Sterility and Mycoplasma Testing — Methods, Detection Limits, and What "Negative" Means

Sterility testing establishes the absence of culturable bacteria and fungi under a defined method. Mycoplasma testing is a separate test for a separate class of organism. The first point to fix, because it causes more confusion than anything else on this page: neither test covers the other, and a certificate showing one does not imply the other.

Mycoplasmas are wall-less bacteria. They pass through the 0.22 µm filters used to sterilise media, they are invisible under routine microscopy, they do not turn a culture turbid, and many species grow poorly or not at all on standard sterility media. A culture can be simultaneously sterile by the compendial test and heavily contaminated with mycoplasma.

Why mycoplasma is the contaminant that matters

It is prevalent and it is silent. A survey of NCBI’s RNA-seq archive covering 9,395 rodent and primate samples across 884 series found that 11 percent of series were contaminated, defined as at least 100 reads per million mapping to mycoplasma in one or more samples. That is an unbiased estimate across hundreds of laboratories, drawn from data those laboratories chose to publish. The same analysis found 61 host genes significantly associated with mycoplasma-mapped read counts in a single-cell dataset, which is the point: contamination is not a cosmetic issue, it changes the biology you are measuring.

The consequences for a buyer are three. Your results shift in ways that look like real biology. The contamination spreads through your incubator to everything else you own. And because it is invisible, you find out at the point where a large amount of work has already been done.

Property	Bacteria / fungi	Mycoplasma
Visible in culture	Yes — turbidity, pH shift, visible colonies	No. Cultures look normal at high contamination loads
Passes 0.22 µm filter	No	Yes, frequently — filter-sterilised media is not a barrier
Detected by sterility test	Yes, that is what the test is for	Not reliably. Many species are fastidious or non-culturable on those media
Killed by common antibiotics in culture medium	Penicillin/streptomycin covers many	No — mycoplasmas lack a cell wall, so β-lactams are ineffective
Effect on the culture	Rapid, obvious, culture is lost	Gradual and systemic: altered gene expression, nutrient depletion, changed drug responses
Spread within a facility	Limited, obvious	Extensive and silent, via aerosols and shared reagents

Sterility testing: the compendial method

The pharmacopoeial sterility test is a presence/absence test with a specific design.

Element	Specification
Methods	Membrane filtration (preferred where the sample can be filtered) or direct inoculation of the culture medium
Media	Fluid thioglycollate medium, primarily for anaerobic bacteria but also detecting aerobes; and soybean–casein digest medium, for fungi and aerobic bacteria. Both are inoculated in parallel
Incubation temperatures	Fluid thioglycollate at 30–35 °C; soybean–casein digest at 22.5 ± 2.5 °C
Duration	Not less than 14 days, with the media examined at intervals during incubation and at its conclusion
Method suitability	Must be established for the specific sample first, demonstrating that the sample does not inhibit growth of indicator organisms
Result	No evidence of microbial growth means the product complies with the test. Growth means it does not comply, unless the test can be demonstrated invalid for reasons unrelated to the product

Three things follow that buyers routinely miss.

The 14 days exist because of slow growers. A shorter incubation is a different, weaker test. If a CoA states a sterility result obtained in three days, that is not the compendial test.

Method suitability is not optional and is not a formality. If the sample matrix inhibits growth, a negative result means the medium was suppressed rather than the sample was clean. A sterility claim on a sample matrix in which suitability was never demonstrated is not evidence.

The test is inferential, not exhaustive. It samples a portion of a lot and infers about the lot. Low-level, non-uniformly distributed contamination can be missed by a sampling plan entirely legitimately. Sterility testing establishes that contamination was not detected under the stated conditions in the portion tested — which is the strongest claim any sterility test of any design can make.

Endotoxin is a separate matter again. Bacterial lipopolysaccharide is heat-stable and persists after the bacteria that produced it are dead. A sample can pass sterility and carry substantial endotoxin. If endotoxin matters to your application, it needs its own line on the certificate with a numerical limit.

Mycoplasma testing: the four methods

Method	What it detects	Turnaround	Detection sensitivity	Practical limits
Culture (broth and agar)	Culturable mycoplasma species, by colony formation	Long — the compendial culture method runs for weeks	Historically the reference method for culturable species	Misses non-culturable and fastidious species entirely. Slow enough that it cannot gate a short-shelf-life product
Indicator cell culture with DNA stain	Extranuclear DNA of mycoplasma adhering to indicator cells, visualised with a fluorescent DNA stain such as Hoechst 33258	Days	Complements culture by catching non-culturable species	Subjective read; requires an experienced microscopist; poor at very low loads
PCR / NAT (endpoint, real-time, digital)	Mycoplasma nucleic acid, typically conserved rRNA regions	Hours	Under the revised Ph. Eur. framework, a NAT method replacing culture must demonstrate a limit of detection of ≤10 CFU/mL or <100 genomic copies/mL	Detects DNA, not viability — dead organisms and residual nucleic acid give positives. Inhibitors in the sample matrix cause false negatives, which is why internal controls are mandatory
Enzymatic / biochemical assays	Mycoplasma-specific enzyme activity	Under an hour	Convenient for routine screening	A screening tool. Generally regarded as less sensitive than NAT; not a compendial release method

The current regulatory direction is worth understanding even if you buy only research-grade material, because it defines what a rigorous answer looks like. The revised European Pharmacopoeia chapter on mycoplasmas establishes defined sensitivity requirements where NAT replaces culture — the ≤10 CFU/mL or <100 genomic copies/mL limit above — introduces genomic copies as a unit so that NAT and culture results are comparable, requires that both culturable and non-culturable mycoplasmas be detected, and requires that testing include both cells and supernatant, because mycoplasmas adhere to and reside within cells. It also requires method validation in the user’s own product matrix even when a validated commercial kit is used, and specifies a genomic-copy-to-colony-forming-unit ratio below 10 for reference materials.

