Hereditas 140: 129 /133 (2004)
Further cytogenetical studies on diploid and polyploid species of Eryngium L. (Saniculoideae, Apiaceae) from Argentina ˜ O1, EDUARDO GREIZERSTEIN1, NATALY O’LEARY1, CAROLINA I. CALVIN 2 ´ SUSANA MARTINEZ and LIDIA POGGIO1 1
Laboratorio de Citogene´tica y Evolucio´n, Departamento de Ecologı´a, Gene´tica y Evolucio´n, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina 2 Departamento de Biodiversidad y Biologı´a Experimental, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina O’Leary, N., Calvin˜o, C. I., Greizerstein, E., Martı´nez, S. and Poggio, L. 2004. Further cytogenetical studies on diploid and polyploid species of Eryngium L. (Saniculoideae, Apiaceae) from Argentina. */ Hereditas 140: 129 /133. Lund, Sweden. ISSN 0018-0661. Received September 8, 2003. Accepted December 19, 2003 Meiotic studies are carried out in 7 species of Eryngium L. (Saniculoideae, Apiaceae), belonging to both sections Foetida and Panniculata. The chromosome number of E. dorae Norm. (n/8) (Foetida) is reported for the first time, while the gametic chromosome number of E. nudicaule Lam. (n/7) (Foetida) and E. eburneum Decne. (n/8), E. horridum Malme (n/8), E. megapotamicum Malme (n /16), E. mesopotamicum Pedersen (n/24), and E. pandanifolium Cham. et Schlechtd. (n/24) (all belonging to Panniculata) is confirmed in several natural populations. Whereas in section Foetida all species are diploids and two basic chromosome numbers are present (x/8 and x /7), in section Panniculata all species are x/8 but there are three different ploidy levels (diploid, tetraploid, hexaploid). This study reveals that meiosis in all species is normal, with regular bivalent formation in all studied cells. Furthermore, the pollen stainability is above 80% in all cases. These data, together with the previous karyotype analyses, will contribute to the clarification of the relationships between members of both sections, where different mechanisms of speciation have been postulated. Eduardo Greizerstein, Laboratorio de Citogene´tica y Evolucio´n, Departamento de Ecologı´a, Gene´tica y Evolucio´n, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Pabellon II (1428), Buenos Aires Argentina. E-mail:
[email protected]
The genus Eryngium (family Apiaceae) includes about 300 species divided into 34 sections (WOLFF 1913) and is represented in Argentina by about 30 species (MARTI´NEZ 1999; MARTI´NEZ and CALVIN˜O 2000). The greatest specific diversity is concentrated in the northeastern and center of the country (MARTI´NEZ and CALVIN˜O, op. cit. ). Previous cytological information for Argentinean species belonging to section Foetida (BELL and CONSTANCE 1957, 1966; CONSTANCE et al. 1971, 1976; VIANNA and IRGANG 1971; CALVIN˜O et al. 2002), agrees that all studied members of this section are diploids. Although the most common basic chromosome number in Eryngium is x /8 (CONSTANCE 1977) some taxa show a basic chromosome number x /7 (E. nudicaule 2n /2x /14) (CONSTANCE et al. 1971; VIANNA and IRGANG 1971; CALVIN˜O et al. 2002). Meanwhile, all studied species in section Panniculata show the same basic chromosome number (x /8) but different ploidy levels (CALVIN˜O et al. 2002). In the latter section the delimitation and identification of the species offers difficulties due to their morphological similarities and the probable occurrence of hybridization and even some degree of
introgression in transitional zones (CONSTANCE 1977; O’LEARY 2002). As part of a multidisciplinary study to obtain information about the evolutionary history of the genus, we have analyzed both the meiotic behavior and pollen staining of 7 species growing in northeastern Argentina belonging to section Foetida: E. nudicaule Lam. (2n /2x /14) and E. dorae Norm. (a species poorly collected and insufficiently known) and to section Panniculata: E. eburneum Decne. (2n /2x/ 16), E. horridum Malme (2n /2x/16), E. megapotamicum Malme (2n /4x/32), E. mesopotamicum Pedersen (2n /6x/48) and E. pandanifolium Cham. et Schlechtd (2n /6x /48). The results are analyzed in relation to the role of chromosome change in the evolution of both sections of Eryngium .
