Two-dimensional spreads of synaptonemal complexes from solanaceous plants

Two-dimensional spreads of synaptonemal complexes from solanaceous plants. V. Tomato (Lycopersicon esculentum) karyotype and idiogram JAMIED. SHERMAN AND STEPHEN M. STACK Department of Biology, and Cell and Molecular Biology Program, Colorado State University, Fort Collins, CO 80523, U.S.A.

Genome Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/03/13 For personal use only.

Corresponding Editor: P. B. Moens Received October 22, 1991 Accepted October 25, 1991 SHERMAN, J. D., and STACK,S. M. 1992. Two-dimensional s'preads of synaptonemal complexes from solanaceous plants. V. Tomato (Lycopersicon esculentum) karyotype and idiogram. Genome, 35: 354-359. We used a hypotonic bursting technique to spread synaptonemal complexes (SCs) from tomato primary microsporocytes. Based on these spreads, we have prepared a karyotype and an idiogram for tomato SCs. Each SC was related to its corresponding chromosome by observing SCs from tomatoes that were trisomic for each chromosome. We have compared this SC karyotype with previously published pachytene chromosome karyotypes for tomato. Key words: synaptonemal complex, tomato, karyotype, idiogram.

J. D., et STACK,S. M. 1992. Two-dimensional spreads of synaptonemal complexes from solanaceous plants. SHERMAN, V. Tomato (Lycopersicon esculentum) karyotype and idiogram. Genome, 35 : 354-359. Des etalements de complexes synaptonemaux (SCs) d'androsporocytes primaires de la tomate ont ete obtenus au moyen d'une technique d'eclatement par hypotonie. Un caryotype de la tomate et un idiogramme ont ete produits a partir de ces etalements. Chaque SC a ete relie a son chromosome correspondant, en examinant les SCs de plantes qui etaient trisomiques pour chaque chromosome. Ce caryotype a ete compare avec ceux deja publies pour la tomate et etablis a l'aide de chromosomes et stade pachytene. Mots clks : complexe synaptonemal, tomate, caryotype, idiogramme. [Traduit par la redaction]

Introduction Tomato (Lycopersicon esculentum, 2n = 24) is an important experimental plant that has been used in many genetic and cytogenetic studies (Rick 1991). While tomato's mitotic chromosomes are small and difficult to differentiate, each of the 12 pachytene bivalents has been distinguished on the basis of its relative length, arm ratio, and pattern of pericentric heterochromatin (Barton 1950; Ramanna and Prakken 1967; Khush and Rick 1968). Here we report that each of tomato's 12 synaptonemal complexes (SCs) likewise can be distinguished on the basis of these criteria. Also, we have related each SC to its corresponding chromosome by examining spreads of SCs from plants trisomic for each of the 12 chromosomes. A reliable karyotype for tomato SCs is a necessary prerequisite for preparing a physical map of recombination nodules on SCs and for mapping genes along SCs by in situ hybridization.

Materials and methods Diploid tomato seeds (var. Cherry) were planted monthly to maintain a supply of healthy, young plants. Flowers were used from plants that were between 2 and 4 months old. Trisomic seeds (var. San Marzano) were obtained from Charles Rick at the University of California in Davis. Trisomic seeds were treated with halfstrength household bleach for 15 min at room temperature and then washed with at least six changes of distilled water to improve germination (Rick and Hunt 1961). The trisomic seeds were planted in small pots of sterile soil and kept under a grow light until they were established. Seedlings were first selected by morphological characteristics for trisomy (Rick 1987). When trisomy was verified by examining root-tip squashes, the seedlings were grown to flowering. Subsequently these plants were propagated asexually from cutPrinled in Canada / imprime au Canada

