Microbiological testing for mineral ingredients
Microbiological testing counts the organisms that grow from a sample and confirms whether a named pathogen is present. The standard panel for a mineral powder is an aerobic plate count, yeast and mould, Escherichia coli and Salmonella. The result is a count or a presence statement, and it should be read as one.
In short
Enumeration and presence/absence tests that report colony counts, not the identity of everything in the sample.
How the samples are tested
A weighed portion, typically 10 g or 25 g, is homogenised in a buffered diluent and plated in serial dilution on a general-purpose medium. The aerobic plate count is incubated at 30–35 °C for 48–72 hours or the period stated in the method, and the yeast and mould count at 20–25 °C for five to seven days, following the compendial chapters or the equivalent ISO standards. Plates in the countable range are read, and the result is expressed as colony-forming units per gram.
Presence or absence tests follow a different logic: an enrichment step lets a small number of stressed cells recover and multiply, and selective media and confirmation tests then establish whether the named organism was present in the tested mass. Salmonella, Escherichia coli and, where specified, Staphylococcus aureus or Pseudomonas aeruginosa are handled this way. The result is reported as detected or not detected in 10 g, not as a count.
Method suitability, negative controls and a positive control organism are run with each batch of samples, because a medium that fails to grow a stressed isolate produces a false pass. Analyst technique, incubation temperature and the age of the media all affect the count. The laboratory's accreditation scope and the method version should appear on the report.
What the result proves
The test reports the number of viable organisms that grow under the specified conditions, or the presence of a named organism in a stated mass of sample. A lot below the specification limit is consistent with that limit at the time of sampling. That is a statement about the sample, and it is the statement a release decision is built on.
Trend data across lots is often more useful than a single pass. A count that rises from a few hundred to near the limit, or a species that appears repeatedly, points to a cleaning, drying or storage problem before the specification is breached. That is a supplier-monitoring signal rather than a release criterion.
The standard panel covers the organisms a supplement specification normally names. When a buyer adds a specific organism, the method has to be validated for that matrix, because a generic compendial method may not recover it from a powder with antimicrobial properties. The validation belongs in the supplier file, not in the footnotes.
What it cannot tell you
A count is only a count. The aerobic plate count enumerates mesophilic organisms that grow on the chosen medium under the chosen conditions; it does not identify them, and it does not detect viable-but-non-culturable cells, ungerminated spores or endotoxins. A low count is not a sterility result.
Sampling statistics limit what absence means. Not detected in 10 g does not establish absence in a 25 kg drum, and a low count in one sample can coexist with a contaminated pocket. The result is a probability statement about the lot, and the sampling plan is part of its interpretation.
Injured cells, clumping, residual antimicrobial compounds in the matrix and incubation conditions can all suppress growth and produce a false low result. A passing microbiological panel also says nothing about heavy metals, residual solvents, identity or the chemical form of the magnesium. Those are separate methods with separate limits.
What it proves
- An aerobic plate count, reported as colony-forming units per gram.
- Yeast and mould counts under the specified incubation conditions.
- Presence or absence of named organisms such as E. coli and Salmonella in a stated mass.
- Whether a lot is consistent with the microbiological limits in its specification.
What it does not prove
- The identity of the organisms counted, beyond the organisms specifically tested for.
- Viable-but-non-culturable cells, ungerminated spores and endotoxins.
- Lot homogeneity, because the result covers the sample that was drawn.
- Chemical contaminants, identity or the form of the magnesium compound.
What to ask the laboratory
- What sample mass was tested, and how many units were composited?
- Which method and incubation conditions were used for each organism?
- Was method suitability run with a stressed isolate in this matrix?
- What were the last six lot results, so the trend can be seen?
Read the sample mass and the method with the result. Not detected in 10 g and not detected in 25 g are different statements, and a count of 900 cfu/g against a 1,000 cfu/g limit is a pass with very little margin. Check the incubation conditions, the method suitability record and whether the laboratory is accredited for the specific test, then look at the trend across lots rather than at the single certificate.
Where the method is used
- Magnesium Bisglycinate — 1:2 magnesium–glycine chelate, anhydrous basis.
Sources
The method pages sit behind the COA page. The terms used here are defined in the buyer glossary, and the regulatory limits are on the compliance page.
Related documents
Send the certificate or specification you are reviewing and we will name the method behind each figure.
Request a specification review