I IUL. I U1l. /ILUU. 3(1.
V J1
Vol. 88, pp. 3877-3881, May 1991
Biochemistry
Mutations in conserved yeast DNA primase domains impair DNA replication in vivo (Saccharomyces cerevisiae/conditional mutants/DNA synthesis)
STEFANIA FRANCESCONI, MARIA PIA LONGHESE, ANNA PISERI, CORRADO SANTOCANALE, GIOVANNA LUCCHINI*, AND PAOLO PLEVANI* Dipartimento di Genetica e di Biologia dei Microrganismi, Via Celoria 26, Milano 20133, Italy
Communicated by Gerald R. Fink, January 31, 1991
ABSTRACT To assess the role of eukaryotic DNA primase in vivo, we have produced conditional and lethal point mutations by random in vitro mutagenesis of the PRII and PRI2 genes, which encode the small and large subunits of yeast DNA primase. We replaced the wild-type copies of PRII and PRI2 with two pril and two pri2 conditional alleles. When shifted to the restrictive temperature, these strains showed altered DNA synthesis and reduced ability to synthesize high molecular weight DNA products, thus providing in vivo evidence for the essential role of DNA primase in eukaryotic DNA replication. Furthermore, mapping of the mutations at the nucleotide level has shown that the two pril and twopri2 conditional alleles and onepri2 lethal allele have suffered single base-pair substitutions causing a change in amino acid residues conserved in the corresponding mouse polypeptide.
assessment ofthe primase function in vivo in different aspects of DNA metabolism. Since pril and pri2 mutants were previously not available, we have pursued the goal of producing such mutants by random in vitro mutagenesis of the cloned genes. This paper describes the mapping at the nucleotide level of lethal and conditional-lethal mutations and the preliminary characterization of the primary defect of two pril and two pri2 conditional mutants as an impairment in DNA replication.
MATERIALS AND METHODS Strains and Plasmids. S. cerevisiae strain BS5A, [YEpBS
possible to address several questions about the role of the complex subunits, their interactions, and modulation of their expression (10-15). Characterization of some poll temperature-sensitive mutants has provided information about the functional significance of amino acid sequences that are conserved among DNA polymerases (13-15); these studies have supported previous structural and functional work in vitro. Moreover, comparison of the sequences of cloned genes has revealed a high degree of conservation between the amino acid sequences of yeast and mouse primase subunits (refs. 5, 9, and 16; B. Tseng, personal communication). Although the evidence for the role of DNA primase in replication of the eukaryotic genome is based on in vitro studies, the isolation of conditional lethal mutants allows the
URA3 PRII] MATa ura3-52 trpJAJ priJAJ, has undergone a 1006-base-pair (bp) chromosomal deletion of the PRIJ coding region between the EcoRI and Nde I sites. Strain DAN7/2a, [pAN3 URA3 PRI2] MATa ura3-52 trplAl pri2AI, has undergone a 1013-bp chromosomal deletion of the PRI2 coding region between the Bgl II and Nco I sites. Strain TD28, MA Ta ura3-52 can1-100 inol (12), was used for replacement of either PRII or PRI2 chromosomal alleles with the corresponding conditional alleles. The prilAl and pri2Al alleles are lethal and cell viability is maintained, respectively, by plasmids YEpBS and pAN3, two YEp24 derivatives (17) carrying either the 2050-bp Nru I-Ava I fragment or the 3300-bp Ava I fragment, respectively containing the wildtype PRII or PRI2 allele. Plasmids pLAN2 and pCS20, used for mutagenesis in vitro, were ARSI TRPI CEN6 plasmids carrying, respectively, the Nru I-Ava I and the Ava I fragments containing the wild-type PRII and PRI2 genes. Plasmids pLAN72, pLAN33, pA16, and pF402, recovered by transformation of Escherichia coli strain DB6507, leu- prothr- r- m- recA- pyrF74::Tn5 (18), were pLAN2 and pCS20 mutagenized plasmids carrying, respectively, the pril-1, pril-2, pri2-1, and pri2-2 alleles. Plasmids YIpTS, YIpCS, YIpA16, and YIpF402 were URA3 yeast integrating plasmids, carrying the mutated EcoRI-Ava I or Ava I fragment derived, respectively, from pLAN72, pLAN33, pA16, and pF402. In Vitro Mutagenesis, Plasmid Shuffling, and Gene Replacement. Hydroxylamine mutagenesis was performed as previously described (12). Plasmid shuffling (19) was performed with yeast strains BS5A and DAN7/2a, after transformation with the mutagenized pLAN2 and pCS20 plasmids, by selecting for uracil-requiring (Ura-) clones on medium containing 5-fluoroorotic acid (5-FOA). Five independent transformations were performed for each strain and uracilindependent tryptophan-independent (Ura+ Trp+) transformants were selected on SD medium (20) at 25°C. About 6000 transformants for each plasmid, randomly sampled from the independent transformations, were assayed for the ability to
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviations: 5-FOA, 5-fluoroorotic acid; ts, temperaturesensitive; cs, cold-sensitive. *To whom reprint requests should be addressed.
The essential function of DNA polymerase a and the tightly associated DNA primase during replication of eukaryotic chromosomes is supported by several lines of biochemical and genetic evidence (1, 2). The catalytic properties of eukaryotic DNA primases, either free or in association with DNA polymerase, are almost identical in yeast and in more complex eukaryotic cells (1). The enzyme catalyzes the synthesis of discrete-length oligoribonucleotides when RNA synthesis is coupled with chain elongation by DNA polymerase, but the mechanism determining the switch from primer synthesis to DNA chain elongation is not yet understood. In Saccharomyces cerevisiae, DNA primase activity is associated with two polypeptides of 58 and 48 kDa (p58 and p48, respectively) (3). Studies with anti-p48 and anti-p58 antibodies and an affinity labeling procedure have shown that both polypeptides are needed for DNA primase activity (4-6). Cloning and characterization of the single essential genes POLI (7, 8), PRII (4, 9), and PRI2 (5), which encode, respectively, DNA polymerase a and the two primase sub-
units p48 and p58 of the S. cerevisiae complex, have made it
3877
3878
Proc. Natl. Acad. Sci. USA 88 (1991)
Biochemistry: Francesconi et al.
