Glucose solution in cell culture

D-glucose (dextrose) has a molecular weight of 180.16 g/mol. The conversion that underlies every calculation on this page is therefore:

1 g/L D-glucose = 5.55 mM, and 1 mM = 0.180 g/L.

Concentrated stock solutions are sold and prepared at several strengths. The percentages are weight per volume:

A worked example: to take a 1 L bottle of low-glucose DMEM (1 g/L) up to high-glucose equivalent (4.5 g/L), you need to add 3.5 g/L, which is 11.7 mL of a 300 g/L stock or 8.75 mL of a 400 g/L stock.

Osmolality is the constraint that limits how much you can add. Glucose is a non-electrolyte, so each mole contributes roughly one osmole - a 300 g/L stock is on the order of 1,700 mOsmol/kg. Adding 10 mL of that stock to a litre of medium raises osmolality by something like 17 mOsmol/kg. One addition is trivial against a medium sitting at 260-320 mOsmol/kg. Ten feeds over a two-week fed-batch run are not, and osmolality creep from repeated feeding is a genuine and frequently overlooked cause of declining viability late in a culture. If you feed repeatedly, either use the most concentrated stock available so the added volume is smallest, or track osmolality alongside glucose.

Basal media differ widely in glucose content, and the difference is deliberate rather than arbitrary. Human blood glucose sits around 5.5 mM, so a "low glucose" formulation is the physiological one and "high glucose" is roughly four to five times physiological.

High glucose supports faster proliferation and lets a culture run longer between feeds, at the cost of more lactate: cells running aerobic glycolysis convert glucose to lactate rather than oxidising it fully, and the lactate acidifies the medium. Low glucose limits that lactate burden and is closer to physiological metabolism, at the cost of needing attention sooner.

Glucose-free formulations exist for a specific and increasingly common reason: replacing glucose with galactose forces cells to rely on oxidative phosphorylation rather than glycolysis, which is the standard manipulation for studying mitochondrial function and for sensitising cells in metabolic assays. Glucose-free media are also the starting point for defined feeding studies where glucose is the variable under test.

A basal medium is formulated to support a culture from seeding to the next scheduled feed. Supplementation is warranted when that assumption stops holding:

High-density and fed-batch culture. This is the main case. A CHO or HEK fed-batch run reaching high viable cell density will exhaust the glucose in the starting medium well before the run ends. Glucose is fed either as part of a complete feed supplement or as a separate concentrated glucose solution, precisely because it is consumed faster than the other components and so has to be dosed independently.

Long cultures between medium changes. Organoids, extended differentiation protocols, and any culture left over a long weekend can run into glucose limitation. Adding glucose is a lighter-touch intervention than a full medium change when the culture is fragile or the schedule does not allow one.

Metabolically demanding cells. Hybridomas, activated lymphocytes and some primary cells consume glucose at rates that outstrip standard formulations.

Building or customising a medium. When making medium from powder, or moving a process to a custom formulation, glucose is dosed as a separate component so that it can be varied without reformulating everything else. Glucose-free feed supplements exist for the same reason: they let the process developer set the glucose independently of the amino acid and vitamin feed.

Recovering an under-fed culture. If an assay shows glucose approaching depletion mid-run, a bolus addition is the immediate fix. It is a correction, not a strategy, and the strategy should then be revised.

Supplementation is not warranted simply because more glucose sounds better. Excess glucose in a culture that cannot use it produces more lactate, not more cells.

The reason glucose feeding has a literature of its own is that the amount of glucose available changes what cells do with it, not just how long they last.

The lactate problem. Cells in glucose-rich medium run aerobic glycolysis, converting glucose to lactate rather than oxidising it completely, and lactate accumulates. Reported thresholds vary with cell line and process, but growth inhibition above roughly 4 g/L lactate is widely described in CHO culture. Lactate also drags the pH down, which the bicarbonate buffer must absorb, and which in turn requires base addition in a controlled bioreactor - adding sodium and pushing osmolality up further.

Two feeding philosophies. They lead to different processes:

Glucose-sufficient feeding holds glucose comfortably above zero, typically by adding a bolus back to around 6 g/L whenever measured glucose falls below about 4 g/L, or by running a controlled feed of a concentrated glucose solution (400 g/L is a commonly reported feed stock) to hold a setpoint. This is simple, robust and forgiving of infrequent sampling. It also maximises lactate production.

