The library is distributed as 18 individual pools in the form of DNA. You will be sent about a microgram of each. Transform a suitable amount into E. coli (use any kanamycin- and tetracycline-sensitive strain suitable for making plasmid preps). Select transformants with 40 ug/ml kanamycin and/or 3ug/ml tetracycline (allow at least an hour for expression following transformation). Obtain 50,000 colonies for each pool. Elute colonies from plates in LB; make a -70^o C stock of this eluate. Dilute eluate into LB plus antibiotic to give a culture with an almost saturated density. Grow at 37^o C for a few hours. Make miniprep or midiprep DNA.

OVERVIEW: Mutagenized DNA from the library is excised from the bacterial vector. It is then transformed into a ura3 strain of yeast. This procedure is outlined in this figure . The best strategy is to screen a few thousand transformants from each pool. Use of a circle-zero strain will prevent recovery of insertions in the 2-micron plasmid. Screening 30, 000 transformants should give you 95% coverage of the yeast genome.

To minimize double integrants, transformations should contain the lowest amount of DNA practicable. We therefore recommend that a pilot experiment be performed to determine transformation efficiency of the strain, and conditions then be scaled up as appropriate. The pilot protocol given below uses a modified version of the method of Chen et al. (1992). . You should use whatever transformation protocol works best in your hands.

1. Plasmid DNA from pools of the mTn-3xHA/GFP-mutagenized genomic library is digested with Not I. A 2.1-kb band from the vector should be very apparent, together with a broad band in the 8-kb region, representing inserts. Because sized genomic DNA was used to make the library, the insert bands are not very heterogeneous in size.
2. A 10-ml culture of the yeast host strain is grown to a density of 10⁷ cells/ml (O.D. 600 of 1). Use of such logarithmically-dividing cultures increases transformation efficiency.
3. Cells are pelleted and washed once with 5 volumes of One Step buffer (0.2M LiAc, 40% PEG 4000, 100 mM [[beta]]-mercaptoethanol). This wash is especially important when culture volumes are increased.
4. Cells are resuspended in 1 ml of One Step buffer containing 1 mg of denatured salmon sperm DNA. 100 ul aliquots of this suspension are then added to tubes containing from 0.1 to 1 ug of Not I digested plasmid DNA.
5. Tubes are vortexed to mix the contents thoroughly, then incubated at 45^o C for 30 minutes.
6. Cells are pelleted and resuspended in 400 ul of SC-ura. 200 ul is plated onto SC-ura medium. Plates are incubated at 30^o C for 3 to 4 days.

Screening for in-frame GFP fusions in yeast

We have not done assays of GFP activity in yeast.

See Niedenthal et al (1996) for their methods.

Analyzing GFP fusion protein localization in yeast

We tested mTn-3xHA/GFP by mutagenesis of BDF1 , which encodes a chromatin-associated protein. We grew individual bdf1 ::mTn-3xHA/GFP transformants to a density of 10⁷ cells/ml in SC-ura. The last four hours of growth were at room temperature, to allow formation of the GFP chromophore. Then we examined cells directly using a Leitz microscopy with a system 13 filter (this may not be optimal). In 4 of 38 transformants, we saw green fluorescence of the nucleus. Fixation and spheroplasting of the cells improved the signal-to-noise ratio.

To determine the site of transposon insertion, genomic DNA imediately adjacent to the transposon sequences must be rescued. We have not yet constructed a rescue vector for mTn-3XHA/GFP. If demand for the library is high we will construct one. Otherwise, we will be happy to provide reagents and information to another laboratory who wishes to construct it. Inverse PCR on genomic DNA could be used to recover the site of insertion. Alternatively, Carl Friddle has developed a 'vectorette PCR' rescue protocol for lacZ-based transposons. I have transcribed Carl's protocol and modified the suggested enzymes and primers, to make it suitable for vectorette PCR of the mTn-3xHA-based transposons .

When transposon insertion has created an in-frame fusion to GFP in the gene of interest, the transposon can be excized to leave a 274 bp insertion (sequence given below) containing the 3xHA tag. With the 5 base pair duplication caused by transposon insertion, this gives an in-frame 93-amino acid insertion in the protein. The popout event is mediated by cre recombinase and requires induction of the GAL1-10 promoter on galactose. Our strains grow poorly on galactose but give 80 to 100% popouts.

The HA triple tag can be detected by mouse monoclonal antibodies 12CA5 (Boehringer) or MMS101R (BAbCo, Richmond, California). These antibody recognise cross-reacting yeast proteins of about 55kD or110kD, respectively, and can give a spotty background on immunofluorescence. Despite this drawback, the 3xHA tag has been used extensively and successfully in yeast. A rabbit polyclonal antisera is also available (101c500; BabCo) but this was less reactive in the one instance we tried. Protocols for yeast immunofluorescence can be found here , or in Methods in Enzymology 194 (1991).

1. Transform strain with pB227/GAL-cre , selecting on SC-leu.
2. To derepress the GAL promoter, inoculate transformants into 2 mls SC-ura-leu with 2% raffinose as carbon source and grow to saturation.
3. Dilute 1/100 into SC-leu with 2% galactose as carbon source (control: SC-leu with 2% glucose as carbon source). Grow for 2 days (some strains induce without growing).
4. If grown, dilute 1/100. Spot a 10ul drop onto an FOA plate and streak it for singles (non-quantitative approach!). Or plate dilutions onto SC media and replica to identify ura- colonies. The induced cultures should give 100x more Ura- cells than the control.
5. PCR primers designed using the sequence given below can be used to determine position of the tag. The IR elements and palindromic loxR region should be avoided.

N.B. When tagging essential genes, the original strain transformed should obviously be diploid. You can dissect the popped-out version to see if the tagged gene is functional. Only believe a tag is lethal if it is complemented by the wild-type gene, and if several popout events give the same phenotype.

Sequence of HAT tag (3xHA):

TR in upper case. loxR in bold .

GGGGTCTGAC GCTCAGTGGA ACGAAAACTC ACGTTAAGgc ggccattgaa ggtagaagag aaaatttgta cttccaaaga aagaaggccg ctatcgcttc ggataactcc tgctatacga agttat gggc ggccgtttac ccatacgatg ttcctgacta tgcgggctat ccctatgacg tcccggacta tgcaggatcc tatccatatg acgttccaga ttacgctccg gccgcCCTTA ACGTGAGTTT TCGTTCCACT GAGCGTCAGA CCCC

GenBank accession

• mTn-3xHA/GFP is U54830

Antibiotics used: