A. Screening of Cosmid Libraries
1. Preparation of Nitrocellulose Filters
i) Wash the desired number of nitrocellulose filters (we use 82 mm diameter
Schleicher and Schuell) in two changes of distilled water. This removes detergent and
other impurities which can retard colony growth and/or reduce the titer.
ii) Stack the filters, separating them with sheets of Whatman 3MM paper. Put two sheets of blotter paper on each end of the stack and wet the blotter sheets with distilled water. Package the resulting stack in aluminum foil and autoclave 20 minutes. This helps prevent shrinkage of the filters from occurring during the hybridization and washing process, which can make aligning signals with colonies on the master plates difficult.
iii) Number the filters and place face down on 30µg/ml ampicillin plates. This equilibrates the filters with the media. After a few minutes, flip the filters.
iv) Plate the titered library at a density of 2-4000 colonies per plate. At 37°C it typically takes about 10 hr for the colonies to reach the proper size for replica plating.
2. Replica Plating
i) Allow colonies to grow to about 0.5 mm in diameter. Larger colonies are more
likely to smear, smaller ones don't always transfer evenly.
ii) Pre-equilibrate filters to be used as replicas as in step 3 above. Up to two replicas can be reliably made with one master filter. Additional replicas can be made from a master after the colonies are re-grown.
iii) Replica plate by setting the pre-equilibrated filters onto a couple sheets of the sterile Whatman 3MM paper (generated in step 2 under "Preparation of Nitrocellulose Filters"), laying the master filter on top of the replica, and putting your weight on the filters using a replica plating tool. Use a 25 ga needle to put holes through the filters while they are stuck together. This will allow alignment of signals with colonies later. It is a good idea to press on the filters with a pair of forceps near where the holes are being made with the needle to prevent the filters from slipping against each other.
iv) Grow the colonies back up. The masters usually take 3-5 hr at 37°C, the replicas anywhere from 4-8 hr. The colonies can then be lysed directly or the cosmid DNA amplified by placing the filters on 100µg/ml chloramphenicol plates for 16 hr. Chloramphenicol amplification generally improves signal to noise.
3. Lysis of Colonies, Hybridization and Washing
i) Lyse the colonies by laying the filters on blotter saturated with 10% SDS 3
minutes, then 0.5M NaOH/1.5M NaCl 5-10 minutes, followed by 1M Tris/3M NaCl,
pH 7.4 for 5-10 minutes. The SDS step is reputed to sharpen signals.
ii) Lay the filters on top of, and then submerge them in 2xSSCP/0.1% SDS. Wipe the colony debris off the submerged filter gently with a folded Kimwipe. Excessive "polishing" will remove the cosmid DNA, the right amount will reduce background. Just wipe hard enough to remove the debris.
iii) Bake the filters, separated by sheets of Whatman 3MM, under vacuum at 80 degrees Celsius for 2-4hr. Longer baking times help reduce DNA loss from filters that are to be repeatedly hybridized, washed, and boiled.
iv) Prehybridize the filters in 3xSSCP/10xDenhardts/250µg/ml carrier DNA/0.1% SDS. A 1hr prehybridization is usually sufficient, unless an unusually large amount of probe is to be added or the hybridization is to last unusually long.
v) Hybridize the probe to the filters in the same type of buffer as used for the prehybridization. We usually hybridize nick translated 32P probes at a concentration of about 2x105 cpm Cherenkov per ml for 12-16 hr.
vi) Wash the filters 5 minutes in 2xSSCP/0.1% SDS at room temperature, then 15 minutes in 2xSSCP/0.1% SDS at 65oC, then in 2xSSCP/0.1% SDS until background cannot be heard on the filters with a monitor. We usually let this last wash go about 2hr. Sandwich the filters between two pieces of Saran wrap and apply to X-ray film.
4. Isolation of positive colonies
i) Align the autoradiogram with the filters and mark the positions of the holes.
ii) Ethanol wipe a piece of cellophane (or something similar). Spot about 0.2 ml of L-broth onto this in a number of droplets. Lay the master filter (with colonies) onto the L-broth, set the cellophane on the autoradiogram. Align the holes, sometimes background on the autoradiogram helps in the alignment. You never thought background could be useful, right? Get the alignment as precise as possible.
iii) If you're lucky you'll have one good candidate for the signal. If not, pick each colony individually with a sterile toothpick and transfer to a plate in a grid pattern. If there are double colonies, etc. pick these and streak them to get single colonies which can be put in a grid. Grow up the grid, lift the colonies from the plate using a nitrocellulose filter (make holes in the filter with an india ink filled hypodermic needle before removing the filter to help align it with the plate later) and treat the filter as above.
iv) Cosmid DNA from the resulting positive colonies can be isolated using a number of methods. The most rapid I've used is to make a boiling miniprep (see Quigley and Holmes, Analyt. Biochem. 114, 193 (1981)) of 1.5ml of a saturated culture (I use L-broth with 100µg/ml ampicillin). This generally yields roughly 5-10 µg of DNA. I've heard that alkaline lysis miniprep methods can give better yields but after determining that the cosmid is what I want using this miniprep DNA I just go ahead and grow a 500ml or 1 liter culture and purify the cosmid by alkaline lysis followed by a CsCl gradient.
B. Isolation of cosmid end fragments by subcloning
The idea here is to isolate cosmid end fragments as rapidly as possible when
chromosome walking. Many cosmid libraries are constructed using vectors where end
fragments from either end of the cosmid can be easily subcloned. For example, the
cosPneo vector has a polylinker containing a BamH1 site flanked on one side by an
Eco R1 site and on the other side by an Xba 1 site. If the cosmid library in this
example is made by inserting genomic DNA into the Bam H1 site, the two ends of the
cosmid can be isolated by intramolecular ligation of complete Xba 1 and Eco R1
digests.
Miniprep DNA is sufficient for this procedure as long as you can digest it to
completion. Digest 0.5-1.0 µg of cosmid DNA in a 50 µl reaction. Check half of the
digest on a gel to be sure it is complete. Heat kill the enzyme if possible. Otherwise,
bring the reaction to 50 µl and add diethylpyrocarbonate (0.5 µl of 10% DEP in
EtOH), heat to 37° for 20 min. Add 2.5 µl 1M Tris (pH 8.0) and heat to 65°C about
10 min. to ensure that the DEP is inactivated. The digest can be run over a P10 spin
column at this point to ensure high cloning efficiency, but this is usually unnecessary.
In any case, bring the reaction to 100 µl and add ligase buffer and 100 units T4 DNA
ligase. Transform competent cells as usual. The resulting endclones can be hybridized
to cosmids from the previous step in the walk to determine which one extends the
walk. They can also be used for in situ hybridization to polytene chromosomes to get
an accurate picture of the progress of the walk.