Ó Springer-Verlag 1998
Mol Gen Genet (1998) 260: 470±474
ORIGINAL PAPER
J. Fajkus á J. FulnecÏkova á M. HulaÂnova á K. Berkova K. RÏõ ha á R. MatyaÂsÏ ek
Plant cells express telomerase activity upon transfer to callus culture, without extensively changing telomere lengths
Received: 31 July 1998 / Accepted: 21 September 1998
Abstract Changes in telomere lengths and telomerase activity in tobacco cells were studied during dedierentiation and dierentiation; leaf tissues were used to initiate callus cultures, which were then induced to regenerate plants. While no signi®cant changes in the range of telomere lengths were observed in response to dedierentiation and dierentiation, there was a conspicuous increase in telomerase activity in calli compared to the source leaves, where the activity was hardly detectable. In leaves of regenerated plants, the telomerase activity fell to almost the same level as in the original plant, showing on the average 0.04% of the level in callus. The process was then repeated using the regenerants as the source material. In the second round of dedierentiation and dierentiation, telomerase activity showed a similar increase in calli derived from regenerated plants and a drop in plants regenerated from these calli. Telomere lengths remained unchanged both in calli and in leaves of regenerants. The conservation of telomere lengths over repeated rounds of dedierentiation and dierentiation, which are associated with dramatic changes in cell division rate and corresponding variation in telomerase activity may re¯ect the function of a regulatory mechanism in plant cells which controls telomerase action to compensate for replicative loss of telomeric DNA.
Communicated by H. Saedler J. Fajkus (&) á K. RÏõ ha á R. MatyaÂsÏ ek Institute of Biophysics Academy of Sciences of the Czech Republic KraÂlovopolska 135, 612 65 Brno, Czech Republic e-mail:
[email protected] Tel.: +4205-41517199, Fax: +4205-41211293 J. Fajkus á J. FulnecÏkova á M. HulaÂnova á K. Berkova Department of Analysis of Biologically Important Molecular Complexes, Masaryk University, KotlaÂrÏ ska 2, 61137 Brno Czech Republic
Key words Plant telomere dynamics á Telomerase activity á Telomere lengths á Telomerase regulation á Plant tissue culture
Introduction Tolemeres are specialised nucleoprotein structures forming the ends of eukaryotic chromosomes. Their DNA component is made up of a tandemly repeated oligonucleotide motif (e.g., [TTAGGG]n is conserved in mammals and [TTTAGGG]n in most plants; Fuchs et al. 1995). Telomeres are essential for stable chromosome maintenance, protecting chromosome ends from degradation and end-to-end fusions. Due to the inability of the primary cellular replication machinery to replicate the 30 -ends of the parental DNA strands completely, in the absence of an independent mechanism for telomeric DNA synthesis each round of DNA replication would be accompanied by chromosome shortening. This shortening can ultimately lead to loss of the telomere's ability to cap chromosomes, and results in activation of the DNA damage response pathway that causes cell cycle arrest. The most common mechanism for replenishing telomeres is the enzyme telomerase, a specialised ribonucleoprotein enzyme that synthesises telomeric DNA by reverse transcription of a template region of its own RNA molecule (Greider and Blackburn 1985; Morin 1997). Telomerase activity, however, is repressed in almost all dierentiated normal human somatic cells and, consequently, the replicative capacity of these cells is limited. On the other hand germline tissues, non-differentiated tissues (e.g. bone marrow progenitor cells) and tumour cells are found to be telomerase-positive, which makes these cells virtually immortal. The composition of the telomerase complex was ®rst studied in Tetrahymena (Greider and Blackburn 1985) and more recently in Euplotes (Lingner and Cech 1996; Lingner et al. 1997) and other organisms including man (Feng et al. 1995; Harrington et al. 1997a; Harrington et al. 1997b).
471
The observation that telomere lengths are stable in telomerase-positive animal cells has led to the isolation of telomere-binding regulatory proteins from several organisms e.g., TRF1 in human cells (Chong et al. 1995). TRF1 behaves as a negative regulator of telomere maintenance, probably by inhibiting the activity of telomerase at the ends of individual telomeres (van Steensel and de Lange 1997). Although telomeric research is now focused mainly on human cells in view of the promising prospects for cancer diagnostics and therapy, plant cells represent a very useful experimental system in this ®eld: unlike the case in animals, no germline is set aside during early plant development (both vegetative and reproductive organs dierentiate from dividing meristem cells during plant growth), and, consequently, genetic changes that accumulate in somatic cells (e.g., telomere shortening) may be transmitted to sexual progeny. In combination with the well known totipotency of plant cells, one would expect these features to be re¯ected in plant telomerase regulation. The totipotency of plant cells enables naturally immortal cell lines to be obtained from meristems of dierentiated plants, while whole plants can be regenerated from cells cultivated (and optionally manipulated) in vitro, usually by adjusting the plant hormone ratio. Recently, we reported the direct detection of telomerase activity in plant cells (Fajkus et al. 1996) and the telomere-repeat ampli®cation protocol (TRAP) was then independently applied to plant cell extracts by two other groups (Heller et al. 1996; Fitzgerald et al. 1996). In this paper we present our results on telomere dynamics in plant cells. We have measured the lengths of telomeres in relation to telomerase activity in tobacco cells in the course of repeated cycles from the dierentiated state (leaves) to undierentiated callus cultures and back to leaves in regenerated plants.
Preparation of plant cell extracts
Materials and methods
Preparation of internal standard/control template
Plant material
To identify false-negative results in samples containing Taq polymerase inhibitors and to enable quantitation of telomerase activity in positive samples, a control template (ROMAN2) was prepared from a 219-bp monomeric unit of the tobacco repetitive DNA sequence NTRS (Z50790) (MatyaÂsÏ ek et al. 1997) inserted between the TS(21) and TP primers. The ampli®cation of the ROMAN2 construct using TS(21) and TP primers yields a 274-bp PCR product when added to TRAP reactions or heat-treated controls. In each 50-ll TRAP reaction 0.1 amol of ROMAN2 was used.
Leaf blades of adult plants of Nicotiana tabacum L. cv. VielblaÈttriger were cut into pieces and placed on modi®ed agar MS medium containing 2 mg/l NAA (1-Naphthylacetic Acid) and 0.2 mg/l BAP(6-Benzylaminopurine) to induce callus formation. The resulting calli were further subcultured on the above medium in petri dishes and after the third 4-week passage, high-molecularweight DNA and cell extracts were prepared from the major portions of individual calli. The remaining portion of each callus was divided into two halves and one of these was subcultured further on the same medium, while the other was transferred to the MS agar medium with the inverse ratio of NAA to BAP (0.2 mg/l NAA and 2 mg/l BAP) to induce shoot formation. Shoots were excised and after a short period of hydroponic culture, the regenerants were grown to maturity in the greenhouse. The process of callus formation and plant regeneration was repeated once in the same manner. At each stage, several independently grown plants or calli were analysed in parallel.
Cell extracts from leaf tissue or callus cultures of N. tabacum were prepared according to Fitzgerald et al. (1996). Brie¯y, 1 g of material was ground in a mortar and pestle under liquid nitrogen, suspended in 4 ml of buer W [50 mM TRIS-acetate pH 7.5, 5 mM MgCl2, 100 mM potassium glutamate, 20 mM EGTA, 1 mM DTT, 0.1 mM PMSF, 0.6 mM vanadyl ribonucleoside complex (NEB), 1.5% (wt/vol) polyvinylpyrrolidone, 10% glycerol] and centrifuged at 16,000 g at 4° C for 15 min. The supernatant was supplemented with PEG 8000 (Sigma, ®nal concentration 10%), mixed for 30 min at 4° C and centrifuged at 20,000 ´g at 4° C for 5 min. The pellet was resuspended in 1 ml of buer W for 30 min at 4° C and was centrifuged at 20,000 ´g at 4° C for 2 min. The supernatant was frozen in liquid nitrogen and stored at )70° C until use. The protein concentration in extracts was determined according to Bradford (1976). It ranged from 0.3 to 3.0 mg/ml. Detection of telomerase activity A modi®ed version of the telomere repeat ampli®cation protocol (TRAP) was used (Fitzgerald et al. 1996). Reaction buer contained 50 mM TRIS-acetate (pH 8.3), 50 mM potassium glutamate, 0.1% Triton X-100, 1mM spermidine, 1 mM DTT, 50 lM of each dNTP, 5 mM MgCl2, 10 mM EGTA, 100 mg/ml BSA and 500 nM T4 gene 32 protein (USB). The telomerase substrate primer (CaMV35S) 50 -CGTCTTCAAAGCAAGTGGATT-30 , was heatdenatured for 3 min at 95° C and chilled on ice prior to addition to the reaction mix (10 pmol per reaction). Alternatively, the substrate primer TS(21)(50 -GACAATCCGTCGAGCAGAGTT-30 ; Fitzgerald et al. 1996) was used. The reaction mix (46 ll) was combined on ice with 2 ll of extract adjusted using buer W (see above) to the desired protein concentration and then incubated at 26° C for 45 min in thermocycler (Progene, fy Techne). Controls for falsepositive results were run in parallel using heat-treated extracts (90° C, 10 min). After the telomerase elongation step, the temperature was increased to 95° C for 5 min and PCR was hot-started at 80° C by adding 2 ll of mixture containing 2 U of Taq polymerase (Promega), and 10 pmol of the C-telomeric primer TP (50 -CCGAATTCAACCCTAAACCCTAAACCCTAAACCC-3 0 ; Kilian and Kleinhofs 1992) to each reaction. Ampli®cation was performed by 35 cycles of 94° C/30 s, 65° C/30 s and 72° C/90 s and the products were separated on a 12.5% polyacrylamide gel. After electrophoresis, the gel was stained with Vistra Green (Amersham) and photographed on a UV transilluminator or scanned on a STORM PhosphoFluorImager (Molecular Dynamics) and evaluated using ImageQuant software.