MATERIAL AND METHODS Plant material All the material studied was collected by the authors from different populations in Argentina, as detailed in Table 1. Voucher specimens have been deposited at the
130
N. O’Leary et al.
Hereditas 140 (2004)
Table 1. Origin of the samples used for cytogenetic analyses. Section
Species
Origin
Voucher
Foetida
E. dorae
Entre Rı´os. Dept. Concordia. Parque Rivadavia
E. nudicaule
Entre Rı´os. Dept. Concordia. Parque Rivadavia
E. eburneum
Buenos Aires. Dept. Campana. Loc. Otamendi Entre Rı´os. Dept. Colo´n. Km. 170 Entre Rı´os. Dept. Gualeguaychu´, Loc. Me´danos Entre Rı´os. Dept. Concordia. Parque Rivadavia. Santa Fe. Dept. San Martı´n. Entre Rı´os. Dept. Concordia. Parque Rivadavia. Entre Rı´os. Dept. Gualeguaychu´, Km. 25 Entre Rı´os. Dept. Gualeguaychu´, Km. 25. Entre Rı´os. Dept. Colo´n. Km. 170 Entre Rı´os. Dept. Colo´n. Km. 170
Cal & O’Lea 207 Cal, Mar & Gre 327 Cal, Mar & Gre 329 Cal, Mar & Gre 330 Cal & O’Lea 102 Cal & O’Lea 103v O’Lea 27 Cal, Mar & Gre 344 O’Lea & Cal 214 O’Lea 12Na Hunziker 24497 O’Lea 11 Cal, Mar & Gre 349 Cal, Mar & Gre 350 Cal, Mar & Gre 343 Cal, Mar & Gre 347
Panniculata
E. horridum E. megapotamicum E. mesopotamicum E. pandanifolium
Abbreviations: Cal/Calvin˜o, Gre/Greizerstein, Mar/Martı´nez, O’Lea /O’Leary.
Herbarium of Instituto de Bota´nica Darwinion (SI) except a voucher specimen of E. megapotamicum that is deposited at the Herbarium of Universidad de Co´rdoba (CORD).
differences between populations of the same species were not registered, data were pooled together. E. nudicaule (2n /2x/14) showed 7 bivalents at early anaphase I (Fig. 1a) and the percentage of pollen stainability for this species was 86%. E. dorae showed 8 bivalents at metaphase I (Fig. 1b), therefore the diploid chromosome number of this species is 2n/ 2x/16; the percentage of pollen stainability was 90%. Eryngium eburneum and E. horridum (both 2n /2x/16), showed 8 bivalents at diakinesis (Fig. 1c, d) and the pollen staining was high (80% and 93% respectively). In both diploids a single bivalent was always observed associated to the nucleolus. E. megapotamicum , (2n /4x /32) presented 16 bivalents (Fig. 1e), and also high pollen staining (83%). Two bivalents were always observed in association to the nucleolus. E. pandanifolium and E. mesopotamicum (2n /6x/48), both showed 24 bivalents (Fig. 1f, g), and the pollen staining was high, 92% and 93%, respectively. In both these species, there were generally three bivalents associated to the nucleolus (Fig. 1f, g).
Cytological analysis For meiotic studies flower buds were fixed in 3 parts of absolute alcohol: 1 part of glacial acetic acid. The anthers were squashed in a drop of acetic haematoxilin 2% (NU´N˜EZ 1968). Pollen stainability was determined using Alexander stain (ALEXANDER 1969). For meiotic chromosome counts 10/30 cells of at least three individuals from each population were analyzed. Pollen stainability was performed on about 300 /500 pollen grains. RESULTS Chromosome numbers, meiotic configurations and the percentage of pollen staining of the seven studied species of Eryngium are indicated in Table 2. Since
Table 2. Somatic chromosome number (2n), basic chromosome number (x), ploidy level, meiotic configurations and pollen stainability of the seven species of Eryngium . Section
Species
No. of cells
2n
x
Ploidy level
Meiotic configuration
Pollen stainability (%)
Foetida
E. E. E. E. E. E. E.
98 52 58 52 65 68 54
/ 14 16 16 32 48 48
8 7 8 8 8 8 8
2x 2x 2x 2x 4x 6x 6x
8II 7II 8II 8II 16II 24II 24II
90 86 80 93 83 92 93
Panniculata
dorae* nudicaule eburneum horridum megapotamicum mesopotamicum pandanifolium
* Chromosome count reported for the first time.
