Journal of Experimental Botany, Vol. 52, No. 359, pp. 1203±1208, June 2001
Dormancy and proliferation in Saccharum officinarum 3 S. spontaneum hybrids which differ in the number of the introgressed S. spontaneum chromosomes R. Acevedo1,4, S. Moreno DõÂaz de la Espina1, M. E. FernaÂndez-GoÂmez1, A. Cuadrado2, N. Jouve2 and C. de la Torre1,3 1 2
Centro de Investigaciones BioloÂgicas, CSIC. VelaÂzquez, 144. E-28006-Madrid, Spain Departamento de BiologõÂa Celular y GeneÂtica, Universidad de Alcala de Henares, Madrid, Spain
Received 29 July 2000; Accepted 05 February 2001
Abstract Proliferating cells remain transiently blocked at different cycle compartments until specific stressors are removed or until the cells become adapted to their presence. This paper investigates the efficiency of cycle blocks in three sugarcane hybrids with the full noble cane (Saccharum officinarum) genome (2n 8x 80) but differing in the number of introgressed S. spontaneum (2n 8x 64) chromosomes. The My5514, B42231 and C236-51 cultivars possess 20, 30 and 40 additional S. spontaneum chromosomes, respectively. Flow cytometry showed that over 90% of cells were accumulated with a 2C DNA content in their dormant primordia. The presence of S. spontaneum chromosomes decreased the low stringency of the 4C block. The greater the number of these chromosomes, the lower was the number of quiescent cells with a 4C DNA content ( P-0.05). Shortly after stimulation of the primordia (85% relative humidity and 30 8C), i.e. in the 2 mm long roots, a negative correlation was found between the number of introgressed S. spontaneum chromosomes and the frequency of cells undergoing replication and mitosis. On the other hand, when roots were already proliferating under steady-state conditions (15 mm long roots) the more S. spontaneum chromosomes the cells possessed, the longer the relative time it took for all chromosomes to replicate and segregate, and the longer the relative time they spent in G2, with the 4C DNA content. The presence of S. spontaneum chromosomes seems to be recognized by these 3 4
proliferating cells as a stressor which preferentially activates checkpoint pathways operating at the second half of the cycle, but not at its onset. Key words: Saccharum officinarum 3 S. spontaneum hybrids, dormancy, proliferation, flow cytometry.
Introduction One hundred years ago, shortly after it was discovered that sugarcane could reproduce sexually, interspeci®c hybridization was begun to improve the resistance of the noble cane Saccharum of®cinarum to pathogens and other stressors. The low resistance to stress displayed by S. of®cinarum was improved through hybridization with the wild sugarcane S. spontaneum. Similar results were obtained with other interspeci®c crosses. Many of these hybrids were further subjected to `nobilization' by about four backcrosses in which S. of®cinarum was used as the female parent. In this way, acquired resistance traits were ®xed (PeÂrez et al., 1997). As there is a single nucleolar organizer per genome and this is distal in S. of®cinarum but interstitial in S. spontaneum (D'Hont et al., 1998), it has been de®ned that the My5514, B42231 and C236-51 cultivars have 20, 30 and 40 chromosomes of the S. spontaneum genome in addition to the whole S. of®cinarum genome (an autooctoploid with 2n 80) (Acevedo et al., unpublished results). The response of proliferating cells to any stressor is made possible by the operation of transient blocks
To whom correspondence should be addressed. Fax: q34 91 562 75 18. E-mail:
[email protected] Permanent address: Instituto de Investigaciones de la CanÄa de AzuÂcar, Ave. Vantroi No. 17 203, Rancho Boyeros, Ciudad de La Habana, Cuba.