Glucose-limited feeding deliberately holds residual glucose low, typically targeting 1-2 g/L with frequent small boluses every 12-24 hours based on measured residual glucose. Cells starved of surplus glucose shift toward oxidative metabolism and, in some cases, begin consuming the lactate they produced earlier. Lactate reductions of 60-80% relative to glucose-sufficient feeding have been reported. The cost is that it demands frequent, accurate glucose measurement: the margin between "limited" and "depleted" is narrow, and running out of glucose entirely will crash a culture.

Practical rules that hold in both cases. Measure rather than assume - glucose consumption rate is not constant across a run, and it rises steeply as density increases. Feed on the measurement, not on the calendar. Use the most concentrated stock that is practical, to keep added volume and osmolality creep down. And track lactate and osmolality alongside glucose, because glucose alone will not tell you why viability is falling.

Glutamine is the other half of the picture. L-glutamine degrades spontaneously in solution at 37 degrees C with a half-life of around six days, releasing ammonia, which is toxic. A culture being fed glucose over two weeks is also a culture whose glutamine has been decomposing the whole time. For runs of more than three to five days between medium changes, the stable dipeptide forms - L-alanyl-L-glutamine or equivalent - avoid both the loss of glutamine and the ammonia.

Do not autoclave glucose together with amino acids. Reducing sugars react with amino groups on heating - the Maillard reaction - producing brown, chemically complex products, some of which are cytotoxic. Glucose also caramelises on its own at autoclave temperatures. A medium autoclaved with its glucose in place is a different medium from the one on the formulation sheet.

The correct approaches are either to sterile-filter the glucose solution through a 0.22 micron membrane, which is the standard for cell culture and avoids the problem entirely, or to autoclave the glucose solution separately from the amino-acid-containing components and combine them aseptically afterwards. Filtration is simpler and is what commercial sterile glucose solutions use.

Making a 300 g/L stock. Dissolve 300 g of cell-culture-grade D-glucose in roughly 700 mL of cell-culture-grade water, warming gently to help it dissolve, then bring to 1 L and filter at 0.22 micron. Concentrated glucose is viscous and slow to filter; a pre-filter saves membranes.

Storage. Sterile glucose solutions are typically stored at 2-30 degrees C protected from light with a shelf life of around 12 months. The specific hazards are: crystallisation at high concentration if the solution gets cold, which is reversible by warming gently but should be fully redissolved before use, since a partly crystallised stock delivers less glucose than you calculated; and microbial growth, because a concentrated sugar solution is an excellent growth medium if sterility is ever broken. A cloudy glucose stock is contaminated until proven otherwise - discard it.

Endotoxin and water quality. Glucose stock goes directly into medium at a meaningful volume, so the water carries straight through to the cells. Use cell-culture-grade water with a stated low endotoxin specification, particularly for primary cells and immune cells.

Animal-origin considerations. D-glucose for cell culture is produced by starch hydrolysis and is animal-origin-free, which is one fewer thing to document in a regulated process.

Glucose depletion. The culture stops growing, viability falls, and the medium may not look particularly acidic because the cells have stopped producing lactate too. This is entirely preventable with routine glucose measurement, and it is the failure that glucose-limited feeding strategies risk if sampling is not frequent enough.

Lactate accumulation. Growth slows while glucose is still plentiful, the medium turns yellow fast, and pH control in a bioreactor requires increasing base addition. The answer is not more glucose. It is either a glucose-limited feeding strategy or a lower-glucose basal formulation.

Osmolality creep. Late-run viability decline in a fed-batch process that looks nutritionally fine on paper. Every bolus of concentrated feed, and every millilitre of base added to counter lactate acidification, adds osmolality. Measure it rather than calculating it.

Ammonia from glutamine. Often blamed on the glucose feed because it appears at the same point in the run. The cause is spontaneous glutamine decomposition at 37 degrees C, and the fix is a stable dipeptide glutamine source, not a change to the glucose regime.

Browning of medium. A brown or amber tint developing in a medium that was autoclaved with its glucose, or in a glucose solution heat-treated too aggressively, indicates Maillard or caramelisation products. Filter-sterilise instead.

High-glucose artefacts in biology. Cells cultured at 25 mM glucose are cultured at four to five times physiological concentration, which alters metabolism, glycation and stress responses. If your experiment is about metabolism, insulin signalling, diabetes, or anything downstream of nutrient sensing, the glucose concentration is an experimental variable and should be stated in the methods and justified - not inherited from whatever bottle was on the shelf.