Updated November 19, 2026
At a glance: what the numbers say
- Total amino acid load differs 2.8-fold across common basal media. Summing the published formulations: DMEM (high glucose) starts at 10.68 mM total amino acids, DMEM/F-12 at 7.26 mM, RPMI-1640 at 6.41 mM, and Ham's F-12 at just 3.83 mM. Same cells, very different starting pools. (Our sum of manufacturer formulation sheets; see Table 1.)
- Glutamine is 26 to 38% of the entire starting pool in every formulation (DMEM 37.5%, DMEM/F-12 34.5%, RPMI-1640 32.0%, Ham's F-12 26.1%), so glutamine kinetics dominate any spent-media profile.
- As an illustration under one published CHO consumption rate (not a prediction for your culture), the glutamine clock runs out in about 1 to 4 days at 1×10⁶ cells/mL. Applying the maximum glutamine consumption rate reported by Avilan Garzon et al. (Frontiers in Bioengineering and Biotechnology, 2025), 1.008 pmol/cell/day, as a constant-density batch model: Ham's F-12 reaches zero in 0.99 days, RPMI-1640 in 2.04 days, DMEM/F-12 in 2.48 days, DMEM in 3.97 days. Asparagine runs out faster still where it exists at all: 0.13 days in DMEM/F-12, 0.26 days in Ham's F-12, 0.98 days in RPMI-1640, and DMEM contains no asparagine to begin with.
- Tryptophan is the smallest molar pool in all four media (0.010 to 0.078 mM), 51 to 100 times smaller than glutamine in the same bottle. Small pools move first in percentage terms, which is why tryptophan is a sensitive spent-media discriminator even though its absolute consumption is low.
- DMEM starts with five amino acids at exactly zero: alanine, asparagine, aspartic acid, glutamic acid, and proline are not in the DMEM formulation. Anything measured for those analytes in spent DMEM was made by the cells, not left over from the medium.
- Stoichiometric ammonia ceiling: if every glutamine in 1 L of DMEM (4.00 mM) deamidated, it would release 4.00 mmol of ammonia. That is a ceiling computed from the formulation, not a measured accumulation rate.
- So what: If you run Ham's F-12 or RPMI-1640, your feed timing is being set by asparagine and glutamine inside the first 1 to 2 days at moderate density, not by the branched-chain amino acids people usually watch. A spent-media result only means something against the correct starting formulation.
- Why it matters: Most "low" spent-media readings are uninterpretable without the starting concentration for that exact medium. Teams waste feed, or miss limitation, because they compare a residual against the wrong baseline. This atlas is the baseline table.
Scope limit, stated plainly: This article uses published formulations only. Commercial CHO production media formulations are partly proprietary and are not included. Depletion timing is computed only where a published consumption rate could be verified (glutamine and asparagine, one CHO study); for the other 18 amino acids we publish starting-concentration rankings and say so, rather than inventing rates.
Introduction
A spent-media result is a residual: what is left of what the cells were given. Read on its own it says almost nothing, because the same low reading can mean heavy consumption from a rich medium or modest consumption from a lean one. This atlas supplies the missing half of every spent-media comparison: the starting concentrations of the common basal media, harmonized onto one scale, and the order in which their amino acid pools run out. Total starting load alone differs 2.8-fold across the four media covered.
The baseline matters because feed timing is set by the first pools to empty, not by the analytes teams usually watch. A feed schedule inherited from one medium can be days late in another, and an amino acid that starts at zero in a formulation measures production by the cells, not leftover medium. Knowing which situation a reading describes is the difference between feeding on evidence and feeding on habit.
The starting concentrations are computed from the published formulation sheets of each medium, with salt forms converted to amino acid molarity. Depletion timing is modeled only where a published consumption rate could be verified, for glutamine and asparagine in one published CHO batch study, run as a constant-density batch bound; for the remaining amino acids, the article publishes starting concentrations and rankings and says so, rather than inventing rates. Read the formulation table as the specification, and the depletion table as an ordering, not a calendar.
Glutamine depletion: the clock runs out in about 1 to 4 days
Published model lines under one CHO consumption rate at constant density: each medium starts at its formulated glutamine concentration and falls to zero at its days-to-zero from Table 2. End labels show the starting concentration of each medium.