tings. Diploid and trisomic plants were grown in 5-gallon pots in a greenhouse with the temperature controlled between 20 and 25°C. The SC spreads were produced essentially as described by Stack (1982) with the following modifications: 0.7 M mannitol rather than 0.8 M sorbitol was used in the digestion medium. Twenty to 40 anthers with pollen mother cells in pachynema were placed in a depression slide with 200 pL of digestion medium. The anthers were bisected transversely and the pollen mother cells were gently squeezed out. Fifteen microlitres of enzyme solution was then added. The enzyme solution was 50 mg desalted sulfatase (Sigma Chemical Co., St. Louis, Mo.), 0-glucuronidase (Sigma), or cytohelicase (IBF biotechnics) in 1 mL of 50% glycerol. The depression slide was placed on a "V"-shaped glass rod in a Petri dish containing 1 mL of water. The closed Petri dish was placed on top of an inverted glass culture dish filled with ice. Cooling the cells during digestion improved their viability. After 45 min, walls of cells were digested, and the rods of protoplasts were friable. At this time the protoplasts were drawn into a micropipette. A drop of protoplast suspensian was placed into 10 pL of spreading solution on a glow discharged, 0.6 or 0.9% Falcon plastic-coated slide. The spreading solution consisted of one part digestion medium and three parts water. A siliconized cover glass was added and 2.5 pL of 0.05% Nonidet P40 was drawn under the cover glass to burst the cells. Cover glasses were removed by the dry ice method and the slides were air dried before staining with silver or with uranyl acetate - lead citrate (Stack and Anderson 1986a, 1986b). Sets of SCs were examined and photographed by electron microscopy. Absolute magnifications of electron micrographs were determined using a carbon replica grating. Negatives were printed and SCs were traced on acetate sheets with the location of kinetochores and heterochromatin indicated. These tracings were measured on a Hewlett Packard 91 11A Graphics Tablet, using a Hewlett Packard 86B personal computer and a computer program that was specifically written for measuring SCs. By this means the relative length, arm ratio, and the percent SC in heterochromatin were determined for each SC.

Genome Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/03/13 For personal use only.

SHERMAN AND STACK

355

FIGS. 1 and 2. Silver-stained tomato synaptonemal complexes (SCs) spread by hypotonic bursting. Fig. 1. A complete set of SCs from diploid tomato. Each SC is numbered at its kinetochore. Note the arrow at the broken short arm of SC 2. The lateral elements in euchromatin and kinetochores are thicker and more densely stained than the lateral elements in pericentric heterochromatin. Transverse lines mark heterochromatin-euchromatin borders. Fig. 2. A trivalent from a plant trisomic for chromosome No. 4. Note the three lateral elements associated by triple synapsis and partner exchanges (arrowheads). Small arrows indicate kinetochores and the large arrow indicates a foldback. Bar = 3 pm.

Results and discussion Tomato SC karyotype To determine the karyotype of tomato SCs, 100 complete diploid sets of SCs in mid to late pachynema were measured. Fifty sets were silver stained and 50 sets were UP stained. Only sets with no apparent stretching of the SCs were used. Stretching of SCs is indicated when the normal separation between synapsed lateral elements is obviously narrowed. One of the sets we used is shown in Fig. 1. Kinetochores are visible as fuzzy, round structures roughly 1 pm in diameter. Lateral elements within kinetochores are densely stained (Stack and Anderson 1986a, 1986b). All tomato chromosomes have pericentric heterochromatin, the location of

which corresponds to the thinner, lighter staining lateral elements of the SC, while the lateral elements in euchromatin are thicker and darker staining (Stack 1982; Stack and Anderson 1986a, 1986b). Absolute lengths of these complete sets of tomato SCs range from 153.2 to 296.5 pm with a mean of 207.4 pm ( + 25.70 pm). Because of this variability in length between sets, relative lengths of SCs within a set were more useful than absolute lengths of SCs for comparing SCs between sets. The relative length of each SC was expressed as a percentage of the total length of the complete set of SCs in which the SC was found. Arm ratios were determined by dividing the length of the long arm by the length of the short arm. Metacentric SCs will have arm ratios near 1.O, while the more acrocentric SCs will have larger

GENOME, VOL. 35, 1992

TABLE1. Tomato SC and pachytene chromosome karyotypes % length of heterochromatin

% length

of set

Arm ratio (long/short)

Short arm

Long arm

Total

Genome Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/03/13 For personal use only.

SC or chromosome (CH) number

NOTE:Numbers in parentheses are standard deviations and numbers in brackets are coefficients of variation expressed as a percent (i.e., x 100). 'SC karyotype for SCs based on measuring every SC in 100 complete diploid sets. Fifty of these were silver stained and 50 were UP stained. These two groups were statistically indistinguishable for all characters. b~~ karyotype based on measuring every SC in 10 trisomic sets for each chromosome. 'Averages from Barton's (1950) karyotype based on 10 measurements of each pachytene chromosome. d ~ v e r a g e taken s from Ramanna and Prakken's (1967) karyotype based on 25 measurements of each pachytene chromosome. eOur measurements of Khush and Rick's (1968) idiogram for tomato pachytene chromosomes. h he short arm (the nucleolus organizer) of SC 2 is broken or asynapsed so only the long arm was measured. Therefore, arm ratio and total percent heterochromatin could not be determined. gThe short arm of SC 2 was determined to be 100% heterochromatic because of the staining characteristics of the lateral elements in the short arm. h ~ l35 l sets of SCs from plants trisomic for chromosome No. 5 always contained 12 bivalents and 1 univalent. Therefore, SC 5 could not be identified based on trivalent association.