ß Society for Experimental Biology 2001
1204
Acevedo et al.
which result from the activation of the corresponding checkpoint mechanisms, for example, the ®ring of both early and late replication origins, the G2 to mitosis transition, the spindle checkpoint, etc. Proliferating cells remain transiently blocked until the stressor is removed or until they become adapted to its presence. For these reasons, an investigation was made into the ef®ciency of the corresponding cycle blocks in three hybrids differing in the number of chromosomes provided by the S. spontaneum genome. Cell distribution was studied in the dormant root primordia of the hybrid stems, followed by an investigation into how proliferation was initiated after their stimulation. Finally, the relative durations of the different cycle phases were estimated in roots proliferating under steady-state kinetics as, under these conditions, the cellular frequency in a compartment is always proportional to the compartment relative duration.
ribonuclease A (Serva, Hiedelberg) were then added. Salmo trutta erythrocytes were used as a standard to evaluate the total DNA content of the four cultivars. Samples were analysed with an EPICS XL (Coulter, FL., USA) ¯ow cytometer. Though it was intended to study a sample of 10 000 cells per histogram, some samples were formed by only half this value. As a consequence, the mean number was of 8000 cells per histogram. The experiments were repeated at least three times. Feulgen technique Primordia and proliferative roots (1, 2, 4, 5, and 15 mm long) were ®xed in ethanol±acetic acid (3 : 1) for 24 h. Fixed samples were rehydrated in water for 15 min and incubated in 5 N HCl solution at 20 8C for 15 min and then washed in water and maintained in Schiff 's reagent for 45 min. They were then washed again in 5 N HCl containing 10% sodium metabisulphite. Finally, two water washes were made and samples digested with the same enzyme mixture used for ¯ow cytometry samples. Root tips were squashed onto clean microscope slides in a drop of 45% acetic acid. De®nitive preparations were made following the technique described earlier (Conger and Fairchild, 1953).
Materials and methods Plant material Four sugarcane (Saccharum spp.) cultivars from Cuba were used: Cristalina, My5514, B42231, and C236-51, supplied by INICA (Instituto Nacional de Investigaciones de la CanÄa de AzuÂcar). Cristalina is a typical S. of®cinarum cultivar and produces the highest sugar yield. However, its high susceptibility to sugar cane mosaic virus (SCMV) precludes its agronomical use. At present it is used as a gene reservoir in hybridization programmes. The My5514 cultivar is resistant to SCMV and Ustilago scitaminea Sydow fungus. B42231 and C236-51 cultivars are used in many phytopathological studies with U. scitaminea and SCMV because of their respective susceptibility to these pathogens. Culture Stem sugarcane cuttings containing three radical bands were sent to us by air mail from Cuba. They were cultured on wet ®lter paper and cotton in a Refritherm-5 (Struers) incubator at 30 8C and 85% relative humidity in the dark. Samples of quiescent root primordia were excised from the radical bands before soaking. Mitotic indices and the frequency of 4C cells were relatively constant in the root from different stems of the same cultivar. Flow cytometry
Primordia and proliferating roots (2, 5 and 15 mm long) were ®xed in 1% paraformaldehyde in TRIS buffer (10 mM TRIS, 10 mM Na2EDTA, 100 mM NaCl, pH 7.5). The samples were then washed in the same buffer and digested with an enzyme mixture: 2% cellulase (Serva, Hiedelberg), 1% pectinase, 0.05% macerozyme (Serva, Hiedelberg), and 0.4 M mannitol (Merck, Darmstadt). Nuclei were isolated from the quiescent roots by using an Ultra Turrax homogenizer and 500 ml lysis buffer (15 mM TRIS, 2 mM Na2EDTA, 80 mM KCl, 20 mM NaCl, 0.1% Triton X-100, pH 7.5). The homogenate was ®ltered through a 30 mm nylon mesh and centrifuged at 600 g for 20 min at 4 8C. Pellets were resuspended in 500 ml lysis buffer. Propidium iodide (0.1 mg ml 1) and DNase-free
Mitotic index determination
Slides were observed by optical microscopy and the mitotic index estimated as the percentage of mitotic cells in relation to total meristematic cells. Five roots per cultivar were used and 1500 cells per sample were counted. Determination of cell frequency in different interphase compartments
Using data obtained by ¯ow cytometry, the frequency of cells in the different stages of the cycle were compared by automatic integration of all cells accumulated in a cycle segment between two chosen DNA contents by using the software XL version System II analysis. For example, all replicating cells will possess DNA contents between 2C and 4C. Also, the frequency of cells found between 2.5 and 3.5 were taken as representing mid-S cells, with the advantage of preventing the overlapping of recording for G1 and early S cells and for late S and G2 cells. Statistical analysis The data were analysed by testing equality between percentages compared two by two and by ANOVA followed by the Newman±Keuls test (Sokal and Rohlf, 1969).
