Molecular analysis of T-DNA insertion mutants identified putative regulatory elements in the AtTERT gene.

Analysis of plants bearing a T-DNA insertion is a potent tool of modern molecular biology, providing valuable information about the function and involvement of genes in metabolic pathways. A collection of 12 Arabidopsis thaliana lines with T-DNA insertions in the gene coding for the catalytic subunit of telomerase (AtTERT) and in adjacent regions was screened for telomerase activity [telomere repeat amplification protocol (TRAP) assay], telomere length (terminal restriction fragments), and AtTERT transcription (quantitative reverse transcription-PCR). Lines with the insertion located upstream of the start codon displayed unchanged telomere stability and telomerase activity, defining a putative minimal AtTERT promoter and the presence of a regulatory element linked to increased transcription in the line SALK_048471. Lines bearing a T-DNA insertion inside the protein-coding region showed telomere shortening and lack of telomerase activity. Transcription in most of these lines was unchanged upstream of the T-DNA insertion, while it was notably decreased downstream. The expression profile varied markedly in mutant lines harbouring insertions at the 5' end of AtTERT which showed increased transcription and abolished tissue specificity. Moreover, the line FLAG_385G01 (T-DNA insertion inside intron 1) revealed the presence of a highly abundant downstream transcript with normal splicing but without active telomerase. The role of regulatory elements found along the AtTERT gene is discussed in respect to natural telomerase expression and putative intron-mediated enhancement.

Introduction

Telomerase is a ribonucleoprotein complex able to add telomeric repeats to chromosome ends and thus to elongate telomeres. Telomerase consists of the catalytic subunit TERT (TElomerase Reverse Transcriptase), and the TR (Telomerase RNA) subunit which serves as a template for the telomere motif elongation. Both telomerase subunits have been characterized in many model organisms (yeast, protozoa, and humans). In plants, the TERT gene has been cloned in Arabidopsis (Fitzgerald ), Oryza (Heller-Uszynska ), and Asparagales species (Sykorova ), and more plant TERT genes were identified in silico in sequenced genomes (reviewed in Podlevsky ; Sykorova and Fajkus, 2009). The plant TERT genes consist of 12 exons (Fitzgerald ; Oguchi ) (Fig. 1C), and all the basic functional domains reported previously for TERT genes from other model organisms (Nakamura ) have been identified in Arabidopsis thaliana TERT (AtTERT) (Fig. 1A, C). The original identification of the AtTERT gene was achieved using a T-DNA insertion line with the insertion inside the telomerase-specific T motif. T-DNA insertion in exon 9 of the AtTERT gene (SALK_061434, Fig. 1A, C) resulted in disruption of telomerase activity in telomerase-positive tissues and progressive telomere shortening during propagation of the mutant line, but without apparent phenotypic defects at least in early plant generations (Fitzgerald ; Ruckova ).

Fig. 1.

Description of the AtTERT gene and of the experimental strategy. (A) Conserved motifs in the AtTERT protein are highlighted: reverse transcriptase motifs (1, 2, A, B', C, D, E); telomerase-specific motives (T2, CP, QFP, T); NLS, nuclear localization-like signal. Details are given in Sykorova and Sykorova and Fajkus (2009). (B) Scheme of the experimental strategy. T-DNA insertion lines obtained from the respective collections were propagated (generation 1; G1 plants represent T3 progeny of the original accession) and genotyped in each generation. Two homozygous (–/–) individual gene lines and four homozygous individual upstream lines were propagated up to the fourth generation (G4); pool, seeds from all plants of the respective individual mutant line were pooled for subsequent cultivation. (C) Structure of the AtTERT gene and positions of T-DNA insertion lines. Exons (boxes, grey) and introns (lines) are depicted approximately true to scale according to their lengths. Positions of conserved regions are highlighted as in A. Triangles indicate the positions of T-DNA insertion; names of mutant lines are shortened as noted in the text. Lines marked by black triangles (S_575 and S_110) were excluded from the analysis. Positions of primers (Supplementary Table S1 at JXB online) used in analysis of AtTERT transcription are given below the gene map. (D) Splicing variant of the AtTERT gene. Open boxes indicate introns retained in the mRNA sequence. The asterisk in exon 1 (C, D) shows the position of the start codon according to Rossignol ; the cross in intron 6 (and in exon 12) shows the position of the stop codon.

The detailed molecular mechanisms of telomerase regulation at both the cellular and organism levels are far from being elucidated. Analysis of tobacco suspension cell cultures showed low telomerase activity except in early S phase (Tamura ), demonstrating cell cycle-dependent telomerase regulation. In both plant and animal models, a correlation between the level of TERT mRNA and telomerase activity was reported, pointing to regulation of TERT transcription as an important factor (Meyerson ; Fitzgerald ; Oguchi ). Moreover, post-translational regulation of telomerase activity by phosphorylation was reported in mammals (Liu ) and in rice (Oguchi ). The dynamic structure of telomeres represents another level of regulation of telomerase activity (reviewed in Blackburn, 2000) in which telomere binding- and telomere-associated proteins play a crucial role.

Alternatively spliced TERT transcript variants and their specific function in the regulation of telomerase activity and telomere homeostasis have been described in both human and plant systems. In humans, two basic TERT deletion variants have been identified: an α deletion (Colgin ; Yi ) and a β deletion (Kilian ). These hTERT forms are enzymatically inactive; moreover, an overexpressed α deletion variant can act as a strong telomerase inhibitor. Apart for these splicing events, their combination (α+β), insertions, and intron retentions were described (for a review, see Sykorova and Fajkus, 2009); expression of the respective variants was quantified and correlated with distinct cell types, developmental stage, or pathological processes (Ulaner , 2001; Kaneko ). Reconstitutions of human telomerase reveal ‘minimal’ hTERT regions indispensable for the formation of an active enzyme complex which differ under in vitro and in vivo experimental conditions (Beattie ).

