Gateway and recombination cloning overview

Gateway recombination, a trademark of its owner, is one commercial class of att-site cloning. The sites and the two reaction directions are older than any one mix: an attachment-site pair recombines and swaps which sites flank the insert. You are deciding whether an entry clone, a destination clone, or both are what this project needs, and which strain is allowed to see a ccdB cassette. This is a decision-level protocol overview. It does not copy a vendor volume table. Follow the enzyme mix you actually open.

The colony you keep still has to sit on a mapped plasmid, as in plasmid cloning from insert to colony. How the host is handled is in bacterial culture and transformation.

What the sites exchange

Four site names describe the logic. attB and attP recombine to produce attL and attR. attL and attR recombine to produce attB and attP. The first direction is the BP class. The second is the LR class. The enzyme mix for BP carries the integration activities used for that direction. The mix for LR also carries the excision activity that direction needs. They are not interchangeable bottles. Times and volumes belong on the sheet in the box you have.

In the cloning use of this class, a fragment flanked by attB sites is recombined into a donor vector whose attP sites flank a ccdB cassette. The product is an entry clone: the insert now flanked by attL sites, in the donor backbone, and the cassette has been swapped out onto a byproduct. An LR reaction then recombines that entry clone with a destination vector whose attR sites flank another ccdB cassette. The product is the destination clone: the insert flanked by attB sites in the destination backbone.

The site pairs are specific. attB1 is meant to recombine with attP1, not with attB2. That asymmetry is what sets orientation. If you place the same site on both ends, orientation is no longer enforced and some molecules will not recombine usefully. Mark left and right on the map before you order primers or a synthetic fragment.

ccdB is the counterselection. Cells that still carry a functional cassette and are sensitive to it do not form colonies. Cells in which the insert replaced the cassette can form colonies if they have the product's resistance gene. That is a cloning-strain issue. Propagate the parent donor or destination only in a strain the vector note names as able to maintain the cassette. Transform the BP or LR reaction into a sensitive strain so unrecombined parents die. Do not treat ccdB as a protein to prepare, express, or purify. You choose a host. You do not make a reagent out of the cassette product.

What you gather before a reaction

You need the insert sequence with the correct att tails and a reading frame you have written out, a donor vector if you are making an entry clone, a destination vector whose promoter and marker match the experiment, the BP or LR enzyme-mix class, and two kinds of competent cells if you must both maintain parents and select products. Antibiotics follow each backbone. A donor and a destination often carry different markers so you can select the product and not the parent. Write those names down. Guessing from a gel photo is how the parent antibiotic gets used on the product plate.

The insert can be a PCR product with attB tails or a synthetic fragment with those tails already in the file. Either way the bases have to be checked against the map. The custom gene synthesis enquiry reference is only a prompt if the fragment is a specification rather than a plasmid you hold. Enzyme mixes and cells are classes to ask about from the molecular biology catalogue.

A staged path with places to stop

Stage one is the map. Draw att sites, the antibiotic of each backbone, and the amino acids the sites will add if they are inside a fusion. If that scar is unacceptable, stop. Restriction ligation or a different assembly class may fit better. Do not start a recombination to discover the scar in a protein gel.

Stage two is the entry clone, if you need one. Run the BP class. Transform a sensitive strain. Select on the donor marker. Pick more than one colony. A digest or a PCR can show that the cassette-sized fragment was replaced by something nearer the insert length. Sequence the entry clone if other destinations will depend on it. A mutation here is copied into every later LR. If the plate is empty, check the strain's sensitivity, the marker, and whether the att tails are the pair the donor expects. If the plate is a lawn of parent, you selected the wrong drug or the strain tolerates the cassette.

Stage three is the destination clone. Run the LR class with the verified entry clone and the destination vector. Transform a sensitive strain. Select on the destination marker. Screen several colonies and sequence the att junctions and enough of the insert to cover the frame. Survival means a cell was not killed by the cassette and could resist the drug. It does not mean the frame is intact. A cassette that has mutated can slip through counterselection. The sequence is the filter.

Branch when you will never reuse the insert. You can still make an entry clone and stop archiving it after the destination is verified, or you can follow a one-fragment route if the mix you have is documented for that path. Archive the plasmid you will actually open next year, plus the sequence file. Do not archive three unnamed intermediate tubes.

Branch when the destination must be a low-copy or tightly regulated backbone because the insert is burdensome. The recombination class does not fix toxicity. It only moves the fragment. Choose the destination on purpose. An empty destination reaction, without entry clone, should not yield colonies on a sensitive strain. If it does, counterselection has already failed and the real plate cannot be trusted.

When the plate lies

Empty plate, and a control plasmid in the same sensitive cells also fails: competence, heat, or the antibiotic. Fix that before you blame the sites.

Empty plate, control plasmid fine: the att tails may be incomplete, the wrong mix was used for the direction, or the insert is toxic once it lands in that destination. An entry clone that grows, followed by an LR that does not, points at the destination context rather than at the BP chemistry.

Many colonies that still look like the parent on a digest: counterselection did not happen. Check that the strain is sensitive and that ccdB on the parent has not been lost in a previous sloppy propagation. Parents should live in the tolerant strain and nowhere else.

Right-sized clone, wrong protein later: the att scar or a frameshift at the junction. This is why the amino acids were supposed to be written in stage one. Sequence now. Do not repeat the LR on the same unverified entry.

Safety stays with the strain and the insert

The cassette is a cloning tool used so the wrong plasmid fails to form a colony. Keep that as a strain choice. Do not culture a sensitive strain "to see the protein," and do not write a purification around it. The insert you are moving may have its own hazard. Containment is your institution's decision. The WHO Laboratory biosafety manual, 4th edition is background, not permission. Antibiotic plates and recombinant waste follow the same committee.

Labels in a shared freezer

A tolerant strain and a sensitive strain look the same in a cryobox. In a busy room, boxes get rearranged and caps get wet. Write the strain genotype and the words ccdB parent or product on the vial. Transforming a destination parent into the tolerant strain and selecting the parent's marker will grow the parent happily, recombination or not. That tube will be mistaken for a finished clone if the label only says the gene name.

What to send when you ask for the mix or the fragment

State BP or LR, the donor and destination markers, whether you need a tolerant strain for the parents, the insert length, and whether the att sites sit in a protein frame. Ask whether a quotation is possible on the quote request. A junction read, once colonies exist, belongs with the Sanger sequencing enquiry reference. Send the map. A reaction class without the sites is not a specification.