Curr Genet (1995) 28:437-440

9 Springer-Verlag 1995

T. Hails 9 O. Huttner 9A. Day

Isolation of a Chlamydomonasreinhardtii telomere by functional complementation in yeast

Received: 10 January / 21 April 1995

We attempted to determine whether Chlamydomonas reinhardtii telomeres, which do not form G-quartet structures readily in vitro, are able to nucleate telomere addition in S a c c h a r o m y c e s cerevisiae. Restricted C. reinhardtii genomic DNA was ligated to a linear S. cerevisiae vector lacking a telomere. A C. reinhardtii telomere ligated to this unprotected end allowed vector replication as a linear DNA molecule in S. cerevisiae. DNA sequencing revealed common [T4AG3] n and variant T6AG 3 and TsAG 3 C. reinhardtii telomere repeats capped by S. cerevisiae telomere repeat units. The recognition of a C. reinhardtii telomere by the telomere maintenance machinery of S. cerevisiae is consistent with a common theme for telomere structure in organisms with divergent telomere repeats.

Abstract

Key words Telomere - Functional complementation C. reinhardtii. S. cerevisiae

Introduction Telomeres confer stability to the ends of chromosomes by protecting them from degradation and recombination and by allowing complete replication of their ends. Compared with other structural components of chromosomes, such as origins of replication and centromeres, telomeres have been relatively well conserved during evolution. Most telomeres are composed of 5-8-bp tandem repeats with oligoguanine blocks on the protruding 3' DNA strand (Pardue 1994). In vitro experiments show that the G-rich singlestranded DNA extensions of telomeres are capable of formHuttner Genetics Laboratory, Biochemistry Department, Oxford University, South Parks Road, Oxford OX1 3QU, UK A. Day (t~) School of Biological Sciences, 3.614 Stopford Building, Manchester University, Oxford Road, Manchester M13 9PT, UK T. H a i l s 9 O.

Communicated by J.-D. Rochaix

ing tetraplex structures involving guanine tetrads (Murchie and Lilley 1994). The isolation of a protein that binds these G-quartet structures supports a role for these structures in vivo (Fang and Cech 1993). Of the telomere repeats tested C h l a m y d o m o n a s reinhardtii telomere repeats are the poorest at forming G-quartet structures in vitro (Petracek and Berman 1992). The high salt concentrations required for G-quartet structure formation in C. reinhardtii telomere oligonuclcotides are unlikely to exist in vivo (Petracek and Berman 1992). Given these differences in vitro, C. reinhardtii telomeres provide a possible exception to the idea that all telomeres composed of 5-8-bp tandem repeats will share a common folded structure in vivo. The isolation of heterologous telomeres in S a c c h a r o m y c e s cerevisiae (Szostak and Blackburn 1982) provided early evidence that telomeres have a common structural theme despite differences in nucleotide sequence. If C. reinhardtii telomeres do have distinct structures in vivo, then they may prove to be recalcitrant to isolation in S. cerevisiae. Here we attempt to isolate a C. reinhardtii telomere in S. cerevisiae.

Materials and methods Cloning in S. cerevisiae. High-molecular-weight C. reinhardtii 137c genomic DNA was fractionated by CsC1/Hoescht 33258 gradient centrifugation (Manuelidis 1977). Fractions enriched for C. reinhardtii telomeres were located using a 32p-labelled C. reinhardtii telomere oligonucleotide, d(TTTTAGGG)3,probe. These fractions produced diffuse (fuzzy) bands of hybridization by Southern-blot analysis (Southern 1975) and were located immediately above the bulk of nuclear DNA. Fuzzy bands are diagnostic of telomere-containing end-fragments (Szostak and Blackburn 1982). A Sau3A digest of a fraction enriched for telomeric DNA (0.5 gg) was ligated, in a total volume of 30 ~tl, to 2 gg of pTV2 (Brown 1989) previously digested with BamHI and NotI and treated with calf intestinal phosphatase. This ligation mix was used to transform S. cerevisiae AB 1380 (MAT a ura3 trpl ade2-1 canl-lOOlys2-1 methis5 [I/t+])(Burke et al. 1987) according to Burgers and Percival (1987) and Ura+Trp+ transformants were selected on SD medium with supplements (Sherman et al. 1986). After colony hybridization (Sherman etal. 1986), one Ura+Trp+ colony hyybridized strongly to a 32P-labelled C. reinhard-

438 tii telomere oligonucleotide, d(TTTTAGGG)3, out of a total of 87 transformants screened. S. cerevisiae total DNA was prepared according to Cryer et al. (1975) except that lyticase was used to remove cell walls. Bal31 digestion. Total high-molecular-weight DNA (0.2 ~tg/gl) was digested with nuclease Bal31 (final concentration of 0.01 units/gl) at 30 ~ for different lengths of time. For each time-point (except at time zero) 5 gl of 1 xBal31 buffer containing 0.5 units of nuclease Bal31 was added to 8 gg of total DNA in 40 gl of Bal31 buffer. The tubes were incubated at 30 ~ for the indicated lengths of time and then placed on ice before adding phenol/chloroform to remove nuclease Bal31. Nuclease Bal31-treated DNA was digested with SalI and fractionated by agarose-gel electrophoresis. Cloning in E. coli. The linear pTV2 C. reinhardtii telomere plasmid present in S. cerevisiae needs to be circularized before it is transferred into E. coli. Fractions containing the linear pTV2 C. reinhardtii telomere plasmid were collected from salt gradients (Kaiser and Murray 1985), treated briefly with Bal31, XhoI digested to remove the left telomere of pTV2 (see Fig. 3), incubated with T4 DNA polymerase (0.05 units/gl) and dNTPs for 2 min, T4 DNA ligase was added and the mixture left to ligate overnight at 14 ~ fresh T4 DNA ligase was then added and incubation continued for a further 6 h. The ligation mix was transformed into E. coli DH5c~ MCR by electroporation (Dower et al. 1988). Treatment of telomere ends with nuclease BaI31 and T4 DNA polymerase ensures removal of possible single-stranded extensions that may prevent intramolecular ligation of the C. reinhardtii telomere to the filled-in XhoI site. After each step the DNA mix was proteinized with phenol, DNA pelleted from 70% v/v ethanol/0.2 M NH 4 acetate and re-suspended in the relevant buffer. Colonies were screened (Sambrook et al. 1989) with 32p-endlabelled d(TTTTAGGG) 3. Plasmid inserts from positive colonies were excised with SalI and Asp700 and transferred to SalI plus SmaIcut pBluescript (Stratagene). DNA manipulations. Methods for DNA extraction, electrophoresis, blotting toGene Screen (DuPont) in alkali (Khandjian 1987), hybridization and sizing (Southern 1979), have been described (Day et al. 1988). Pulsed-field gel electrophoresis and methods for preparing high-molecular-weight DNA have also been described (Hails et al. 1993). Pulsed-field agarose gels (1.5% w/v) were run for 36 h at 16 ~ using a switch time of 20 s. Restriction fragments were labelled with [c~-32P]dCTP according to Feinberg and Vogelstein (1983). Oligonucleotides were labelled with [~-3zP]ATP and polynucleotide kinase. Blots were washed at 50~ in 0.1 xSSC, 0.1% (w/v) sodium dodecyl sulphate for restriction fragment DNA probes (1 x SSC = 0.15 M NaC1, 0.015 M trisodium citrate). The salt was increased to 1 xSSC for oligonucleotide probes. Restriction enzymes and nucleases were used according to the supplier (BoehringerMannheim). DNA sequencing. Sequence overlaps were obtained by making nested deletions using exonuclease III (Henikoff 1987), nuclease Bal31, or suitable restriction enzymes. Single and double-stranded DNA templates were prepared and both DNA strands sequenced as described (Sharpe and Day 1993). The DNA sequence has the accession number X77590 and is designated CRTELREP in the EMBL nucleic acid data library.

