Recording contamination in tissue culture: what to write down when a vessel turns

Why “it went bad” is not a record

The note is familiar. A tray comes off the shelf, a few jars have turned, someone writes “contam” on the transfer sheet, and the vessels go to the autoclave. The loss gets counted. The note says nothing about why it happened, so the next loss arrives with no more context than this one.

Contamination has more than one way in. A review of microbial hazards in plant tissue and cell cultures names three: contaminants can arrive with the explant, during manipulations in the laboratory, or carried by mites and thrips. The same review notes that contaminants may express themselves immediately or remain latent for long periods, which often makes the source difficult to identify. A jar that turns in week six may have carried the problem since initiation, or picked it up at the last transfer. “It went bad” cannot tell those apart.

A consistent record is the difference between a loss and a lead. Written the same way every time, contamination events stop looking like isolated bad luck and start lining up with things a lab can change: a media batch, a hood, a transfer session, a mother plant, an initiation round. The point is being able to answer “what do the turned jars have in common?” from the record instead of from memory.

Reading what you see, provisionally

Most contamination is first spotted by eye. A university plant propagation teaching resource describes yeasts, fungi and bacteria as usually appearing on the agar surface within a few days to a week, as white or opaque slime or variously colored colonies, sometimes with black spores and mycelium. Bacterial growth often reads as a wet, milky film, frequently where the explant meets the medium, or as clouding in liquid medium. Fungal growth tends to look fuzzy or cottony and spread across the surface. Yeast often forms smooth, creamy colonies that can be hard to tell from bacteria by eye.

Timing is part of the picture. A proceedings paper on hygiene in tissue culture propagation notes that fungal infection can become apparent within a few days of culture, that some bacteria are slow to grow and may take a few weeks to show, and that latent bacteria are sometimes detected only after several months. It also points out that the medium affects what you see: a contaminant may show easily in liquid medium and stay hidden on solid medium, or be held back by a compound in one medium and appear after transfer to a medium without it.

So visual identification belongs in the record as an observation first and a category second. “Cream-colored slime at the explant base, medium clouding” is something two people can agree they saw. “Bacterial” is a provisional call, and worth marking as one; naming the organism takes a lab result. If the lab indexes cultures or sends samples out, that result joins the same record when it comes back.

Write down what you saw before you write down what you think it was. The verdict can wait. The observation cannot.

When contamination shows up subcultures later

The hardest losses to trace show up several subcultures after their cause. Bacteria that propagate along with plant material without obvious symptoms are described in the literature as latent, internal, endophytic or endogenous. A study of bacterial contamination in in vitro plant cultures notes that bacterial contaminants are typically detected by direct observation, so cultures without visible symptoms are assumed to be bacteria free. The same study found contaminating bacterial DNA in extracts from plants that showed no symptoms.

One published case shows how far it can run. Grape cultures maintained continuously in vitro for six to seven years showed declining root and shoot growth, and indexing the culture medium on bacteriological media revealed covert bacteria in 75 to 100 percent of cultures. That is one cultivar in one study, not a figure for any other lab or crop. It does show why a clean-looking line is an observation with a date, and nothing more.

This changes what a record has to carry. If a jar turns at the fourth subculture, the useful question is whether its siblings and cousins from the same initiation round or the same mother plant are turning too. Answering it needs every vessel traceable up its lineage: which transfer it came from, which explant, which stock plant. Without lineage, a slow-emerging problem looks like scattered bad luck across the growth room.

The minimum useful record

The record does not need to be long. It needs to be complete the same way every time, and quick enough to fill in before the shift ends. For each event, capture the vessel or batch ID, scanned from its label where possible rather than copied by hand; the date the problem was first seen; the stage, whether initiation, multiplication or rooting; and the days since the last transfer.

Then capture what links the event to a possible cause. That is the media batch the vessel was poured from, not only the recipe name, because two batches of one recipe are two separate events in the prep room. It is who did the transfer, at which hood or workstation, and on which date. And it is the mother plant or initiation round the line traces back to.

