t-loops at trypanosome telomeres
2001; Springer Nature; Volume: 20; Issue: 3 Linguagem: Inglês
10.1093/emboj/20.3.579
ISSN1460-2075
Autores Tópico(s)CRISPR and Genetic Engineering
ResumoArticle1 February 2001free access t-loops at trypanosome telomeres Jorge L. Muñoz-Jordán Jorge L. Muñoz-Jordán Laboratory of Molecular Parasitology, 1230 York Avenue, New York, NY, 10021 USA Laboratory of Cell Biology and Genetics, Box 159, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author George A.M. Cross George A.M. Cross Laboratory of Molecular Parasitology, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author Titia de Lange Corresponding Author Titia de Lange Laboratory of Cell Biology and Genetics, Box 159, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author Jack D. Griffith Jack D. Griffith Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, 27599 USA Search for more papers by this author Jorge L. Muñoz-Jordán Jorge L. Muñoz-Jordán Laboratory of Molecular Parasitology, 1230 York Avenue, New York, NY, 10021 USA Laboratory of Cell Biology and Genetics, Box 159, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author George A.M. Cross George A.M. Cross Laboratory of Molecular Parasitology, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author Titia de Lange Corresponding Author Titia de Lange Laboratory of Cell Biology and Genetics, Box 159, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author Jack D. Griffith Jack D. Griffith Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, 27599 USA Search for more papers by this author Author Information Jorge L. Muñoz-Jordán1,2, George A.M. Cross1, Titia de Lange 2 and Jack D. Griffith3 1Laboratory of Molecular Parasitology, 1230 York Avenue, New York, NY, 10021 USA 2Laboratory of Cell Biology and Genetics, Box 159, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA 3Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, 27599 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (2001)20:579-588https://doi.org/10.1093/emboj/20.3.579 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Mammalian telomeres form large duplex loops (t-loops) that may sequester chromosome ends by invasion of the 3′ TTAGGG overhang into the duplex TTAGGG repeat array. Here we document t-loops in Trypanosoma brucei, a kinetoplastid protozoan with abundant telomeres due to the presence of many minichromosomes. These telomeres contained 10–20 kb duplex TTAGGG repeats and a 3′ TTAGGG overhang. Electron microscopy of psoralen/UV cross-linked DNA revealed t-loops in enriched telomeric restriction fragments and at the ends of isolated minichromosomes. In mammals, t-loops are large (up to 25 kb), often comprising most of the telomere. Despite similar telomere lengths, trypanosome t-loops were much smaller (∼1 kb), indicating that t-loop sizes are regulated. Coating of non-cross-linked minichromosomes with Escherichia coli single-strand binding protein (SSB) often revealed 3′ overhangs at both telomeres and several cross-linked minichromosomes had t-loops at both ends. These results suggest that t-loops and their prerequisite 3′ tails can be formed on the products of both leading and lagging strand synthesis. We conclude that t-loops are a conserved feature of eukaryotic telomeres. Introduction The advent of linear chromosomes in eukaryotes was accompanied by the acquisition of specialized terminal structures that preserve chromosome ends. Most eukaryotic telomeres feature a tandem array of short repeats and a 3′ overhang (Wellinger and Sen, 1997). These telomeres are maintained by telomerase, which adds telomeric repeats to the 3′ end of the chromosome (Greider and Blackburn, 1985; Nugent and Lundblad, 1998). Telomerase-mediated telomere elongation is the predominant mechanism by which eukaryotes compensate for the failure of lagging strand synthesis to complete the replication of terminal sequences. Telomeres protect chromosome ends against degradation and end-to-end fusion, and they prevent inappropriate activation of checkpoint pathways that respond to chromosome breaks (Muller, 1938; McClintock, 1941; Sandell and Zakian, 1993; van Steensel et al., 1998; Karlseder et al., 1999). This capping function is mediated by telomere-associated proteins. The stability of Saccharomyces cerevisiae telomeres depends on Cdc13p, a single-stranded telomeric DNA binding protein that protects telomeres from degradation and prevents activation of the RAD9 checkpoint pathway (Garvik et al., 1995; Lin and Zakian, 1996; Nugent et al., 1996). Similarly, hypotrichous ciliates have short [ 5 Mbp, which carry the essential genes, and ∼100 minichromosomes, of 25–150 kb, which are predominantly composed of tandem 