Stabilization of yeast artificial chromosome clones in a
rad54-3
recombination-deficient host strain
Stabilization of yeast artificial chromosome clones in a rad54-3 recombination-deficient host strain
Yunzheng
Le
+
and
Melanie J.
Dobson*
Department of Biochemistry, Faculty of Medicine, Sir Charles Tupper Medical
Building, Dalhousie University,
Halifax
, Nova Scotia B3H 4H7,
Canada
Received November 1, 1996;
Revised and Accepted January 28, 1997
ABSTRACT
The cloning and propagation of large fragments of DNA on yeast artificial
chromosomes (YACs) has become a routine and valuable technique in genome
analysis. Unfortunately, many YAC clones have been found to undergo
rearrangements or deletions during the cloning process. The frequency of
transformation-associated alterations and mitotic instability can be reduced in a
homologous recombination-deficient yeast host strain such as a
rad52
mutant.
RAD52
is one member of an epistatic group of genes required for the recombinational
repair of double-strand breaks in DNA.
rad52
mutants grow more slowly and transform less efficiently than
RAD
+ strains and are therefore not ideal hosts for YAC library construction. We
have investigated the ability of both null and temperature-sensitive alleles of
RAD54
, another member of the
RAD52
epistasis group, to prevent rearrangements of human YAC clones containing
tandemly repeated DNA sequences. Our results show that the temperature-sensitive
rad54-3
allele blocks mitotic recombination between tandemly repeated
DYZ3
satellite sequences and significantly stabilizes a human
DYZ5
satellite-containing YAC clone. Yeast carrying the
rad54-3
mutation can undergo meiosis, have growth and transformation rates comparable
with
RAD
+ strains, and therefore represent improved YAC cloning hosts.
INTRODUCTION
The development of yeast artificial chromosome (YAC) technology has been a major advance in genome research. YACs have been used to clone DNA
inserts >1.5 Mbp, to isolate DNA segments that were unclonable in conventional
bacterial systems, to physically map megabase scale DNA fragments and to
facilitate positional cloning of genes (for reviews see
1
-
3
). This technology also provides a basis for the development of mammalian artificial chromosomes (
4
). However, there have been problems associated with YACs. Many YAC inserts have
been found to contain deletions or rearrangements and the frequency of chimeric
clones observed has been as high as 59% in some libraries (
5
). Recent studies have suggested that the rearrangements and chimeras arise from
recombination during or after transformation of the yeast host in YAC library
construction (
6
). The major group of genes involved in recombination in yeast belong to the
RAD52
epistasis group and were originally identified by their requirement for the repair of ionizing radiation-induced DNA damage (
7
). An alternative pathway for mitotic recombination involves the product of the
RAD1
gene, which is required for excision repair of UV-induced DNA damage and has also been shown to play a role in recombination
between repeated DNA sequences, perhaps by mediating DNA sister strand exchange (
8
). To reduce the frequency of chimeric YAC clones, yeast DNA repair mutants such as
rad1
or
rad52
have been used as alternative yeast hosts for YAC cloning (
9
). While a
rad1
null mutation did not reduce the instability of YAC clones, a
rad52
mutation was shown to significantly reduce the level of chimeric clones and the frequency of clone
rearrangement (
9
,
10
). Unfortunately, a
rad52
null mutation has pleiotropic effects, reducing both growth rate and transformability of the
yeast, making it difficult to construct libraries in these strains.
To explore whether other homologous recombination-deficient mutants in the
rad52
epistasis group might provide better features for YAC cloning, we have
introduced both null and temperature-sensitive alleles of the
RAD54
gene (
11
,
12
) into strains carrying the genetic markers required for selection of YACs. Yeast carrying
rad54
mutations are phenotypically similar to
rad52
mutants, being highly sensitive to ionizing radiation, slightly sensitive to UV
radiation and displaying reduced rates of spontaneous and induced mitotic
recombination (
7
,
12
).