Results Dormant root primordia
Flow cytometry was performed on the dormant root primordia of each hybrid and also in the Cristalina cultivar, a representative of the noble cane which has the auto-octoploid S. of®cinarum genome only. As shown by ¯ow cytometry, the stronger the presence of the S. spontaneum genome in these hybrids, the fewer the cells accumulated with 4C DNA contents in the dormant primordia (P-0.05) (Fig. 1). These cells should be at a quiescent G2 as the whole set of histograms
Dormancy and proliferation in sugarcane hybrids
(see below) discard any possible switching to G1 of an endocycle. Cells that were not stagnated with 4C DNA contents were blocked with 2C instead. These data also show that the presence of S. spontaneum chromosomes selectively weakens the stringency of the G2 checkpoint block leading to dormancy, while the higher stringency of the G1 checkpoint mechanism allows it to remain effective when suboptimal environmental conditions are present. Comparison of the images of cell populations belonging to dormant (Fig. 2A) and activated primordia (Fig. 2B) reveals the lack of mitotic cells in the former to be an obvious distinguishing feature. Onset of proliferation in root primordia
In order to determine the kinetics of entry into mitosis of cells retained before division in dormant primordia from all three hybrids, the evolution of the mitotic index was followed in roots which had reached different lengths following the stimulation of their growth (Fig. 3A). The frequency of cells having entered mitosis in the short term re¯ected the kinetics at which these cells made the transition from G2 to mitosis. The greater the number of S. spontaneum chromosomes present in the hybrid, the later the steady maximum level of cell frequencies was reached. Thus, the My5514 population achieved its ®nal mitotic index even in 2 mm long roots, while steady indices were only achieved by the B42231 and the C236± 51 hybrids when their roots had reached 5 mm and 15 mm, respectively. The statistical signi®cance of the differences at each root length among the hybrids was estimated (P-0.05) (Fig. 3B). In order to follow the progress of cells through interphase, ¯ow cytometry was performed with the different sized roots. As seen in Fig. 4, cell distribution is compatible with cycle progression even in the 2 mm long roots. Integration of the number of cells at selected DNA contents (abscissae in Fig. 4) was performed in order
Fig. 1. Histogram showing the frequency of cells with 4C DNA content present in the dormant primordia of the Cristalina cultivar (S. of®cinarum L.) and in the three hybrid cultivars possessing increasing numbers of the introgressed S. spontaneum chromosomes. Lettering on top of the bars represents statistical differences at the P-0.05 level.
1205
to achieve a clearer picture of changes in proliferation dynamics. In 2 mm long roots, before steady-state growth had been achieved for a replicative compartment located between 2.5 and 3.5 C DNA (Fig. 5A), a negative correlation was found between the number of cells having begun replication and the number of S. spontaneumchromosomes possessed (®rst histogram to the left of Fig. 5A). The interval between these values represents the mid half-portion of the S period. This window avoids values close to 2C±4C where the readings for cells in G1and G2 may overlap. Finally, the integrated frequency of cells with DNA contents over 3.5C was also estimated for the three hybrid cultivars (histogram to the left of Fig. 5B). The frequency of cells in late interphase was higher in dormant My5514 than in the other cultivars (see Fig. 1 for frequencies with 4C DNA contents). However, cell frequencies above the 3.5C value in B42231 and C236-51 were not signi®cantly different (P-0.05). Proliferation under steady-state kinetics
When a constant distribution of cells in the different cycle compartments was achieved, the rates of entry into, progression through and exit from these intervals were in
Fig. 2. Micrographs after Feulgen staining showing portions of the squashed meristems present in the dormant primordia (A) and in the proliferating 5 mm long roots of the My5514 cultivar (B). Notice the dense chromatin in the nuclei of the dormant primordia and the presence of mitotic ®gures in those proliferating. Bar represents 10 mm.
1206
Acevedo et al.