Finally, capture the evidence: the signs in plain words (color, texture, where in the vessel, medium clear or cloudy), how many vessels in the batch are affected out of how many, and a photo with the label in frame, which settles questions weeks after the vessel has been through the autoclave. If a vessel is only suspect, record it as suspect. A log that allows “watching” as a status is more honest than one that forces a yes or no.

Verify this

Before you call it and discard a batch

Confirm that the signs are microbial growth and not something that can pass for it at a glance, such as browning that leaches from the explant, condensation, or callus. Check the vessel against its label so the event is recorded on the right line. Count how many vessels in the batch actually show signs rather than assuming all of them do, record suspect vessels separately from confirmed ones, and compare with earlier photos of the same line if there are any.

If the lab indexes cultures or sends samples out for identification, decide whether a sample is needed before anything goes to the autoclave, and take it only where your biosafety rules allow, never in the transfer area. Then check your SOP for who is authorized to discard a batch. That call belongs to the procedure, not to whoever happened to find the jar.

Quarantine first, then dispose

A record is only half of the response. A contaminated or suspect vessel should come off the growth shelf and go somewhere it cannot be mistaken for clean stock, clearly marked. The hygiene paper cited above advises keeping all contaminated cultures away from the culture area and destroying them by autoclaving. It also describes mites carrying fungal spores from vessel to vessel until a whole room of cultures is affected, which is why regular checks and prompt removal matter.

Two habits are worth making non-negotiable. Suspect vessels are not opened in the transfer area or at a hood used for clean work, because opening a sporulating jar is how one problem becomes several. And contaminated vessels are autoclaved before anything is emptied or discarded, loaded and run the way your lab’s own biosafety rules and local waste requirements specify. Those rules protect the people in the room as well as the cultures, and they take precedence over general advice, this article included.

The quarantine step deserves its own line in the log: when the vessel was pulled, where it went, and when it was autoclaved. That closes the loop, so nobody later wonders whether a jar marked contaminated is still sitting on a shelf.

Finding the pattern without finding someone to blame

With events recorded the same way, patterns are simple to look for. Sort by media batch: if most of two weeks’ losses sit on one batch, its preparation and autoclave run are the first place to look. Sort by hood: a cluster at one station points at the station and its maintenance before it points at anyone who worked there. Sort by mother stock or initiation round: losses that follow a lineage across sessions and media suggest something carried in with the plant material, which the hygiene paper names as the main source of contaminants. Then look across the calendar, in case initiation losses rise and fall with when and where explants were collected.

Sorting by person is the sensitive one, and it is also where good records protect people. A cluster under one name is first a question about the conditions of those sessions: which hood, which media, which stock, which day of the week. Often the cluster dissolves once the other fields are read. When it does not, the fix is a refresher or a paired transfer session, framed as a process change. People record honestly when the log is used to fix systems, and stop recording when it is used to assign fault.

Every pattern should end in a written, dated corrective action: a media batch held and remade, a hood taken out of use until its filter is checked, a stock plant retired. Then the log can answer the question that matters afterwards: did losses on that line drop after the change? If not, the cause is elsewhere, and the record shows where to look next.

What a good log lets a lab do

A consistent contamination log helps a lab trace a loss to its likely source, quarantine the relatives of an affected line sooner, and spend less time guessing. It also changes the conversation with customers. A lab that can show which lot a shipment came from, the lineage behind it and the contamination history of that line has something concrete to offer when a buyer asks about provenance. That is a statement about the records kept, not a promise about the plants.

In xPlant Pro, contamination events are recorded against the affected plant or explant with notes, severity, status and photos. Explants trace back to their source material and mother plant, transfers to fresh media and culture location moves are logged, and QR labels print for plants, explants and stages. On the plans that include them, quarantined items are managed in the same place and an activity history records who changed what. Contamination events can be filtered across the whole collection and the history exported. None of that keeps a jar clean. It helps a lab answer, from the record, what the turned jars have in common.