SHERMAN AND STACK

Genome Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/03/13 For personal use only.

arm ratios. Percent heterochromatin in each arm was determined by dividing the length of the SC arm in heterochromatin by the total length of the SC arm. The percent heterochromatin for a whole SC was determined by dividing the length of SC in heterochrornatin by the length of the whole SC. Standard deviations were calculated for the relative length of each SC, the arm ratio of each SC, and the percent heterochromatin for each SC arm and each SC. To compare the relative amount of variation of these characters from SCs of different sizes, the coefficient of variation was determined for each character. The coefficient of variation is the standard deviation divided by the mean ( x 100 for percent, Simpson et al. 1960). Silver-stained and UP-stained sets were analyzed separately. Because sets of SCs stained by these two methods are not significantly different, data from these two groups were merged (Table 1). Relative lengths and arm ratios were the most useful characters for describing the karyotype because they showed less variability than percent heterochromatin, and these two characters alone were usually sufficient to identify an SC. Coefficients of variation for relative length varied from 5 to 9% (Table 1). Coefficients of variation for arm ratio varied from 4 to 12Vo (excluding SC 2, see below and Table 1). In comparison, coefficients of variation for percent heterochromatin varied from 16 to 23% in the short arm (excluding SC 2, see below), 21 -30% in the long arm, and 15-25% in whole SCs (Table 1). The basis for the variability in heterochromatin is uncertain. It may be due to differences in packaging heterochromatin from one cell to another or to indistinct borders between eu- and hetero-chromatin. In support of the latter explanation, the variability of SC length within heterochromatin had little effect on relative lengths or arm ratios, i.e., the length of SC in euchromatin compensated for the differences in length of SC in heterochromatin. On the basis of the foregoing considerations, the SCs of tomato were differentiated in the following manner (Table 1). SC 1 is the longest (12.9%), being 30% longer than the next longest SC (No. 3). SC 2 is the third longest (9.1%) based only on its long arm. It has a broken and (or) asynapsed short arm that appears to be completely heterochromatic. SC 3 is the second longest (9.9%) and has the third largest arm ratio (3.25). SC 4 is the fourth longest (8.8%) and has the smallest arm ratio (2.62) of the four longest SCs. SC 5 has a relative length of 7.3% making it shorter than SCs 1-8. With an arm ratio of 1.05, it is the longest metacentric SC. SC 6 is indistinguishable on the basis of length (7.9%) from SCs 7 and 8. However, SC 6 has .the largest arm ratio (3.48) other than SC 2 and the smallest percentage of heterochromatin (30%) of any SC. SC 7 is similar in length (8.0%) to SCs 6 and 8, but SC 7 is more nearly metacentric (1.86) than SCs 6 or 8. SC 8 is similar in length (8.0%) to SCs 6 and 7, but its arm ratio (2.38) is intermediate between the arm ratios of SCs 6 and 7. SC 9 is tied with SC 10 for ninth longest SC (7.3%), but SC 9 has a smaller arm ratio (1.79) than SC 10. SC 10 is similar in length to SC 9 (7.2%), but SC 10 has a larger arm ratio (2.12).

3 57

SC 11 is the 1lth longest SC (7.0%), but it is only slightly longer than SC 12. It differs from SC 12 in that it has a larger arm ratio (1.22). SC 12 is the shortest SC (6.8%), but it is only slightly shorter than SC 11. It differs from SC 11 in that it is more metacentric (1.06). All of the SCs are in the proper order when ranked according to relative length, except for SCs 2 and 5. SC 2's short arm is heterochromatic and carries the nucleolus organizer (Moens and Butler 1963). As a result, the short arm is generally broken and (or) asynapsed. Because of the variability of the shbrt arm, only the long arm was measured. If SC 2's short arm was consistently present and included in the total length, it would probably be the second longest and the most acrocentric SC. Although SC 5 is also out of order, it undoubtedly corresponds to chromosome 5 based on our trisomic analysis (see below). Pachytene chromosome 5 is similarly out of order in Ramanna and Prakken's (1967) karyotype (Table l), but chromosome 5 is in the correct position according to length in Barton's (1950) and Khush and Rick's (1%8) pachytene chromosome karyotype (Table 1). However, all the karyotypes agree that SC 5 and chromosome 5 are the longest metacentrics. In most organisms examined so far, all the SCs cannot be distinguished because kinetochores are not visible and (or) two or more of the SCs have similar relative lengths and arm ratios. For example, Croft and Jones (1986) could identify 6 of 11 SCs from Locusta migratoria and 2 of 11 SCs from Schistocerca gregaria. Albini and Jones (1988) could identify three of eight SCs from both Allium cepa and Allium fistulosum. In contrast, tomato SCs are dissimilar enough that all 12 can be identified when found in the proper combination. The only SC that can be identified unambiguously when found alone is SC 2 based on its broken and (or) asynapsed short arm. The other SCs can be identified in incomplete sets if the most similar SCs are also present for comparison. As described above, all tomato SCs can be identified unambiguously in complete sets.