177 bp repeats (Weiden et al., 1991). Minichromosomes have canonical telomeres at both ends and some telomeres carry silent VSG genes that can contribute to the expressed VSG repertoire through transposition to ESs in the large chromosomes (van der Ploeg et al., 1984; Weiden et al., 1991). Minichromosomes represent an abundant source of telomeres, which were used in this study to address the evolutionary conservation of t-loops. Results Trypanosoma brucei telomeres carry ∼10–20 kb of TTAGGG repeats and have a 3′ overhang It was previously shown that T.brucei telomeres are composed of TTAGGG repeats and that most TTAGGG repeats in the trypanosome genome occupy terminal sites, based on their sensitivity to exonuclease treatment of intact genomic DNA (Blackburn and Challoner, 1984; van der Ploeg et al., 1984). In order to determine the median length of the TTAGGG repeat arrays in the T.brucei line used in this study, we applied a technique previously used to measure the length of human telomeres (Saltman et al., 1993). The rate at which the exonuclease Bal31 removes TTAGGG repeat hybridization signal is compared with the rate at which the enzyme shortens terminal DNA fragments. In T.brucei, this approach is facilitated by the availability of probes for subtelomeric VSG genes and the detailed knowledge of the restriction maps of these loci, allowing precise measurements of Bal31 digestion rates on well defined terminal restriction fragments. Digestion of DNA from BF and PF T.brucei with frequently cutting enzymes yielded telomeric fragments in the 10–20 kb range (Figure 1A), suggesting that the telomeres contain long arrays of TTAGGG repeats. To measure the length of the telomeric repeat array directly, intact PF DNA was treated with Bal31 exonuclease for increasing times, digested with HinfI–AluI–RsaI and hybridized to a TTAGGG repeat probe (Figure 1B). Quantification of the TTAGGG repeat signal at each time point indicated that the exonuclease removed ∼2.6% of the TTAGGG repeat signal per minute. The rate at which Bal31 shortened the telomeric fragment that carries VSG 221 was determined in parallel (Figure 1B; see Figure 1C for restriction map). The same shortening rate was found for the telomeric restriction fragment carrying VSG 121 (Figure 1B; see Figure 1C for restriction map). As expected, Bal31 did not affect chromosome-internal restriction fragments, such as those carrying non-telomeric copies of VSG 121 (Figure 1B). Comparison of the two rates (Figure 1D) showed that Bal31 removed ∼10% of the TTAGGG repeat signal in the time needed to shorten the telomere by 1.5 kb, implying that the average length of the TTAGGG repeat array was ∼15 kb. This value for the length of the TTAGGG repeats array was in agreement with the median length for the telomeric fragments observed in DNA digested with HinfI and RsaI (Figure 1A), which are expected to remove most of the subtelomeric sequences from the terminal fragments. Figure 1.Analysis of T.brucei telomeric DNA. (A) Southern blotting analysis of telomeric restriction fragments in DNA from PF and BF trypanosomes. The DNAs were digested with RsaI (R), HinfI (H), RsaI–HinfI (RH), MboI (M) or AluI (A). The left panel shows the ethidium bromide; the right panel shows a Southern blot probed with a (TTAGGG)27 probe. (B) Bal31 digestion of trypanosome telomeric DNA. Intact genomic DNA from BF trypanosomes was treated with Bal31 for the indicated times (in minutes) and digested with AluI–HinfI–RsaI (left panel), EcoRI (middle) or XmnI (right panel) and probed as indicated below the panels. The arrows indicate three non-telomeric VSG 121 fragments. (C) Restriction maps of the telomeric VSG 221 and 121 loci and the chromosome internal 121 genes. E, EcoRI; X, XmnI. (D) Graph of the rate at which Bal31 removed the TTAGGG repeat signal in (B) plotted against the rate at which the exonuclease shortened the VSG 221 telomeric EcoRI fragment in (B). (E) Overhang assay. DNA from BF and PF was digested with AluI–HinfI–RsaI after treatment with (+) or without (−) E.coli exonuclease I. DNA was incubated with radiolabeled single-stranded [TTAGGG]4 or [AATCCC]4 probes as indicated and separated by agarose gel electrophoresis (van Steensel et al., 1998). The gel was dried and exposed on a PhosphorImager (Molecular Dynamics). The signal at the front represents free probe. Download figure Download PowerPoint We next determined whether T.brucei chromosome ends carry an overhang of the TTAGGG repeat strand. Mammalian chromosomes have up to 200 nt of single-stranded TTAGGG repeats at their 3′ termini (Makarov et al., 1997; McElligott and Wellinger, 1997; Wright et al., 1997; van Steensel et al., 1998; Huffman