Results p T V 2 l i n e a r i z e d with B a m H I and NotI possesses one telo m e r e and will only survive in S. cerevisiae b y circulation, the addition o f S. cerevisiae t e l o m e r e s ( c h r o m o s o m e healing), or the addition of a heterologous t e l o m e r e at its BamHI site. Vector survival b y circularization w o u l d be e x p e c t e d to p r e d o m i n a t e (Riethman et al. 1989). B y ligating S a u 3 A -

Fig. 1A, B Analysis of total DNA (S.c. DNA) from a S. cerevisiae transformant containing C. reinhardtii telomere repeats. A ethidium bromide (EtBr) fluorescence of high-molecular-weight DNA (arrowed) fractionated on a 0.8% (w/v) agarose gel. B blot of gel probed with a C. reinhardtii d(TTTTAGGG)3 telomere oligonucleotide. The sizes of unlabelled and 32p-end-labelled linear DNA size standards in the marker lanes are indicated

d i g e s t e d C. reinhardtii g e n o m i c D N A to the B a m H I site of p T V 2 w e o b t a i n e d one Ura+Trp + S. cervisiae t r a n s f o r m a n t containing C. reinhardtii T 4 A G 3 t e l o m e r i c repeats (see M a t e r i a l s and methods). B l o t analysis o f h i g h - m o l e c u l a r w e i g h t D N A ( a r r o w e d in lane 2 o f Fig. 1 A) isolated f r o m this t r a n f o r m a n t r e v e a l e d a single b a n d ( p T V 2 - C R T E L ) o f 10 - l l k bp that hyb n"d l"z ed to a 32P - l a b e l l e d C. reinhardtii t e l o m e r e repeat o l i g o n u c l e o t i d e ( a r r o w e d in lane 3 o f Fig. 1 B). The relative m i g r a t i o n o f this 10-11 k b p band with r e s p e c t to linear size standards r e m a i n e d the same on p u l s e d - f i e l d gels and under a variety o f c o n v e n t i o n a l elect r o p h o r e s i s conditions (data not shown). This indicates that it is a linear D N A molecule. C i r c u l a r D N A m o l e c u l e s exhibit altered m o b i l i t y with c h a n g i n g e l e c t r o p h o r e t i c conditions. A size o f 10-11 k b p indicates that the 8-kbp NotIBamHI-digested p T V 2 vector had acquired 2 - 3 k b p o f DNA. F i g u r e 2 shows a S o u t h e r n - b l o t analysis of total S. cerevisiae D N A containing p T V 2 - C R T E L , treated with nuclease Bal31 and then d i g e s t e d with SalI. N u c l e a s e BaI31 will d e g r a d e linear, but not circular, D N A templates. T h e 2.3-kb SalI f r a g m e n t b e a r i n g the C. reinhardtii t e l o m e r e (Fig. 2) is p r o g r e s s i v e l y shortened b y nuclease Bal31 (Fig. 2 A) suggesting that p T V 2 - C R T E L is linear. The hyb r i d i z a t i o n signal o b t a i n e d with the C. reinhardtii telom e r e - r e p e a t o l i g o n u c l e o t i d e was r a p i d l y r e d u c e d b y nuclease Bal31 (Fig. 2 B). R e m o v a l o f a p p r o x i m a t e l y 0.5 kbp from the 2.3-kbp SalI f r a g m e n t results in signal loss. This indicates that C. reinhardtii t e l o m e r e repeats do not extend to m o r e than a distance o f around 0.5 kbp f r o m the end. The m a j o r b a n d v i s i b l e at one minute in Fig. 2 A lacks

439 CREND

pTV2 (8 kbp)

I

I

~

r162

X

Xhol ~ i~u . A =o E

~ I

~ _L r ...........

Not

9

Fig. 2A-C Bal31 sensitivity of the pTV2 C. reinhardtii telomere plasmid. Total high-molecular-weight DNA from S. cerevisiae was digested with Bal31 exonuclease for 0-32 rain and then digested with SalI. Digests were fractionated on a 1% (w/v) agarose gel, blotted and probed sequentially with CREND in A, a C. reinhardtii d(TTTTAGGG)3 telomere oligonucleotide in B, and a 1.8-kbp BamHI fragment encoding HIS3 from pYAC4 (Burke et al. 1987) in C. The location of CREND is shown in Fig. 3. DNA size markers are shown on the left of the blot. The 2.3-kb C. reinhardtii band at time zero is arrowed

