CHROMOSOMAL POLYMORPHISM IN 12 POPULATIONS OF Mikania micrantha ( COMPOSITAE )

Mikania micrantha is a climbing perennial weed of the family Asteraceae, with a vast distribution from South America to south of the United States. This species is widely distributed throughout Brazil, where it shows little morphological variation. Mitotic chromosomes of 12 populations of M. micrantha derived from several Brazilian sites were studied using Feulgen staining and C-banding. The populations included eight diploid (2n = 36 and 42) and four tetraploid (2n = 72) cytotypes. Chromosome numbers of 2n = 36 and 2n = 42 are reported for the first time for M. micrantha. These populations had a secondary constriction in the middle of the larger arm of chromosome pair 1, following the same pattern described for all Mikania species analyzed so far. Numerical and structural variation of the chromosomes was quite common among the karyotypes and nearly all cytotypes differed from each other in some aspect. Most of the chromosomal differentiation may be attributed to inversions and addition or deletion of DNA fragments. C-banding, applied to three of the 12 populations, also revealed polymorphism in the distribution of heterochromatin. Additionally, one to 14 supernumerary or B-chromosomes were observed. The Bs were detected in six of the 12 populations and varied in size, number, and structure among karyotypes and also among cells of the same root meristem. The B chromosomes were also heterochromatic, showing a C-banding pattern similar to the A chromosomes, and suggesting that they may be derived from the chromosomes of the A complement. Departamento de Biologia Geral, CCB, Universidade Estadual d Londrina, 86051-990 Londrina, PR, Brasil. Send correspondence to P.M.R. Departamento de Biologia da Universidade Estadual Paulista, Assis, SP, Brasil. Departamento de Botânica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brasil.


INTRODUCTION
Mikania micrantha H.B.K. is a climbing perennial weed of the family Asteraceae distributed throughout tropical and subtropical regions of the American continent.It is a pioneer species, frequently found in disturbed environments or in changing communities where the original vegetation has been destroyed for crop introduction.It holds an advantage over many other pioneer species because of its vigorous vegetative and sexual reproduction (Swamy and Ramakrishnan, 1987).M. micrantha is of great ecological importance due to its participation in the reposition of degraded or newly open habitats.It is widespread throughout many Brazilian regions where it grows in environments such as forest borders, roadsides, along fences, and in newly changed habitats.
Numerous cytotypes have been reported in M. micrantha, including n = 19 in a population from Colombia and n = 19 and 20 in a West Indian population (Powell and King, 1969a,b).Chromosome numbers of n = 17 and n = 19 were reported for populations from Ecuador (King et al., 1976) and Argentina (Turner et al., 1979 andWaisman et al., 1984).Nineteen bivalents were found in one population of M. micrantha from Mexico (Strother, 1983) and another from Jamaica (Keil et al., 1988).The karyotype of a polyploid population (2n = 72) from Londrina, Brazil, was analyzed by Ruas and Ruas (1987).The present study examines the karyotypes of 12 populations of M. micrantha collected from different Brazilian sites to help provide a better understanding of chromosome variation in this species.

MATERIAL AND METHODS
The specimens of M. micrantha were obtained from 12 sites in Brazil (Table I) and cultivated in a greenhouse.At least five samples were collected from each population.Root tips were collected from potted plants, pretreated with 0.002 M 8-hydroxyquinoline for 4 h at 8 o C, fixed in 3:1 ethanol-glacial acetic acid overnight, transferred to 70% alcohol, and stored in a refrigerator until used.The conventional Feulgen method with modifications described by Nogueira et al. (1995) was used for chromosome preparations.C-banding was obtained with the method of Schwarzacher et al. (1980).
The morphological chromosome data used in this study included: 1) absolute length of individual chromosomes and haploid chromosome length, both measured in µm; 2) relative length of each chromosome; 3) arm ratio for each chromosome (long arm/short arm).
The chromosomes were classified according to the nomenclature of Levan et al. (1964).Diploid numbers were determined by counting the chromosomes in at least 10 metaphases of every sample of each population.Supernumerary or B chromosomes were analyzed in 25 cells of each population.The average of the chromosome lengths for each chromosome pair was obtained from the measurements of five well-spread metaphases and was used for the analysis and for the construction of the idiograms.
Karyotype asymmetry was determined by using the index total form percent (TF%) according to Huziwara (1962).The asymmetry index was also analyzed by using a plot of the values generated by the Zarco index (Zarco, 1986).The TF%, the size of the largest and smallest chromosomes, the haploid chromosome length and relative length, and the arm ratio values were compared by oneway analysis of variance and Tukey's test (Steel and Torrie, 1960).