Fig. 4. Number of nuclei distributed in relation to their DNA content after propidium iodide staining, as recorded by ¯ow cytometry using roots of different lengths from each hybrid. Nuclei to the left and right of the peaks corresponding to 2C and 4C DNA contents were integrated to estimate the 4C : 2C DNA ratio in the 15 mm long roots (bottom row). Each histogram contains over 5000 nuclei, typically about 8000. The ordinate scales were from 0 to 107.
Fig. 3. Mitotic indices for different sized roots. In (A), the statistical signi®cance for each value compared to others in the curve (P-0.05) is represented by differential lettering. Notice how the 2 mm long My5514 roots had already achieved a constant MI, while this only occurred in 5 and 15 mm long roots in B42231 and C236-51, respectively. In (B) the same statistical level of signi®cance is seen when looking at the three different cultivars in roots of the samelength.Values for B42231 and C236-51 were not signi®cantly different in either the 2 or 5 mm long roots, but were so at all other root lengths (P-0.05).
equilibrium. Under these conditions only, the higher the frequency of cells in a compartment, the longer the relative time taken to complete it. As for mitosis, steadystate kinetics were achieved in the C236-51 hybrid only in 15 mm long roots. This was therefore taken to represent steady-state conditions for all hybrids. By looking at the different histograms for the 15 mm long roots, it can be seen that the frequencies of cells in the S period (Fig. 5A), in the G2 period (Fig. 5B) and in mitosis (Fig. 3B) become higher as the number of S. spontaneum chromosomes increases. Thus, the greater
Fig. 5. (A) Integration of cell frequencies in a replicative compartment (from 2.5C to 3.5C DNA) to show how cells enter (2 mm long roots) and progress through the S period (5 and 15 mm long roots). Cells in the 2 mm long roots went earlier into the S compartment, the fewer the S. spontaneum chromosomes their nuclei contained. However, the higher the number of these introgressed chromosomes their nuclei possessed the higher the frequency and, then, the longer relative duration taken to complete this replication subperiod. (B) Integration of cell frequencies over 3.5C DNA content in the same meristems. The same trend shown in mid-S was maintened by the G2 cells.
Dormancy and proliferation in sugarcane hybrids
the number of these heterologous chromosomes present in the hybrid, the longer the relative duration of its S, G2 and mitosis. In order to estimate the ratio between cells with 2C and 4C DNA contents in meristems proliferating under steady-state kinetics, the integrated frequencies of cells to the left of the mean 2C line, and those to the right of the mean 4C line, which contain the purest concentrations of true 2C and 4C DNA nuclei, respectively, were evaluated in 15 mm long roots (see hatched regions in the lower panels of Fig. 4). In the Cristalina cultivar the ratio was 0.19 when no chromosomes from S. spontaneum were present (pure S. of®cinarum genome, data not shown). However, the 4Cu2C DNA ratio increased up to 0.26 in My5514 and 0.47 and 0.45 in the B42231 and C236-51 cultivars, respectively. Hence, the addition of S. spontaneum chromosomes diminished the relative duration of G1 in the hybrid cultivars, though not in a linear fashion.