In plants, a number of alternatively spliced TERT isoforms have been identified in Oryza (Heller-Uszynska ; Oguchi ) and Asparagales (Sykorova ). Detailed analysis of the A. thaliana TERT V(I8) isoform (Fig. 1D; GenBank accession no. AM384991; Rossignol ) revealed its ability to bind the telomere-binding protein AtPOT1a effectively, pointing to a biological significance of the splicing variant in the plant system.

The identification and description of plant gene functions are mainly based on analysis of T-DNA insertion lines. Insertion into either an exon or an intron in the protein-coding region is equally effective in knocking out the target gene (86%), while insertions before the start codon or after the stop codon are slightly less effective (Wang, 2008). However, when a transcript is produced due to expression of a fusion transcript driven by a strong T-DNA promoter or is truncated due to transcription termination by the insertion, it can still be translated into a functional protein (Ren ; Dohmann ; Ohtomo ; Wilmoth ) or into a truncated protein lacking essential domains (Kim ; Henderson ; for a review, see Wang, 2008). To investigate structure–function relationships in TERT subdomains in plants, a collection of A. thaliana T-DNA insertion lines has been utilized (Fig. 1C; Supplementary Fig. S1A available at JXB online) in which the insertion was in distinct AtTERT regions encompassing the N-terminal part with telomerase-specific motifs, the central part (reverse transcriptase motifs), and the C-terminal extension (Fig. 1A), or in the upstream and downstream sequences. Telomerase activity was abolished and telomeres were shortened in all the mutant lines with a T-DNA insertion in the AtTERT gene regardless of the T-DNA position. On the other hand, telomerase function was maintained in the mutant lines with a T-DNA insertion within the AtTERT upstream region, and a correlation between increased transcription level and telomerase activity was observed. T-DNA insertions at the 5' end of the AtTERT gene or in the region upstream of the ATG start codon led to the activation of putative regulatory elements. Possible outcomes of these observations are discussed.

Materials and methods

Plant material and genotyping

Arabidopsis thaliana lines with a T-DNA insertion in the gene coding for telomerase reverse transcriptase AtTERT (At5g16850; Supplementary Fig. S1A at JXB online) were selected from the public T-DNA Express database established by the Salk Institute Genomic Analysis Laboratory accessible at the SIGnAL website http://signal.salk.edu. Seeds were obtained from the Nottingham Arabidopsis Stock Centre [SALK and SAIL lines, Columbia wild type (wt) (Sessions ; Alonso ) and Versailles INRA collection (FLAG lines, Wassilevskija-4 wt (Brunaud )]; for the T-DNA positions, see Fig. 1C. Note that the lines FLAG_490F05 and FLAG_492C08 (F_490/492 in Fig. 1C) represent flanking sequence tags (FSTs) of the same accession (Brunaud ).

Seeds were placed onto half-strength Murashige and Skoog (MS; Duchefa) agar plates and grown under cycles of 8 h light, 21 °C and 16 h dark, 19 °C (except root elongation analysis; see Supplementary methods at JXB online). After 2 weeks, seedlings were potted and grown for 3 weeks under the same conditions until samples of plant material (leaves) for genotyping and genomic DNA were collected. Plants were then grown under 16 h/8 h light/dark cycles suitable for flowering, and seeds from each plant were collected. Individual plants (generation G1) from each T-DNA insertion line were genotyped (see Supplementary Table S1 at JXB online for primer sequences) to select individual segregated wt, heterozygous, and homozygous individual mutant lines. At least two mutant lines for each T-DNA accession (referred to here as individual mutant lines) were propagated up to the fourth generation (G4) and each individual plant was genotyped (Fig. 1B). The seeds from homozygous plants in G2, G3, and G4 were pooled. Heterozygous and wt plants were revealed by genotyping of the line SALK_126201 which is referred to as homozygous for T-DNA insertion in the T-DNA Express database. The lines SALK_110053 and SAIL_575_F07 were excluded from the study because of mis-mapping of their position in the T-DNA Express database. Two control lines were used, the line SALK_061434 previously described in detail (Ruckova ) for the telomere repeat amplification protocol (TRAP) analysis, and the line SAIL_1287_C04 located 3174 bp upstream of the start codon (Fig. 1C) for the analysis of root growth.

DNA isolation and telomere length analysis

Genomic DNA for genotyping was isolated according to the ‘modified IRRI method’ (Collard ) from leaves of 4- to 5-week-old plants. DNA for analysis of telomere length by the terminal restriction fragments (TRFs) method was isolated according to Dellaporta from rosette leaves of 5- to 7-week-old plants. The quality of DNA was checked and the concentration estimated using electrophoresis on 0.8% (w/v) agarose gels stained with ethidium bromide. The Gene Ruler 1-kb DNA Ladder (Fermentas) was used as a standard and data were analysed by Multi Gauge software (FujiFilm).

Telomere length was assessed as the length of TRFs resulting from the digestion of genomic DNA by a frequently cutting restriction endonuclease whose recognition site is not located within the telomeric region. DNAs (500 ng) isolated from individual plants were cleaved overnight at 65 °C by Tru1I (MseI) (Fermentas) and separated in 0.8% (w/v) agarose gels at 1.5 V cm−1 for 16 h. Agarose gels were alkali blotted onto Hybond XL membranes (GE Healthcare). Membranes were hybridized overnight at 55 °C with the radioactively labelled telomeric oligonucleotide (GGTTTAG)4 and washed twice at 55 °C for 30 min in 2× SSC (1× SSC=150 mM NaCl, 15 mM Na3 citrate, pH 7.0), 0.1% SDS. Signals were visualized using a FLA7000 phosphorimager (FujiFilm). Evaluation of fragment lengths was done by using the Gene Ruler 1-kb DNA Ladder (Fermentas) as a standard; Multi Gauge software (FujiFilm) was used to analyse hybridization patterns. Mean telomere lengths were calculated by the Telometric tool (Grant ).