Karyotype characterization
The cytotypes of M. micrantha derived from 12 collection sites were studied (Table I).Their karyotypes and respective idiograms are in Figures 1-4.Eight cytotypes were diploid, seven with 2n = 2x = 36 and one with 2n = 2x = 42 chromosomes; four were tetraploid with 2n = 4x = 72.Numbers of 2n = 36 and 2n = 42 are documented here for the first time.Comparisons among the diploid cytotypes (2n = 36) showed that whereas the karyotypes followed similar patterns, minor structural differences were detected in almost all corresponding chromosomes (Table II; Figures 1 and 3).Pair 1 was the most inconstant, showing variation in both size and structure.The largest chromosomes were in cytotypes from Estrela do Norte and Praia de Ipanema, with haploid sets of 32.5 ± 1.62 µm and 31.3 ± 0.36 µm, respectively (Table II; Figure 1E,F).The population from Praia Grande, on the other hand, exhibited the smallest chromosomes and a haploid set of 19.9 ± 1.55 µm (Figures 1C and 3C).The differences in chromosome sizes encompass almost all chromosome pairs, reflecting a gain of genetic material that may have derived through mechanisms of DNA amplification.Two other diploid cytotypes (populations from Alfredo Guedes and Campinas) with 2n = 36 showed the same karyotype formula and similar haploid sets (Figures 1D,G and 3D,G;Tables I, II), but they differed slightly in the arm-ratio values, which were probably modified by inversions.Similar conclusions can be reached in the analysis of the four tetraploid cytotypes with 2n = 4x = 72 chromosomes (Tables III, IV; Figures 2  and 3).In this group, only the population from Apucarana (PR) differed in the value of the haploid chromosome length (Table III).Most of the variation among the tetraploids was represented by centromeric shifts (Table V) even though the general karyotype pattern, described for the genus (Ruas and Ruas, 1987;Ruas and Aguiar-Perecin, 1997), was maintained.
The longest chromosome in the karyotypes of all populations of M. micrantha studied had a secondary constriction located in the middle of the long arm (Figures 1-3).This is a conservative pattern which has been described in many Mikania species.Ruas and Ruas (1987) studied six species of Mikania and suggested that the secondary constriction might be considered a cytological marker for the genus, which could help in the identification of the species.Those results were fully supported by the work of Ruas and Aguiar-Perecin (1997), in which the same karyotype pattern was revealed in 10 other Mikania species.
Karyotype asymmetry was also determined for all cytotypes.The TF% and Zarco index (Tables II, III; Figure 4) showed that the population from Campinas had the karyotype with the highest degree of symmetry, whereas the karyotypes of the populations from Praia de Ipanema, Joinville, and Petrópolis were the most asymmetrical.The highest value of interchromosomal asymmetry (Tables II, III) was found in the population from Estrela do Norte.This value reflects the huge size of chromosome 1, which may have resulted from duplication of parts of this chromosome pair.Less significant differences in the Zarco in-    dex were also present in the other populations.The tetraploid cytotypes had similar values of TF%.However, the Zarco index permitted the detection of slight differences among them.Chromosome variation among populations of the same species has been observed in many plant groups.In the genus Serjania (Nogueira et al., 1995), two populations of S. laruotteana, two of S. fuscifolia and three of S. gracilis with the same chromosome number (2n = 24) had different chromosome rearrangements as observed in M. micrantha (Tables II, III; Figures 1-3).Whereas the presence of chromosome races in many groups certainly involves biological adaptation to different habitats, nearly every population of M. micrantha studied showed differences in the karyotypes (Tables IV and V).These differences do not seem to be related to adaptive variables, since populations occupying similar environments bear distinct karyotypes.Similarly, karyotype variation was not accompanied by modification in plant morphology.For example, the polyploid cytotype from Apucarana had the largest leaves (14.0 cm in length, 7.0 cm in width) while the polyploid from Salto Apucaraninha had small leaves (5.0 cm long and 3.5 cm in width).The diploid population from Campinas (2n = 36) had leaves 10.0 cm long and 7.0 cm in width.These data suggest no correlation between leaf size and ploidy level.Inflorescence types were also constant among all cytotypes, with only small variation in size.