To determine the utility of these alternative yeast host strains, we have
assessed their ability to reduce the frequency of rearrangements of two human
YAC clones containing tandemly repeated sequences. YAC clones containing these
repeats are unstable in
RAD
+
recombination-proficient yeast strains, giving rise to YACs with smaller inserts due to
loss of repeat units in the tandem array (
9
). The YAC clone YOR 1B9F contains a
DYZ3
(alphoid satellite DNA) repeat array originating from the centromeric region of
the human Y chromosome (
9
). The
DYZ3
unit repeat size is ~170 bp, with a higher order repeat of ~5.7 kb found as a tandem array block that on the Y chromosome varies
in size from ~250 to ~1400 kb between different individuals (
13
). The
DYZ3
-containing YAC clone YOR 1B9F is extremely unstable in
RAD
+
yeast and only partially stabilized in a
rad52
host strain (
9
). A second human YAC clone analysed in this study, YOR 2B6H, contains a
different Y chromosome tandem repeat,
DYZ5
(
9
).
DYZ5
has a unit repeat size of ~20 kb and is found in tandem arrays of between ~300 and ~800 kb in length in the proximal region of the human Y chromosome
short arm (
14
). The
DYZ5
-containing YAC clone YOR 2B6H is mitotically unstable in
RAD
+
,
rad1
,
rad52
and
rad1/rad52
yeast hosts (
9
). We have asked whether
rad54
mutations can stabilize YACs containing either of these human Y chromosome
tandemly repeated DNA sequences.
Our results show that the temperature-sensitive
rad54-3
allele blocks mitotic recombination between tandemly repeated
DYZ3
sequences and significantly stabilizes
DYZ5
-containing YAC clones. In addition, yeast carrying the
rad54-3
mutation have growth and transformation rates comparable with
RAD
+
strains, can undergo normal meiosis and therefore represent improved YAC cloning hosts.
MATERIALS AND METHODS
Yeast strains
Genotypes of yeast strains used in this study are shown in Table
1
. Yeast strains were constructed and propagated by standard techniques (
15
). All
rad54
alleles were introduced into strains by meiotic recombination, with haploid
progeny being scored for the segregation of radiation sensitivity, amino acid
requirements and mating type. Deletion of the
RAD52
gene in AP1 was achieved by replacing a 78 bp
Kpn
I-
Bgl
II fragment within the
RAD52
open reading frame on plasmid pSM21 (provided by D.Schild) with a 2.0 kb
Kpn
I-
Bam
HI fragment containing the
LEU2
gene. The
Bam
HI fragment containing the disrupted
RAD52
gene was then used to transform DBY747-ade2[Delta] to leucine prototrophy. Radiation sensitivity of the strains and
all transformants was confirmed by irradiation with a 40 krad X-ray dose (
7
). Yeast carrying the
rad54-3
mutation were incubated at 37oC after irradiation, since this allele confers a temperature-sensitive phenotype; highly X-ray sensitive at 37oC while only modestly so at 23oC (
16
). Yeast were routinely grown at 30oC in liquid or on solid SC medium containing 0.67% yeast nitrogen base
without amino acids, 2.0% dextrose, 1.0% casamino acids, 20 [mu]g/ml adenine to select for the presence of YACs. Media were solidified with
2.0% DIFCO agar.
YAC clones
The human YAC clones YOR 1B9F and YOR 2B6H in AB1380 (
9
) were gifts from Chris Tyler-Smith (Department of Biochemistry, University of Oxford, Oxford, UK). The original AB1380 transformants
carrying these clones display multiple YAC-hybridizing bands in a Southern blot analysis due to the instability of the tandem
repeats in a
RAD
+
strain (
9
). To ensure that the appearance of multiple sizes of YACs after transfer to a
new host was not merely the consequence of introducing multiple DNA molecules
into the yeast during transformation, individual colonies of AB1380 containing
YOR 1B9F were screened to find one containing primarily a single large sized YAC DNA. YOR 1B9F/8, containing predominantly a YAC of 140 kb (Fig.
1
, lane 2) corresponding to the largest observed hybridization signal of YACs
prepared from YOR 1B9F (Fig.
1
, lane 3), was used as the source of
DYZ3
-containing YAC DNA for this study. Individual colonies of AB1380
containing YOR 2B6H were screened to identify one containing a single large
sized YAC clone. YOR 2B6H/8, containing mainly a 190 kb YAC (Fig.
2
, lane 2) corresponding to the largest hybridization signal of YOR 2B6H (Fig.
2
, lane 3), was used as the source of
DYZ5
-containing YAC DNA for this study.