Analysis of telomerase activity (TRAP assay)

Protein extracts from undeveloped buds and 7-day-old seedlings were prepared as described (Fitzgerald ; Sykorova ) and tested for telomerase activity according to the protocol in Fajkus . Briefly, 1 μl of 10 μM CAMV or TS21 substrate primer (Supplementary Table S1 at JXB online) was mixed with 1 μl of telomerase extract (protein concentration 50 ng μl−1). Primer elongation proceeded in 25 μl of the reaction buffer at 26 °C for 45 min. After extension, telomerase was heat inactivated for 10 min at 95 °C and the sample was cooled to 80 °C. A 1 μl aliquot of 10 μM TELPR reverse primer (Supplementary Table S1) and 2 U of DyNAzymeII DNA Polymerase (Finnzymes) were added to start the PCR step of the TRAP assay (35 cycles of 95 °C/30 s, 65 °C/30 s, 72 °C/30 s) followed by a final extension (72 °C/5 min). Products of the TRAP reactions were analysed by electrophoresis on a 12.5% polyacrylamide gel in 0.5× TBE buffer, the gel was stained with GelStar Nucleic Acid Gel Stain (LONZA), and signals were visualized using the LAS-3000 system (FujiFilm).

The quantitative variant of the TRAP analysis was performed as described in Herbert using FastStart SYBR Green Master (Roche) and TS21 and TELPR primers. Samples were analysed in triplicate. A 1 μl aliquot of extract diluted to 50 ng μl−1 protein concentration was added to the 20 μl reaction mix. Ct values were determined using the Rotorgene3000 (Qiagen) machine software, and relative telomerase activity was calculated by the ΔCt method (Pfaffl, 2004).

RNA isolation and analysis of AtTERT transcription

Total RNA was isolated from undeveloped buds, 7-day-old seedlings, and mature leaves with the RNeasy Plant Mini Kit (Qiagen) followed by treatment by DNase I (TURBO DNA-free, Applied Biosystems/Ambion) according to the manufacturer's instructions. The quality and quantity of RNA were checked by electrophoresis on 1% (w/v) agarose gels and by absorbance measurement (NanoPhotometr IMPLEN). cDNA was prepared by reverse transcription of 1 μg of RNA using the M-MuLV (NEB) reverse transcriptase and Random Nonamers (Sigma). Quantification of the AtTERT transcript level relative to the ubiquitin reference gene was done using FastStart SYBR Green Master (Roche) by the Rotorgene3000 (Qiagen) machine. A 1 μl aliquot of cDNA was added to the 20 μl reaction mix; the final concentration of each forward and reverse primer (sequences are given in Supplementary Table S1) was 0.25 μM. Reactions were done in triplicate; the PCR cycle consisted of 15 min of initial denaturation followed by 40 cycles of 20 s at 94 °C, 20 s at 56 °C, and 20 s at 72 °C. SYBR Green I fluorescence was monitored consecutively after the extension step. The amount of the respective transcript was determined for at least two individual mutant lines, and AtTERT transcription was calculated as the fold increase/decrease relative to the wt plant tissue (ΔΔCt method; Pfaffl, 2004).

Analysis of alternatively spliced AtTERT transcripts was done by PCR using DyNAzyme II (Finnzymes) polymerase. The 20 μl reaction mix consisted of 1 μl of cDNA, 1×DyNAzyme II reaction buffer (Finnzymes), 0.25 μM forward and reverse primers, 1 mM MgCl2, 350 μM of each dNTP, and 1 U of DyNAzyme II polymerase; PCR conditions were the same as for quantitative reverse transcription-PCR (qRT-PCR). Reaction products were analysed by electrophoresis on a 2% (w/v) agarose gel, stained by ethidium bromide, and visualized using the LAS3000 (FujiFilm).

Results

Characterization of T-DNA insertion lines

Publicly accessible A. thaliana lines with insertions mapped inside the AtTERT gene sequence (At5g16850) and adjacent regions were searched for and selected. The AtTERT gene is located on chromosome 5; upstream of its start codon is the At5g16860 gene coding for a putative pentatricopeptide repeat-containing protein (PPR), and downstream of the stop codon is the At5g16840 gene coding for a putative RNA recognition motif-containing protein (RRM) (Fig. 1C; Supplementary Fig. S1A at JXB online). The set of T-DNA insertion lines included accessions from three T-DNA libraries derived from Columbia (Col wt, SALK, and SAIL lines) and from Wassilevskija ecotypes (WS4 wt, FLAG lines). They covered the AtTERT gene region (six lines), the upstream region (five lines), and the downstream region (one line) (Fig. 1C). The lines FLAG_490F05 and FLAG_492C08 from the INRA collection were supplied as one accession because they represent FSTs of the same T-DNA insertion (Supplementary Fig. S1A, B; Brunaud ). Genotyping of the insertion lines (primer sequences are given in Supplementary Table S1 at JXB online) showed that the lines SALK_110053 and SAIL_575_F07 did not contain T-DNA insertions at the mapped positions (the RRM gene and PPR gene, respectively) and they were excluded from this study.

The region upstream of the AtTERT start codon was covered by lines SALK_126201, SALK_048471, FLAG_490F05/FLAG_492C08, and SAIL_1287_C04 (designated here as S_126, S_048, F_490/492, and S_1287; ‘upstream lines’, Fig. 1C) and the gene region from the start to the stop codon was covered by lines FLAG_385G01, FLAG_493F06, SALK_061434, SALK_041265, SALK_050921, and SAIL_284_B07 (designated here as F_385, F_493, S_061, S_041, S_050, and S_284; ‘gene lines’, Fig. 1C). For subsequent analysis, at least two individual mutant lines derived from each gene line and four individual mutant lines from each upstream line classified as homozygous for T-DNA insertion from each T-DNA accession were propagated (Fig. 1B). These individual mutant lines were cultivated up to the fourth generation (G4) which was originally described as critical for telomere shortening but not for plant survival (Ruckova ).