Table 1. Starting concentrations (mM), computed from published formulations
Salt forms were converted to amino acid molarity. Cystine/cysteine row combines the sulfur amino acid form each formulation actually uses; see the note under the table.
| Amino acid | DMEM (high glucose) | RPMI-1640 | Ham's F-12 | DMEM/F-12 |
|---|---|---|---|---|
| Glutamine | 4.000 | 2.053 | 1.000 | 2.500 |
| Asparagine | 0 | 0.378 | 0.100 | 0.050 |
| Arginine | 0.398 | 1.148 | 1.000 | 0.699 |
| Leucine | 0.802 | 0.381 | 0.100 | 0.451 |
| Isoleucine | 0.802 | 0.381 | 0.030 | 0.416 |
| Valine | 0.803 | 0.171 | 0.100 | 0.452 |
| Lysine | 0.798 | 0.219 | 0.200 | 0.499 |
| Threonine | 0.798 | 0.168 | 0.100 | 0.449 |
| Phenylalanine | 0.400 | 0.091 | 0.030 | 0.215 |
| Serine | 0.400 | 0.285 | 0.100 | 0.250 |
| Glycine | 0.400 | 0.133 | 0.100 | 0.250 |
| Tyrosine | 0.398 | 0.110 | 0.030 | 0.214 |
| Histidine | 0.200 | 0.097 | 0.100 | 0.150 |
| Methionine | 0.201 | 0.101 | 0.030 | 0.116 |
| Cystine/Cysteine* | 0.201 | 0.208 | 0.200 | 0.200 |
| Proline | 0 | 0.174 | 0.300 | 0.150 |
| Alanine | 0 | 0 | 0.100 | 0.050 |
| Aspartic acid | 0 | 0.150 | 0.100 | 0.050 |
| Glutamic acid | 0 | 0.136 | 0.100 | 0.050 |
| Tryptophan | 0.078 | 0.024 | 0.010 | 0.044 |
| Total | 10.679 | 6.408 | 3.830 | 7.255 |
* DMEM and RPMI-1640 supply cystine; Ham's F-12 supplies cysteine (as cysteine HCl·H₂O, 0.200 mM); DMEM/F-12 supplies both cystine (0.100 mM) and cysteine (0.100 mM), summed here to 0.200 mM. RPMI-1640 also contains hydroxyproline (0.153 mM), which is not in the other three and is excluded from the totals above; including it, RPMI-1640 totals 6.561 mM.
Worked example (Table 1, one full chain). RPMI-1640 arginine: the Corning formulation lists L-arginine (free base) at 200.00 mg/L. Molarity = 200.00 mg/L ÷ 174.20 g/mol (arginine free base) = 1.148 mmol/L = 1.148 mM. Total for RPMI-1640 = 2.053 + 0.378 + 1.148 + 0.381 + 0.381 + 0.171 + 0.219 + 0.168 + 0.091 + 0.285 + 0.133 + 0.110 + 0.097 + 0.101 + 0.208 + 0.174 + 0 + 0.150 + 0.136 + 0.024 = 6.408 mM.
Glutamine starting concentration by medium
Formulated glutamine in mM (Table 1). Glutamine is the largest single pool in every formulation in this atlas.
Table 2. The depletion model (only where a published rate exists)
This model is an illustration. It assumes a constant cell density of 1×10⁶ cells/mL, no feed, and one published consumption rate from a single CHO study. Real cultures grow, are fed, and consume at different rates, so read the table as an ordering of which pools empty first, not as a calendar.
Model: batch culture, constant viable density 1×10⁶ cells/mL (1×10⁹ cells/L), no feeding, consumption at the maximum specific rate reported in Avilan Garzon et al. 2025 for a mAb-producing CHO batch culture (glutamine −4.2×10⁻¹¹ mmol/cell/h; asparagine −1.6×10⁻¹¹ mmol/cell/h on day 2). This is an illustrative bound, not a prediction for your cell line.
| Amino acid | Specific rate (pmol/cell/day) | Volumetric use at 10⁹ cells/L (mM/day) | Days to zero: DMEM | RPMI-1640 | Ham's F-12 | DMEM/F-12 |
|---|---|---|---|---|---|---|
| Glutamine | 1.008 | 1.008 | 3.97 | 2.04 | 0.99 | 2.48 |
| Asparagine | 0.384 | 0.384 | absent (0 mM) | 0.98 | 0.26 | 0.13 |
Worked example (Table 2). Rate conversion: 4.2×10⁻¹¹ mmol/cell/h × (1 pmol / 10⁻⁹ mmol) = 0.042 pmol/cell/h; × 24 h = 1.008 pmol/cell/day. Volumetric: 1.008 pmol/cell/day × 10⁹ cells/L = 1.008×10⁹ pmol/L/day = 1.008 mmol/L/day = 1.008 mM/day. Days to zero for DMEM glutamine = 4.000 mM ÷ 1.008 mM/day = 3.97 days.