Meiotic crosses
A
rad54-3
yeast strain, MD62/8, stably transformed with YOR 1B9F/8 (Fig.
1
, lane 17) was mated with both a recombination-proficient
RAD
+
strain, DBY746, and a
rad54
null, MD61/5. Diploids were selected on minimal medium containing 0.67% yeast
nitrogen base without amino acids, 2.0% dextrose or were picked directly by
micromanipulation. Diploids were induced to sporulate by overnight growth in 1%
yeast extract, 2% peptone, 1% potassium acetate followed by 2 days in 2.0%
potassium acetate at 30oC. Tetrads were picked by micromanipulation and scored for presence of the
YAC, for radiation sensitivity and for segregation of other markers in the
cross.
Yeast transformation
YACs were transferred into new host strains by spheroplast transformation using
total yeast DNA prepared in agarose plugs and melted prior to the transformation as described by McCormick
et al.
(
17
), with the following alteration: spheroplasts were prepared by harvesting cells
at the lower density of 1.0-1.5 * 10
7
cells/ml, washing in 1.0 M sorbitol and incubation for 1 h at 30oC in 1.0 M sorbitol, 2% glusulase (Dupont) with gentle shaking.
Transformation frequencies were calculated based on the number of colonies
formed on SC medium lacking both uracil and tryptophan when 1.0 [mu]g of a 27 kb circularized YAC vector was used to transform a 100 [mu]l aliquot of competent spheroplasts (~7.5 * 10
7
spheroplasts).
Transformation efficiencies of strains used in this study
Strain
RAD
+
RAD
+
rad52
[Delta]
rad54
[Delta]
rad54
[Delta]
rad54-3
(DBY747-ade2[Delta])
(W303/a)
(AP1)
(MD61/5)
(MD61/17)
(MD62/8)
Transformants
a
per [mu]g DNA
11300 +- 500
7627 +- 2364
211 +- 60
382 +- 201
354 +- 188
12453 +- 1836
a
Transformation efficiency represents the average of at least three independent
transformations. See Materials and Methods.
Manipulation and analysis of YACs
Yeast chromosomal DNA in low melting point agarose plugs was prepared by the
lithium dodecyl sulphate method (
18
). Pulsed field gel electrophoresis (PFGE) was carried out on a CHEF apparatus (BioRad) in 1% agarose gels in 0.5* TBE at 15oC for 16 h at 6 V/cm, with a switch time of 10 s ramped to 40 s. Chromosome I
in AB1380 was ~230 kb, whereas in MD62/8 (
rad54-3
) it was ~250 kb. This difference allowed the two strains to be readily distinguished on PFGE. The DNA was transferred onto a nylon membrane (Zeta-Probe; BioRad), which was hybridized in 6* SSPE, 5* Denhardt's, 1% SDS, 100 [mu]g/ml sonicated denatured salmon sperm DNA at 65oC for 16 h (1* SSPE = 150 mM sodium chloride, 50 mM sodium
phosphate, 1 mM EDTA). Membranes were then washed twice for 30 min at 65oC in 2* SSC, 0.1% SDS and once in 0.1* SSC, 0.1% SDS before being set up for autoradiography (1* SSC = 150 mM sodium chloride, 15 mM sodium citrate).
The DNA probe used in all hybridizations to detect YAC sequences was
32
P-labeled
Bam
HI-digested pBR322, which recognizes the vector sequences on both arms of all YAC clones used
in this study. Probes were labeled using a random priming kit from Boehringer according to the manufacturers instructions. Autoradiograms and gel photographs were scanned and prepared for figures using Ofoto version 2.0, Adobe Photoshop 5.0 and Aldus Pagemaker 5.0 software.