Furthermore, experiments were conducted to determine if T-DNA insertions upstream of AtTERT and in the gene region may influence the development of roots, the typical telomerase-positive tissue (Fitzgerald ). Analysis of the root growth of all lines via calculation of root elongation dynamics (Supplementary Figs S2, S3 at JXB online) did not reveal any changes related to the position of T-DNA insertions in comparison with the segregated wts (Supplementary Figs S2, S3).

Plant lines with T-DNA insertions inside the AtTERT coding region show telomere shortening independently of the T-DNA position

The telomere length was estimated by measuring TRFs, which involves hybridization of digested genomic DNA on Southern blots with a telomeric repeat probe. The hybridization pattern represents the telomeres plus the region adjacent to the most telomere-proximal recognition site of the restriction enzyme used. The lengths of telomeric tracts differ in A. thaliana ecotypes (Gallego and White, 2001; Riha ; Shakirov and Shippen, 2004), giving a different starting point for telomere shortening. The TRF lengths reach 2–4 kb in the Columbia ecotype (wt of the SALK and SAIL mutant lines) and 3.5–8 kb in the Wassilevskija ecotype (wt of the FLAG mutant lines). The upstream lines S_126, S_048, and F_490/492 showed telomere lengths fully comparable with the corresponding wts (examples are shown in Fig. 2C). Telomeric tracts in all gene lines were remarkably shortened as compared with the wt (as shown for F_385/22 in Fig. 2A, B) using three criteria for evaluation of TRFs: absolute maximum and minimum per generation, and weighted mean TRF length for each individual plant (Fig. 2B, C). Weighted mean TRF length was calculated as the mean TRF adjusted by the decreased signal of the telomere probe in the shorter TRF in relation to the unchanged portion of the non-telomeric part of the TRF (Grant ). Similar shortening of telomeres was reported previously for the original AtTERT mutant line S_061 (shortening by 250–500 bp per generation; Fitzgerald ; Ruckova ).

Fig. 2.

Analysis of the telomere length in T-DNA insertion lines. (A) Example of the telomere length analysis. The lengths of the terminal restriction fragments (TRFs) were determined in four individual plants (numbers above the panel) of each generation (G2, G3, and G4) of the individual mutant line F_385/22 and compared with Wassilevskija-4 wt (Ws, three plants). Note a significant shift of hybridization signals to lower molecular weights in the mutant samples. The arrow delimits the position of the internal telomere-like sequence near the centromere (∼500 bp) which serves as a control (A, B). Marker lengths are shown in kb (1-kb DNA Gene Ruler Ladder, Fermentas). (B) An example of TRF evaluation is shown using lane intensity charts in which the x-axis of the graphs shows a reference value (marker in kb) and the y-axis value corresponds to the pixel intensity at each point along the lane. Open squares show the absolute maximum and minimum per generation in mutant and control plants (as in C). (C) Graphical representation of the TRF results shown in A. Evaluation of TRF lengths from one and two individual mutant lines of upstream and gene lines, respectively, in comparison with the wild types is shown. The TRF lengths were calculated using the Multi Gauge program (FujiFilm) from the hybridization pattern of four individual mutant plants of each generation (G2, G3, and G4) and wt plants (Col, four plants; Ws, three plants). The absolute minimum and maximum (open squares) are shown per generation; the weighted mean telomere length from each line (filled squares) was calculated using the Telometric tool (Grant ).

The splicing variant AtTERT V(I8) (Fig. 1D) with suggested telomeric function as an interaction partner of AtPOT1a (Rossignol ) could possibly be transcribed and translated in the gene lines S_041, S_050, and S_284 that have the T-DNA insertion at the 3' end of the TERT gene and downstream of the AtTERT V(I8) region (see Fig. 1C). However, no difference was observed in telomere length dynamics between gene lines with the T-DNA insertion located at the 5' and the 3' regions of the AtTERT gene (Fig. 2C). These results suggest that (i) shortening of telomeres in gene lines was not influenced by the position of the T-DNA insertion and (ii) telomere maintenance was not affected in upstream lines.

Telomerase activity is disrupted in lines with a T-DNA insertion inside the AtTERT coding region, while insertion upstream of the ATG start codon shows no effect

To investigate whether the distance of the T-DNA insertion from the AtTERT start codon in upstream lines influences telomerase activity, TRAP assays were performed in homozygous, heterozygous, and segregated wt plants of S_126 (insertion at position –271 bp relative to the ATG start codon), S_048 (–360 bp), F_490/492 (–441 bp), and S_1287 (–3 174 bp, not shown) using the CAMV and TELPR primer set (Fig. 3; Supplementary Table S1 at JXB online). Telomerase extracts prepared from buds of the first generation (G1, Fig. 1B) of all upstream lines showed telomerase activity comparable with that in the corresponding wt tissues (Fig. 3, left panels). This result and the position of the T-DNA insertion in S_126 suggest that the region of 271 bp upstream of the ATG start codon is sufficient to act as a putative ‘minimal promoter’ (E Sýkorová et al., unpublished results). Telomerase activities in gene lines harbouring insertions inside the protein-coding region (F_385, F_493, S_061, S_050, S_041, and S_284) were examined using protein extracts prepared from G1 buds of homozygous, heterozygous, and segregated wt individual plants. Extracts from heterozygous plants showed telomerase activity similar to that of wt plants, but no activity was observed in homozygous mutant plants (Fig. 3, right panels).

Fig. 3.

