Western Blot Transfer Buffer: Towbin, Methanol

Q: Can I reuse transfer buffer?

No. Ionic composition and pH shift during a run as H + and OH − accumulate at the electrodes, so a reused buffer transfers progressively worse.

A western blot transfer buffer is not a single recipe. It is a set of four decisions — which buffering system, how much methanol, how much SDS, and matched to which transfer apparatus — and each one is made against the size of your target protein and the membrane you are transferring onto. Copying a labmate’s 20% methanol Towbin recipe onto a 150 kDa target headed for PVDF is not a neutral shortcut; it is an active choice that predicts a specific failure.

This page maps each composition decision to the artefact the wrong choice produces. It is the step before troubleshooting: if you have already run the blot and nothing appeared, work the symptom-first branches in the western blot no-bands decision tree instead. For the surrounding eight-step protocol, see western blot protocol basics, and for the separation step that precedes transfer, SDS-PAGE gel selection and troubleshooting.

What the transfer buffer actually has to do

Electrophoretic transfer has to accomplish two mechanically opposed things in the same tank. Protein has to elute out of the polyacrylamide gel, and it then has to bind and stay on the membrane. Methanol and SDS each help one of those and obstruct the other:

Methanol strips SDS from the SDS–protein complex and improves adsorption of protein onto the membrane. It also prevents the gel swelling during transfer and reduces gel pore size.
SDS keeps the protein soluble and negatively charged and improves elution out of the gel — most importantly for large proteins — but it works against membrane binding.

That is the whole tension. Every formulation below is a different point on that trade-off, and the buffering species (glycine, CAPS, carbonate) sets the pH at which the trade-off is made.

Towbin buffer: the 1979 original, and what it assumes

The standard tris-glycine transfer buffer comes from Towbin, Staehelin and Gordon, Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications, PNAS 76(9):4350–4354 (1979). Its composition:

25 mM Tris, 192 mM glycine, 20% (v/v) methanol, pH approximately 8.3.
Optionally 0.01–0.1% SDS.
Do not pH-adjust it. Made from Tris base and glycine at these concentrations, the buffer arrives at ~8.3 on its own. Titrating it with acid or base adds counter-ions, raises conductivity, and drives heating and electrode-surface damage.

Note what the original paper was doing: transferring onto nitrocellulose, in a tank. Towbin buffer is well-matched to that. Its two structural assumptions — a mid-range, acidic-to-neutral pI protein, and a nitrocellulose membrane that benefits from high methanol — are exactly the assumptions that break when your target is 150 kDa, or strongly basic, or destined for PVDF.

Preparing a 10× stock

Per litre of 10× stock: 30.3 g Tris base and 144.1 g glycine dissolved in ~600 ml water, optionally plus 5–50 ml of 20% SDS solution, made up to 1,000 ml. The working solution is made fresh from 100 ml of stock, 700–800 ml water, and 100–200 ml methanol — which is where the 10–20% methanol decision below is actually made. If you are working out the arithmetic from scratch, molarity and solution calculations for the lab and buffer and solution preparation cover the general mechanics.

Decision 1 — methanol: what it buys, and what it costs

What it buys. Methanol prevents the gel swelling during transfer, which keeps bands sharp, and it improves protein adsorption onto the membrane. For gradient gels it is effectively mandatory — without methanol a gradient gel swells into a trapezium, distorting every lane.

What it costs. Methanol decreases pore size in both gel and membrane, and at high concentration proteins can precipitate inside the gel rather than leaving it. Critically, for large proteins above roughly 100 kDa — particularly onto PVDF — methanol should be reduced or omitted, because transfer efficiency of large SDS-coated proteins is poor in the presence of alcohols. Two mechanisms are given for this: the tighter methanol-set network of gel and membrane, and methanol stripping SDS off the protein, which is what was keeping the large complex mobile in the first place.

The failure this produces. Running 20% methanol on a 150 kDa target gives you a faint or absent high-molecular-weight band with a Ponceau stain that looks fine at mid-range and empty at the top — and, if you check, protein still detectable in the post-transfer gel by Coomassie. That is not an antibody problem, and no amount of longer exposure fixes it.

Native gels and native-protein transfers are the other case for little or no methanol.