Determination of telomere lengths High-molecular-weight DNA samples from leaves were prepared either from protoplasts or cell nuclei according to EspinaÂs and Carballo (1993). To determine the lengths of terminal restriction fragments (re¯ecting the size of the telomeric TTTAGGG repeat), DNA samples embedded in agarose blocks (3±5 lg) were equilibrated with restriction buer and digested with 40 U of the
472 restriction endonuclease HaeIII or TaqI (NEB) in a fresh portion of buer for 16 h. Pulsed-®eld gel electrophoresis was performed on a CHEF-DRII apparatus (Bio-Rad) under the following conditions: 1% Fast Lane agarose gel (FMC), 0.5 ´ TBE buer, 190 V, 15° C; pulse time was ramped from 2 to 25 s over 20 h. After electrophoresis, the gel was stained with ethidium bromide and alkaliblotted onto a Hybond N+ membrane. Terminal restriction fragments were detected by hybridisation with an end-labelled (CCCTAAA)6 probe and visualised on X-ray ®lm or on a Phosphoimager STORM (Molecular Dynamics).
Results and discussion Telomerase activity in plant leaves and calli Plant TRAP assays of cell extracts which were accumulated in the course of experiments on each of the successive plant and callus stages, are illustrated in Fig. 1. It can be seen that there is no detectable telomerase activity in the leaves of the original plant. In calli derived both from the original and regenerated plants (callus I and II; lanes 6±9 and 16±19, respectively), telomerase activity is detectable over the whole range of protein concentrations from 5 ng to 5 lg per 50 ll reaction. In both callus I and II the activities are similar. Weakly positive results were obtained in leaves of regenerated plants using 5 lg of protein, as well as in leaves of plants from the second round of regeneration. Other plants and calli which were grown and tested in parallel gave corresponding results. A semiquantitative evaluation showed that the activity found in leaves of plants after the ®rst and the second round of regeneration represents 0.04 and 0.06%, respectively, of the average activity found in calli. This result agrees with similar observations in barley (Heller et al. 1996) and cauli¯ower (Fitzgerald et al. 1996), where only very weak or no telomerase activity was observed in leaves. The weak activity found in leaves of regenerated plants (lanes 14 and 24) could re¯ect transient disturbance in Fig. 1 Telomerase activity in plant leaves and callus cultures. Activity was assayed using a range of protein concentrations (5, 50, 500 and 5000 ng per assay) is shown for a leaf from the original plant (Orig. Plant), for callus derived from it (Callus I), for the plant regenerated from this callus (Reg. Pl. I), for callus obtained from Reg. Pl. I (Callus II) and for the plant regenerated from Callus II (Reg. Pl. II). Lanes 5, 10, 15, 20 and 25 (marked HI) show reactions with heat-inactivated cell extracts at a protein concentration of 5 lg. The CaMV35S oligonucleotide was used as substrate in all TRAP reactions
regulation of telomere maintenance, which may occur due to the change in dierentiation state and cell division rate. An inhibitory eect was frequently observed at the PCR stage of the TRAP essay when higher protein levels were used, as shown by a diminished intensity of the PCR products of the control template ROMAN2 (see Fig. 2). This inhibition may occasionally lead to underestimation of telomerase levels in the reaction. Both substrate primers-CaMV35S and TS(21)-used in this work gave corresponding results. The amount of protein sucient for the non-radioactive version of TRAP used in this work was found to be about 5 ng per reaction. ROMAN2 had no eect on TRAP results, whether added before or after the elongation step. Interestingly, the most expensive component of the TRAP assay, T4 gene 32 protein (USB), had no eect on telomerase activities (not shown). Telomere lengths Previous results with tobacco telomeres (Suzuki et al. 1994; Fajkus et al. 1995a; KovarõÂ k et al. 1996) have shown that in leaf cells these are very long, and although their lengths vary widely at individual chromosome ends (between 20 and 170 kb) the spectrum of terminal restriction fragments is composed of discrete bands. In the case of tobacco, the telomere-associated DNA has been partially characterised (Fajkus et al. 1995b) and, therefore, appropriate restriction endonucleases could be used to minimize the length of telomere-associated DNA in terminal restriction fragments to