Analysis of telomerase activity in buds collected from plants of the first generation of T-DNA insertion lines. High telomerase activity was detected in all upstream line samples homozygous (–/– lines) or heterozygous (+/–) for T-DNA insertions and in segregated wt plants (+/+) using the CAMV×TELPR primer set. In individuals of gene lines heterozygous for T-DNA insertion and in segregated wt plants, active telomerase was revealed; no telomerase activity was detected in individuals of gene lines homozygous for T-DNA insertion (except for bands of high molecular weight; see Supplementary methods, Supplementary Figs S4, S5 at JXB online). NC, negative control (no protein extract in TRAP reaction); TSR8, control template from the TRAPeze® XL Telomerase Detection Kit (Millipore); S_061/7, control line homozygous for T-DNA insertion in exon 9 of the AtTERT gene (Ruckova ).

The TRAP assay with the TS21 substrate primer (Supplementary Table S1) confirmed the loss of telomerase activity in gene lines, and active telomerase was detected in seedlings and buds of all upstream lines analysed up to the fourth generation (Supplementary Fig. S5). The primer TS21 was chosen as an alternative substrate primer also suitable for the modified quantitative TRAP assay (see below) instead of the CAMV primer which showed the presence of high molecular weight TRAP products (details in Supplementary methods and Supplementary Figs S4, S5 at JXB online).

Quantitative TRAP assays using TS21+TELPR primers showed increased telomerase activity in buds and seedlings of the S_048 upstream line, while activity in S_126 and F_490/492 lines was comparable with that of the wts (Fig. 4). These results point to the presence of a putative regulatory element in the upstream region whose function was changed in the S_048 line by T-DNA insertion.

Fig. 4.

Quantitative telomerase activity assay using the TS21 and TELPR primer set in G4 seedlings (A) and G4 buds (B) of the upstream lines. Two individual mutant lines of the F_490/492 accession, four individual mutant lines of S_126 and S_048, and wts (Col, Columbia; Ws4, Wassilevskija) were analysed. Representative raw data for cycling (A, left panel) grouped curves for individual mutant lines S_048 and curves for wts and other upstream lines as marked by arrows. A slight but reproducible increase of telomerase activity relative to Columbia wt was observed in S_048 tissues. (This figure is available in colour at JXB online.)

Reduced telomerase activity is thus tightly correlated with the shortening of telomeric repeats in mutant lines. In upstream lines with the telomere length comparable with that of the wts, high telomerase activity was observed in telomerase-positive tissues but, on the other hand, no telomerase activity was detected in gene lines in accordance with their progressively shortened telomeres.

AtTERT transcription is comparable with that of the wt in telomerase-positive upstream lines but decreases downstream of the T-DNA insertion in 3′ gene lines

The AtTERT gene was identified via its T-DNA insertion mutant (S_061, insertion in exon 9) in which the loss of telomerase activity and shortening of telomeres was observed (Fitzgerald ; Ruckova ). However, the presence/absence of AtTERT transcripts from regions upstream/downstream of the T-DNA insertion site has not been demonstrated yet. Previous studies of A. thaliana telomerase revealed the presence of the alternatively spliced isoform AtTERT V(I8) (Fig. 1D; Rossignol ) with suggested telomeric function as a putative AtPOT1a binding partner. The presence of TERT transcripts including the variant AtTERT V(I8) from regions located upstream and downstream of T-DNA insertions (Figs 5–7) was investigated using RT-PCR and qRT-PCR in telomerase-positive tissues (seedlings and buds) and telomerase-negative tissues (mature leaves; see below).

In telomerase-positive tissues, upstream lines generally showed levels of transcripts originating from exons 1 and 10 comparable with or slightly higher than those in the wts (Figs 5, 6; Table 1). The transcription of the PPR gene (At5g16860) located upstream of AtTERT (Fig. 1C, Supplementary Fig. S1A) was also examined, and no RT-PCR product in buds, seedlings, or leaves was observed in either upstream lines or in Columbia wt plants (not shown).

Fig. 5.

AtTERT transcription in G4 seedlings of T-DNA insertion plant lines of the Columbia background. (A) Analysis by conventional RT-PCR. (B) Analysis by qRT-PCR; at least two individual mutant lines of each accession were analysed. AtTERT transcription was slightly increased or was comparable with that in Columbia in upstream lines (S_048 and S_126). The expression levels in other gene lines (S_041, S_050, and S_284) correlated with the position of the T-DNA insertion, i.e. it was comparable with the wt in upstream regions and decreased in the region downstream of the T-DNA insertion. N.A., not analysed.

Fig. 6.

AtTERT transcription in G4 seedlings (A and B) and mature leaves (C) of T-DNA insertion lines of the Wassilevskija background (Ws4). (A) Analysis by conventional RT-PCR. Note the strong bands of PCR products in the F_385/22 individual line in the regions downstream of the T-DNA insertion (exons 2–5, 4–5, and 10) using a marginal amount of cDNA template for amplification (weak ubiquitin band). There is no signal in the F_439/11 individual line in the regions downstream of the T-DNA insertion. (B) Analysis by qRT-PCR. Transcription increased markedly in the regions downstream of the T-DNA insertion in gene line F_385 (left panel). In the upstream line F_490/492 (right panel), expression was comparable with that in the wt in both regions tested (exons 1 and 10). In gene line F_493 (right panel), an increase of expression was detected in the region upstream of the T-DNA insertion (exon 1), while in the downstream regions the expression is at the limit of detection. N.A., not analysed. (C) AtTERT transcription was analysed in mature leaves collected from five individual plants of the F_385 line (left panel) and three individual plants of the F_493 line (right panel). In both mutant lines, the AtTERT expression profile in leaves is similar to that in seedlings. Note the relatively high interindividual variability between samples.

Table 1.