Read that list as a checklist for interrogating any mycoplasma claim:

Which method?
What limit of detection, in what units?
Was the sample cells, supernatant, or both?
Was the method validated in this matrix?
Were internal inhibition controls and an external positive control run?

Published practice in a working biobank illustrates what a routine implementation looks like: real-time PCR for mycoplasma alongside 10-STR capillary electrophoresis for authentication, run as a standing programme rather than an occasional check. The pairing is not accidental — authentication and mycoplasma testing are the two tests that the biobanking literature treats as the minimum quality control for a distributed cell line.

What a negative result actually means

This is the section that matters most, and it is short.

A negative mycoplasma result means: no mycoplasma nucleic acid (or growth, or enzyme activity) was detected in the sample tested, by the stated method, at that method’s detection limit, on that date.

It does not mean:

That the vial in your hand is clean. The tested sample was drawn from the lot at release. Your vial is a different vial and has since travelled.
That the material stays clean. Contamination is acquired in culture. A negative at release says nothing about your third passage.
That contamination below the detection limit is absent. Below the limit of detection is not zero. A slow-building contamination can be genuinely undetectable at release and obvious three weeks later.
That the specific organism in your facility would have been detected. NAT primer sets cover a defined species panel. Species outside the panel are not covered.
That the culture is sterile. Different test, different organisms.

The corresponding rule for sterility is the same in structure: a passing sterility test means no growth was observed under the stated method, in the portion tested, over the stated incubation.

Practical procurement and laboratory rules

Test on receipt, before the material meets your incubator. This is the single highest-value habit in the list. Quarantine incoming material, test it, and only then introduce it to shared space. It converts a facility-wide outbreak into a single rejected lot.

Test your own cultures on a schedule, not on suspicion. Because mycoplasma is invisible, suspicion-triggered testing means testing after the damage. Monthly is a common cadence; whatever you choose, make it a calendar item.

Test before you bank and before you publish. A contaminated master bank propagates indefinitely. A contaminated dataset propagates further.

Never rely on antibiotics as prophylaxis. Routine penicillin/streptomycin does not control mycoplasma, and continuous antibiotic use masks low-level bacterial contamination that would otherwise announce itself. Antibiotic-free culture is a detection method as much as a purity preference.

Treat a positive as a facility event. Discard, do not attempt to rescue unless the material is genuinely irreplaceable. Decontaminate the incubator and shared reagents, and test neighbouring cultures. Mycoplasma elimination reagents exist, they work imperfectly, and a rescued culture carries a permanent asterisk.

Ask what the supplier tests, per lot versus per bank. As with everything on a certificate of analysis, a bank-level test inherited by descendant lots is a weaker assurance than a per-lot test, and the certificate rarely distinguishes them unless you ask.

Questions to put to a supplier

Question	What a strong answer contains
What mycoplasma method do you use?	A named method class — real-time PCR, digital PCR, culture, indicator cell — not “we test for mycoplasma”
What is the limit of detection, and in what units?	CFU/mL or genomic copies/mL, with the validation basis
Was the assay validated in this product matrix?	Yes, with a statement about inhibition controls
Is the test performed on cells, supernatant, or both?	Both. Mycoplasmas adhere to and reside within cells, so supernatant-only sampling understates
Per lot or per bank?	Per lot for release; per bank for the deeper characterisation
Is the sterility test compendial, or an in-house observation?	The chapter referenced and the method used, or an honest statement that it is an in-house growth check
Is endotoxin tested, and against what limit?	A number, or a clear statement that it is not tested
Is testing in-house or by an accredited third party?	Either is legitimate; a third-party report attached to the CoA is a stronger document

Boundary cases

“Antibiotic-free culture” is a quality signal, not a test result. It means contamination would be visible rather than suppressed. It is a good sign about a supplier’s confidence and it is not evidence about a lot.

A rapid mycoplasma kit result is not a compendial result. Both are useful; they are not interchangeable, and a CoA that presents a rapid screening result in the position where a compendial result would sit is inviting a wrong inference.

Viral and adventitious agent testing is a third category again. Ordinarily absent from research-grade certificates. Its absence is normal and should still be noticed — see RUO vs GMP.

Testing a frozen vial and testing the culture it produces are different tests. Detection sensitivity is generally better in an actively growing culture, where any contaminant has had the chance to amplify, than in a frozen aliquot. If you can afford to test both, the culture test is the more informative one.

The one-line summary

Sterility and mycoplasma are two separate tests for two separate organism classes, and neither covers the other. A negative result means “not detected by this method, at this detection limit, in this sample, on this date” — nothing more. Ask for the method, the limit of detection and whether cells as well as supernatant were sampled, then quarantine and re-test on receipt regardless of what the certificate says.

Sterility and mycoplasma testing: the methods and what a negative result establishes