Decision 2 — SDS: the mirror-image trade-off

What it buys. SDS improves the elution of large proteins out of the gel in particular. Where a protein is stubbornly difficult to get out of the gel, adding SDS to the transfer buffer is the standard first lever.

What it costs. SDS increases relative current, power and heat generation — a real constraint on an uncooled tank. And for very small proteins and peptides, especially onto nitrocellulose, SDS can drive the peptide straight through the membrane. This is “blowout”: the protein transfers perfectly well and is simply not there afterwards, because it kept going.

The membrane asymmetry matters. Vendor guidance splits cleanly here: on PVDF, SDS may be reduced but should not be omitted, since blotting efficiency onto PVDF is meaningfully enhanced by SDS (0.01–0.05% is the usual supplement). On nitrocellulose, SDS may be omitted entirely, and for small proteins it generally should be — particularly in semi-dry blotting.

The failure this produces. Too much SDS on a <20 kDa target on nitrocellulose gives a blank membrane with clean, complete gel clearance — the mirror image of the methanol failure above, and diagnostically distinguishable from it by staining the post-transfer gel.

Decision 3 — the buffering system: Towbin, Bjerrum, Dunn or CAPS

Glycine is not the only choice, and its pH of ~8.3 is the reason to look elsewhere. A protein at a pH near its isoelectric point carries little net charge and will not move in an electric field, so a basic protein in Towbin buffer can be close to electrically neutral and simply fail to transfer. The general rule vendors give is to use a transfer buffer whose pH sits about two units above the pI of your protein.

System	Composition	pH	Use it when
Towbin (tris-glycine)	25 mM Tris, 192 mM glycine, 10–20% methanol, ±0.01–0.1% SDS	~8.3	Default for tank blotting; mid-range proteins onto nitrocellulose
Bjerrum–Schäfer-Nielsen	48 mM Tris, 39 mM glycine, 0–20% methanol, ±0.01–0.1% SDS	~9.2	Semi-dry continuous systems; native proteins (~0.04% SDS, 0–10% methanol)
Dunn carbonate	10 mM NaHCO₃, 3 mM Na₂CO₃, 10–20% methanol	~9.9	Basic proteins that will not move in Towbin
CAPS	~10 mM CAPS, 10% methanol	10.5–11.0	Basic proteins; high-molecular-weight targets; blots destined for N-terminal sequencing

None of these should be pH-adjusted after preparation except where the protocol explicitly says so — CAPS stock is the exception, adjusted to pH 10.5–11.0 with NaOH before dilution.

Why CAPS for sequencing

CAPS is the buffer of choice when the blot is going on to Edman degradation or N-terminal sequencing, because glycine interferes with those downstream chemistries. This is a case where the buffer decision is driven by what happens after the blot, not by the transfer itself — and it is the most commonly missed reason to move off Towbin.

Discontinuous systems: having it both ways

Semi-dry blotting permits something tank blotting cannot: different buffers at the anode and cathode. A discontinuous Tris–CAPS system published by Bio-Rad places methanol in the anode (membrane-side) buffer and SDS in the cathode (gel-side) buffer — 60 mM Tris / 40 mM CAPS at pH 9.6 with 15% methanol on the anode side, 0.1% SDS on the cathode side. That geometry is the clean resolution of the whole trade-off: SDS acts where elution happens, methanol acts where binding happens, and neither is present where it would do harm. Vendors publish system-specific variants of this; check the bulletin for your own apparatus rather than porting a formulation between instruments.

Decision 4 — matching the buffer to the transfer system

Tank (wet) blotting accepts continuous buffer systems only. Semi-dry blotting accepts both continuous and discontinuous systems, and discontinuous generally transfers better, precisely because the two buffers can be optimised independently. Rapid “turbo” systems ship proprietary concentrates matched to their own cassettes and run times; those are not interchangeable with a hand-made Towbin buffer, and substituting one for the other silently changes both ionic strength and run duration.

The practical consequence: a formulation is only meaningful together with the system it was specified for. A 0.05% SDS figure for a tank transfer becomes 0.01% for the same protein in semi-dry, in the vendor matrix below.

