MEM Alpha (α-MEM) and Minimum Essential Medium

Harry Eagle published Minimum Essential Medium in 1959 in Science, after several years of systematically determining what a mammalian cell actually needs in culture. MEM is deliberately minimal: thirteen essential amino acids, eight vitamins, glucose, and the six inorganic salts of Earle's balanced salt solution (calcium chloride, potassium chloride, magnesium sulfate, sodium chloride, sodium phosphate and sodium bicarbonate). Nothing else. That minimalism was the point — it made MEM a defined baseline against which the requirements of any cell type could be tested.

Alpha-MEM is the "alpha modification", published in 1971 by Clifford Stanners, Gerald Eliceiri and Howard Green while working on ribosomes in mouse–hamster hybrid cells. They needed a richer medium that would support hybrid and fastidious cells without moving to the very high amino acid concentrations of DMEM. Their modification kept MEM's salt base and glucose exactly as they were and added nutrients:

• All the non-essential amino acids — alanine, asparagine, aspartic acid, glutamic acid, glycine, proline and serine, plus cysteine
• Sodium pyruvate at 110 mg/L (1 mM)
• Ascorbic acid (vitamin C) at 50 mg/L, biotin at 0.1 mg/L, vitamin B12 at about 1.3 mg/L
• Lipoic acid at 0.2 mg/L
• Optionally, ribonucleosides and deoxyribonucleosides

The result is a medium that is nutritionally much richer than MEM but shares its ionic composition and CO2 requirement — which is why alpha-MEM behaves differently from DMEM even though both are "enriched MEM".

This is the standard MEM formulation with Earle's balanced salts — the medium sold as "MEM", "EMEM" or "MEM Eagle" without further qualification.

Inorganic salts and other components

Amino acids (mg/L) — the thirteen essential amino acids only

L-arginine·HCl 126; L-cystine·2HCl 31; L-histidine·HCl·H2O 42; L-isoleucine 52; L-leucine 52; L-lysine·HCl 73; L-methionine 15; L-phenylalanine 32; L-threonine 48; L-tryptophan 10; L-tyrosine (as disodium salt dihydrate) 52; L-valine 46; plus L-glutamine 292.

Vitamins (mg/L) — eight

Choline chloride 1; D-calcium pantothenate 1; folic acid 1; niacinamide 1; pyridoxal·HCl 1; thiamine·HCl 1; i-inositol 2; riboflavin 0.1.

That is the whole medium. MEM contains no non-essential amino acids, no nucleosides, no pyruvate, no ascorbate, no biotin, no vitamin B12, no lipoic acid and no trace elements. The minimalism is deliberate — Eagle's purpose was to establish what a mammalian cell minimally requires, and everything else was expected to come from serum. It is also why MEM remains useful today for virology, vaccine production and nutrient-requirement studies, where added components would confound the result.

A MEM with Hanks' balanced salts version also exists, identical except that sodium bicarbonate drops to about 350 mg/L for use at atmospheric CO2 (see the salts section below). Two further common variants are MEM with non-essential amino acids already added, and 2x MEM for overlay work in plaque assays.

Standard alpha-MEM formulation (with nucleosides; the without-nucleosides version is identical minus the eight nucleosides):

Inorganic salts and other components

Amino acids (mg/L)

L-alanine 25; L-arginine·HCl 126; L-asparagine·H2O 50; L-aspartic acid 30; L-cysteine·HCl·H2O 100; L-cystine·2HCl 31; L-glutamic acid 75; glycine 50; L-histidine·HCl·H2O 42; L-isoleucine 52; L-leucine 52; L-lysine·HCl 73; L-methionine 15; L-phenylalanine 32; L-proline 40; L-serine 25; L-threonine 48; L-tryptophan 10; L-tyrosine (as disodium salt dihydrate) 52; L-valine 46.

Vitamins (mg/L)

Ascorbic acid 50; i-inositol 2; choline chloride 1; D-calcium pantothenate 1; folic acid 1; niacinamide 1; pyridoxal·HCl 1; thiamine·HCl 1; riboflavin 0.1; biotin 0.1; vitamin B12 ~1.3.

Nucleosides (mg/L, in the with-nucleosides version)

Adenosine 10; cytidine 10; guanosine 10; uridine 10; thymidine 10; 2'-deoxyadenosine 10; 2'-deoxyguanosine 10; 2'-deoxycytidine·HCl 11.

Two contrasts with DMEM are worth holding onto. Alpha-MEM has one quarter the glucose (1,000 versus 4,500 mg/L) and substantially less bicarbonate (2,200 versus 3,700 mg/L, so 5% CO2 rather than 10%). But it has a broader nutrient set — ascorbate, biotin, B12, lipoic acid and the full non-essential amino acid panel — none of which DMEM contains. Alpha-MEM is not a weaker DMEM; it is a different design.

This is the single most consequential choice when ordering alpha-MEM, and getting it wrong silently breaks selection experiments.

Cells synthesise nucleotides by two routes: the de novo pathway, which builds purines and pyrimidines from scratch and depends on dihydrofolate reductase (DHFR) and thymidylate synthase, and the salvage pathway, which recycles preformed nucleosides using enzymes such as HGPRT and thymidine kinase. Nucleosides in the medium feed the salvage pathway.

Use nucleoside-free alpha-MEM when the salvage pathway must be blocked or when de novo synthesis is your selection marker:

• DHFR-based selection in CHO-DXB11 and CHO-DG44. These lines are DHFR-deficient and cannot make nucleotides de novo. They survive only when nucleosides are supplied — so nucleoside-free alpha-MEM is exactly the selective pressure that identifies cells which have taken up a DHFR-carrying expression vector. Supplying nucleosides defeats the selection entirely. This is the classic use case, and the reason alpha-MEM without nucleosides is a standard bioproduction reagent.
• Methotrexate amplification. The same logic: methotrexate inhibits DHFR, and nucleosides in the medium would rescue the cells you are trying to select against.
• HAT and HT selection in hybridoma work, where the intent is to control which pathway is available.