extracts were loaded onto 4.8-ml 5-20% (wt/vol) sucrose gradients containing 0.5 M NaCI/0.25 M NaOH/1 mM EDTA and cushioned with 0.2 ml of 70% sucrose. Centrifugation was carried out at 16,000 rpm (Beckman SW 55 rotor) for 16 hr at 20°C. Fractions were collected, made 0.5 M in NaOH, and heated at 800C for 30 min to hydrolyze RNA. After the addition of salmon sperm DNA to 100 ,ug/ml and neutralization of the samples, 150 ,ul of each fraction was processed on GF/C glass-fiber squares (Whatman) to measure trichloroacetic acid-insoluble radioactive material (3). Other Techniques. Southern blot analysis, plasmid DNA preparation, and DNA restriction analysis were performed as described (22). Total yeast DNA was prepared by the method of Winston et al. (21) and yeast transformation was performed by the lithium acetate procedure (20).
give rise to Ura- clones at 25°C. A two-step gene replacement procedure (21) was used to replace the wild-type PRIJ and PRI2 chromosomal copies with the corresponding conditional alleles. Strain TD28 was transformed with plasmids YIpTS and YIpCS linearized with HindIlI and with plasmids YIpA16 and YIpF402 linearized with Hpa I. Ura+ transformants selected at 25°C (pril-1, pri2-1, pri2-2) or 28°C (pril-2), were tested by Southern analysis, and single integrants at either PRII or PRI2 were replica-plated on 5-FOA plates to select for loss of the plasmids at the same temperatures. Ten Ura- clones for each integrant were tested for temperature sensitivity or cold sensitivity on YPD plates (20) and for proper excision of the plasmids by Southern blot analysis. Siing of Newly Synthesized DNA Chains by Alkaline Sucrose Sedimentation. Yeast cells were grown in YPD medium at 250C to a concentration of 107 cells per ml and [5,63Hluracil was added to the cultures to a concentration of 27 ,uCi/ml. After 3 hr of incubation at 250C or 37°C, cells were collected, washed, and lysed as described (14). The viscous
RESULTS Isolation of pril and pri2 Mutations. New pril and pri2 mutations were isolated according to the plasmid shuffling
A AMINO ACID CHANGE
CODON
MUTANT ALLELE
NUMER
MUTATION CAA AAA GAA -CAA --. TAA
CGA-K
pril-3*
184 316 380
pri2-1 pri2-2
152 434
GAA-
pril-l pril-2
pri2-3
TGC ATG -a. CAT TCA -CAG -
1
401
pri2-4 pri2-5 pri2-6 pri2-7* pri2-8
41
181 197 376
CAATGG -
CAA TAC ATA CGT TAA TAG TAA TAG
MUTANT PHENOTYPE
Arg -
Gln Glu- . Lys Gln- b. Stop
temperature-sensitive
Glu - Gln Tyr CysMet-.0 Ile
temperature-sensitive temperature-sensitive
cold sensitive lethal
lethal lethal lethal lethal lethal lethal
Hys-b Arg
Ser-er Stop Gln- -o Stop Gln-_- Stop Trp-em- Stop
B pril-1 162 160
|S G R R G|A| |S G R R G|V
311 302
DV R D
D C
E S V
AT D V Q[R N V L D S G I V E S S A V
Yeast Mouse
pril-2
K K
a
CV SX RA[
TQ T I|H L L K APC I S G V N|H L L K SLPJS P
P A P K
NVC R I S
Yeast Mouse
pri2-1
130 102
|S H F I L R L|C FRKIC EEKiV R A]TFiF K Z Q Q QNM DWL R S H YGY F I L R[ R L[G DRWWI
I F
Mouse
R|D W S H R S A F L R|H S D ALWK Q X M
Mouse
Yeast
pri2-2
420 370
L S X P WI L T N P
y
GDY H GCP F D Y H GC P r JQFG R
pri2-4
384
334
N
T
N Z
F| N Z R D G 8N I
P D SKJD
Yeast
R
G|N N Y JJS F K
I N
DY
Yeast
mouse
FIG. 1. Molecular nature of pril and pri2 conditional and lethal alleles. (A) Sequences affected by the mutations. Nucleotide sequencing was performed on centromeric plasmids containing the pril or pri2 mutant allele, by the dideoxy chain termination method (23, 24), using a set of synthetic oligonucleotides spanning their entire coding regions as primers. The asterisks indicate cases in which independent mutant alleles with identical nucleotide substitutions were isolated. (B) Homology with mouse DNA primase subunits of the amino acid sequences altered by the missense mutations. Amino acid substitutions in the mutants are indicated above the sequences.
Biochemistry: Francesconi et al.
