Library transformation using Y190
Y190 = MATa gal4
gal80 his3 trp1-901 ade2-101 ura3-52 leu2-3,-112
+ URA3::GAL-->
lacZ, LYS2::GAL-->HIS3 cyhr
After subcloning
the gene of interest into pAS1-CYH or a similar GAL4 DNA-binding
domain fusion vector, transform this construct into Y190 selecting
for TRP1 in the case of pAS1. It should be noted that expression
of the fusion protein should be verified by western blotting
with anti HA antibodies available from BABCO if specific antisera
are not available for your protein. Occasionally, a bona fide
fusion is made and cannot be detected with HA (levels too
low, but sufficient for screening).
The resulting strain
should be checked for its growth properties on SC-His plates
containing differing concentrations of 3-AT (3-aminotriazole,
SIGMA, A8056) and on its ability to activate the lacZ reporter.
These tests should be carried out relative to strains carrying
pSE1112 alone. We have found that 3-AT concentrations of 25
mM to 50 mM are sufficient to select against pAS1 subclones
that fail to activate transcription on their own. If your
construct activates transcription alone, it cannot be used
in this assay. If it fails to activate transcription, like
most fusions, you may proceed to the library transformation
step. Also check out the RNA transformation protocol we have
included. It works a little more reproducibly in our hands.
1. Use a colony
or overnight to inoculate 200 mls SC-Trp and grow overnight
at 30oC. The use of a saturated overnight from the night before
ensures that the 200 ml culture will be fully grown the next
day.
2. Take OD A600
of above culture and inoculate 500 mls YEPD such that in ~2
generations
(3- 4 hrs) the
A600=0.5 to 0.8 . (Synthetic complete - Trp media is used
to select for pAS1 but YEPD gives best transformation efficiencies)
3. Harvest cells
at 5K for 10 minutes in a Sorvall centrifuge
4. Wash once with
distilled water ~100 mls and resuspend in 50 mls LiSORB and
incubate at 30oC for 15-30'.
5. Spin down as
above and resuspend in ~625 µl LiSORB. Hold on ice.
6. Prepare carrier
DNA mix.
Boil 200 µl
20 mg/ml sheared salmon sperm DNA for 7-10'
add 800 µl
LISORB (room temp, RT)
mix by pipeting
mixture up and down
cool to RT (ice
can be used,but care must be taken to ensure that the Temp.
does not go
lower than RT or
the mixture will gel
add 40µg
library DNA
7. Add ~100µl
of above DNA mix to 100 µl cells from step 6.
8. Incubate at
30oC for 30 minutes (optional)
9. To 100 µl
cells + DNA add 900 µl 40% PEG3350 in 100 mM LiAc/TE
and incubate at 30oC for 30'. Heat Shock at 42°C for 7
minutes. Plate 5 ul of cells to test transformation efficiency
on SC-Trp, Leu.
10. Recovery: Pool
cells and add to ~100ml SC--His,Trp, Leu liquid media, shake
at 30oC for 1-3 hours; harvest cells and resuspend in ~6 mls
of SC--His,Trp,Leu liquid media and plate ~300 µl per
150mm plate (SC--His,Trp, Leu+ 25 mm AT) or 50 mM 3-AT. (Plating
directly from the PEG also works but is more messy.)
11. Colonies that
grow after 3 to 5 days are then tested for b-galactosidase
activity using the X-Gal colony filter assay described in
the accompaning protocol. Blue colonies are taken for further
study, they can often be taken directly from the filters in
addition to the original plate.
Efficiency = ~5x104
to 105 colonies/µg cDNA library.
For determination
of efficiency plate 5 µl before and after recovery on
SC-Trp, Leu
LiSORB=100 mM LiOAC,
10 mM Tris pH 8, 1 mM EDTA, 1 M Sorbitol.
LiAcTE=100 mM LiOAC,
10 mM Tris pH 8, 1 mM EDTA
We have found that
total yeast RNA works more reproducibly as a carrier but it
is more work to prepare the RNA than DNA. That transformation
protocol is included as well. We usually place the transformation
mix in SC-His, Trp, Leu after the heat shock step and allow
it to recover for 3 hours. This allows the transformants to
be established and HIS3 transcription to be activated. The
cells at this stage can be pelleted and resuspended in a smaller
volume and plated directly or made 10% DMSO, and frozen at
-70°C. Cells in PEG seem more fragile and often die when
pelleted so the recovery step is useful. Cells lose less than
half their viability when frozen and can be stored indefinitely.
This is useful because they can be thawed and plated at a
optimal density for screening/selection at your leisure. This
protocol can be scaled up for 1 liter of cells or more.
The HIS3/3-AT selection
sometimes works as a good selection and sometimes looks like
more of an enrichment. We often see many micro colonies on
the original selection plates. Occasionally they are 1% of
the total Leu+Trp+ colonies. In most cases true positives
continue to grow into large colonies while the micro-colonies
seem to stop growing. The secondary screen for blues eliminates
these micro colonies. The majority of His+ blue colonies are
the large colonies that grow out. An enrichment of 100-fold
is very useful because it allows you to screen 100 times as
many colonies on a single plate so a whole library can be
screened in only 20 large plates. We have also developed a
GAL-->URA3 selection system which requires higher levels
of 2 hybrid activated transcription than the His selection.
That strain is available upon request.
