Phenol Red in Cell Culture Media

Phenol red, chemically phenolsulfonphthalein, is a sulfonephthalein dye with a molecular weight of 354.4 as the free acid and 376.4 as the sodium salt used in media formulations. It is included in most classical cell culture media for one reason: it is a pH indicator whose transition range happens to straddle the pH at which mammalian cells are cultured.

It is worth being clear about what it is not. Phenol red is not a buffer. At the 5-15 mg/L found in media it is present at roughly 13-40 micromolar, against 24-44 millimolar of sodium bicarbonate - three orders of magnitude less. It contributes nothing measurable to the pH stability of the medium. It is a read-out, not a control.

It is also not a nutrient, not a supplement, and not required by cells. Every classical medium is available in a phenol-red-free version, and cells grow in those versions exactly as they do in the coloured ones. The dye exists purely so that a researcher walking past an incubator can see, in a second and without opening anything, whether a culture is heading in the wrong direction.

The concentration is not standardised across formulations, which is why the same shade of orange means slightly different things in different media, and why the colour of RPMI never looks quite like the colour of DMEM.

Ham's F-12 is the outlier at 1.2 mg/L, which is why F-12 looks pale and washed out compared with DMEM, and why DMEM/F-12 - the 1:1 blend - sits in between at 8.1 mg/L. If you are comparing colour between flasks, compare only flasks containing the same medium.

Some balanced salt solutions carry phenol red and some do not. HBSS is sold in both forms; PBS and DPBS are usually supplied without it.

Phenol red has a gradual transition from yellow to red across pH 6.8 to 8.2, with a pKa reported between about 7.7 and 8.0. The acid form absorbs maximally near 443 nm and the base form near 570 nm, which is what gives the dye its two very different appearances.

The important habit is comparing against a reference. Hold the flask against a white background next to a bottle of the same fresh medium at the same temperature. Colour judged in isolation, or across two different media, or against a coloured bench, is unreliable. The comparison table below gives the working interpretation.

One caveat before you act on any colour: it is semi-quantitative and it lags. A culture can be metabolically stressed before the medium visibly shifts, and the shift you finally see reflects the accumulated acid load over hours, not the state of the cells right now. Treat the colour as a prompt to look down the microscope, not as a diagnosis on its own.

Yellow means acid, and in cell culture acid almost always means lactate. Cells in culture run a high rate of aerobic glycolysis and export lactate into the medium; the protons that accompany it titrate the bicarbonate buffer down, and once the pH falls below about 6.8 the indicator is fully in its yellow form.

The common causes, in rough order of frequency:

The culture is simply dense. A confluent or over-confluent monolayer, or a suspension culture near its peak density, produces enough lactate to turn medium yellow within a day. This is normal biology and the answer is to feed or passage. If a flask reliably goes yellow before the next scheduled feed, seed fewer cells or feed more often.

The culture has been left too long. Media exhaustion over a weekend is the classic. Glucose falls, lactate accumulates, and the culture sits in acid long enough that viability drops even after fresh medium is added.

Bacterial contamination. Bacteria acidify medium far faster than mammalian cells and usually make it turbid at the same time. The signature is medium that was fine yesterday and is bright yellow and cloudy today, often with visible motile particles at high magnification. Discard, do not attempt to rescue, and check anything handled in the same session.

Too much CO2, or the wrong medium for the incubator. A medium formulated for 5% CO2 run at 10% will sit acid from the start. Match the bicarbonate to the atmosphere: media below about 1.5 g/L sodium bicarbonate suit ambient air or 4% CO2, 1.5-2.2 g/L suits 5%, and above 3.5 g/L suits 10%. DMEM at 3.7 g/L is formulated for 10% CO2 and will run slightly acid in a 5% incubator relative to its design point - which most labs accept without trouble, but it is worth knowing when a flask looks orange from day one.

Pink or fuchsia means alkaline, and in a bicarbonate-buffered medium alkalinity nearly always means CO2 has left the system. The bicarbonate is still there; the carbonic acid that balanced it has escaped as gas, and the pH climbs.

Medium taken out of the incubator. This is the single most common false alarm. A flask carried to the bench begins losing CO2 immediately and visibly pinks within minutes. It means nothing about the culture. Judge colour on flasks straight out of the incubator, or accept that anything you photograph on the bench will look more alkaline than it is.

The incubator is not delivering CO2. An empty cylinder, a closed valve, a failed regulator or a drifting sensor will turn every flask in the incubator pink together. If all your cultures went pink at once, suspect the incubator before the cells - and check the CO2 reading with an independent meter, because a miscalibrated sensor will report the set point while delivering nothing.

The door has been open, or the incubator is overloaded. Frequent or prolonged door openings prevent recovery of the set point. Flasks near the front go pink first.

Loose caps or the wrong vessel. Vented caps are designed to equilibrate with the incubator atmosphere. A sealed cap on a bicarbonate medium in a CO2 incubator prevents equilibration; a vented cap on a bench-top culture lets CO2 out. Match the cap to where the culture lives.

Bicarbonate too high for the atmosphere. The mirror image of the acid case: an Earle's-salts formulation at 2.2 g/L bicarbonate, or DMEM at 3.7 g/L, left in ambient air will go alkaline quickly. Hanks'-salts formulations at 0.35 g/L are the ones designed to hold pH in air.

Some contaminants alkalinise. A minority of contaminating organisms raise pH rather than lowering it. Pink plus turbidity is still contamination until proven otherwise.

Two failure modes matter enough to state plainly.

Mycoplasma usually produces no colour change and no turbidity. Mycoplasma-infected cultures often look entirely normal - clear medium, expected colour, cells that grow - while the organism alters metabolism, gene expression and experimental results. Colour is useless as a mycoplasma screen. The only reliable approach is routine testing by PCR or a dedicated detection assay on a fixed schedule.

A normal colour does not mean the medium is nutritionally intact. Glucose can be substantially depleted, and glutamine can degrade to ammonia, without the indicator moving far. Phenol red reports proton balance, nothing else. A culture that looks fine by colour can still be starved.

There is also a practical limit worth remembering: phenol red does not distinguish between a heavy healthy culture and a contaminated one. Both turn medium yellow. The microscope, not the colour, tells you which.

Phenol red interferes with optical measurements, and that is the main reason to remove it.

Absorbance assays. The dye absorbs strongly near 443 nm and near 570 nm. That second peak sits directly on top of the read wavelengths for several standard colorimetric assays - MTT is commonly read at 570 nm and the Griess nitrite assay at about 540 nm - so residual phenol red inflates the blank and compresses the dynamic range. Even when the assay includes a medium blank, a coloured background costs sensitivity, and because the dye's absorbance is itself pH-dependent, the background moves with the metabolic state of the well. That last point is the one that quietly ruins plates: a busy well and a quiet well have different backgrounds.

Fluorescence measurements. Phenol red contributes broad background across the green to orange range and can quench signal by inner-filter effects. Live-cell imaging, fluorescent plate-reader assays and flow cytometry of cells still carrying medium all benefit from phenol-red-free formulations. For imaging, a phenol-red-free medium with HEPES is the common choice, since it also holds pH on a microscope stage outside a CO2 atmosphere.

Oestrogen-responsive cell work. Discussed in the next section.

Biomanufacturing and downstream processing. Where a medium component has to be cleared from a product, an unnecessary dye is an unnecessary clearance problem, and phenol-red-free formulations are standard in production media.

Some mass spectrometry and metabolomics workflows, where the dye and its impurities appear as contaminating ions.

This is worth stating precisely, because it is widely repeated in a garbled form.

Berthois, Katzenellenbogen and Katzenellenbogen reported in 1986 that phenol red preparations show measurable oestrogenic activity at 15-45 micromolar, which is exactly the range at which the dye is present in culture media. They found binding to the oestrogen receptor of MCF-7 human breast cancer cells with an affinity around 0.001% that of estradiol (Kd about 2 x 10^-5 M), and dose-dependent stimulation of proliferation in oestrogen-receptor-positive MCF-7 cells but not in receptor-negative MDA-MB-231 cells.

Two years later, Bindal and colleagues showed by HPLC and solvent partitioning that the receptor binding and growth-promoting activity does not belong to phenolsulfonphthalein itself but to more lipophilic impurities present in commercial preparations - one identified as bis(4-hydroxyphenyl)[2-(phenoxysulfonyl)phenyl]methane - and that interfering amounts were present in many commercially available preparations. Related work attributed cation transport effects to a lipophilic impurity as well.

The practical conclusion is the same either way, and it does not depend on which molecule is responsible: if you are studying oestrogen responses, or anything downstream of oestrogen receptor signalling, use phenol-red-free medium. The standard combination is phenol-red-free medium with charcoal-stripped serum, which removes both the dye-associated activity and the steroids contributed by the serum. Purity of a given phenol red lot is not something you can verify at the bench, so removing the variable is more reliable than assuming a clean lot.

For cells with no oestrogen receptor and no interest in that pathway, this is not a reason to change medium.

Going phenol-red-free costs you the free, continuous, at-a-glance pH signal, and labs that switch often under-appreciate how much they were relying on it. Contaminated flasks are noticed later, and exhausted medium is noticed later still.

Three compensations are worth putting in place. Fix the feeding schedule rather than feeding by eye, based on a growth curve for that line at that seeding density. Check pH directly when it matters - a calibrated meter on a spent-medium aliquot takes a minute. And keep a coloured control flask of the same line on the same schedule if you are running a long phenol-red-free experiment, so you have a visual reference for when the uncoloured cultures need attention.

Adding HEPES at 10-25 mM is a common companion change for phenol-red-free work, since it holds pH outside a CO2 incubator during imaging or extended handling. HEPES does not replace the indicator - it addresses stability, not visibility - and it has its own considerations, including phototoxicity under illumination.

One last point of confusion to clear up: phenol red also appears in microbiology as the indicator in carbohydrate fermentation broths, which is a different application in a different context. The material on this page concerns phenol red as the pH indicator in mammalian cell culture media.