Proc. Natl. Acad. Sci. USA 88 (1991)
the failure of the transformed clones to grow on 5-FOA. Uraclones derived from transformants able to grow on 5-FOA were then spotted on YPD plates, which were incubated at 19TC, 25TC, 30'C, and 370C, to identify clones carrying pril or pri2 conditional alleles on the mutagenized centromeric plasmids. Plasmids containing putative mutations were then recovered by transformation of E. coli cells with the DNA from the corresponding yeast clones, checked by restriction enzyme analysis to detect gross alterations, and used to transform diploid strains heterozygous for lethal deletions in the PRII or PRI2 loci. The diploids were analyzed by sporulation and tetrad analysis. By these tests seven PRI2 and two PRII independent centromeric plasmids failed to complement the corresponding chromosomal null allele, thus confirming that they carried lethal pril or pri2 mutations. By the same procedure four plasmids were proved to carry, respectively, one temperature-sensitive (ts) and one cold-sensitive (cs) pril allele and two ts pri2 alleles. Sequence Analysis of the Mutant Alleles. The nucleotide sequence of each of the new pril and pri2 alleles was determined. As shown in Fig. 1A, each independent allele had suffered a single base-pair substitution, usually a G-C -+ A-T transition. Although most of the pri2 lethal mutations were due to the formation of stop codons, the pri24 allele causes the replacement of a histidine residue, suggesting a critical role for this residue in the protein; this residue is also conserved in mouse p59. The two independent pril lethal mutations both result in the formation of the same stop codon at position 380, thus generating a truncated protein and identifying a 30-amino acid COOH-terminal region essential for p48 activity. As shown in Fig. 1B, all the missense mutations affect amino acid residues in regions that are highly conserved in the corresponding mouse primase subunit. A nonconserved glutamic residue in a region of extensive homology is replaced by a basic lysine residue in pril-2, and conserved amino acid residues are changed in pril-l, pri2-1, pri2-2, and pri24. Growth Rate, DNA Synthesis in Vivo, and Terminal Phenotype of pril and pri2 Conditional Mutants. Replacement of the chromosomal wild-type alleles with the corresponding mutated copies (see Materials and Methods) produced four isogenic mutant strains that were defective for growth at 37°C (pril-1, pri2-1, pri2-2) or 19°C (pril-2). Their growth defect was fully complemented by the wild-type copy of either PRII or PRI2 on centromeric plasmids, thus confirming that the strains carry recessive pril or pri2 mutations. These mutants were compared to the isogenic wild-type strain for growth rate, cell morphology, and kinetics of DNA and RNA synthesis after a shift to the restrictive temperature (37°C or 19°C) of asynchronous cultures growing exponentially at the permissive temperature (25°C or 28°C) (Fig. 2). No substantial difference between wild-type and pri2-1 or pri2-2 mutant cells at 25°C, and between wild-type and pril-2 strains at 28°C, was observed for any of the parameters analyzed. The pril -I strain, which is leaky for growth at 25°C, was partially defective for DNA synthesis at the same temperature. In all of the mutants, both the growth rate and the kinetics of DNA synthesis were affected by a shift to the restrictive temperature. They quickly became arrested in the
B
A
method described by Boeke et al. (19). Centromeric plasmids carrying the yeast selectable marker TRPI and the complete PRII or PRI2 genes (see Materials and Methods) were mutagenized in vitro with hydroxylamine hydrochloride and used to transform haploid trpl, ura3 yeast strains containing a lethal deletion in either PRII or PRI2, complemented by the wild-type copy of the corresponding gene carried on a 2-,um plasmid with the URA3 marker. About 6000 transformants were analyzed for each gene. Centromeric plasmids carrying lethal mutations in PRII or PRI2 were directly identified by
3879
a
0 CU
0.0ig CL L.o 0
0.
0
0 ;Ir
0 ID
._
CD
(ua CD L-
0 2
4
6 8
Time, hr
0 2 4 6 8 Time, hr
FIG. 2. Growth rate and kinetics of DNA synthesis for wild-type (wt) and pril or pri2 mutant cells. Exponentially growing cultures (250C or 280C) were supplemented with [5,6-3H]uracil at 7 A.Ci/ml and, after 2.5 hr at the permissive temperature, were shifted to the restrictive temperature (370C or 19'C). Cell number (A) and incorporation of [3H]uracil into DNA (B) or RNA were monitored as previously described (12). The increase in [3H]uracil incorporation is given as the ratio between the amount of radioactivity measured at the indicated times and that found after 1 hr of incubation at permissive temperature in the presence of the labeled precursor.
pri2-1 strains at 37TC and were significantly reduced compared with wild-type strain in pril-I or pri2-2 cultures at 37TC and pril-2 cultures at 19TC. Furthermore, all mutant cultures at the restrictive temperature accumulate dumbbell-shaped cells (Fig. 3) with a single nucleus at the isthmus between mother and daughter cells (95% in pri2-1, 60% in pril-J, pril-2, and pri2-2 cultures), a terminal phenotype typical of mutants defective in DNA replication (2). Wild-type cells with similar morphology (large budded cells) showed a normal nuclear division at both permissive and restrictive temperatures. Alkaline sucrose gradient analysis of the DNA synthesized in vivo (Fig. 4) further supported the conclusion that the primary defect in pril or pri2 ts mutants is an impairment in chromosomal DNA replication. In fact, a shift to the restrictive temperature affects the mutant's ability to synthesize high molecular weight DNA products, causing a concomitant accumulation of DNA chains of heterogeneous length. At present, the nature of these DNA products has not been further analyzed due to the inefficiency of in vivo DNA labeling techniques in yeast cells, but they might be nicked DNA replication intermediates and/or anomalously elongated Okazaki fragments.
DISCUSSION The biochemical properties of the eukaryotic DNA primase that is found associated with DNA polymerase a suggest a
3880
Biochemistry: Francesconi et al. wt 370C
Proc. Natl. Acad. Sci. USA 88 (1991)
Dri 1-1 370rJ
c
0
pri2-1 370C
pri2-2 370C
0
CD
wt 190C
pril-2 190C
0m
0.
0 a)
FIG. 3. Terminal phenotype of pril and pri2 conditional-lethal strains. After 3 hr of incubation at the indicated restrictive temperatures, yeast cells were fixed and stained with 4',6-diamidino-2phenylindole (DAPI) (20). wt, Wild-type controls. (x625.)