Trisomic analysis to identify tomato SCs To relate individual SCs to specific chromosomes, SC spreads from plants that were trisomic for all 12 chromosomes were examined. In some sets of SCs from all trisomics except 5, trivalents were observed. Trivalents consisted of three lateral elements associated by partner exchanges and (or) by triple synapsis (Fig. 2; Sherman et al. 1989). The presence of the extra lateral element in trivalents allowed these SCs to be related to particular chromosomes. On the other hand, in the 35 spreads of SCs from plants trisomic for chromosome 5 that we examined, there were 12 bivalents and 1 univalent, i.e., there were no trivalents. Also the univalent was never obviously associated with any bivalent. As a result, the SC associated with chromosome 5 could not be identified directly by trivalent formation, but SC 5 must correspond to chromosome 5, since all of the other SCs have been reliably identified. The trisomic sets of SCs were measured in the same way as the diploid sets. When relative lengths, arm ratios, and amounts of heterochromatin were compared between diploid and trisomic plants, the values were not significantly different, confirming the identity of all the SCs (Table 1). Apparently neither the extra chromosome in trisomics nor the different varieties of tomatoes

Genome Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/03/13 For personal use only.

358

GENOME, VOL. 35, 1992

FIG. 3. Idiogram of tomato synaptonemal complexes (SCs) based on 100 complete sets of SCs. The 12 SCs are shown as line drawings. Thick lines represent the location of euchromatin, thin lines represent the location of heterochromatin, and circles represent kinetochores. The dashed line represents the broken and (or) asynapsed short arm of SC 2. Each SC is identified with a number that corresponds to its somatic chromosome, pachytene chromosome, and linkage group. The length of these SCs is based on the average length of a set of tomato SCs (207 pm). Bar = 10 pm.

(Cherry versus San Marzano) had any observable effect on the characteristic features of the SCs. Subsequently when we refer to "the SC karyotype," we mean the karyotype based on analysis of 100 complete sets of SCs, i.e., the first karyotype in Table 1. Tomato S C idiogram Figure 3 is an idiogram for tomato SCs in which each SC has a number that corresponds to its somatic chromosome, pachytene chromosome, and linkage group (Ramanna and Prakken 1967; Rick and Khush 1968; Tanksley and Mutschler 1990). comparing the tomato S C karyotype with tomato pachytene chromosome karyotypes Since SCs are located between synapsed homologous chromosomes, SC karyotypes should be closely related to pachytene chromosome karyotypes. Three karyotypes based