Hereditas 140 (2004)
Cytogenetical studies species of Eryngium
131
Fig. 1. a /g Meiotic configurations: (a) E. nudicaule (n /7); (b) E. dorae (n /8); (c) E. eburneum n/8; (d) E. horridum n/8; (e) E. megapotamicum n/16; (f) E. pandanifolium n/24; (g) E. mesopotamicum n/24. All the cells are at diakinesis, except (a) early anaphase I and (b) metaphase I. Bar/10 mm. Arrows show the nucleolus.
DISCUSSION Eryngium nudicaule is a widely distributed and highly polymorphic taxon (MARTI´NEZ and GALOTTI 2001) that has already been cytologically studied (in meiosis and mitosis) in several Brazilian and Argentinean populations (CONSTANCE et al. 1971; VIANNA and IRGANG 1971; ALMADA et al. 2002; CALVIN˜O et al. 2002). In the present work the gametic chromosome number n /7 for this diploid species is confirmed. Eryngium dorae is in several aspects a poorly known taxon, probably because it is an annual species with a
short vegetative phase, an infrequent condition in Eryngium where plants are generally perennials. There were no previous cytological studies on this species and this paper represents the first report of the chromosome number and meiotic behaviour for E. dorae (n/8). Both taxa belong to section Foetida, a group where all studied species (eleven) are diploids and where two basic chromosome numbers, x/8 and x /7, are found. The number x /7 is only present in E. nudicaule and E. corallinum Math. and Const., both
132
N. O’Leary et al.
sympatric and morphologically similar taxa (CONSTANCE et al. 1971). In a recent cladistic analysis on this section (CALVIN˜O 2001), mostly based on morphological characters, the basic chromosome number x /7 appears as a atavic number originated in the common ancestor of E. nudicaule and E. corallinum . CALVIN˜O et al. (2002) and ALMADA et al. (2002) have suggested that aneuploidy could explain the reduction of the modal basic number x/8 to x/7. CALVIN˜O et al. (2002) based on karyotype analysis, stated that the most parsimonious explanation for the reduction of the chromosome number seemed to be an unequal reciprocal translocation with loss of nonessential genetic material. The other five studied species belong to section Panniculata, a group where polyploidy seems to have played a major role in speciation. CALVIN˜O et al. (2002) found three ploidy levels: diploid (E. horridum and E. eburneum , 2n /2x/16), tetraploid (E. megapotamicum , 2n /4x /32) and hexaploid (E. pandanifolium and E. mesopotamicum , 2n /6x/48). Most authors agree that the basic chromosome number in all studied taxa belonging to this section is x /8 (CONSTANCE 1977; HORE 1979; CONSTANCE and CHUANG 1982; CALVIN˜O et al. 2002). Eryngium eburneum and E. horridum are two diploid species with 2n/2x /16, which is here confirmed by the presence of 8 bivalents in meiosis. It is worthwhile to point out that ALMADA et al. (2002) found 2n /14 for a population of E. horridum from Misiones, Argentina. Since all previous chromosome counts on this taxon showed 2n /16 (CONSTANCE et al. 1971, 1976; VIANNA and IRGANG 1971; CALVIN˜O et al. 2002) the 2n /14 report could be indicating the presence of polymorphisms (different diploid number within populations) or polytypisms (different diploid number in different populations) for both basic chromosome numbers x /7 and x/8. Eryngium megapotamicum is a tetraploid taxa with 2n /4x /32 (CALVIN˜O et al. 2002). The presence of 16 bivalents is the first meiotic result, and agrees with the previous report in mitotic cells. The diploid species E. eburneum and E. horridum , and the tetraploid species E. megapotamicum grow in sympatry sharing the same ecological niches. Moreover E. megapotamicum is similar to E. eburneum regarding its mericarp morphology and straight up position of caulinar leaves, and to E. horridum in plant size, plantule morphology and capitula form. Based upon its karyotype morphology and the exo-phenotypic intermediate characters, CALVIN˜O et al. (2002) suggested an hybrid origin for this tetraploid, being the diploid species E. eburneum and E. horridum its putative progenitors. The regular meiotic behaviour observed
Hereditas 140 (2004)