major role for this enzyme in chromosomal DNA replication (1). However, the lack of mutants in the genes coding for the two subunits of eukaryotic DNA primase has limited our knowledge of their functions in vivo and of their interactions within the DNA polymerase a-primase complex as well as with other components of the DNA replication machinery. By using the plasmid shuffling procedure (19) on the two cloned S. cerevisiae primase genes, we have been able to identify two lethal, one ts, and one cs pril alleles and seven lethal and two ts pri2 alleles. Each was the consequence of a single point mutation in the corresponding gene. Most mutations causing a null phenotype were due to the formation of nonsense codons, resulting in nonfunctional polypeptides. Among them, the most informative is the stop codon generated in the pril-3 allele, which identifies a 30-amino acid COOH-terminal region essential for p48 activity. All the missense mutations in PRI) or PRI2 cause changes in amino acid residues that fall in regions highly conserved in the corresponding mouse DNA primase subunits. Sequence information on the pril and pri2 mutant alleles might be useful to direct the construction of DNA primase mutations in other organisms by site-directed mutagenesis. Transplacement of the mutant primase alleles into their corresponding chromosomal loci showed that the pril and pri2 conditional mutations affected cell growth and DNA synthesis in vivo at the nonpermissive temperature. A comparable decrease in the rate of RNA synthesis was not observed, indicating that the primary defect in these mutants is an impairment in DNA replication. The almost complete absence of DNA synthesis after a shift of the pri2-1 mutant to 370C confirms previous data on poll ts mutants (14, 15), suggesting that a functional DNA polymerase a-primase complex is required for proper replication ofboth lagging and leading strands. The residual growth and DNA synthesis in pril-), pril-2, and pri2-2 mutants at the restrictive temperature are probably a consequence of the leakiness of the mutations, which could increase the length of S phase without completely blocking DNA replication and subsequent cell division. All mutant cultures at the restrictive temperature
Gradient fraction FIG. 4. Alkaline sucrose gradient analysis of newly synthesized DNA. DNA from asynchronously growing TD28 (o), pril-) (e), pri2-1 (A), or pri2-2 (A) cells was labeled for 3 hr at 250C (A) or 370C (B) and analyzed by alkaline sucrose gradient sedimentation. The percentage of radioactivity found in each fraction is presented relative to the total loaded in each gradient. Sedimentation is from left to right. Intact strands of phage A DNA (48,000 nucleotides in size) run under identical conditions on a parallel gradient peaked in fractions 18-20, and their sedimentation was used to calculate (14) the average size of DNA fragments accumulating in fractions 13-15 (21,500 nucleotides in length).
accumulate dumbbell-shaped cells, a terminal phenotype typical of mutants defective in DNA replication (2). Furthermore, the major role of DNA primase in chromosomal DNA replication is consistent with the observation that a shift to 370C impairs the ability of pril and pri2 ts mutants to synthesize appropriate amounts of high molecular weight DNA products. The concomitant accumulation of DNA chains of heterogeneous size suggests the possibility that some of the DNA products could represent longer than normal Okazaki fragments, as a result of inefficient primer formation on the lagging strand. The analysis of the effect of pairwise combinations of the available poll, pril, and pri2 conditional alleles and the search for extragenic suppressors should help to elucidate the functions and interactions of the DNA polymerase-primase polypeptides in DNA replication. We thank Dr. Ben Y. Tseng (Genta, San Diego) for communicating the sequence of the p59 subunit of mouse DNA primase. S.F. was supported by a fellowship from the Fondazione A. Buzzati-Traverso. This work was partially supported by the Consiglio Nazionale delle Ricerche Target Project on Biotechnology and Bioinstrumentation. 1. Kaguni, L. S. & Lehman, I. R. (1988) Biochim. Biophys. Acta 950, 87-101. 2. Newlon, C. S. (1988) Microbiol. Rev. 52, 568-601. 3. Plevani, P., Foiani, M., Valsasnini, P., Badaracco, G., Cheriathundam, E. & Chang, L. M. S. (1985) J. Biol. Chem. 260, 7102-7106.
Biochemistry: Francesconi et al. 4. Lucchini, G., Francesconi, S., Foiani, M., Badaracco, G. & Plevani, P. (1987) EMBO J. 6, 737-742. 5. Foiani, M., Santocanale, C., Plevani, P. & Lucchini, G. (1989) Mol. Cell. Biol. 9, 3081-3087. 6. Foiani, M., Lindner, A. G., Hartmann, G. R., Lucchini, G. & Plevani, P. (1989) J. Biol. Chem. 264, 2189-2194. 7. Johnson, L. M., Snyder, M., Chang, L. M. S., Davis, R. W. & Campbell, J. L. (1985) Cell 43, 369-377. 8. Lucchini, G., Brandazza, A., Badaracco, G., Bianchi, M. & Plevani, P. (1985) Curr. Genet. 10, 245-252. 9. Plevani, P., Francesconi, S. & Lucchini, G. (1987) Nucleic Acids Res. 15, 7975-7989. 10. Johnston, L. H., White, J. H. M., Johnson, A. L., Lucchini, G. & Plevani, P. (1987) Nucleic Acids Res. 15, 5017-5030. 11. Plevani, P., Foiani, M., Muzi Falconi, M., Pizzagalli, A., Santocanale, C., Francesconi, S., Valsasnini, P., Comedini, A., Piatti, S. & Lucchini, G. (1988) Biochim. Biophys. Acta 951, 268-273. 12. Lucchini, G., Mazza, C., Scacheri, E. & Plevani, P. (1988) Mol. Gen. Genet. 212, 459-465. 13. Pizzagalli, A., Valsasnini, P., Plevani, P. & Lucchini, G. (1988) Proc. Natl. Acad. Sci. USA 85, 3772-3776. 14. Budd, M. E., Wittrup, D., Bailey, J. E. & Campbell, J. L. (1989) Mol. Cell. Biol. 9, 365-376.
Proc. Nail. Acad. Sci. USA 88 (1991)
3881
15. Lucchini, G., Muzi Falconi, M., Pizzagalli, A., Aguilera, A., Klein, H. L. & Plevani, P. (1990) Gene 90, 99-104. 16. Prussak, C. E., Almazan, M. T. & Tseng, B. Y. (1989) J. Biol. Chem. 264, 4957-4%3. 17. Botstein, D., Falco, S. C., Steward, S. E., Brennan, M., Scherer, S., Stinchomb, D. T., Struhl, K. & Davis, R. W. (1979) Gene 8, 17-24. 18. Holm, C., Goto, T., Wang, J. C. & Botstein, D. (1985) Cell 41, 553-563. 19. Boeke, J. D., Trueheart, J., Natsoulis, G. & Fink, G. R. (1987) Methods Enzymol. 154, 164-175. 20. Sherman, F., Fink, G. R. & Hicks, J. H. (1986) Methods in Yeast Genetics (Cold Spring Harbor Lab., Cold Spring Harbor, NY). 21. Winston, F., Chumley, F. F. & Fink, G. R. (1982) Methods Enzymol. 101, 211-228. 22. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular Cloning:A Laboratory Manual (Cold Spring Harbor Lab., Cold Spring Harbor, NY). 23. Hattori, M. & Sasaki, Y. (1986) Anal. Biochem. 152, 232-238. 24. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Nail. Acad. Sci. USA 74, 5463-5467.