Included in the
kit is pSE1111 (SNF4 fused to the activation domain in pSE1107)
and pSE1112 (SNF1 fused to the DNA-binding domain of GAL4
in pAS1). These are both inserted as BamH1 fragments. These
can serve as a positive control for X-gal staining and 3-AT
resistance.
It should be noted
that pAS1 alone can activate lacZ weakly. Therefore it is
not a good negative control. pSE1112 is a better negative
control. The weak activation of pAS1 appears to go away when
genes are cloned into it. We do not understand why it is weakly
activating alone, but we think it is likely due to sequences
beyond the polylinker and which are of no consequence once
cDNAs are cloned into it.
We have recently
made improvements in the system. We have constructed a Y153
derivative, Y190, that is resistant to cycloheximide (2.5
ug/ml) due to a mutation in the CYH2 gene. Y190 is now being
supplied with the kit. This is a recessive drug resistance.
When a plasmid carrying the wild type CYH2 gene is in the
strain, cells become sensitive to cycloheximide. We have constructed
a pAS1-CYH plasmid that contains the CYH2 gene. After a positive
clone has been selected in the system, loss of the pAS1-CYH
plasmid can be achieved by streaking on SC-Leu 2.5 ug/ml cycloheximide
media. It is probably a good idea to streak the colonies out
on SC-Leu before streaking on cycloheximide media to allow
plasmid loss and dilution of the CYH2 gene product. However,
it does work streaking directly from SC-TrpLeu. The colonies
that grow should be Trp-, but they should be checked for loss
of the TRP marker, just to be safe and avoid CYH2 gene conversion
events. This plasmid loss allows one to check for plasmid
dependency of lacZ activation as well as generating a strain
that contains only the library plasmid, facilitating plasmid
recovery into bacteria.
False positives
do occur that appear to be dependent upon both plasmids. This
is a reoccurring problem of the 2-hybrid system. You should
give some thought to secondary criteria for distinguishing
a true positive. False positives will light up many non-related
pAS1 fusions, that is one of the definitions of a false positive.
To eliminate false positives, we usually generate a strain
that has lost the pAS plasmid but retains the library plasmid.
This is done using the CYH trick or growing in YEPD, plating
on SC-Leu and replica plating to SC-Trp to look for loss of
pAS1. This strain (Leu+Trp-) is mated to a strain,Y187 = MATa
gal4 gal80 his3 trp1-901 ade2-101 ura3-52 leu2-3,-112 URA3::GAL-->lacZ,
that is of the opposite mating type but contains different
unrelated fusions in pAS1(Leu-Trp+) such as SNF1(pSE1112),
lamin, p53 (Lamin and p53 are provided by Stan Fields). Once
your Y190 strain of interest has lost the pAS1-fusion plasmid
(by plasmid loss or cycloheximide selection) it is Trp-Leu+
and can be mated to Y187 containing pAS1-X. Diploids can be
selected by growth on SC-TrpLeu and then immediately tested
for b-galactosidase activity in the filter screen assay. Colonies
that activate lacZ expression significantly above background
levels (pSE1112) probably contain Leu plasmids encoding false
positives that non-specifically activate your fusion and should
be disgarded.
We have used several
different tests to detect in vitro binding of positives. One
is to make a PCR primer to the library plasmid that has a
T7 promoter placed in an appropriate position to place the
insert of the library plasmid under T7 control. The PCR-derived
fragments can then be directly added to a coupled transcription-translation
system (TnT from Promega) and radiolabeled protein made. We
usually add 6 ul of a robust 30 cycle PCR reaction to 25 ul
of the translation mix(TnT). 5 ul of this reaction mixture
run on a SDS gel gives a readily detectible signal on an overnight
exposure. This tells you the size of the fused protein and
can be used to detect interaction in vitro with a GST-bait
fusion.
The sequences of
the PCR primers we have used successfully are:
1)
TAA TAC GAC TCA
CTA TAG GGA GAC CAC ATG GAT GAT GTA TAT AAC TAT CTA TTC
T7 Promoter Met
Gal4 Activation Domain
21 AA before the
BglII site
2)
CTA CCA GAA TTC
GGC ATG CCG GTA GAG GTG TGG TCA
In the ADH TerminatorA
second test is to IP your protein out of yeast extracts and
then use antibodies to the activation domain of GAL4 to detect
binding of the fusion protein. Our GAL4 activation domain
antibodies are currently being tested for this purpose. Unfortunately,
they do not IP presently.
A third method
is to switch the bait and prey in their respective plasmids,
i.e. take the library insert out of pACT and insert it into
pAS1, and place the original pAS1 insert into pACT. The majority
of false positives will not interact in this test. It should
be noted that some true positives may not activate for structural
reasons, so only a positive result can be trusted. We have
recently placed the pAS1 polylinker into pACT to facilitate
this transfer, creating pACT2.
For general yeast
protocols we recommend the Methods in Enzymology Vol. 194
"Guide to Yeast Genetics and Molecular Biology"
by Guthrie and Fink and the Red book.
As with all protocols,
this is probably not optimal and I encourage you to add variations.
Let me know if any new variation makes for a simpler or more
efficient protocol. For example, several people have said
that direct plating of the transformation mix in peg onto
selective plates workes well for them. It is probably worth
a try.
Our lab, Wade Harper's
lab, and Stan Fields' lab have all successfully isolated clones
from our l ACT libraries using the HIS3 selection of Y153
and Y190 and have each contributed to these protocols. Good
luck.
|
