With the availability of the complete yeast genomic sequence, techniques which
allow the rapid functional analysis of genes of interest are of increasing
importance. Here we report a technique which allows the initial
characterisation of genes of interest, through the construction of
conditionally expressed mutations for functional analyses and the generation of
epitope-tagged fusion proteins for immuno-localisation and immuno-purification, entirely by PCR.
A PCR-based technique for the creation of chromosomal gene disruptions has been
described (
1
). We extend this technique to allow the rapid creation of conditionally
expressed alleles (
GAL
mutants) and the synthesis of proteins fused to epitope tags. The technique
relies on the PCR amplification of HIS3-p
GAL
or
HIS3-
p
GAL
-TAG cassettes using two primers containing flanking sequences specific to
the target gene followed by the transformation of the PCR product into a
his3
-
strain.
The regions which are complementary to the pTL vectors are underlined. The EMBL
accession numbers of genes
SSB1
,
RRP3
and
RRP41
are M17244, YHRO65c and X82775 respectively.
Four vectors have been designed and tested (Fig.
1
). In these the
HIS3
marker is flanked either only by the
GAL10
promoter (vector pTL26) or by the
GAL10
promoter fused to different epitope tags. The epitope tag sequences are 2* Protein A, 3* c-myc and His
8
(vectors pTL27, pTL28 and pTL32 respectively; see legend of Figure
1
for full description). As an example of the use of these vectors, the
construction of a
GAL
-regulated, ProtA::Ssb1p fusion is outlined in Figure
2
. The
SSB1
flanking sequences present on the 5' and 3' primers target the chromosomal integration of the PCR construct
upstream of, and in frame with, the initiator AUG of
SSB1
(Fig.
2
).
SSB1
is a non-essential gene and transformants were directly plated on 2% glucose
minimal medium lacking histidine (SD-his). For the essential genes
RRP3
(C. L. O'Day, F. Chavanikamannil and J. Ableson, submitted for publication) and
RRP41
(P.
Mitchell, D. Lafontaine and D. Tollervey, unpublished), transformants were
plated under permissive conditions for
GAL
transcription; minimal medium lacking histidine and containing 2% sucrose, 2%
raffinose and 2% galactose. Strains were then streaked on SD-his to check for the effects of depletion.
The expression level of the ProtA::Ssb1p fusion was tested on medium lacking
histidine and containing 4% sucrose + 2% galactose (Fig.
3
, lane 4), 2% sucrose + 2% raffinose + 2% galactose (Fig.
3
, lane 5) and 4% raffinose + 2% galactose (Fig.
3
, lane 6). Galactose functions as a non metabolizable inducer in strains YDL401
and YDL402 since they carry the
gal
[Delta]
108
mutation (
3
) (Table
3
). These strains also carry a mutation in the galactose permease gene (
gal2
).
The effects of galactose addition to medium containing 2% sucrose and 2%
raffinose was tested for the GALHis
8
::SSB1 strain. As expected, the presence of galactose in the medium was found to
have little effect on the level of expression of the fusion protein (data not
shown). A potential problem with the use of
GAL
-regulated constructs is that many proteins are heavily over-expressed when their genes are transcribed from induced GAL
promoters. This, for example, can make the analysis of the sub-cellular localisation of the fusion protein unreliable. Ssb1p is an snoRNP
protein and the level of ProtA::Ssb1p was compared with the level of expression
of another snoRNP protein, ProtA::Nop1p, expressed under the control of its own
promoter (Fig.
3
, lane 1). In these strains the level of ProtA::Ssb1p expressed in medium
containing 4% sucrose + 2% galactose is similar to that of ProtA::Nop1p,
suggesting that its expression is in the same range as expression of endogenous
Ssb1p. During growth on medium containing 2% glucose (Fig.
3
, lane 2) the level of ProtA::Ssb1p was undetectable.
Many
GAL
-regulated mutants show incomplete growth inhibition on glucose medium due
to residual transcription (
4
). The effects of the transcriptional repression can be enhanced at the
translational level through modification of the context of the initiator AUG or
by the introduction of an additional, out of frame upstream AUG sequence. In
the system reported here, such mutants can simply be made by altering the
sequence of the 3' primer (primer 3' in Fig.
2
). The URA3 gene of
Kluyveromyces lactis
is functionally homologous to the
S.cerevisiae URA3
gene and fully complements
ura3
-
strains, but has sufficient sequence divergence to prevent genetic
recombination (
5
). To allow epitope-tagging of more than one protein in the same strain, we are currently
constructing vectors based on the
K.lactis URA3
gene. Templates for the construction of C-terminal fusions are also in preparation.
The ease with which
GAL
-regulated and epitope-tagged alleles of genes of interest can be constructed using this
strategy allows initial functional analyses of the effects of genetic depletion
to be carried out using tagged alleles. This allows the degree of genetic
depletion to be followed at the protein level in the absence of specific
antibodies (Fig.
3
). The construction of such alleles by conventional techniques typically
involves several cloning steps and generally generates only plasmid-borne alleles. In the case of essential genes, these must be transformed
into heterozygous diploid strains and suitable haploid progeny recovered after
sporulation. In contrast, the technique reported here allows mutant alleles of
essential genes such as
RRP3
and
RRP41,
to be simply constructed in haploid strains.
ACKNOWLEDGEMENTS
We are very grateful to M. Johnston for strain YNN72 and to R. Jansen and E.
Hurt for the ProtA-Nop1p strain. We also thank C. L. O'Day, F. Chavanikamannil and J. Ableson
for providing us with data prior to publication and members of the laboratory
for critical reading of the manuscript. D.L. is the recipient of an EMBO long-term fellowship.
REFERENCES
1 Baudin, A., Ozier-Kalogeropoulos, O., Denouel, A., Lacroute, F. and Cullin, C. (1993) Nucleic Acids Res., 21, 3329-3330.MEDLINE Abstract
2 Gietz, D., St. Jean, A., Woods, R. A. and Schiestl, R. H. (1992) Nucleic Acids Res., 20, 1425.MEDLINE Abstract
3 St John, T. P., Scherer, S., Mc Donell, M. W. and Davis, R. W. (1981) J. Mol. Biol., 152, 317-334.
4 Lafontaine, D., Vandenhaute, J. and Tollervey, D. (1995) Genes Dev., 9, 2470-2481.MEDLINE Abstract
5 Längle-Rouault, F. and Jacobs, E. (1995) Nucleic Acids Res., 23, 3079-3081.
6 Sikorski, R. S. and Hieter, P. (1989) Genetics, 122, 19-27.
7 Moks, T., Abrahamsen, L., Österlöf, B., Josephson, S., Östling, M., Enfors, S.O., Persson, I., Nilsson, B. and Uhlén, M. (1987) Bio/Technology, 5, 379-382.
8 Schimmang, T., Tollervey, D., Kern, H., Frank, R. and Hurt, E. C. (1989) EMBO J., 8, 4015-4024.MEDLINE Abstract
9 Clark, M. W., Yip, M. L. R., Campbell, J. and Abelson, J. (1990) J. Cell Biol., 11, 1741-1751.