C-banding analysis
C-banding was applied to three diploid cytotypes of M. micrantha, including the populations from Piracicaba, Campinas (Figure 5A,B), and Praia Grande (data not shown).C-band analysis revealed a variable pattern in the amount and distribution of heterochromatin in these cytotypes.The cytotype from Piracicaba showed a large heterochromatic block near the secondary constriction of chromosome 1.Three other chromosome pairs had small centromeric bands (Figure 5A).Similarly, another cytotype (Campinas) exhibited a block of heterochromatin located near the secondary constriction of the large arm of chromosome 1 and several other chromosomes showed centromeric C-bands (Figure 5B).The cytotype from Praia Grande (data not shown) on the other hand had only three pairs of very small centromeric bands and total absence of heterochromatin in chromosome 1.Variation in heterochromatic blocks has been observed in several groups of plants, such as in populations of Trillium kamtaschaticum (Kurabayashi, 1957), Tulbaghia leucantha (Vosa, 1973) and Gibasis karwinskyana (Kenton, 1991).The differences in the patterns of C-band observed among the three cytotypes of M. micrantha may be associated with the small detectable variation in the size of the haploid set (Table II).Therefore, at least for the diploid cytotypes, the small differences in the haploid chromosome length may reside in unique and repetitive sequences of DNA.
Besides the variation in number from cell to cell and among cytotypes, the B-chromosomes of M. micrantha diverged in size (from micro-size to about 0.8 µm) and morphology (Figure 6).According to Jones (1995), at least 65 plant species have two or more forms of B chromosomes.Metacentric (m) and telocentric (t) B chromosomes were observed in Aegilops mutica (Mochizuki, 1960) and large and micro-sized B chromosomes were verified in Brachycome dichromosomatica (Smith-White and Carter, 1970).Loidl (1982) showed that the B chromosomes can be distinguished by their overall size, arm ratio, or Giemsa C-banding in Allium flavum.In M. micrantha, the six cytotypes with Bs show three morphological types, such as m, submetacentric (sm), and subtelocentric (st) types, which also vary from micro-sized in some cells to larger telocentrics in others (Figure 6 A-D).A variable number of very small m type Bs were found in many cells of the cytotype from Piracicaba.The m type Bs may explain the origin of micro-Bs by centromere misdivision of a single unpaired B, giving rise to two different-sized chromosomes and further derivatives by deletion of parts of the arms.The B-chromosomes of sm type were predominant in the other populations.Jones and Rees (1982) suggested that the variation in the frequency of Bs among populations may be of adaptive value in a stress situation.The distribution of Bs did not follow any specific pattern in the cytotypes of M. micrantha.Their frequency varied among cells of individual plants as well as among populations which occupy different or similar environments such as sea shores (Praia de Ipanema and Praia Grande) and high altitudes (Petrópolis).Therefore, the presence of Bs could not be associated with any adaptive requirement in M. micrantha.
C-banding (Figure 5 A,B) in the diploid cytotypes from Piracicaba, Campinas, and Praia Grande (data not shown) showed almost totally heterochromatic B chromosomes.No detectable differences were observed in the Cband pattern between A and B chromosomes, suggesting that homologous repetitive sequences may be present in both A and B chromosomes of M. micrantha.
column followed by different lowercase letters are significantly different at the 5% level by the Tukey test.Table V -Tukey test for arm ratio and relative length in the chromosome complement of four tetraploid populations of Mikania micrantha with 2n = 72 a .column followed by different lowercase letters are significantly different at the 5% level by the Tukey test.
cells of the same root meristem (Table

Table I -
Sites of origin, collection number, chromosome number, and karyotypic formula of 12 populations of Mikania micrantha.

Table III -
Chromosome length (µm), haploid chromosome length (HCL), ratio of longest/shortest chromosomes (L/S), and karyotype symmetry using total form (TF%) and Zarco index of four tetraploid cytotypes of Mikania micrantha with 2n = 72 chromosomes.a a Means within each column followed by different lowercase letters are significantly different at the 5% level by the Tukey test.

Table II -
Chromosome length (µm), haploid chromosome length (HCL), ratio of longest/shortest chromosomes (L/S), and karyotype symmetry using total form (TF%) and Zarco index of cytotypes of Mikania micrantha, seven with 2n = 36, and one with 2n = 42.a Means within each column followed by different lowercase letters are significantly different at the 5% level by the Tukey test. a

Table IV -
Tukey test for arm ratio values and relative length of the karyotypes of seven populations of Mikania micrantha with 2n = 36 and one with 2n = 42 chromosomes.a