The composition matrix: protein size × membrane × system

The following is vendor guidance (Carl Roth blotting technical brochure, rev. 06/2020), given as approximate starting values, not as a standard. Treat it as the starting point you optimise from, and verify against your own instrument vendor’s bulletin.

Target	Methanol (denaturing)	SDS (denaturing)	Methanol (native)
Small, <~20 kDa	20%	Tank 0.01%; semi-dry 0% on nitrocellulose, 0.01% on PVDF	5–10%
Mid-range, ~20–80 kDa	10%	Tank 0.05%; semi-dry 0.01%	0–5% (PVDF only)
Large, >~80 kDa	10% on nitrocellulose; 0–5% on PVDF	Tank 0.1%; semi-dry 0.05%	0% (PVDF only)

Read the gradient down the methanol column and up the SDS column: as the protein gets larger, methanol comes out and SDS goes in. That single inverse relationship is most of what this page is for.

Membrane-specific rules

	PVDF	Nitrocellulose
Equilibration before use	Obligatory, in methanol	In water
SDS in transfer buffer	May be reduced, not omitted — SDS increases blotting efficiency onto PVDF	May be omitted; small proteins in particular should be blotted without SDS (blowout risk)
Methanol in transfer buffer	May be omitted	May be reduced, not omitted — methanol increases protein adsorption

PVDF’s methanol pre-wetting step is a membrane activation requirement, not a buffer decision, and is separate from the methanol content of the transfer buffer itself. Skipping it leaves the membrane hydrophobic and unable to wet in aqueous buffer at all.

Failure-mode map: which wrong choice causes what

What you see	Likely composition cause	Change
High-MW band faint or absent; mid-range bands fine; protein still in the post-transfer gel	Methanol too high and/or SDS too low for a large target	Drop methanol to 0–5% (PVDF) or 10% (nitrocellulose); raise SDS to 0.1% tank / 0.05% semi-dry
Small target absent; gel completely cleared; membrane blank	Blowout — SDS too high for a small peptide, typically on nitrocellulose	Omit SDS; keep methanol at 20%; consider a 0.2 µm membrane
Basic protein does not transfer at all, at any voltage	Buffer pH too close to the protein’s pI — it is near-neutral in Towbin	Move to Dunn carbonate (~9.9) or CAPS (10.5–11.0)
Blurred or smeared bands; gel visibly larger after transfer	Insufficient methanol, or gel not pre-equilibrated, so it swelled during transfer	Restore methanol; pre-equilibrate the gel 15–60 min in transfer buffer
Gradient gel distorted into a trapezium	Methanol omitted from a gradient-gel transfer	Add methanol; equilibrate the gradient gel until it stops changing size
Excessive heat, browned “burnt” blotting paper, damaged electrodes	Buffer pH adjusted after preparation, raising conductivity; or SDS raising current	Remake without titrating; cool the buffer and the unit; reduce power and extend time
Weak transfer that worsens across repeated runs	Transfer buffer reused, or too little blotting paper in the stack	Never reuse transfer buffer; keep ≥2 mm of paper each side of the sandwich
Poor N-terminal sequencing after an otherwise clean blot	Glycine from Towbin buffer interfering with the sequencing chemistry	Transfer in CAPS instead

Preparation rules that are not optional

Do not adjust the pH unless the formulation explicitly calls for it. Titration raises conductivity, generates heat, and can damage electrode plates.
Never reuse transfer buffer. Ionic composition shifts during a run as H⁺ forms at the anode and OH⁻ at the cathode.
Use analytical-grade (p.a. / Ph. Eur.) methanol. Metallic contaminants in lower grades plate onto the electrodes and raise buffer conductivity. Low ionic strength and low conductivity are what you are aiming for.
Pre-equilibrate the gel in transfer buffer for 15–60 minutes. This removes residual running-buffer salts (which would otherwise raise conductivity and heat), lowers the SDS concentration in the gel that would hinder membrane adsorption, and lets the gel reach its final size before the run rather than swelling during it.
Do not use a cathode buffer below pH 8.3 in a discontinuous semi-dry system — frequent use damages the electrodes.
Prefer lower power and longer time. Better transfers generally come from reducing the power setting and extending the run than from pushing current.
Keep tank buffer stirring during the run, and cool both buffer and unit.