in E. megapotamicum is the expected for an allopolyploid of hybrid origin. Although a regular meiotic behaviour could also be characteristic of an autopolyploid with a diploidized behavior, the intermediate morphology observed strongly suggests that E. megapotamicum would be an allopolyploid. In E. pandanifolium and E. mesopotamicum the presence of 24 bivalents in meiosis agrees with the previous data of CONSTANCE et al. (1971), HUNZIKER et al. (1985) and CALVIN˜O et al. (2002), suggesting that both are hexaploid taxa with 2n /6x/48. However, VIANNA and IRGANG (1971) and ALMADA et al. (2002) report 2n /64 for Argentinean and Brazilian populations of both species. MATHIAS and CONSTANCE (1971) suggest that E. pandanifolium and related entities constitute a polyploid complex, so the presence of taxa with different ploidy levels and little morphological variation is not unexpected, given that several processes of hybridization and polyploidy could have originated the different chromosome numbers that are found in this taxa complex. In relation to the basic chromosome number, CERCEAU-LARRIVAL (1971) suggests that E. pandanifolium and E. mesopotamicum (sub. nom E. lassauxii ) are octoploids with x/6, without giving further evidence for it and apparently based upon chromosome counts of CONSTANCE et al. (1971) and BELL and CONSTANCE (1966). Even when the basic chromosome number x /6 could have originated by a aneuploidy process, it is more parsimonious to suppose that both entities are hexaploids with x /8. Furthermore, the presence of three bivalents associated to the nucleolus in E. pandanifolium and E. mesopotamicum would support an hexaploid level with a basic chromosome number x/8, since diploid taxa show one bivalent associated to the nucleolus. O’LEARY et al. (2002) by means of studies with FISH using the ribosomal probe pTa 71 have found 2, 4 and 6 rDNA zones in the diploid, tetraploid and hexaploid species respectively, which coincides with the observations performed in the present work. Since E. pandanifolium and E. mesopotamicum differ morphologically only in the colour of their inflorescences (purple in the first case and green in the second) and minor differences in the spination of their leaves, some authors treat them as varieties of a single species (MATHIAS and CONSTANCE 1971). No cytological differences were found between both the two entities with regard to their meiotic behaviour, their pollen stainability, or their karyotype morphology (CALVIN˜O et al. 2002; O’LEARY 2002). The hexaploid level of E. pandanifolium and E. mesopotamicum could be understood as an allopolyploid event between a diploid and a tetraploid taxon. In their area of
Hereditas 140 (2004)
distribution the only known tetraploid species is E. megapotamicum , which could be one of the putative ancestors. Moreover the diploids E. eburneum and E. horridum which were postulated as progenitors of E. megapotamicum are abundant species and morphologically similar to these hexaploids. It is plausible that they have played a role in the origin of the hexaploid taxa (O’LEARY 2002). So far the mitotic and meiotic studies performed on Eryngium suggest that the different basic chromosome numbers would probably be the result of aneuploid changes (ALMADA et al. 2002; CALVIN˜O et al. 2002) and that polyploidy would have played an important role in the evolution of the group of species belonging to section Panniculata. Since E. paniculatum and E. mesopotamicum are morphologically and cytologically very similar, it can be postulated that they had originated from a unique polyploid event, and that they diverged later through mutation and/or introgression. More studies from ecological, morphological and molecular viewpoints will allow us to precise the degree of divergence between both taxa. Nevertheless, the nature of the chromosome rearrangements involved in the aneuploid process, as well as the recognition of the parentals of the polyploid taxa, need further molecular phylogenetic and cytogenetic analyses. Acknowledgements / We thank Dr. Alba Papeschi for critical revision of the manuscript. This work has been supported by grants from the Agencia Nacional de Promocio´n Cientı´fica, Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas (CONICET) and Universidad de Buenos Aires (UBA) to Lidia Poggio. We also thank the Universidad de Buenos Aires for the fellowship to Carolina Calvin˜o.
