/g, the promise is about where in that sequence the final blend lands.
Drying and milling then act on the figure: more residual water lowers the units per gram even when the catalytic capacity is untouched, and heat from a mill can destroy activity before the sieve is reached. Colour tracks the same history and is the least quantitative signal on the sheet; a darker lot may have met every limit.
The method is the last variable. Change the substrate lot, the buffer strength, the incubation time or the standard curve, and two certificates headed ALU/g stop being comparable; such drift shows as a shift in the mean across lots, not as one outlier.
The independent assay, and when to buy one
Four occasions justify the fee: a new source, a disputed result, any claim that rests on the activity figure, and a first shipment into a market where the buyer has no history. The panel is short: protein, moisture, microbiology, heavy metals by ICP-MS, a protein pattern by gel, and activity run by the buyer's own method.
For the comparison to mean anything, the laboratory has to run the specification's substrate and conditions, not its own house method. Metal results need a detection limit beside each "not detected", because the limit describes the method and not the powder. The sample tested should be the retained sample, and the lot code should appear on the test request as well as on the report.
Two quotations, six lines
| Line | What to ask for |
|---|---|
| Unit and definition | ALU/g with the written definition and the method reference |
| Assay conditions | Substrate concentration, pH, temperature, incubation time |
| Measured activity | The release result for the lot, not only the limit |
| Protein | Protein content and ALU per milligram of protein |
| Physical data | Loss on drying and the sieve fraction |
| Carrier | Identity and allergen status of the standardising agent |
Offers that match on ALU/g and diverge on the rest are not the same material, and the gap appears at the first blend trial. The paper trail matters as much: one lot code across certificate, pack and retained sample, and a named organism and strain, since the source is part of what a food enzyme is.
The catalog figures are listed on the lactase product page; the difference between ALU and an FCC lactase unit is set out on the unit comparison page; the release sequence behind the certificate is described on the quality page; and the terms are defined in the COA glossary entry.
Related documents
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