V J1
Vol. 88, pp. 3877-3881, May 1991
Biochemistry
Mutations in conserved yeast DNA primase domains impair DNA replication in vivo (Saccharomyces cerevisiae/conditional mutants/DNA synthesis)
STEFANIA FRANCESCONI, MARIA PIA LONGHESE, ANNA PISERI, CORRADO SANTOCANALE, GIOVANNA LUCCHINI*, AND PAOLO PLEVANI* Dipartimento di Genetica e di Biologia dei Microrganismi, Via Celoria 26, Milano 20133, Italy
Communicated by Gerald R. Fink, January 31, 1991
ABSTRACT To assess the role of eukaryotic DNA primase in vivo, we have produced conditional and lethal point mutations by random in vitro mutagenesis of the PRII and PRI2 genes, which encode the small and large subunits of yeast DNA primase. We replaced the wild-type copies of PRII and PRI2 with two pril and two pri2 conditional alleles. When shifted to the restrictive temperature, these strains showed altered DNA synthesis and reduced ability to synthesize high molecular weight DNA products, thus providing in vivo evidence for the essential role of DNA primase in eukaryotic DNA replication. Furthermore, mapping of the mutations at the nucleotide level has shown that the two pril and twopri2 conditional alleles and onepri2 lethal allele have suffered single base-pair substitutions causing a change in amino acid residues conserved in the corresponding mouse polypeptide.
assessment ofthe primase function in vivo in different aspects of DNA metabolism. Since pril and pri2 mutants were previously not available, we have pursued the goal of producing such mutants by random in vitro mutagenesis of the cloned genes. This paper describes the mapping at the nucleotide level of lethal and conditional-lethal mutations and the preliminary characterization of the primary defect of two pril and two pri2 conditional mutants as an impairment in DNA replication.
MATERIALS AND METHODS Strains and Plasmids. S. cerevisiae strain BS5A, [YEpBS
possible to address several questions about the role of the complex subunits, their interactions, and modulation of their expression (10-15). Characterization of some poll temperature-sensitive mutants has provided information about the functional significance of amino acid sequences that are conserved among DNA polymerases (13-15); these studies have supported previous structural and functional work in vitro. Moreover, comparison of the sequences of cloned genes has revealed a high degree of conservation between the amino acid sequences of yeast and mouse primase subunits (refs. 5, 9, and 16; B. Tseng, personal communication). Although the evidence for the role of DNA primase in replication of the eukaryotic genome is based on in vitro studies, the isolation of conditional lethal mutants allows the
URA3 PRII] MATa ura3-52 trpJAJ priJAJ, has undergone a 1006-base-pair (bp) chromosomal deletion of the PRIJ coding region between the EcoRI and Nde I sites. Strain DAN7/2a, [pAN3 URA3 PRI2] MATa ura3-52 trplAl pri2AI, has undergone a 1013-bp chromosomal deletion of the PRI2 coding region between the Bgl II and Nco I sites. Strain TD28, MA Ta ura3-52 can1-100 inol (12), was used for replacement of either PRII or PRI2 chromosomal alleles with the corresponding conditional alleles. The prilAl and pri2Al alleles are lethal and cell viability is maintained, respectively, by plasmids YEpBS and pAN3, two YEp24 derivatives (17) carrying either the 2050-bp Nru I-Ava I fragment or the 3300-bp Ava I fragment, respectively containing the wildtype PRII or PRI2 allele. Plasmids pLAN2 and pCS20, used for mutagenesis in vitro, were ARSI TRPI CEN6 plasmids carrying, respectively, the Nru I-Ava I and the Ava I fragments containing the wild-type PRII and PRI2 genes. Plasmids pLAN72, pLAN33, pA16, and pF402, recovered by transformation of Escherichia coli strain DB6507, leu- prothr- r- m- recA- pyrF74::Tn5 (18), were pLAN2 and pCS20 mutagenized plasmids carrying, respectively, the pril-1, pril-2, pri2-1, and pri2-2 alleles. Plasmids YIpTS, YIpCS, YIpA16, and YIpF402 were URA3 yeast integrating plasmids, carrying the mutated EcoRI-Ava I or Ava I fragment derived, respectively, from pLAN72, pLAN33, pA16, and pF402. In Vitro Mutagenesis, Plasmid Shuffling, and Gene Replacement. Hydroxylamine mutagenesis was performed as previously described (12). Plasmid shuffling (19) was performed with yeast strains BS5A and DAN7/2a, after transformation with the mutagenized pLAN2 and pCS20 plasmids, by selecting for uracil-requiring (Ura-) clones on medium containing 5-fluoroorotic acid (5-FOA). Five independent transformations were performed for each strain and uracilindependent tryptophan-independent (Ura+ Trp+) transformants were selected on SD medium (20) at 25°C. About 6000 transformants for each plasmid, randomly sampled from the independent transformations, were assayed for the ability to
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviations: 5-FOA, 5-fluoroorotic acid; ts, temperaturesensitive; cs, cold-sensitive. *To whom reprint requests should be addressed.
The essential function of DNA polymerase a and the tightly associated DNA primase during replication of eukaryotic chromosomes is supported by several lines of biochemical and genetic evidence (1, 2). The catalytic properties of eukaryotic DNA primases, either free or in association with DNA polymerase, are almost identical in yeast and in more complex eukaryotic cells (1). The enzyme catalyzes the synthesis of discrete-length oligoribonucleotides when RNA synthesis is coupled with chain elongation by DNA polymerase, but the mechanism determining the switch from primer synthesis to DNA chain elongation is not yet understood. In Saccharomyces cerevisiae, DNA primase activity is associated with two polypeptides of 58 and 48 kDa (p58 and p48, respectively) (3). Studies with anti-p48 and anti-p58 antibodies and an affinity labeling procedure have shown that both polypeptides are needed for DNA primase activity (4-6). Cloning and characterization of the single essential genes POLI (7, 8), PRII (4, 9), and PRI2 (5), which encode, respectively, DNA polymerase a and the two primase sub-
units p48 and p58 of the S. cerevisiae complex, have made it
3877
3878
Proc. Natl. Acad. Sci. USA 88 (1991)
Biochemistry: Francesconi et al.