Discussion The differences in the number of S. spontaneum chromosomes introgressed into the full genome of S. of®cinarum belonging to the three sugarcane hybrids studied here, provide a simple means of looking at the role of interspeci®c chromosomal introgression in dormancy and the onset of proliferation in roots, and for investigating cycle kinetics after steady-state levels have been reached. The presence of S. spontaneum chromosomes lowered the already low stringency of the G2 block to environmental stresses. Thus, the greater the number of S. spontaneum chromosomes, the higher the number of cells accumulating at G0 (with 2C DNA contents) in dormant primordia. Checkpoint pathways in proliferating plant cells (Del Campo et al., 1997; Pelayo et al., 2001), like their homologues in mammalian cells (Hartwell and Kastan, 1994; Lukas et al., 1996; SaÂnchez et al., 1997), are complex enzymatic processes. When a stress is sensed they magnify its signal, transduce it to relevant cellular targets and prevent the activation of the corresponding cyclindependent kinases (Walworth et al., 1993; Weinert, 1997). The cycle block produced as a result of the checkpoint mechanisms will be operative until the stress is over or until the cells become adapted to it. In this way, checkpoint mechanisms halt the main irreversible cycle transitions, among them G1 to S, the ®ring of late replicating origins in the last part of the S period, the G2 to mitosis transition, and the mitotic transitions affecting nuclear envelope breakdown and centromere division. For instance, the proliferating cells of Allium cepa L. root meristems have, in G2, two controls for monitoring DNA integrity (Pelayo et al., 2001), and at least one other for
1207
checking chromosome condensation (GimeÂnez-AbiaÂn et al., 2000). Interspeci®c hybridization is a well-known source of stress at the level of the somatic cell cycle (McClintock, 1983), and this seems to be the case for the three hybrid sugarcane cultivars analysed here. Though lengthening of the cycle duration occurs with increasing DNA content, both in monocotyledons and dicotyledons (Evans et al., 1972; Ivanov, 1978), the hybrid sugarcane cultivars present a peculiarity: the stress produced by the excess chromosomes selectively lengthens the relative time taken by their cells to complete their replication and segregation during mitosis. Further, it took longer for them to complete the G2 period, while the relative duration of G1 decreased. Trigonelline is a metabolite in the pyridine nucleotide pathway which promotes retention at G2 rather than G1 (Evans and Van't Hof, 1974 Evans et al., 1984). Oxidative stress produced in plant cells by DNA-strand breakage apparently leads to increased trigonelline (Berglund, 1994; Berglund et al., 1996). This may also be the signal operating in these hybrids since they show chromosomal instability and recombinant chromosomes (D'Hont et al., 1996). The lengthening of the S period, G2 and mitosis is probably related to the delay produced when dealing with the subsequent intra-S, G2 and mitotic spindle checkpoints. The relative shortening of G1 in relation to G2 in the presence of additional S. spontaneum chromosomes makes the cycle of these proliferating cells more analogous to that of Schizosaccharomyces pombe than to Saccharomyces cerevisiae. This analogy is also supported by features of their cytokinesis, such as the cellular transversal division they share, which is different to that of S. cerevisiae and mammalian cells. Interspeci®c hybridization in somatic cells leads to a gradual loss of the chromosomes of one of the partnersÐthe basis of somatic cell genetics developed in the 70s (Harris, 1974). In the case of these sugarcane hybrids, which are mostly reproduced vegetatively, the loss of a fraction of the 64 chromosomes of the autooctoploid S. spontaneum genome is evident, while the S. of®cinarum genome remains intact. This situation is probably due to the fact that the S. of®cinarum genome was used as the maternal partner in the hybridization crosses. However, at variance with the progressive chromosomal loss observed in mammalian interspeci®c crosses (Harris, 1974), the selective loss of S. spontaneum chromosomes is not a direct function of the similar time elapsed since their ®rst interspeci®c crosses (c. 1893), or of their time of stabilization: in 1955 for My5514, 1942 for B42231 and 1951 for C236-51 cultivars (Stevenson, 1965; PeÂrez et al., 1997) Homeostatic resources of eukaryotic cells are unveiled when studying their behaviour during proliferation, as
1208
Acevedo et al.
shown in sugarcane cultivars believed to be adapted to interspeci®c hybridization stress. Acknowledgements This work was ®nancially supported by the Spanish DGES (Projects PB96-0909 and PB98-0647) and by the CSICuCITMA Agreement (Project 99CU0010). Ricardo Acevedo is a Mutis fellow ®nanced by the Spanish Agency of International Cooperation (AECI). We wish to thank the Instituto de Investigaciones de la CanÄa de AzuÂcar (INICA) for providing the sugarcane used in this work. Our thanks to Adrian Burton for linguistic assistance.