RESULTS
Transformation efficiency of
rad
strains
In preliminary experiments, yeast carrying a
rad54
null mutation,
rad54
[Delta]::
LEU2
, were tested for their ability to stably maintain the human
DYZ3
-containing YAC clone YOR 1B9F. The YAC clone did not show any instability
when transferred into this genetic background (data not shown), which indicated
that
rad54
mutants might be good candidates for YAC cloning hosts. An equally important
consideration in YAC library construction is the need to transform the yeast
host at a reasonable level of efficiency. To assess this parameter, the
transformation efficiencies of two
rad54
null strains, the temperature-sensitive
rad54-3
mutant and a
rad52
null mutant, were measured and compared with those of the two
RAD
+
parents used in construction of the
rad
strains (Table
1
). All strains were grown at 30oC, the semi-permissive temperature for the
rad54-3
allele. The results are shown in Table
2
. Comparison of the isogenic pair AP1 (
rad52
[Delta]::
LEU2
) and DBY747-ade2[Delta] (
RAD
+
) shows that disruption of the
RAD52
gene reduces transformation efficiency ~50-fold. Similarly, both
rad54
null strains (MD61/5 and MD61/17) transform 30-fold less efficiently than their
RAD
+
parental strain, DBY747-ade2[Delta]. In contrast, the temperature-sensitive
rad54-3
strain transforms with the same efficiency as the
RAD
+
strains. There was no significant difference in the survival rate of
spheroplasts for any of these strains (data not shown). This indicates that the
difference in transformation efficiency is not due to a reduction in
spheroplast viability during the transformation procedure.
Stability of the
DYZ3
YAC in
RAD
+ and
rad54-3
strains
To determine whether the
rad54-3
mutation would reduce homologous recombination-mediated instability of YAC clones to a level comparable with that
observed in either
rad52
or
rad54
null strains, the YAC containing the human
DYZ3
repeats, YOR 1B9F/8, was transferred to MD62/8 by transformation. As a control,
the same YAC was also transformed into a
RAD
+
yeast, AB1380. Chromosome-sized DNA was isolated from 10 randomly picked transformants for both strains and analysed by PFGE and Southern
blotting (Fig.
1
). As has previously been reported (
9
), the majority (seven) of the recombination-proficient AB1380 transformants had YACs that were smaller than the original 140 kb transforming DNA and in two cases displayed two sizes of YAC within one
transformant. In contrast, all MD62/8 (
rad54-3
) transformants contained a single YAC the same size as the transforming DNA
(lanes 14-23). These results indicate that the
DYZ3
-containing YAC is stabilized in a
rad54-3
strain grown and transformed at the semi-permissive temperature of 30oC.
Stability of the
DYZ5
YAC in
RAD
+ and
rad54-3
strains
Long-term stability of YACs in a
rad54-3
strain
The instability of YAC clones containing tandemly repeated DNA sequences
transformed into
RAD
+
yeast strains could be due to recombination events induced by the
transformation process or result from mitotic recombination events during the
30 generations or more of growth required before a transformed colony can be
analysed (this assumes that the generation of an average sized colony will
require 20-24 cell divisions, while an overnight culture prepared from the colony
undergoes a further 10-12 generations of growth before the plug DNA is prepared). To determine
the stability of YACs during mitotic growth, YAC-transformed strains were streaked on selective plates and chromosomal DNA
was prepared from single colonies. Each DNA preparation should therefore
contain the YACs obtained from a single cell after it has gone through ~30 generations of mitotic growth. The DNA was analysed by PFGE and Southern
blotting (Fig.
3
). For YOR 1B9F
RAD
+
transformants (lanes 1-5), all colonies showed YAC instability, whereas for the
rad54-3
transformants (lanes 6-10), no rearrangements were observed after this ~30 generations of mitotic growth or even after 100 generations of
further growth (data not shown). This analysis is not sensitive enough to detect relatively infrequent YAC rearrangements, but the absence of detectable rearrangements does set a minimum number of
generations for which the YAC is stable in the
rad54-3
background. In contrast, for both the
RAD
+
(lanes 12-16) and
rad54-3
(lanes 17-21) YOR 2B6H YAC transformants, a few colonies showed some degree of YAC instability, although to a lesser extent than that observed immediately after transformation, particularly in the
RAD
+
background. The
rad54-3
mutation therefore stabilizes the
DYZ3
-containing YAC both during transformation and subsequent mitotic growth,
whereas its most significant effect on stabilizing the
DYZ5
-containing YAC appears to occur at the transformation stage.