extracts were loaded onto 4.8-ml 5-20% (wt/vol) sucrose gradients containing 0.5 M NaCI/0.25 M NaOH/1 mM EDTA and cushioned with 0.2 ml of 70% sucrose. Centrifugation was carried out at 16,000 rpm (Beckman SW 55 rotor) for 16 hr at 20°C. Fractions were collected, made 0.5 M in NaOH, and heated at 800C for 30 min to hydrolyze RNA. After the addition of salmon sperm DNA to 100 ,ug/ml and neutralization of the samples, 150 ,ul of each fraction was processed on GF/C glass-fiber squares (Whatman) to measure trichloroacetic acid-insoluble radioactive material (3). Other Techniques. Southern blot analysis, plasmid DNA preparation, and DNA restriction analysis were performed as described (22). Total yeast DNA was prepared by the method of Winston et al. (21) and yeast transformation was performed by the lithium acetate procedure (20).
give rise to Ura- clones at 25°C. A two-step gene replacement procedure (21) was used to replace the wild-type PRIJ and PRI2 chromosomal copies with the corresponding conditional alleles. Strain TD28 was transformed with plasmids YIpTS and YIpCS linearized with HindIlI and with plasmids YIpA16 and YIpF402 linearized with Hpa I. Ura+ transformants selected at 25°C (pril-1, pri2-1, pri2-2) or 28°C (pril-2), were tested by Southern analysis, and single integrants at either PRII or PRI2 were replica-plated on 5-FOA plates to select for loss of the plasmids at the same temperatures. Ten Ura- clones for each integrant were tested for temperature sensitivity or cold sensitivity on YPD plates (20) and for proper excision of the plasmids by Southern blot analysis. Siing of Newly Synthesized DNA Chains by Alkaline Sucrose Sedimentation. Yeast cells were grown in YPD medium at 250C to a concentration of 107 cells per ml and [5,63Hluracil was added to the cultures to a concentration of 27 ,uCi/ml. After 3 hr of incubation at 250C or 37°C, cells were collected, washed, and lysed as described (14). The viscous
RESULTS Isolation of pril and pri2 Mutations. New pril and pri2 mutations were isolated according to the plasmid shuffling
A AMINO ACID CHANGE
CODON
MUTANT ALLELE
NUMER
MUTATION CAA AAA GAA -CAA --. TAA
CGA-K
pril-3*
184 316 380
pri2-1 pri2-2
152 434
GAA-
pril-l pril-2
pri2-3
TGC ATG -a. CAT TCA -CAG -
1
401
pri2-4 pri2-5 pri2-6 pri2-7* pri2-8
41
181 197 376
CAATGG -
CAA TAC ATA CGT TAA TAG TAA TAG
MUTANT PHENOTYPE
Arg -
Gln Glu- . Lys Gln- b. Stop
temperature-sensitive
Glu - Gln Tyr CysMet-.0 Ile
temperature-sensitive temperature-sensitive
cold sensitive lethal
lethal lethal lethal lethal lethal lethal
Hys-b Arg
Ser-er Stop Gln- -o Stop Gln-_- Stop Trp-em- Stop
B pril-1 162 160
|S G R R G|A| |S G R R G|V
311 302
DV R D
D C
E S V
AT D V Q[R N V L D S G I V E S S A V
Yeast Mouse
pril-2
K K
a
CV SX RA[
TQ T I|H L L K APC I S G V N|H L L K SLPJS P
P A P K
NVC R I S
Yeast Mouse
pri2-1
130 102
|S H F I L R L|C FRKIC EEKiV R A]TFiF K Z Q Q QNM DWL R S H YGY F I L R[ R L[G DRWWI
I F
Mouse
R|D W S H R S A F L R|H S D ALWK Q X M
Mouse
Yeast
pri2-2
420 370
L S X P WI L T N P
y
GDY H GCP F D Y H GC P r JQFG R
pri2-4
384
334
N
T
N Z
F| N Z R D G 8N I
P D SKJD
Yeast
R
G|N N Y JJS F K
I N
DY
Yeast
mouse
FIG. 1. Molecular nature of pril and pri2 conditional and lethal alleles. (A) Sequences affected by the mutations. Nucleotide sequencing was performed on centromeric plasmids containing the pril or pri2 mutant allele, by the dideoxy chain termination method (23, 24), using a set of synthetic oligonucleotides spanning their entire coding regions as primers. The asterisks indicate cases in which independent mutant alleles with identical nucleotide substitutions were isolated. (B) Homology with mouse DNA primase subunits of the amino acid sequences altered by the missense mutations. Amino acid substitutions in the mutants are indicated above the sequences.
Biochemistry: Francesconi et al.