Handling methanol

Transfer buffer at 10–20% methanol is a flammable, acutely toxic mixture — methanol is toxic by inhalation, ingestion and skin absorption, with optic-nerve damage as the characteristic effect. Prepare buffer in a fume hood, wear nitrile gloves, and collect used transfer buffer as organic solvent waste rather than pouring it down the drain. The product’s own safety data sheet is the authoritative source for exposure limits and incompatibilities; this page is a method reference, not a substitute for it.

Frequently asked questions

What is the standard western blot transfer buffer recipe?

Towbin buffer — 25 mM Tris, 192 mM glycine, 20% methanol, pH ~8.3 — from Towbin et al. (1979). But “standard” only means it is the common default for tank blotting of mid-range proteins onto nitrocellulose. There is no single universal recipe: methanol and SDS content should both change with protein size, membrane and transfer system.

Why is there methanol in transfer buffer?

It stops the gel swelling during transfer, which keeps bands sharp, and it improves adsorption of protein onto the membrane by stripping SDS off the protein complex. The same SDS-stripping is why it hurts large-protein transfer.

Can I make transfer buffer without methanol?

Yes, and for some targets you should. Large proteins above ~100 kDa onto PVDF transfer better with methanol reduced or omitted, and native-protein transfers use little or none. On nitrocellulose, though, methanol may be reduced but should not be omitted — it is what drives protein adsorption onto that membrane.

Should I add SDS to my transfer buffer?

Add it if your target is large or difficult to elute from the gel — typically 0.05–0.1% for tank blotting, less for semi-dry. Omit it for small proteins and peptides on nitrocellulose, where it causes blowout through the membrane. On PVDF, keep at least 0.01–0.05% in all cases.

What is the difference between Towbin buffer and CAPS buffer?

Towbin is tris-glycine at pH ~8.3; CAPS runs at pH 10.5–11.0 with about 10% methanol. Use CAPS when the target is basic and near-neutral at 8.3, when the target is high-molecular-weight, or when the blot is going on to N-terminal sequencing, where glycine interferes with the chemistry.

Do I use the same buffer for wet and semi-dry transfer?

Not necessarily. Tank blotting requires a continuous system, so Towbin, Bjerrum or Dunn. Semi-dry permits discontinuous systems with different anode and cathode buffers, which generally transfer better. SDS concentrations also differ between the two for the same protein — roughly 0.05% tank versus 0.01% semi-dry for a mid-range target. Rapid turbo systems use vendor concentrates matched to their own cassettes and are not interchangeable with hand-made buffers.

Does the transfer buffer differ for PVDF and nitrocellulose?

Yes, and in opposite directions. PVDF needs SDS present (reduce but do not omit) and tolerates methanol being dropped. Nitrocellulose needs methanol present (reduce but do not omit) and tolerates SDS being dropped. PVDF also requires obligatory methanol pre-wetting of the membrane itself, which is separate from the buffer’s methanol content.

Can I reuse transfer buffer?

No. Ionic composition and pH shift during a run as H⁺ and OH⁻ accumulate at the electrodes, so a reused buffer transfers progressively worse.

Sources

Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. PNAS 76(9):4350–4354, 1979. doi:10.1073/pnas.76.9.4350
Bjerrum OJ, Schäfer-Nielsen C (1986) — origin of the 48 mM Tris / 39 mM glycine, pH ~9.2 semi-dry formulation.
Dunn MJ (1986) — origin of the sodium carbonate, pH ~9.9 formulation for basic proteins.
Carl Roth, Transfer Buffers and General Tips on Blotting Procedures (ROTI Blot technical brochure, rev. 06/2020) — source of the protein-size × membrane × system matrix, the membrane-specific rules, and the preparation constraints. Retrieved directly.
Bio-Rad, Types of Western Blot Transfer Buffers and Transfer Buffer Formulations (Bulletin 6211) — source of the CAPS pH 11 / 10% methanol formulation and the discontinuous Tris–CAPS semi-dry system. Bio-Rad’s servers block automated retrieval, so the discontinuous Tris–CAPS figures here were not independently fetched from the bulletin; consult Bio-Rad’s own bulletin for the instrument-matched formulation before adopting it. The CAPS composition itself is independently corroborated by the Carl Roth brochure above.