on squashed pachytene chromosomes from tomato are included in Table 1 for comparison with the tomato SC karyotype. Barton (1950) prepared a karyotype based on 10 measurements of each tomato pachytene chromosome, while Ramanna and Prakken (1967) based their karyotype on 25 measurements of each pachytene chromosome. They both determined averages for the absolute length in micrometres of each chromosome, each chromosome arm, and heterochromatin in each arm. For the purpose of comparison, we converted these measurements into relative lengths, arm ratios, and percent heterochromatin. The third pachytene chromosome karyotype is based on our measurements of an idiogram published by Khush and Rick (1968). This idiogram was taken after Barton (1950), with some modifications based on Khush's extensive experience with squashes of tomato pachytene chromosomes (C. Rick, personal communication). Therefore, as one might expect, the Barton (1950) and Khush and Rick (1968) karyotypes are the most similar. However, the three pachytene chromosome karyotypes are at some variance with one another and with the SC karyotype, especially in the amount of heterochromatin. In this regard, the SCs have up to 50% more heterochromatin than the pachytene chromosomes. Some of the variation between the SC karyotype and the pachytene chromosome karyotypes may be due to the following differences in preparation: (i) fixing in formaldehyde versus acetic acid - ethanol (1:3), (ii) spreading by hypotonic bursting versus squashing, (iii) staining with silver versus acetocarmine, (iv) observing by electron microscopy versus light microscopy, and (v) analyzing complete sets of SCs versus partial sets of pachytene chromosomes. However, all the variation cannot be explained on the basis of these differences in preparation because the variation between the pachytene chromosome karyotypes is often as great as the variation between SC and pachytene chromosome karyotypes (Table 1). In spite of the differences, the SC karyotype and pachytene chromosome karyotypes are in general agreement. Each SC and its corresponding pachytene chromosome have similar relative lengths and arm ratios. Also, every SC and pachytene chromosome have pericentric heterochromatin. Both sorts of karyotypes are complementary ways of describing chromosomes. The tomato pachytene chromosome karyotype will continue to be useful for studies by light microscopy, while the tomato SC karyotype is a new tool for high-resolution analysis by electron microscopy. Acknowledgements We thank Dr. Charles Rick of the Department of Vegetable Crops at the University of California at Davis for supplying trisomic tomato seeds; Lorrie Anderson, Lisa Herickhoff, and Dan Peterson for reading the manuscript; and Mark Anderson for writing the computer program for measuring SCs. The work was supported in part by grants DCB-87 18170 and DCB-8918613 from the National Science Foundation and grant 1878-670 from the Colorado Agricultural Experiment Station. The work was done as partial fulfillment of Ph.D. requirements for J .D. Sherman. Albini, S., and Jones, G. 1988. Synaptonemal complex spreading in Allium cepa and Allium fistulosum. 11. Pachytene observations: the SC karyotype and the correspondence of late recombination nodules and chiasmata. Genome, 30: 399-410.

SHERMAN AND STACK

Genome Downloaded from www.nrcresearchpress.com by Nanjing University of Posts and Telecommunications on 06/03/13 For personal use only.

Barton, D. 1950. Pachytene morphology of the tomato chromosome complement. Am. J. Bot. 37: 639-643. Croft, J., and Jones, G. 1986. Surface spreading of synaptonemal complexes in locusts. I. Pachytene observations. Chromosoma, 93: 483-488. Khush, G., and Rick, C. 1968. Cytogenetic analysis of the tomato genome by means of induced deficiencies. Chromosoma, 23: 452-484. Moens, P., and Butler, L. 1963. The genetic location of the centromere of chromosome no. 2 in the tomato (Lycopersicon esculenturn). Can. J. Gen. Cytol. 5: 364-370. Ramanna, M., and Prakken, R. 1967. Structure of and homology between pachytene and somatic metaphase chromosomes of the tomato. Genetica (The Hague), 38: 115-133. Rick, C. 1987. Seedling traits of primary trisomics. Tomato Genet. COOP.37: 60-61. Rick, C. 1991. Tomato paste: a concentrated review of genetic highlights from the beginnings to the advent of molecular genetics. Genetics, 128: 1-5. Rick, C., and Hunt, D. 1961. Improved seed germination with the use of sodium hypochlorite. Tomato Genet. Coop. 11: 22.

359

Sherman, J., Stack, S., and Anderson, L. 1989. Two-dimensional spreads of synaptonemal complexes from solanaceous plants. IV. Synaptic irregularities. Genome, 32: 743-753. Simpson, G., Roe, A., and Lewontin, R. 1960. Quantitative Zoology. Harcourt, Brace & World, Inc., New York. pp. 89-95. Stack, S. 1982. Two-dimensional spreads of synaptonemal complexes from solanaceous plants. I. The technique. Stain Technol. 57: 265-272. Stack, S., and Anderson, L. 1986a. Two-dimensional spreads of synaptonemal complexes from solanaceous plants. 11. Synapsis in Lycopersicon esculenturn. Am. J. Bot. 73: 264-281. Stack, S., and Anderson, L. 1986b. Two-dimensional spreads of synaptonemal complexes from solanaceous plants. 111. Recombination nodules and crossing over in Lycopersicon esculentum (tomato). Chromosoma, 94: 253-258. Tanksley, S., and Mutschler, M. 1990. Linkage map of tomato (Lycopersicon esculentum). In Genetic maps. Locus maps of complex genomes. Book 6 . Plants. Edited by S. O'Brien. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. pp. 4-15.