Proc. Natl. Acad. Sci. USA 88 (1991)
the failure of the transformed clones to grow on 5-FOA. Uraclones derived from transformants able to grow on 5-FOA were then spotted on YPD plates, which were incubated at 19TC, 25TC, 30'C, and 370C, to identify clones carrying pril or pri2 conditional alleles on the mutagenized centromeric plasmids. Plasmids containing putative mutations were then recovered by transformation of E. coli cells with the DNA from the corresponding yeast clones, checked by restriction enzyme analysis to detect gross alterations, and used to transform diploid strains heterozygous for lethal deletions in the PRII or PRI2 loci. The diploids were analyzed by sporulation and tetrad analysis. By these tests seven PRI2 and two PRII independent centromeric plasmids failed to complement the corresponding chromosomal null allele, thus confirming that they carried lethal pril or pri2 mutations. By the same procedure four plasmids were proved to carry, respectively, one temperature-sensitive (ts) and one cold-sensitive (cs) pril allele and two ts pri2 alleles. Sequence Analysis of the Mutant Alleles. The nucleotide sequence of each of the new pril and pri2 alleles was determined. As shown in Fig. 1A, each independent allele had suffered a single base-pair substitution, usually a G-C -+ A-T transition. Although most of the pri2 lethal mutations were due to the formation of stop codons, the pri24 allele causes the replacement of a histidine residue, suggesting a critical role for this residue in the protein; this residue is also conserved in mouse p59. The two independent pril lethal mutations both result in the formation of the same stop codon at position 380, thus generating a truncated protein and identifying a 30-amino acid COOH-terminal region essential for p48 activity. As shown in Fig. 1B, all the missense mutations affect amino acid residues in regions that are highly conserved in the corresponding mouse primase subunit. A nonconserved glutamic residue in a region of extensive homology is replaced by a basic lysine residue in pril-2, and conserved amino acid residues are changed in pril-l, pri2-1, pri2-2, and pri24. Growth Rate, DNA Synthesis in Vivo, and Terminal Phenotype of pril and pri2 Conditional Mutants. Replacement of the chromosomal wild-type alleles with the corresponding mutated copies (see Materials and Methods) produced four isogenic mutant strains that were defective for growth at 37°C (pril-1, pri2-1, pri2-2) or 19°C (pril-2). Their growth defect was fully complemented by the wild-type copy of either PRII or PRI2 on centromeric plasmids, thus confirming that the strains carry recessive pril or pri2 mutations. These mutants were compared to the isogenic wild-type strain for growth rate, cell morphology, and kinetics of DNA and RNA synthesis after a shift to the restrictive temperature (37°C or 19°C) of asynchronous cultures growing exponentially at the permissive temperature (25°C or 28°C) (Fig. 2). No substantial difference between wild-type and pri2-1 or pri2-2 mutant cells at 25°C, and between wild-type and pril-2 strains at 28°C, was observed for any of the parameters analyzed. The pril -I strain, which is leaky for growth at 25°C, was partially defective for DNA synthesis at the same temperature. In all of the mutants, both the growth rate and the kinetics of DNA synthesis were affected by a shift to the restrictive temperature. They quickly became arrested in the
B
A
method described by Boeke et al. (19). Centromeric plasmids carrying the yeast selectable marker TRPI and the complete PRII or PRI2 genes (see Materials and Methods) were mutagenized in vitro with hydroxylamine hydrochloride and used to transform haploid trpl, ura3 yeast strains containing a lethal deletion in either PRII or PRI2, complemented by the wild-type copy of the corresponding gene carried on a 2-,um plasmid with the URA3 marker. About 6000 transformants were analyzed for each gene. Centromeric plasmids carrying lethal mutations in PRII or PRI2 were directly identified by
3879
a
0 CU
0.0ig CL L.o 0
0.
0
0 ;Ir
0 ID
._
CD
(ua CD L-
0 2
4
6 8
Time, hr
0 2 4 6 8 Time, hr
FIG. 2. Growth rate and kinetics of DNA synthesis for wild-type (wt) and pril or pri2 mutant cells. Exponentially growing cultures (250C or 280C) were supplemented with [5,6-3H]uracil at 7 A.Ci/ml and, after 2.5 hr at the permissive temperature, were shifted to the restrictive temperature (370C or 19'C). Cell number (A) and incorporation of [3H]uracil into DNA (B) or RNA were monitored as previously described (12). The increase in [3H]uracil incorporation is given as the ratio between the amount of radioactivity measured at the indicated times and that found after 1 hr of incubation at permissive temperature in the presence of the labeled precursor.
pri2-1 strains at 37TC and were significantly reduced compared with wild-type strain in pril-I or pri2-2 cultures at 37TC and pril-2 cultures at 19TC. Furthermore, all mutant cultures at the restrictive temperature accumulate dumbbell-shaped cells (Fig. 3) with a single nucleus at the isthmus between mother and daughter cells (95% in pri2-1, 60% in pril-J, pril-2, and pri2-2 cultures), a terminal phenotype typical of mutants defective in DNA replication (2). Wild-type cells with similar morphology (large budded cells) showed a normal nuclear division at both permissive and restrictive temperatures. Alkaline sucrose gradient analysis of the DNA synthesized in vivo (Fig. 4) further supported the conclusion that the primary defect in pril or pri2 ts mutants is an impairment in chromosomal DNA replication. In fact, a shift to the restrictive temperature affects the mutant's ability to synthesize high molecular weight DNA products, causing a concomitant accumulation of DNA chains of heterogeneous length. At present, the nature of these DNA products has not been further analyzed due to the inefficiency of in vivo DNA labeling techniques in yeast cells, but they might be nicked DNA replication intermediates and/or anomalously elongated Okazaki fragments.
DISCUSSION The biochemical properties of the eukaryotic DNA primase that is found associated with DNA polymerase a suggest a
3880
Biochemistry: Francesconi et al. wt 370C
Proc. Natl. Acad. Sci. USA 88 (1991)
Dri 1-1 370rJ
c
0
pri2-1 370C
pri2-2 370C
0
CD
wt 190C
pril-2 190C
0m
0.
0 a)
FIG. 3. Terminal phenotype of pril and pri2 conditional-lethal strains. After 3 hr of incubation at the indicated restrictive temperatures, yeast cells were fixed and stained with 4',6-diamidino-2phenylindole (DAPI) (20). wt, Wild-type controls. (x625.)