et al., 2000) and this overhang is presumed to be important for the formation of t-loops. The presence of single-stranded TTAGGG repeats can be assessed by annealing labeled C-strand-specific oligonucleotides to genomic DNA. Using this approach, we found that native DNA from T.brucei contained single-stranded TTAGGG repeats (Figure 1E). The signal was not detected in DNA digested with E.coli exonuclease I, which is specific for 3′ single-stranded tails, or in a control hybridization with an oligonucleotide representing the G-rich telomeric strand, consistent with the signal being derived from 3′ single-stranded TTAGGG tails. Furthermore, the signals were present on large fragments (>10 kb) in DNA that was digested with AluI–HinfI–RsaI, as would be expected if the G-tails are present at the ends of the trypanosome telomeres. The presence of single-stranded telomeric tails was corroborated by EM analysis of minichromosomes coated with E.coli SSB (see below). These data indicate that T.brucei telomeres resemble human telomeres in both the length of the TTAGGG repeat array and the presence of a 3′ [TTAGGG]n overhang. t-loops in T.brucei telomeric DNA from procyclic and bloodstream forms The telomeric repeat arrays of T.brucei are sufficiently long to allow their isolation by differential size fractionation after digestion of genomic DNA with frequently cutting restriction endonucleases (see Figure 1A). We previously employed this approach to isolate telomeric DNA from the human genome, for cloning and for EM visualization (de Lange et al., 1990; Griffith et al., 1999). Furthermore, the TTAGGG sequence of trypanosome telomeres, like their human counterparts, lends itself to cross-linking with psoralen (AMT) and UV light, which cross-links T residues on opposite strands at AT steps, potentially stabilizing t-loops during their isolation. Accordingly, PF or BF T.brucei were permeabilized with digitonin and treated with AMT and UV light. Following deproteinization, the DNA was cleaved with AluI–HinfI–RsaI and size fractionated on a Bio-Gel A-15m matrix. Fractions containing large DNA fragments were then prepared for EM by spreading on a denatured film of cytochrome c protein, followed by rotary shadowcasting. EM examination revealed the presence of long DNA molecules is the early eluting (high molecular weight) fractions of the Bio-Gel column, and many of these molecules contained loops at one end. In five experiments, the fraction of long linear DNAs (≥10 kb) containing a loop at one end varied from 8% to as much as 25% (n ≥100 for each experiment), which is a frequency of t-loops very similar to that observed in mammalian telomeric restriction fragments. The structure of trypanosome t-loops appeared to be similar to that of t-loops from mammalian cells, containing a single terminal loop of variable size and a variably sized unforked tail. Examples of trypanosome t-loops are shown in Figure 2, where the loops vary in size from 6.3 (A) to 0.63 kb (F). No difference was observed in t-loop frequency or structure in DNA from BF and PF trypanosomes. Figure 2.Visualization of t-loops from T.brucei DNA photo-cross-linked with AMT and UV light. Trypanosomes were permeabilized with digitonin and treated with AMT and UV light, followed by endonuclease cleavage of purified DNA and isolation of the telomeric restriction fragments by gel filtration. DNA fragments were prepared for EM by spreading on a denatured film of cytochrome c protein and rotary shadowcasting with platinum:paladium. Shown in reverse contrast. t-loops shown in (A–F) measured 6.3, 1.75, 1.5, 1.2, 0.99 and 0.63 kb, respectively. Bar is equivalent to 1 kb. Download figure Download PowerPoint Trypanosome t-loops are small Measurement of loop contour lengths of t-loops in enriched telomeric restriction fragments showed a wide variety of sizes ranging from as small as 0.3 kb to as large as 8 kb. However, >65% of the loops were quite small (<1.5 kb) and the median length of the 48 t-loops from telomeric restriction fragments was ∼1.1 kb. Similarly, t-loops at the ends of isolated minichromosomes (see below) showed a range in loop sizes from as small as 570 bp to as large as 8.4 kb, with a median value of 1.0 kb for 21 loops analyzed. The combined data on the size range of the t-loops in both types of DNA preparations are given in Figure 3. Overall, the median size of the loops was 1.1 kb, and 42 out of 69 t-loops analyzed were very small, ranging between 0.5 and 1.0 kb. Furthermore, a number of trypanosome t-loops measured 3 kb) were rare in both the enriched telomeric restriction fragments and in isolated minichromosomes. Figure 3.Size distribution of trypanosome t-loops. Bar