Summary of analyses of AtTERT transcription in T-DNA insertion lines. Transcription in the seedlings (S), buds (B), and old mature leaves (L) is related to that in the respective wt tissue (except for F_385 and F_493 old leaves where transcription is related to Ws-4 seedlings). Data were taken from analyses of at least two individual mutant lines.

T-DNA insertion line	Tissue/generation	Exon 1	Exon 10	Exon 12
FLAG490F05/FLAG_492C08	SG4	1.04±0.12	0.98±0.22	NA
	BG4	1.13±0.26	1.08±0.17	NA
	LG4	ND	NA	NA
SALK_126201	SG4	1.78±0.73	1.92±0.85	NA
	BG4	2.25±0.21	2.48±0.83	NA
	LG4	ND	NA	NA
SALK_048471a	SG4	3.85±1.21	3.29±0.83	NA
	BG4	4.09±0.86	3.83±0.56	NA
	LG4	ND	NA	NA
FLAG_385G01	SG2	306±32	3417±434	NA
	SG3	331±34	3841±446	NA
	SG4	344±29	3290±560	NA
	BG4	424±53	3541±520	NA
	LG4	440±120	3079±615	NA
FLAG_493F06	SG3	22.5±2.3	0.03±0.01	NA
	SG4	37.8±4.1	0.12±0.02	NA
	BG4	51.1±12.5	0.20±0.05	NA
	LG4	39.3±17.3	ND	NA
SALK_041265b	SG2	1.19±0.32	NA	NA
	SG3	1.12±0.30	NA	0.33±0.11
	SG4	1.42±0.63	0.9±0.29	0.28±0.10
	BG4	1.49±0.22	1.23±0.36	0.21±0.08
	LG4	ND	NA	NA
SALK_050921	SG4	1.05±0.03	0.89±0.05	0.15±0.07
	LG4	ND	NA	NA
SAIL_284_B07	SG4	1.25±0.15	1.2±0.01	0.47±0.07
	LG4	N.D.	NA	NA

NA, not analysed; ND, not detected (expression below the detection limit).

Data shown for individual mutant lines S_048/3, 8, 33; transcription in G4 seedlings of S_048/23 was significantly different from the others (exon 1, 26.3±2.5; exon 10, 21.1±3.8).

Complete analysis of the individual mutant line S_041/3 is presented in the Supplementary data at JXB online.

All gene lines showed a level of the exon 1 transcript (upstream of the T-DNA insertion for all lines) similar to that in the wt (except for F_385 and F_493; see below), while transcription from exon 12 (downstream of T-DNA insertions for all lines) was significantly decreased (except for F_385; see below) (Figs 5, 6; Table 1). The level of transcription from exon 10 was markedly decreased in F_493 (Fig. 6) and similar to that in the wt in S_041/2, S_050, and S_284 (Fig. 5), thus corresponding to the relative position of the T-DNA insertion. In some cases, differences were observed in the transcript level among individual mutant line representatives; for example, line S_048/23 showed a more pronounced increase in expression of exons 1 and 10 (by more then one order of magnitude; Table 1) in comparison with other individual mutant lines of the same accession (S_048/2, 8, 33; Table 1), and a similar difference was observed between individual mutant lines of S_041 (see Supplementary methods for analysis of the individual mutant line S_041/3; Supplementary Table S2 at JXB online). Previous results revealed that transcription was not reduced upstream of T-DNA insertion sites and there is a possibility that the alternatively spliced isoform of AtTERT could be present and even functional in mutant lines with insertions positioned downstream of the AtTERT V(I8) poly(A) site (Fig. 1C, D). In addition to these 3' gene lines, the position of T-DNA insertions in upstream lines might influence the transcriptional pattern of this variant. The transcription of the alternatively spliced isoform AtTERT V(I8) was investigated using primers derived from unique features of this variant; a combination of primers designed from spliced borders of exons 5+6 (5-6F), exons 7+8 (7-8R), and forward or reverse primers from intron 6 (6iF, 6iR) was used in qRT-PCR (not shown) and conventional RT-PCR. The respective transcripts were detected using conventional RT-PCR in seedlings and buds in all upstream lines and also in gene lines with the T-DNA insertion located downstream of the putative poly(A) signal of the AtTERT V(I8) isoform (Fig. 7A, B). However, the amount of alternatively spliced transcript was too low (in the range of a few percent as compared with the full-length transcript in Col and WS4 wts) to perform fully reproducible quantitative assays by the qRT-PCR technique (M Fojtová, unpublished results).

Fig. 7.

Presence of the alternatively spliced AtTERT variant in seedlings and buds of the fourth generation (G4) of mutant lines of Wassilewskija (A) and Columbia (B) background. A faint band (∼500 bp of length) evidencing the presence of an alternatively spliced product was detected in all samples tested (primer set 5-6F and 7-8R). Ws4, Col, wt; –, negative control. (C) Presence of the RT-PCR product of regular length along the AtTERT gene in G4 seedlings of the F_385 gene line. No RT-PCR product was obtained using primers surrounding the T-DNA insertion site (1F+7-8R). Strong bands were detected when analysing regions downstream of the T-DNA insertion (2F+7-8R; 9F+12R). The primer 2F is located in a region containing a natural ATG codon inside of exon 2.

A specific pattern of AtTERT transcription in the 5' gene lines F_385 and F_493

AtTERT transcription in gene lines F_385 and F_493 with T-DNA insertions at the 5′ end of the AtTERT gene was completely different from that in other gene lines. They showed an ∼300- and 30-fold increased transcription, respectively, of exon 1 in G4 seedlings relative to WS4 wt (Fig. 6B; Table 1) and a similar increase in transcription was revealed in G3 seedlings and G4 buds from both lines (Table 1).