Meiotic stability of the
DYZ3
YAC in a
rad54-3
strain
Manipulation of yeast often involves mating and sporulation (meiosis) as a
method of altering genetic background. These approaches are used to transfer
YAC clones to strains of a different genotype, to allow selection for a broader
range of targeting vectors (
19
) or to generate larger YAC clones by
in vivo
recombination of smaller overlapping YAC clones (
20
). The
Rad52
gene product is essential for meiotic recombination (
21
). A homozygous
rad52
null is unable to go through meiosis (
7
) and the use of such a mutation in all YAC library hosts might preclude the use
of standard genetic techniques for manipulation of YAC clones. Unlike
rad52
mutants, some
rad54
homozygous mutants are capable of undergoing successful meiosis (
7
). The role of Rad54 in meiosis is not clear. Although
RAD54
expression is induced during meiosis, promoter mutations blocking induction of
the
RAD54
gene have not been reported to have a significant effect on meiotic
recombination (
22
). To investigate whether the temperature-sensitive
rad54-3
mutation would allow stable maintenance of YAC clones through meiosis, one isolate of MD62/8 transformed with the 140 kb
DYZ3
-containing YAC YOR 1B9F was mated to both a
RAD
+
and a
rad54
null strain. Both the
RAD
+
/
rad54-3
and the
rad54
[Delta]/
rad54-3
diploids sporulated efficiently. Chromosomal sized DNA was isolated from
cultures grown from spores to which the YAC had segregated for both crosses,
fractionated by PFGE and analysed by Southern blotting as before. The results
are shown in Figure
4
. Three of four spores from the
rad54
[Delta]/
rad54-3
cross show a single band with the same mobility as the starting YAC, whereas
one has a single band of lower molecular weight, indicating that in most YAC-containing spores the YAC did not undergo rearrangement by passage through
meiosis. In the
RAD
+
/
rad54-3
cross, two spores contain single bands with lower molecular weights than the
starting YAC, whereas the other two spores show more than one band, indicating
that all spores contain deleted YACs.
DISCUSSION
The data presented here show a correlation between the level of recombinational
DNA repair activities in a yeast strain and the strain's capacity to be
transformed with exogenous DNA. We have shown that null mutations in either the
RAD52
or the
RAD54
gene significantly decrease transformation efficiency, whereas a temperature-sensitive
rad54
mutant grown at the semi-permissive temperature of 30oC transforms with the same efficiency as a
RAD
+
strain. The reduced growth rate of the
rad
null mutants may affect the composition of their cell walls and membranes or
some other aspect of cell physiology, thereby altering their capacity to take
up DNA or be converted to competent spheroplasts. Alternatively, DNA repair and
recombination activities may be required for some aspect of successful
transformation, such as conversion of the naked incoming DNA to chromatin or
maintenance of the integrity of the transforming DNA. A study by Larinov
et al.
(
23
) provides support for this latter explanation. They studied the fate of
plasmids carrying nicks or double-strand breaks during transformation in yeast and found that the survival
of linear plasmids in a
rad52
mutant was lower than in a recombination-proficient strain. If the inability of plasmids or YACs to be repaired by
a recombination pathway reduces their ability to transform yeast, it seems
likely that any yeast strain which is completely deficient in homologous
recombination will not be capable of being transformed at a frequency high
enough to support the easy construction of representative YAC libraries. The
conditional nature of the
rad54-3
allele overcomes this problem by reducing recombination-mediated rearrangements of YAC clones at the semi-permissive temperature of 30oC without a concommitent drop in transformation frequency. It
is possible that higher transformation frequencies than those reported here
might be obtained by incubating the
rad54-3
strain at the fully permissive temperature of 23oC prior to transformation. Similarly, YAC stability might be further
improved if
rad54-3
transformants were grown at the non-permissive temperature of 37oC. These possibilities remain to be investigated, but these options
increase the utility of
rad54-3
yeast as hosts for YAC cloning and for other experimental approaches that require blocks in mitotic recombination.