C. reinhardtii telomere repeats and is not visible in Fig. 2 B. The hybridization visible in lanes 2 and 3 of Fig. 2 B represents large D N A molecules, a minor fraction of the population, that have retained their C. reinhardtii telomere repeats. Internal D N A sequences would be sensitive to Bal31 digestion if the source DNA had a low average size. Nuclease Bal31 digestion has little effect on a high-molecular-weight band containing a chromosomal copy of the HIS3 gene (Fig. 2 C). This confirms the integrity of our source DNA. The detailed structure of pTV2-CRTEL is shown in Fig. 3 and was determined following circularization, transfer to E. coli and DNA sequencing (see Materials and methods). The vector was circularized by intramolecular ligation between the XhoI site, filled in with dNTPs and T4 DNA polymerase, and the Bal31-treated right end (Fig. 3). A 2-kbp region of C. reinhardtii DNA was found ligated to the B a m H I site of pTV2. The distal end, relative to the original B a m H I site (lost by S a u 3 A ligation), was composed of 32 telomere repeats of C. reinhardtii capped by [TGI_3] n telomere repeats of S. cerevisiae. Repeats 1 (T6AG3), 11 and 19 (TsAG3) contained one or two additional thymine bases and are variations of the predominant T4AG 3 repeat unit. The orientation of C. reinhardtii and S. cerevisiae telomere repeats in Fig. 3 was deduced by sequencing outwards from the Asp700 site of pTV2 (Fig. 3) towards the XhoI site and into the ligated S. cerevisiae telomere repeats. DNA sequencing revealed 150 bases of S. cerevisiae [TGI_3] n telomere repeats beyond the C. reinhardtii telomere repeats. Since the sequenced product involved nuclease Bal31 digestion from the right end prior

Fig. 3 The structure of the pTV2 C. reinhardtii telomere clone. Black arrowheads labelled TEL represent telomeric sequences from Tetrahymena (left) or S. cerevisiae (right). The DNA sequence at the junction of the C. reinhardtii telomere (CRTEL) with S. cerevisiae telomeric sequences is shown. Shaded regions represent C. reinhardtii DNA. The BamHI site of pTV2 was lost on ligation to Sau3A C. reinhardtii fragments. Ura3, Trpl, ARS1 and CEN4 allow propagation in S. cerevisiae. Amp and ori function in E. coli. CREND=the DNA probe used in Fig. 2 A

to cloning in E. coli this is a minimum estimate. blot analysis of terminal restriction fragments that the actual length of S. cerevisiae telomere pTV2-CRTEL was approximately 250 bases shown).

Southernindicated repeats in (data not

Discussion

We have shown that a C. reinhardtii telomere can be isolated in S. cerevisiae by functional complementation of a centromere plasmid lacking a telomere. The 262-base C. reinhardtii telomere isolated was similar in size to those previously isolated in E. coli (Petracek et al. 1990). Actual C. reinhardtii telomeres are likely to be larger because heterologous telomeres are truncated during cloning in S. cerevisiae (Brown 1989). Variant TsAG 3 repeats were found embedded within the common T4AG 3 telomere repeats. Unlike other repeat variants, such as T4AGGA (Petracek et al. 1990), which prevent the formation of G-quartet structures, the additional thymine residue in T s A G G G can be accommodated in the loops separating the blocks of three G4 tetrads and would not greatly disrupt telomere structure (Murchie and Lilley 1994). A previous report (Petracek et al. 1990) found only one variant T4A2GGG repeat unit embedded within 160 telomere repeat units. Most repeat variants were found predominantly at the junction where telomere repeats meet telomere-associated DNA (Petracek et al. 1990). Our finding of two variant repeats interspersed amongst 32 repeat units may indicate greater telomere heterogeneity than previously reported or else may represent a cloning artefact. The isolation of a C. reinhardtii telomere in S. cerevisiae is a rare event (1/87 transformants). Low frequencies are also obtained with human telomeres, 2/240 transformants (Cross et al. 1989), and Arabidopsis telomeres,