Independent corroboration, not used in the model: the same study reports glutamine, asparagine, and cystine as the limiting substrates, depleted almost simultaneously, with asparagine completely consumed after about 150 hours. Our model independently puts asparagine first or tied-first in every formulation that contains it. Gap: the study's cystine consumption rate was not extractable in a form we could verify, so cystine gets no modeled days-to-zero here, even though the study flags it as co-limiting. That is a finding, not an omission to paper over.
Perspectives
Process: feed timing. Read Table 2 as an ordering, not a calendar. In three of four media, asparagine is the first pool the model empties; in Ham's F-12 it empties in about 6 hours at the modeled density and rate (0.26 days). If your feed is scheduled on a fixed day-3 or day-4 rhythm regardless of medium, the formulation table says that rhythm was probably inherited from a DMEM process.
Analytical: which analytes discriminate a spent sample. The analytes that move are the ones with high consumption relative to pool size: glutamine, asparagine, and the small-pool analytes (tryptophan, methionine, phenylalanine in F-12 at 0.030 mM each). An analyte that starts at zero (five of them in DMEM) measures production, not consumption. A useful spent-media panel has to cover both directions, plus ammonia, which is why our panel reports ammonia on every run.
Formulation: over- and under-supply. Arginine is the clearest mismatch: RPMI-1640 and Ham's F-12 start arginine at 1.148 and 1.000 mM, roughly 2.5 to 2.9 times DMEM's 0.398 mM, while their total pools are smaller. F-12's branched-chain supply is the thinnest (isoleucine 0.030 mM, 27 times lower than DMEM's 0.802 mM). Whether that matters depends on your cells' rates, which is exactly what a spent-media time course measures.
QC: the starting point is a specification, not a constant. These tables are the published nominal formulations. Lot-to-lot and supplier-to-supplier differences (salt forms, glutamine vs stabilized dipeptide forms) shift the true starting point. Measuring the fresh medium alongside the spent sample removes that ambiguity for the cost of one extra sample.
Check your own medium
If you want your spent sample read against its actual starting point, send the spent sample and, ideally, an aliquot of the fresh medium it came from.
AA-1 ($150 per sample) reports 19 amino acids plus ammonia in µM and mg/L on spent media and on the fresh medium. Cystine, the form of the sulfur amino acid in most media, is not in AA-1; it needs the AA-3 workup, priced after a free feasibility call. Send 200 µL per sample, 400 µL preferred. For a bioreactor time course, RUN-1 is one run with up to 24 time points and AA-1 on each, $1,990 (about $83 per time point, against $150 each when ordered singly), with a combined table and trend chart.
Order a RUN-1 package Order samples See an example report
Overnight turnaround available, with expedited service on request. Standard results in 3 to 5 business days from receipt.
Methods & Sources
- Formulations (accessed November 5, 2026): Thermo Fisher Scientific, DMEM high glucose (cat. 11995) formulation table; Corning cellgro RPMI-1640 formulation as published by Fisher Scientific; Lonza Ham's F-12 formulation sheet (12-615); Thermo Fisher Scientific DMEM/F-12 (cat. 11039) formulation table.
- Consumption rates (accessed November 5, 2026): Avilan Garzon et al., "mAb production kinetics in CHO batch culture," Frontiers in Bioengineering and Biotechnology 13:1546105 (2025). Only the glutamine and asparagine rates were used; see the cystine gap above.
- Formulas: molarity = (mg/L) ÷ molecular weight (g/mol); totals are column sums; days to zero = starting mM ÷ volumetric use (mM/day). Ammonia ceiling = glutamine mM × 1 mol NH₃ per mol glutamine deamidated (stoichiometric bound only).
- Assumptions: constant density 10⁶ cells/mL; constant maximum rate; no feeding, no production of glutamine/asparagine by cells. Real cultures violate all three, which is why Table 2 is labeled a model bound.
- Limitations: published formulations only; no proprietary CHO production media; one consumption study, one cell system. Rates from other lines and phases will differ.