Use nucleoside-containing alpha-MEM when you simply want a rich medium: routine mesenchymal stromal cell expansion, bone marrow culture, primary cell growth, and maintenance of DHFR-deficient lines before selection begins. The nucleosides reduce the metabolic burden of nucleotide synthesis and generally improve growth of demanding primary cells.

If your protocol does not mention nucleosides at all and you are not doing selection, the with-nucleosides version is the safer default. If your protocol involves DHFR, methotrexate, HAT or HT, read it carefully — the answer is usually "without", and the two products are frequently confused because their catalogue names differ by one word.

MEM is supplied on two different salt bases, and the difference is entirely about where you intend to keep the culture.

MEM with Earle's balanced salts contains 2,200 mg/L sodium bicarbonate and is designed for a 5% CO2 incubator. This is the standard laboratory form and is what "MEM" means without further qualification. Alpha-MEM uses this base.

MEM with Hanks' balanced salts contains only about 350 mg/L sodium bicarbonate and is designed for use in a sealed vessel or at atmospheric CO2 — historically for transport of tissue, for closed-flask culture, and for procedures performed on the open bench. Put Hanks'-based MEM in a 5% CO2 incubator in a vented flask and the pH will fall below the buffer's capacity.

The rule generalises beyond MEM: bicarbonate concentration and incubator CO2 are a matched pair, because pH in these media is set by the ratio of bicarbonate to dissolved CO2. Whenever you see an unexpectedly low bicarbonate figure on a datasheet, check whether the product is a Hanks'-based or sealed-vessel formulation before putting it in an incubator.

Mesenchymal stromal cells (MSCs). Alpha-MEM is the standard base for human and animal bone-marrow- and adipose-derived MSC expansion, usually with 10–20% FBS or a defined supplement. Its low glucose is a positive here — high glucose has documented effects on MSC proliferation and differentiation potential.

Bone and haematopoietic cultures. Osteoblasts, osteoclast differentiation assays, bone marrow stromal cultures and long-term marrow cultures conventionally use alpha-MEM. Its ascorbate content is directly relevant to osteoblast work, since ascorbate is required for collagen hydroxylation and matrix deposition.

CHO-DXB11 and CHO-DG44. As described above, alpha-MEM with and without nucleosides is the standard pairing for DHFR selection and methotrexate amplification in these production lines.

Primary cells generally. Chondrocytes, dental pulp cells, keratinocyte feeder layers, primary fibroblasts and many other primary cultures do better in alpha-MEM than in plain MEM because of the non-essential amino acids, pyruvate, ascorbate and nucleosides.

Plain MEM remains in use where a minimal, well-defined base is wanted — virology and vaccine production on Vero and MRC-5, plaque assays, and nutrient-requirement studies where added components would confound the result.

Note the calcium: alpha-MEM contains 1.8 mM, the same as DMEM. This suits adherent and differentiating cultures but makes it a poor base for keratinocyte protocols that specify low calcium.

CO2. Alpha-MEM's 2,200 mg/L (26 mM) bicarbonate is matched to 5% CO2. This is the standard incubator setting, so alpha-MEM holds pH more predictably out of the box than DMEM does.

Glutamine. 2 mM (292 mg/L) in glutamine-containing formulations. It degrades in solution to ammonia, so glutamine-free medium plus fresh addition, or the stable dipeptide L-alanyl-L-glutamine, gives more consistent long-term culture — particularly relevant for MSC expansion, where cultures run for weeks.

Serum. 10% FBS is typical; MSC protocols often specify 15–20%, and increasingly a defined or human-platelet-lysate supplement for clinical-grade work.

Ascorbate stability. Alpha-MEM's 50 mg/L ascorbic acid is genuinely useful but genuinely unstable — it oxidises in solution over days to weeks. For osteogenic differentiation, where ascorbate drives collagen deposition, supplement with fresh ascorbate-2-phosphate rather than relying on what is left in a medium bottle that has been open for a month. Ascorbate-2-phosphate is the stabilised form used in differentiation protocols for exactly this reason.

Storage. 2–8 °C in the dark, typically 12–24 months unopened. After glutamine addition, about four weeks. Alpha-MEM is more light-sensitive than plain MEM because ascorbate, riboflavin and folate all photodegrade. Keep it in a closed refrigerator.

Powdered alpha-MEM is supplied without sodium bicarbonate; add 2.2 g/L at reconstitution, adjust pH about 0.2 units below target to allow for the rise during filtration, and sterile-filter at 0.22 µm.

Ordering the wrong nucleoside version. The two products differ by one word in the catalogue name and completely in behaviour under DHFR or methotrexate selection. Check the datasheet, not the shelf label.

Assuming alpha-MEM is a low-glucose DMEM. It is not. Different bicarbonate (2,200 vs 3,700 mg/L), different CO2 requirement, different vitamin set, and it contains nucleosides, ascorbate, lipoic acid and pyruvate that DMEM lacks.

Relying on the bottle's ascorbate for differentiation experiments. Ascorbic acid in solution degrades. Use ascorbate-2-phosphate as a fresh supplement for osteogenic work.

Putting a Hanks'-salts MEM in a CO2 incubator. Its 350 mg/L bicarbonate cannot buffer at 5% CO2 and the culture acidifies.

Switching MSC cultures between alpha-MEM and DMEM mid-study. Glucose, calcium and micronutrient content all change; proliferation rate and differentiation potential shift with them.