440 8/1272 transformants (Richards et al. 1992), isolated in S. cerevisiae. These comparisons between functional complementation experiments are complicated by differences in genome sizes and experimental procedures, as well as the small number of events studied. Sub-terminal DNA sequences can promote the addition of S. cerevisiae telomere repeats to non-telomeric DNA sequences at the ends of linear DNA (Kramer and Haber 1993). It can be argued that sub-terminal DNA sequences, resembling S. cerevisiae [TGI_3] n telomere repeats, rather than foreign telomeres, are responsible for rescuing linear plasmids in S. cerevisiae. This possibility does not explain the selective enrichment of telomeres from other genome components that is obtained by functional eomplementation. Telomere repeats have been located at internal chromosome sites in some species (for example, Southern 1970). Although it is likely that the C. reinhardtii telomere repeats isolated in S. cerevisiae were derived from the end o f a C. reinhardtii chromosome, an internal origin cannot be excluded. The C. reinhardtii telomere was capped by approximately 250 bases of S. cerevisiae telomere repeats. This length is similar to the sizes of endogenous S. cerevisiae telomeres (Shampay et al. 1984) indicating that C. reinhardtii telomere repeats cannot replace S. cerevisiae telomere repeats. The function of C. reinhardtii telomere repeats in S. cerevisiae probably does not extend beyond seeding the formation of S. cerevisiae telomere repeats. The observation that C. reinhardtii telomere repeats can be recognized by the telomere maintenance machinery of S. cerevisiae is consistent with a common theme for telomere structure. This theme would presumably include conservation of the folded structures involving the singlestranded 3' overhangs of the G-rich strand. Although C. reinhardtii telomeres do not from G-quartet structures readily in vitro it is possible that association of these overhangs with proteins may promote their formation in vivo (Petracek and Berman 1992). A protein that promotes G-quartet formation has recently been identified in Oxytricha (Fang and Cech 1993). If G-quartets were important for stabilization of the C. reinhardtii telomere in S. cerevisiae, then their formation could conceivably have been promoted by telomere-binding proteins present in S. cerevisiae.

Acknowledgements We thank Drs W. R. A. Brown for plasmid pTV2, C. Tyler-Smith and R. Fisher for S. cerevisiae strains, S. Taylor for help with electroporation of E. coli and Mr K. Johnson for photography. This work was supported by the Agricultural and Food Research Council, UK.

References Brown WRA (1989) Nature 338:774-776 Burgers PMJ, Percival KJ (1987) Anal Biochem 163:391-397 Burke DT, Carle GF, Olson MV (1987) Science 236:806-812 Cross SH, Allshire RC, McKay SJ, McGill NI, Cooke HJ (1989) Nature 338:771-774 Cryer DR, Eccleshall R, Marmur J (1975) Methods Cell Biot 12:39-44 Day A, Schirmer-Rahire M, Kuchka MR, Mayfield SP, Rochaix J-D (1988) EMBO J 7:1917-1927 Dower WJ, Miller JF, Ragsdale CW (1988) Nucleic Acids Res 16:6127-6145 Fang G, Cech TR (1993) Cell 74:875-885 Feinberg AR Vogelstein B (1983) Anal Biochem 137:266-267 Hails T, Jobling MJ, Day A (1993) Chromosoma 102:500-507 Henikoff S (1987) Methods Enzymol 155:156-165 Kaiser K, Murray NE (1985) In Glover DM (ed) DNA cloning - a practical approach, vol 1. IRL Press, Oxford, pp 1-47 Khandjian EW (1987) Biotechnology 5:165-167 : Kramer KM, Haber JE (1993) Genes Dev 7:2345-2356 Manuelidis L (1977) Anal Biochem 78:561-568 Murchie AIH, Lilley DM (1994) EMBO J 13:993-1001 Pardue ML (1994) Curr Opin Genet Dev 4:845-850 Petracek ME, Berman J (1992) Nucleic Acids Res 20:89-95 Petracek ME, Lefebvre PA, Silflow CD, Berman J (1990) Proc Natl Acad Sci USA 87:8222-8226 Riethman HC, Moyzis RK, Meyne J, Burke DT, Olson MV (1989) Proc Natl Acad Sci USA 86:6240-6244 Richards EJ, Chao S, Vongs A, Yang J (1992) Nucleic Acids Res 20:4039-4046 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual, 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York Shampay J, Szostak JW, Blackburn EH (1984) Nature 310:154-157 Sharpe JA, Day A (1993) Mol Gen Genet 237:134-144 Sherman F, Fink GR, Hicks JB (1986) Laboratory course manual for methods in yeast genetics. Cold Spring Harbor Laboratory, Cold Spring Harbour, New York Southern EM (1970) Nature 227:794-798 Southern EM (1975) J Mol Biol 98:503-518 Southern EM (1979) Anal Biochem 100:319-323 Szostak JW, Blackburn EH (1982) Cell 29:245-255