Transcription of exon 10 downstream of T-DNA insertions was convincingly reduced in F_493, while in F_385 it was ∼3500-fold higher in the G4 seedlings and buds in comparison with the corresponding wt tissues (Fig. 6B; Table 1). RT-PCR and qRT-PCR analyses with primer sets amplifying regions from exons 4 to 5 and exons 2 to 5 revealed that vigorous transcription had started in exon 2 closely downstream of the T-DNA insertion in the line F_385 (Fig. 6A, B). Moreover, analysis of transcripts downstream of the T-DNA insertion site (primer sets 2F–7,8R and 9F–12R) revealed bands of regular length in G4 seedlings of the F_385 gene line (Fig. 7C), suggesting that splicing of the overexpressed transcripts in both regions is identical to that in wt plants with the functional AtTERT mRNA variant. To exclude the possibility that transcription was driven by a strong T-DNA promoter (Supplementary Fig. S1B, C at JXB online), lines F_385 and F_493 were genotyped in more detail. RT-PCRs with T-DNA-specific primers and AtTERT-specific primers confirmed that no chimeric transcript was produced, and thus the increased transcription of exons 1 and 10 in line F_385 was not caused by a leaking terminator in the T-DNA in any of these lines (Supplementary Fig. S1C).

In addition, the levels of exon 1 and 10 transcripts were investigated in mature leaves which represent typical telomerase-negative tissue. Surprisingly, the transcription level and expression pattern were similar to those in seedlings and buds (Fig. 6C; Table 1) in both F_385 and F_493 lines. In other insertion lines, AtTERT transcription in mature leaves was under the detection limit (Table 1), i.e. silenced, as expected for plant mature tissues.

Discussion

Two major features of AtTERT function were monitored by analysing T-DNA insertion lines of A. thaliana for telomerase activity and telomere length. In accordance with previous reports on telomere dynamics in the T-DNA insertion mutant S_061 (Fitzgerald ; Ruckova ) telomere shortening by 100–500 bp per generation and an absence of telomerase activity was found in lines with a disrupted AtTERT gene sequence downstream of the start codon (Fig. 8).

Fig. 8.

Overview of results and analyses performed using T-DNA insertion lines. The position and location of T-DNA lines with respect to the ATG start codon and the AtTERT gene region are shown in the 5'–3' direction (not to scale). The lines excluded from the study are marked by filled triangles and by a cross (mis-maping of position); the position of the alternatively spliced isoform AtTERT V(I8) is marked above the line representing the genomic AtTERT region. Regulatory regions and affected processes are depicted below the line. The minimal promoter region (arrow) is defined by a T-DNA insertion site in the line S_126; the T-DNA insertion line S_048 revealed an increase in transcription and telomerase activity (line with two arrows) in comparison with lines S_126 and F_490/492. Disruption of the regulatory region downstream of the start codon (oval) in lines F_385 and F_493 influenced tissue specificity and the level of AtTERT expression upstream of T-DNA insertion in both lines (cooperation with an unknown part of the promoter region, dashed line with arrows) and activated overexpression downstream of the insertion site in F_385 (arrow). The summary of analyses incorporates measurement of telomere shortening and telomerase activity, transcription analyses including the AtTERT V(I8) variant, the root phenotype study, and genotyping of T-DNA insertions. The control line S_061 was characterized in detail in Ruckova (asterisk). The T-DNA arrangement is marked as IRRB (inverted repeat along the right border) and LBRB (5'–3' orientation) of T-DNA borders. Note that the arrangement was studied using PCR with the T-DNA primers and specific genotyping primers and it does not exclude the possibility of a tandem repeat in LBRB arrangement. n.a. = not analysed.

Unchanged telomerase function in upstream lines

Lines with the T-DNA insertion upstream of the start codon showed positive telomerase activity and telomere length comparable with those in wt samples. These results and the position of the T-DNA insertion in the S_126 mutant line suggest that a region as short as 271 bp upstream of ATG can act as a putative ‘minimal promoter’ able to drive sufficient transcription of the telomerase protein subunit gene, resulting in normal telomerase function.

Nevertheless, an increase of AtTERT transcription was observed in both seedlings and buds of the S_048 mutant line, pointing to a possible involvement of a putative regulation element in the region upstream of the minimal promoter. On the other hand, the amount of transcript in mature leaves in upstream lines was reduced convincingly, in contrast to gene lines F_385 and F_493. This suggests that the respective regulatory element may function only at the level of transcription, while it is not involved in tissue-specific regulation.

Negative telomerase regardless of the position of T-DNA insertion in gene lines

Experiments to determine the minimal length of the telomerase protein subunit which can reconstitute telomerase activity indicated that different regions of hTERT are sufficient to function in vitro and in vivo (Beattie ). Deletion of the 20 C-terminal amino acids reduced the activity of human telomerase reconstituted in rabbit reticulocyte lysate (RRL), and truncations of hTERT that lacked the last 205 amino acids (including part of the E motif; see Fig. 1A) remained active when transfected into 293T cells. N-terminal truncations of 300 amino acids were active in both systems, and a truncation spanning amino acids 201–927 was active in 293T cells but not in an RRL. The position of T-DNA insertion in lines S_050 and S_284 (inside exon 11 and 12, respectively) and the unaffected transcription level upstream of the T-DNA insertion might hypothetically give rise to AtTERT products containing a ‘minimal length subunit’ analogous to that of hTERT which possess all essential enzyme motifs (Figs 1A, C, 8) However, both lines showed telomere shortening and absence of telomerase activity. Similarly, line F_385 with the putative AtTERT gene product shortened from the N-terminal end did not show any telomerase activity. There is a possibility that mRNA originating from the gene disrupted by T-DNA insertion is recognized as aberrant and is not effectively translated or produces a non-functional protein, and this could explain the possible absence of translated products of the alternatively spliced isoform AtTERT V(I8) in the gene lines S_061, S_041, S_050, and S_284. Also, it is possible that the AtTERT V(I8) isoform is not directly involved in telomere maintenance. Unfortunately, these possibilities are difficult to test in plant models because antibodies specific for the respective TERT functional domains are not available.

