Gel extraction of a DNA band

Gel extraction answers one question: can I take this length of DNA off the gel and leave the other lengths behind? A column cleanup of the whole reaction answers a different question: can I remove primers, nucleotides, and salt while keeping every long DNA species that was in the tube? This page compares those two cleanups. The photograph is the agarose gel under an orange safety shield, which is the moment you either cut or decide not to. How that gel was run, and why the band position is only a size, is agarose gel electrophoresis for DNA. What you do with a recovered insert is restriction, ligation, and Gibson assembly.

Cut, dissolve, bind

You separate on a percentage that puts daylight between the band you want and the neighbour you refuse. The molecular weight ladder has to flank that size on the same gel. Loading dye tells you when to stop. It is not the thing you extract.

Under illumination you cut a tight slice around the band. Ultraviolet damages the DNA you hope to clone, so the exposure is as short as the cut allows, and the orange shield stays between the lamp and your eyes. A blue-light transilluminator with a dye matched to it reduces that nicking. It does not cancel the stain's own safety data sheet. Trim agarose you do not need. Extra gel dilutes the dissolve step and carries more carbohydrate into the bind.

The slice goes into a chaotropic buffer that melts or dissolves the agarose at the temperature the kit states, and the DNA binds a silica column or a silica surface. Washes remove salt. A low-salt elution brings the DNA off. That is the same bind-wash-elute idea used in many nucleic-acid preps, aimed here at one slice. Low-melt agarose is a grade that softens at a lower temperature, used when a protocol wants a gentle melt. The certificate states the range. It is not required for every silica kit, and it is not a reason to skip the shield.

What a column cleanup leaves in the tube

A PCR or digest cleanup column never sees the gel. You bind the reaction, wash, and elute. Primers and primer-dimers are the species this is built to reduce. A second PCR product of a similar length to the amplicon comes through with it. So does a partial digest that still contains the parent plasmid, if both pieces are large enough to bind. The eluate is cleaner salt and still a mixture. If the next step cannot tolerate that mixture, the column was the wrong tool.

Size-selective bead cleanups are a third class. They can deplete short species without a slice. Two fragments that are close in length still travel together. When the contaminating band sits near the one you want, the gel is the separation. When the contaminant is a primer, the column or the beads are kinder and usually return more DNA.

Yield, carryover, and ligation

The band looked abundant because stain and camera compress a wide range of mass into "bright". Recovery is lower than that impression. You lose DNA in the slice edges you did not cut, on the silica, and in a volume that evaporates or sticks to the tip. Expect to quantify the eluate when the ligation or the assembly ratio matters. A second gel of a few microlitres of eluate tells you the length survived. It still will not give a reliable mass from a saturated photo.

Agarose that was not fully removed, and chaotrope that the washes left behind, inhibit ligation. The failure looks like a cloning problem. The map can be perfect. A260/A230 often falls when chaotrope or carbohydrate remains, which is a hint, not a proof of identity. If the ratio is poor, wash and elute again under the kit's instruction before you redesign the overlap. Do not pile more ligase onto a salty eluate and call it optimisation.

Gibson-style assembly and a classical ligation both dislike that carryover. They dislike it for different enzyme reasons and the same practical one. A dirty nanogram is not the nanogram you calculated.

When not to extract

Do not extract a single-band PCR if the next step is only a cleanup before sequencing or a digest. The column keeps the yield higher and skips the ultraviolet. Do not extract a band that is already at the edge of visibility if you need a ligation. The losses will leave you with a story and an empty plate. Redesign the PCR toward one product, or pool several lanes only if the size match is real on each gel.

Do not extract two bands that were never separated. A slice that covers both is a column cleanup with extra damage. Change the percentage or the run length first. Do not extract to "prove identity". The slice has the length you cut. Sequence or a diagnostic digest is the identity check, and a digest protocol is a separate decision from the slice.

A transfer membrane is the wrong object entirely. Protein blots do not yield clonable DNA. If the goal was a protein band, you are in a different method.

Failure modes

No DNA on a check gel of the eluate. The slice missed the band, the dissolve was incomplete, the wash contained ethanol so elution failed, or the fragment was outside the kit's size habit. The kit sheet states that habit. This page does not copy a volume table.

Ligation fails and the check gel shows a band. Suspect damage from the lamp and salt or agarose carryover before you suspect the map. A parallel ligation of a control insert that never saw the transilluminator tells those apart.

The wrong band was cut because the ladder was on a different gel or the loading dye was mistaken for a small product. Cut only on a gel that still shows the ladder in the same exposure, shield in place.

Safety and the bench

Chaotropic salts are chemical hazards on their own sheets. Ultraviolet is an eye, skin, and DNA hazard. The shield in the photograph is the control for the person. Short exposure is the control for the fragment. Stained gel waste follows the institutional rule for the dye you used. The WHO laboratory biosafety manual is background, not a permit. A slice from an infectious amplicon stays under the containment of that sample.

In a hot room a dissolved slice left on the bench can re-set or sit long enough for damage you did not plan. Elute and move the DNA to the condition the next enzyme expects. A power cut with the UV lamp on is a reason to finish the cut and get the gel off the box, not a reason to wait in front of it.

What the enquiry should name

EVRINTH can take a sourcing question. State the fragment length, whether a second band must be excluded, the stain and the lamp, and whether the eluate must ligate or sequence. The molecular biology catalogue is the reagent list, and the quote request carries the specification. The nucleic acid analysis pathway is the context when extraction sits between a PCR and a clone. Ask whether a quotation is possible. "Gel extraction kit" without the size and the downstream enzyme is not yet a requirement.