graph depicting the size distribution of T.brucei t-loops. The data were obtained from measurements of 48 t-loops in enriched telomeric restriction fragments (examples shown in Figure 2) and from measurements of 21 t-loops at the ends of minichromosomes (examples shown in Figure 5). Download figure Download PowerPoint The level of resolution of the surface spreading method employed here is such that circles of <150–200 bp would frequently appear as balls rather than small loops or donuts. Examination of the minichromosomes by directly adsorbing the samples to carbon supports and rotary shadowcasting in the absence of denatured protein allowed a higher resolution inspection of the DNA ends. However, even using this technique, no examples of circles smaller than those detected by surface spreading were observed (data not shown). Nonetheless, it remains possible that some very small loops were present and not scored in these experiments. t-loops and single-stranded tails at both ends of T.brucei minichromosomes We next asked whether t-loops can occur at both ends of a chromosome. Trypanosome minichromosomes are sufficiently small to allow their visualization as intact molecules, allowing inspection of both ends of each chromosome (Weiden et al., 1991). To isolate minichromosomes for this purpose, permeabilized trypanosomes were treated with psoralen and UV, gently lysed, deproteinized, and sedimented through a 5–20% sucrose gradient. Gradient fractions containing minichromosomes were identified by gel electrophoresis, under conditions that separate minichromosomes from larger chromosomes, followed by detection of telomeric DNA with a TTAGGG repeat probe. Using this approach, fractions were identified that were highly enriched for minichromosomal DNA (Figure 4A and B). These fractions appeared to lack DNA derived from the larger chromosomes because they did not contain detectable amounts of an abundant 50 bp repeat element that is present upstream of ESs on the larger chromosomes (Melville et al., 1998) (compare fractions to total DNA in Figure 4C). Figure 4.Isolation and analysis of minichromosomes. (A) Southern blot of fractions from sedimentation of BF trypanosome chromosomal DNA through a linear 5–20% sucrose gradient. Fractions were analyzed by rotating agarose gel electrophoresis (RAGE) and probed with [TTAGGG]27. Two fractions from BF chromosomes (10 and 11) and two fractions from a parallel sedimentation of PF chromosomes (11 and 12) are compared alongside input PF DNA prior to sedimentation (T). Blots were probed with TTAGGG repeats (B) or a 50 bp repeat probe (C). Download figure Download PowerPoint The enriched minichromosomal fractions were analyzed by EM and found to contain linear DNA molecules ranging from 20 to 50 kb. This size range is about half that expected from measurements by EM and gel electrophoresis (Figure 4), suggesting that some of the minichromosomal DNAs were broken during isolation. However, EM analysis showed that 14 out of 144 molecules contained a small t-loop at one end. In a second experiment, 23 out of 115 large molecules showed a loop at one end. Thus, overall, ∼15% of the ends had a t-loop. In these experiments we found four minichromosomal DNAs that carried t-loops at both ends (Figure 5) and a fifth double-looped minichromosome was found in a third experiment. If the t-loop frequencies in our preparations were primarily determined by the extent to which t-loops were lost during DNA isolation due to incomplete cross-linking or breakage, we would expect that the frequency of double-looped molecules would be ∼2.25% (15% of 15%), predicting approximately six double-looped molecules in the 259 DNAs that were examined. This number is in reasonable agreement with the four double-looped molecules that were observed, suggesting that t-loops often occur at both ends of minichromosomes. Figure 5.t-loops in T.brucei minichromosomes. Minichromosomes enriched by sucrose gradient sedimentation were prepared for EM as described in Materials and methods. The minichromosome in (A) measures 28.7 kb, and the loops at the left and right ends measure 650 and 710 bp, respectively. The minichromosome in (B) is 30.5 kb, with loops of 1310 and 790 bp at the left and right ends, respectively. The molecule in (C) is 21.1 kb (most likely a broken minichromosome) and the loop at the left end is 1470 bp. Shown in reverse contrast. Bar is equivalent to 5 kb. Download figure Download PowerPoint A 3′ overhang of single-stranded telomeric repeats is likely to be a prerequisite for t-loop formation. However, based on the mechanism of DNA replication, 3′ overhangs are not expected to occur at chromosome ends formed by leading