References Berglund T. 1994. Nicotinamide, a missing link in the early stress response in eukaryotic cells: a hypothesis with special reference to oxidative stress in plants. FEBS Letters 351, 145±149. Berglund T, Kalbin G, Strid A, Rydstrom J, Ohlsson AB. 1996. UV-B- and oxidative stress-induced increase in nicotinamide and trigonelline and inhibition of defensive metabolism induction by poly(ADP-ribose)polymerase inhibitor in plant tissue. FEBS Letters 380, 188±193. Conger AD, Fairchild LM. 1953. A quick-freeze method for making smear slides permanent. Stain Technology 28, 281±283. Del Campo A, GimeÂnez-MartõÂn G, LoÂpez-SaÂez JF, De la Torre C. 1997. Frailty of two cycle checkpoints which prevent entry into mitosis and progression through early mitotic stages in higher plant cells. European Journal of Cell Biology 74, 289±293. D'Hont A, Grivet L, Feldman P, Rao S, Berding N, Glaszmann JC. 1996. Characterization of the double genome structure of modern sugarcane cultivars (Saccharum spp.) by molecular cytogenetics. Molecular and General Genetics 250, 405±413. D'Hont A, Ison D, Alix K, Roux C, Glaszmann JC. 1998. Determination of basic chromosome numbers in the genus Saccharum by physical mapping of ribosomal RNA genes. Genome 41, 221±225. Evans GM, Rees H, Snell CL, Sun S. 1972. The relationship between nuclear DNA amount and the duration of the mitotic cycle. Chromosomes Today 3, 24±31. Evans LS, Tramontano WA, O'Connor RF, Gill R, Ciancaglini E. 1984. Metabolism of nicotinic acid and nicotinamide in
cultured and seedling roots of Pisum sativum in relation to promotion of cell arrest in G2 by trigonelline. Environmental and Experimental Botany 24, 283±292. Evans LS, Van't Hof J. 1974. Promotion of cell arrest in G2 in root and shoot meristems in Pisum by a factor from the cotyledons. Experimental Cell Research 87, 259±264. GimeÂnez-AbiaÂn JF, Clarke DJ, Devlin J, GimeÂnez-AbiaÂn MI, De la Torre C, Johnson RT, Mullinger AM, Downes CS. 2000. Premitotic chromosome individualization monitored by a G2 topoisomerase II-dependent checkpoint. Chromosoma 109, 235±244. Harris H. 1974. Nucleus and cytoplasm. London, New York: Oxford University Press (Clarendon). Hartwell LH, Weinert TA. 1989. Checkpoints: controls that ensure the order of cell cycle events. Science 246, 629±634. Ivanov VB. 1978. DNA content in the nucleus and rate of development in plants. Soviet Journal of Developmental Biology 9, 28±40. Lukas J, Bartkova J, Bartek J. 1996. Convergence of mitogenic signalling cascades from diverse classes of receptor at the cyclin-D-dependent kinase-pRb-controlled G1 checkpoint. Molecular and Cell Biology 16, 6917±6925. McClintock B. 1983. The signi®cance of responses of the genome to challenge. In: Fedoroff N, Botstein D, eds. The dynamic genome. New York: Cold Spring Harbor Laboratory Press. 174±193. Pelayo HR, Lastres P, De la Torre C. 2001. Replication and G2 checkpoints. Their response to caffeine. Planta 212, 444±453. PeÂrez G, De Prada F, Abrantes EI. 1997. Origen, posicioÂn taxonoÂmica y desarrollo del mejoramiento de la canÄa de azuÂcar en Cuba. In: PeÂrez G, Bernal N, Chinea A, O'Really F, De Prada F, eds. Recursos geneÂticos de la canÄa de azuÂcar, La Habana: IMAGO, 10±12. SaÂnchez Y, Wong C, Thoma RS, Richman R, Wu A, Piwnica-Worms H, Elledge SJ. 1997. Conservation of the Chk1 checkpoint pathway in mammals: linkage of DNA damage to Cdk regulation through Cdc25. Science 277, 1497±1501. Sokal RR, Rohlf FJ. 1969. Biometry. San Francisco: WH Freeman and Co. Stevenson GC. 1965. Genetics and breeding of sugarcane. London: Longman. Walworth N, Davey S, Beach D. 1993. Fission yeast chk1q protein kinase links the rad checkpoint pathway to cdc2q. Nature 363, 368±371. Weinert T. 1997. A DNA damage checkpoint meets the cell cycle engine. Science 277, 1450±1505.