REFERENCES Alexander, M. P. 1969. Differential staining of aborted and non-aborted pollen. / Stain Technol. 2: 117 /137. Almada, R. D., Seijo, G. and Davin˜a, J. 2002. Karyological studies in Argentinian species of Eryngium (Apiaceae). / Cytologia 205: 205 /211. Bell, R. and Constance, L. 1957. Chromosome numbers in Umbelliferae I. / Am. J. Bot. 44: 565 /572. Bell, R. and Constance, L. 1966. Chromosome numbers in Umbelliferae III. / Am. J. Bot. 53: 512 /520. Calvin˜o, C. 2001. Relaciones filogene´ticas entre las especies de Eryngium L. (Umbelliferae), Section Foetida Wolff. / Seminario de Licenciatura en Ciencias Biolo´gicas. Facultad de Ciencias Exactas y Naturales. Universidad de Buenos Aires.
Cytogenetical studies species of Eryngium
133
Calvin˜o, C., O’Leary, N., Greizerstein, E. et al. 2002. Chromosome studies on eight species of Eryngium L. (Saniculoideae, Apieaceae) from Argentina. / Caryologia 55: 315 /321. Cerceau-Larrival, M. T. 1971. Plantules et pollens dans le Genre Eryngium L. Relations avec la bioge´ographie. / C. r. Soc. Bioge´ogr. 424: 103 /112. Constance, L. 1977. Some problems in new world Eryngium . Actes du 2me. Symp. Int. sur les Ombelliferes, Cauvet, M. and Carbonnier, J. (eds). Perpignan. Constance, L. and Chuang, T. 1982. Chromosome numbers of Umbelliferae (Apiaceae) from Africa south of the Sahara. / Bot. J. Linn. Soc. 85: 195 /208. Constance, L., Chuang, T. and Bell, R. 1971. Chromosome numbers in Umbelliferae IV. / Am. J. Bot. 58: 577 /587. Constance, L., Chuang, T. and Bell, R. 1976. Chromosome numbers in Umbelliferae V. / Am. J. Bot. 63: 608 /625. Hore, A. 1979. Structure and behavior of chromosomes as an aid to the study of phylogeny of Umbelliferae with special reference to the tribe Apieae (Ammineae) and Saniculeae. / Cytologia 43: 389 /402. Hunziker, J. H., Xifreda, C. C. and Wulff, A. F. 1985. Estudios cromoso´micos en Angiospermas de Sudame´rica. / Darwiniana 26: 7 /14. Martı´nez, S. 1999. Apiaceae. / In: Zuloaga, F. and Morrone, O. (eds), Cata´logo de las Plantas Vasculares de Argentina II. Monogr. Syst. Missouri Bot. Gard. 74: 46 /64. Martı´nez, S. and Calvin˜o, C. 2000. Contribucio´n al conocimiento del Ge´nero Eryngium L. (Umbelliferae) en Argentina. / Gayana Bot. 57 (Supl.): 60. XXVII Jornadas Argentinas de Bota´nica y XII Reunio´n Anual de la Soc. Bota´nica de Chile. Concepcio´n, Chile. Martı´nez, S. and Galotti, L. D. 2001. Las especies de Eryngium Section Foetida (Apiaceae) de Argentina. / Darwiniana 39: 155 /169. Mathias, M. E. and Constance, L. 1971. New taxa and new combinations in the Umbelliferae of Santa Catarina, Brasil. / Sellowia 23: 45 /51. Nun˜ez, O. 1968. An acetic-haematoxilin squash method for small chromosomes. / Caryologia 21: 115 /119. O’Leary, N. 2002. Estudios de citogene´tica cla´sica y molecular en especies del ge´nero Eryngium L. (Umbelliferae), Section. Panniculata Wolff. / Seminario de Licenciatura en Ciencias Biolo´gicas. Facultad de Ciencias Exactas y Naturales. Universidad de Buenos Aires. O’Leary, N., Greizerstein, E., Martinez, S. et al. 2002. Determinacio´n de zonas rDNA en 5 especies de Eryngium L. (Umbelliferae), Section Panniculata Wolff. / BAC 15 (Suppl): 87. Vianna, F. M. S. and Irgang, B. E. 1971. Levantamento do numero cromosoˆmico em espe´cies do Gen. Eryngium L. (Umbelliferae) do R. G. S. Iheringia. / Bota´nica 15: 49 / 51. Wolff, H. 1913. Umbelliferae, Saniculoideae. / In: Engler, Das Pflanzenreich 4 (228): 1 /350.