strand DNA synthesis and there are conflicting reports on whether both ends of human chromosomes have a single-stranded tail (Makarov et al., 1997; Wright et al., 1997). To address this issue we used E.coli SSB to query the status of the DNA at the ends of trypanosome minichromosomes (Figure 6). Trypanosoma brucei minichromosomes were prepared by lysis of PF cells and sucrose gradient sedimentation in the presence of sarcosyl without AMT and UV treatment. Aliquots were then chromatographed over Bio-Gel A-15m to remove the detergent and the minichromosomes incubated with E.coli SSB protein to bind any single-stranded DNA. Single tetramers or octamers of SSB bound along the length or at the ends of otherwise duplex DNA can be distinguished by EM, and represent the presence of ∼75 (single tetramer) or 150 nt (octamer) of single-stranded DNA (Chrysogelos and Griffith, 1982). The minimum length of a single-stranded DNA overhang that will allow binding of an SSB tetramer has not been established. Thus, overhangs less than ∼75 nt may be missed using this approach. Following preparation of the complexes for EM, examination of 138 minichromosomes judged to be ≥50 kb revealed that 70% showed no SSB on either end, 23% showed SSB binding at one end and 7% had SSB at both ends (Figure 6). Figure 6.Trypanosoma brucei minichromosomes with single-strand overhangs stained with SSB. Non-cross-linked minichromosomes enriched by sucrose gradient sedimentation were incubated with E.coli SSB and prepared for EM as described in Materials and methods, including adsorption to thin carbon foils, dehydration and rotary shadowcasting with tungsten. (A) A 31 kb minichromosome with single-stranded overhangs at both ends. The overhangs on this molecule are longer than most and were selected for greater visibility at low magnification. (B–E) Individual minichromosome ends with SSB bound. The size of the particle in (E) corresponds to a single SSB tetramer. Shown in reverse contrast. Bar equals 0.5 μm (A) and 0.27 μm (B–E). Download figure Download PowerPoint To evaluate the length of the overhang, the number of SSB tetramers bound at an end was counted. For the minichromosomes with SSB bound at just one end or at both ends, 71 and 70%, respectively, of the ends showed from one to three tetramers bound, suggestive of overhangs in the range of 75–225 nt. The remaining 30% of the ends showed longer SSB-bound tracts ranging up to ∼500 nt. Thus, consistent with the annealing data in Figure 1, trypanosome chromosome ends contain substantial regions of single-stranded DNA and these overhangs can occur at both ends of the same chromosome. Discussion This report documents the presence of t-loops at chromosome ends in trypanosomes. Although they have the same sequence and overall length, trypanosome telomeres had loops that were significantly smaller than those of mammalian telomeres, indicating that t-loop sizes are determined by a specific mechanism. A significant fraction of isolated intact trypanosome minichromosomes had two t-loops and carried single-stranded overhangs at both ends, showing that telomeres generated by both leading and lagging strand synthesis can be remodeled into t-loops. Generation of the 3′ overhang for t-loop formation at the end duplicated by leading strand synthesis must involve post-replicative modification since the replication product is predicted to be blunt. Our findings, together with the demonstration of t-loops in mammals and ciliates, indicate that they are a conserved feature of eukaryotic telomeres and their presence at both ends of a chromosome is consistent with a requirement for t-loops in the function of all telomeres. Trypanosome t-loops are relatively small The trypanosome telomeres analyzed in this study are composed of 10–20 kb of TTAGGG repeats. Human telomeres have a very similar structure, containing a duplex TTAGGG repeat array in the 5–20 kb range. Despite these similarities, the size distribution of the t-loops observed in these two species was significantly different. Trypanosome telomeres often had small t-loops (median size 1.1 kb), whereas human telomeres very rarely showed t-loops in that size range. For instance, HeLa cells with telomeres in the 20 kb range had t-loops with a median size of 14 kb and 500 million years ago, long before the origin of their metazoan hosts (Stevens and Gibson, 1999). The molecular biology of these highly diverged protozoa is quite distinct from the perceived norm as represented by yeast, plants and mammals. For instance, trypanosomes have a specialized organelle for glycolysis, their mitochondria contain an unusual network of small circular DNAs, mitochondrial RN
Referência(s)