major role for this enzyme in chromosomal DNA replication (1). However, the lack of mutants in the genes coding for the two subunits of eukaryotic DNA primase has limited our knowledge of their functions in vivo and of their interactions within the DNA polymerase a-primase complex as well as with other components of the DNA replication machinery. By using the plasmid shuffling procedure (19) on the two cloned S. cerevisiae primase genes, we have been able to identify two lethal, one ts, and one cs pril alleles and seven lethal and two ts pri2 alleles. Each was the consequence of a single point mutation in the corresponding gene. Most mutations causing a null phenotype were due to the formation of nonsense codons, resulting in nonfunctional polypeptides. Among them, the most informative is the stop codon generated in the pril-3 allele, which identifies a 30-amino acid COOH-terminal region essential for p48 activity. All the missense mutations in PRI) or PRI2 cause changes in amino acid residues that fall in regions highly conserved in the corresponding mouse DNA primase subunits. Sequence information on the pril and pri2 mutant alleles might be useful to direct the construction of DNA primase mutations in other organisms by site-directed mutagenesis. Transplacement of the mutant primase alleles into their corresponding chromosomal loci showed that the pril and pri2 conditional mutations affected cell growth and DNA synthesis in vivo at the nonpermissive temperature. A comparable decrease in the rate of RNA synthesis was not observed, indicating that the primary defect in these mutants is an impairment in DNA replication. The almost complete absence of DNA synthesis after a shift of the pri2-1 mutant to 370C confirms previous data on poll ts mutants (14, 15), suggesting that a functional DNA polymerase a-primase complex is required for proper replication ofboth lagging and leading strands. The residual growth and DNA synthesis in pril-), pril-2, and pri2-2 mutants at the restrictive temperature are probably a consequence of the leakiness of the mutations, which could increase the length of S phase without completely blocking DNA replication and subsequent cell division. All mutant cultures at the restrictive temperature
Gradient fraction FIG. 4. Alkaline sucrose gradient analysis of newly synthesized DNA. DNA from asynchronously growing TD28 (o), pril-) (e), pri2-1 (A), or pri2-2 (A) cells was labeled for 3 hr at 250C (A) or 370C (B) and analyzed by alkaline sucrose gradient sedimentation. The percentage of radioactivity found in each fraction is presented relative to the total loaded in each gradient. Sedimentation is from left to right. Intact strands of phage A DNA (48,000 nucleotides in size) run under identical conditions on a parallel gradient peaked in fractions 18-20, and their sedimentation was used to calculate (14) the average size of DNA fragments accumulating in fractions 13-15 (21,500 nucleotides in length).
accumulate dumbbell-shaped cells, a terminal phenotype typical of mutants defective in DNA replication (2). Furthermore, the major role of DNA primase in chromosomal DNA replication is consistent with the observation that a shift to 370C impairs the ability of pril and pri2 ts mutants to synthesize appropriate amounts of high molecular weight DNA products. The concomitant accumulation of DNA chains of heterogeneous size suggests the possibility that some of the DNA products could represent longer than normal Okazaki fragments, as a result of inefficient primer formation on the lagging strand. The analysis of the effect of pairwise combinations of the available poll, pril, and pri2 conditional alleles and the search for extragenic suppressors should help to elucidate the functions and interactions of the DNA polymerase-primase polypeptides in DNA replication. We thank Dr. Ben Y. Tseng (Genta, San Diego) for communicating the sequence of the p59 subunit of mouse DNA primase. S.F. was supported by a fellowship from the Fondazione A. Buzzati-Traverso. This work was partially supported by the Consiglio Nazionale delle Ricerche Target Project on Biotechnology and Bioinstrumentation. 1. Kaguni, L. S. & Lehman, I. R. (1988) Biochim. Biophys. Acta 950, 87-101. 2. Newlon, C. S. (1988) Microbiol. Rev. 52, 568-601. 3. Plevani, P., Foiani, M., Valsasnini, P., Badaracco, G., Cheriathundam, E. & Chang, L. M. S. (1985) J. Biol. Chem. 260, 7102-7106.
Biochemistry: Francesconi et al. 4. Lucchini, G., Francesconi, S., Foiani, M., Badaracco, G. & Plevani, P. (1987) EMBO J. 6, 737-742. 5. Foiani, M., Santocanale, C., Plevani, P. & Lucchini, G. (1989) Mol. Cell. Biol. 9, 3081-3087. 6. Foiani, M., Lindner, A. G., Hartmann, G. R., Lucchini, G. & Plevani, P. (1989) J. Biol. Chem. 264, 2189-2194. 7. Johnson, L. M., Snyder, M., Chang, L. M. S., Davis, R. W. & Campbell, J. L. (1985) Cell 43, 369-377. 8. Lucchini, G., Brandazza, A., Badaracco, G., Bianchi, M. & Plevani, P. (1985) Curr. Genet. 10, 245-252. 9. Plevani, P., Francesconi, S. & Lucchini, G. (1987) Nucleic Acids Res. 15, 7975-7989. 10. Johnston, L. H., White, J. H. M., Johnson, A. L., Lucchini, G. & Plevani, P. (1987) Nucleic Acids Res. 15, 5017-5030. 11. Plevani, P., Foiani, M., Muzi Falconi, M., Pizzagalli, A., Santocanale, C., Francesconi, S., Valsasnini, P., Comedini, A., Piatti, S. & Lucchini, G. (1988) Biochim. Biophys. Acta 951, 268-273. 12. Lucchini, G., Mazza, C., Scacheri, E. & Plevani, P. (1988) Mol. Gen. Genet. 212, 459-465. 13. Pizzagalli, A., Valsasnini, P., Plevani, P. & Lucchini, G. (1988) Proc. Natl. Acad. Sci. USA 85, 3772-3776. 14. Budd, M. E., Wittrup, D., Bailey, J. E. & Campbell, J. L. (1989) Mol. Cell. Biol. 9, 365-376.
Proc. Nail. Acad. Sci. USA 88 (1991)
3881
15. Lucchini, G., Muzi Falconi, M., Pizzagalli, A., Aguilera, A., Klein, H. L. & Plevani, P. (1990) Gene 90, 99-104. 16. Prussak, C. E., Almazan, M. T. & Tseng, B. Y. (1989) J. Biol. Chem. 264, 4957-4%3. 17. Botstein, D., Falco, S. C., Steward, S. E., Brennan, M., Scherer, S., Stinchomb, D. T., Struhl, K. & Davis, R. W. (1979) Gene 8, 17-24. 18. Holm, C., Goto, T., Wang, J. C. & Botstein, D. (1985) Cell 41, 553-563. 19. Boeke, J. D., Trueheart, J., Natsoulis, G. & Fink, G. R. (1987) Methods Enzymol. 154, 164-175. 20. Sherman, F., Fink, G. R. & Hicks, J. H. (1986) Methods in Yeast Genetics (Cold Spring Harbor Lab., Cold Spring Harbor, NY). 21. Winston, F., Chumley, F. F. & Fink, G. R. (1982) Methods Enzymol. 101, 211-228. 22. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular Cloning:A Laboratory Manual (Cold Spring Harbor Lab., Cold Spring Harbor, NY). 23. Hattori, M. & Sasaki, Y. (1986) Anal. Biochem. 152, 232-238. 24. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Nail. Acad. Sci. USA 74, 5463-5467.