A putative regulatory element in the region of T-DNA insertion in F_493 and F_385 gene lines

The position of the T-DNA insertion in the AtTERT gene could efficiently uncouple the transcription unit from its putative regulatory elements. Delimitation of regulation elements away from their original positions could provide additional information about their putative role in regulation of transcription, tissue specificity or in collaboration with other regulatory elements (Saracco ; Son ), as demonstrated here for the lines F_385 and F_493. Interestingly, in these lines and in the S_048 upstream line, the AtTERT expression in distinct regions was significantly increased, suggesting a possible disruption of the suppressive function of the putative upstream regulatory elements (Fig. 8). The T-DNA insertions inside intron 1 (F_385) and exon 2 (F_493) revealed several interesting features of these regions. Transcription upstream of both insertion sites was increased, but to different extents: the lines F_385 and F_493 showed a 300- and a 30-fold increase, respectively. Moreover, transcription downstream of the insertion sites revealed completely different patterns: an ∼3500-fold increase of AtTERT transcription was observed in line F_385, but transcription in line F_493 was completely disrupted similarly to the other gene lines. The differences in AtTERT expression in these two lines are probably not related to the T-DNA insertions which both have the same orientation and vector type, and increased AtTERT transcription was not caused by a leaking terminator in the T-DNA (Supplementary Fig. S1B, C at JXB online), or to a wt background, because the upstream line F_490/492 showed an expression pattern fully comparable with that of the Wassilevskija wt. These results provide evidence for the presence of a putative regulatory network or at least regulation motifs around the F_385 insertion site. The respective T-DNA insertions are located inside intron 1 (F_385) and exon 2 (F_493) at a distance of 190 bp. The marked differences in AtTERT expression both between lines F_385 and F_493 and in comparison with the Wassilevskija wt suggest disruption of some regulatory element in both lines (high expression in the region upstream of T-DNA insertion). This upstream suppressor element could be located (at least partially) between the respective insertion sites (Fig. 8). Nevertheless, the disruption of the putative suppressor function by T-DNA insertion in the second exon in the F_493 mutant line is manifested only by increased transcription in the region upstream of the T-DNA insertion, while T-DNA insertion inside the first intron in the F_385 mutant line resulted in extremely increased transcription of the regions downstream of the insertion site. The fate of this AtTERT mRNA is unclear; although the extremely up-regulated transcription downstream of T-DNA insertion seems to result in mRNA of regular length (Fig. 7C), there is no active telomerase in this line and telomeres are progressively shortened. This line thus clearly shows disruption of a tight coupling between AtTERT transcription and telomerase activity which occurs under natural conditions.

Both gene lines display disruption of the tissue-specific regulation of AtTERT transcription (Fig. 8), because a significant increase in the amount of transcripts was observed in telomerase-positive tissues (seedlings and buds) and also in mature leaves (Fig. 6C) which have previously been reported as typical telomerase-negative tissues (Fitzgerald ). This specific expression pattern can be attributed to the persisting activity of a strong regulatory element. The 35S enhancer/promoter in the transgenic cassette could be a promising candidate, but obviously no chimeric transcript is produced (Supplementary Fig. S1C at JXB online), the expression pattern is different among FLAG lines, and the transcription level was unaffected in line F_490/492. The present state of knowledge does not allow determination of whether the putative element involved in the disruption of tissue-specific regulation of telomerase expression in F_493 and F_385 lines is a natural regulatory factor, or a DNA sequence acting as a strong regulatory element in the specific sequence context after the T-DNA insertion was randomly activated (Fig. 8). There are no data available on a possible translation of transcripts in the lines F_385 and F_493; however, the dynamics of telomere shortening and lack of telomerase activity in the F_385 gene line support the hypotheses that the aberrant nature of mRNA prevents its effective translation or results in production of a non-functional protein. It should also be mentioned that this property of line F_385 might be an example of switching on of a cryptic promoter, and/or of the more general phenomenon termed intron-mediated enhancement (IME; Mascarenhas ; E Sýkorová et al., unpublished results). The specific features of line F_385, the position of the T-DNA insertion inside intron 1, and the extremely increased transcription level downstream of T-DNA match features of some IME constructs (Jeon ; Chaubet-Gigot ) which were used for enhancing transgene expression in crops (Wang ; Upadhyaya ). Although the mechanism of IME is largely unknown, 5'-proximal introns can increase the expression of transgenes in plants, although at a lower level as observed in line F_385 (for a review, see, for example, Rose ). Several reports showed expression downstream of T-DNA in Arabidopsis mutant lines which harbour an insertion at the 5' end of the gene (Bertrand ; Xu ; for a review, see Wang, 2008) but a detailed description of the transcription level and profile is missing in most cases. In conjunction with the present results, these observations suggest a new potential in publicly available T-DNA lines, for example searching for cryptic promoters, regulation sites, or IME elements. Nevertheless, the data presented here reveal additional levels of complexity of the regulation of telomerase activity which form a basis for further research.

Supplementary data

Supplementary data are available at JXB online.

Supplementary methods and results.

Table S1. Primers used in analysis of telomerase activity (TRAP), genotyping of mutant lines, and analysis of AtTERT transcription.

Table S2. Analysis of AtTERT transcription in the individual mutant line S_041/3.

Figure S1. Description of T-DNA lines.

Figure S2. Root growth analysis in S_050/4 (segregated wt) and S_050/13 (gene line).

Figure S3. Root lengths of 13-day-old plants of individual mutant lines.

Figure S4. Sequence alignment of cloned products of the TRAP assay using the CAMV substrate primer in telomerase-negative gene lines.

Figure S5. RNA origin of TRAP products obtained using the CAMV substrate primer in telomerase-negative gene lines.

Figure S6. Detailed analysis of telomerase activity in the individual mutant line S_041/3.