Several studies have shown that many of the rearrangements and deletions
observed in YAC clones are produced by the yeast recombinational DNA repair
mechanism during the cloning process (
6
,
9
,
23
). Our results with the
DYZ5
-containing YAC YOR 2B6H, in which the highest level of rearrangements was observed immediately after transformation, support these findings. With the
DYZ3
-containing YAC YOR 1B9F instability was manifested both at the transformation stage and during subsequent mitotic growth in the
RAD
+
background, indicating that there may be some sequence or secondary structure
specificity that contributes to this homologous recombination-mediated instability. The
rad54-3
mutation, unlike a
rad52
null mutation (
9
), completely blocked recombination between the human
DYZ3
tandem repeats on the YAC clone and also partially stabilized the
DYZ5
tandem repeat in most randomly picked transformants. The differences between
these two
rad
mutants may reflect their different roles in the recombination process. The
Rad52 protein is thought to play a direct role in recombination; it is found in
the cell complexed with the yeast RecA homologue Rad51 and may regulate the
latter's strand exchange activity (
24
). The predicted sequence of the Rad54 protein suggests it may function as a DNA
helicase, although its precise role in the recombinational DNA repair pathway
is unknown (
25
). Rad54 shares sequence similarity with the yeast Snf2 protein, a
transcriptional activator proposed to have a role in remodelling chromatin (
25
). This has led to the suggestion that Rad54 may also be involved in providing
access to regions of chromatin, thereby facilitating recombination, rather than
being directly involved in the recombination process (
25
,
26
). It is possible that tandem repeat arrays such as those found in the YAC
clones used in this study have a chromatin structure that requires a Rad54
unwinding function in order to become accessible to the recombination complex.
Thus, for some sequences a
rad54
strain might represent a better cloning host than a
rad52
strain with respect to the faithful maintenance of the DNA. Since a
rad54-3
mutant strain can more stably maintain tandem arrays of
DYZ3
and
DYZ5
repeats,
this strain may allow the cloning of regions that have been unclonable in other
systems. For example, centromeric regions of human chromosomes are primarily
composed of repetitive DNA sequences and have been difficult to isolate in an
unaltered form (
9
). The significance of these findings for YAC cloning is the implication that
rad54
mutants may stabilize a different and perhaps broader spectrum of clones than
rad52
mutants or
RAD
+
strains.
The
rad54-3
mutant was able to undergo normal meiosis after mating with a
rad54
null. The original YAC present in the
rad54-3
strain used in these crosses could be recovered afterwards, unaltered in most products of the cross. Diploid yeast strains homozygous for
the
rad52
mutation are completely blocked in meiosis, as are some
rad54
null homozygotes, so it appears that the
rad54-3
mutation is unique in allowing all standard genetic manipulations that are
available for
RAD
+
yeast hosts but with the advantage of a reduction in the level of homologous
recombination.
ACKNOWLEDGEMENTS
We thank John Game (Lawrence Berkeley Laboratory, Berkeley, CA) for his gift of
the
rad
yeast strains, Chris Tyler-Smith for providing the human YAC clones, Ying Zhang for technical
assistance, Andrew Pickett for the construction of AP1 and Grant MacNevin and Edward DeZeeuw (Cancer Treatment and Research Foundation of Nova Scotia) for irradiating yeast strains. This work was
supported by a grant (GO-12404) from the Canadian Genome Analysis and Technology Program.
11 Calderon,I.L., Contopoulou,C.R. and Mortimer,R.K. (1983) Curr. Genet., 7, 93-100.
12 Ho,K.S. and Mortimer,R.K. (1975) Mutat. Res., 33, 157-164.MEDLINE Abstract
13 Tyler-Smith,C. and Brown,W.R.A. (1987) J. Mol. Biol., 195, 457-470.
14 Tyler-Smith,C., Taylor,L. and Muller,U. (1988) J. Mol. Biol., 203, 837-848.
15 Rose,M.D., Winston,F. and Hieter,P. (1990) Methods in Yeast Genetics: A Laboratory Course Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
16 Ho,K.S.Y. and Mortimer,R.K. (1975) Mutat. Res., 33, 157-164.
20 Sears,D., Hegemann,J.H. and Hieter,P. (1992) Proc. Natl. Acad. Sci. USA, 89, 5296-5300.MEDLINE Abstract
21 Game,J.C. (1983) In Spencer,J.F.T., Spencer,D. and Smith,A.R.W. (eds), Yeast Genetics: Fundamental and Applied Aspects. Springer-Verlag, New York, NY, pp. 109-139.
*To whom correspondence should be addressed. Tel: +1 902 494 7182; Fax: +1 902
494 1355; Email: dobson@is.dal.ca
+
Present address: Laboratory of Biochemistry and Metabolism, National Institute
of Diabetes and Digestive and Kidney Disease, National Institutes of Health,
Bethesda, MD 20892-1800, USA
