Introduction

Solanaceae posses about 90 genera and 3,000 species, which have great economical importance, as Solanum tuberosum L. (potato), Solanum melogena L (eggplant) and several peppers (Capsicum L.), used as food in several cultures (D’Arcy 1991). Members of the family Solanaceae show basic chromosome number varying from x = 7 to x = 13 and occurrence of symmetrical karyotypes with predominance of meta- and submetacentric chromosomes. These features are quite evident in Capsicum L. (Moscone 1993; Moscone et al. 1993) and Nicotiana L. (Nakamura et al. 2001; Kitamura et al. 2001). The genus Solanum L., with about 1,400 species of herbs, subtrees and trees (Nee 1999; Child and Lester 2001; Bohs 2005), possess species with 2n = 2x = 24, besides polyploids with 2n = 4x = 48, 2n = 6x = 72 and 2n = 8x = 96 (Hunziker 2001).

Chromosome banding and FISH studies show that the species of Solanaceae have a wide variation of repetitive DNA occurrence and distribution. We can mention the occurrence of AT-rich heterochromatin in the largest chromosomes of Nicotiana kawakamii Y. Ohashi (Nakamura et al. 2001), the predominance of centromeric C-bands in some species of Capsicum in relation to other that are more terminal (Moscone et al. 1993), and different heterochromatin types in Cestrum L. (Fregonezi et al. 2006). The positioning of the rDNA segments has been described in S. bulbocastanum Dunal (Dong et al. 2000) and S. tuberosum L. (Komarova et al. 2004), but scarcely documented in other native species of Solanum.

Molecular phylogeny studies have changed substantially the taxonomic organization in groups close to Solanum, e.g., through the inclusion of Cyphomandra Mart. ex Sendtn. as a section of Solanum L., at the same time that genera which are morphologically closer [such as Lycianthes (Dunal) Hassl.] have been maintained out of Solanum (Olmstead and Palmer 1997; Bohs and Olmstead 1997, 1999; Olmstead et al. 1999; Bohs 2004, 2005). The cytogenetic data do not support the above mentioned taxonomic arrangements, which are mainly based in sequences of few genes. In order to contribute to a better understanding of the taxonomic relationships among Solanum and related genera, this study aimed the karyotype characterization of some species of Solanum, as compared to Lycianthes australe and Vassobia breviflora (Sendtn.) Hunz.

Materials and methods

Plant material

Seeds and samples of Solanum, Lycianthes and Vassobia were collected in different localities of Brazil: Mato Grosso do Sul, Paraná and Rio Grande do Sul states (Table 1), and cultivated in pots at Laboratório de Biodiversidade e Restauração de Ecossistemas––LABRE, CCB, UEL, Londrina, Brazil. At least three individuals from each species were used for the karyotype analyses. Vouchers are kept in the herbarium of the Universidade Estadual de Londrina (FUEL). The species were organized in sections according to Nee (1999).

Table 1 Karyotype features of Solanum species (according Nee 1999), Lycianthes australe and Vassobia breviflora
Full size table

Conventional staining

The method described by Guerra (1983) with modifications was used for the study of somatic chromosomes. Roots were pretreated with 2 mM 8-hydroxyquinoline for 24 h and fixed in absolute ethanol: glacial acetic acid (3:1, v:v) for 12 h and kept at −20°C until used. Root tips were digested for 1 h in a mixture of 4% (w/v) cellulase and 40% pectinase (v/v), further hydrolyzed in 1 M HCl at 60°C for 11 min, dissected in a drop of 45% acetic acid and squashed. The cover slips were removed after freezing in liquid nitrogen. The material was stained with 2% Giemsa and permanent slides mounted in Entellan. The chromosome measurements performed using the MicroMeasure 3.3 software (http://www.biology.colostate.edu), allowed to calculate the: (1) haploid set length and (2) centromere position and (3) karyotype formulae. Chromosome types were classified according to Guerra (1986).

Chromosome banding

The banding was performed as described by Vanzela et al. (2002), based in the procedure of Schwarzacher et al. (1980). Root tips were digested in an enzyme solution composed of 4% cellulase (w/v) and 40% pectinase (v/v) at 37°C and dissected in a drop of 45% acetic acid. After removal of the cover slips, the slides were aged for 3 days, and then incubated in 45% acetic acid for 10 min at 60°C, 5% barium hydroxide for 10 min at room temperature and 2 × SSC, pH 7.0, for 10 min at 60°C. The samples were washed in distilled water, air dried and stained with the fluorochromes: 0.5 mg/mL CMA3 for 1.5 h and 2 μg/mL DAPI for 30 min. Slides were mounted with a medium composed of glycerol/McIlvaine buffer (pH 7.0) 1:1, plus 2.5 mM MgCl2.

FISH

The in situ hibridization was performed according Heslop-Harrison et al. (1991) and Cuadrado and Jouve (1994), with modifications. Slides were prepared as described for banding and immediately used for FISH. The pTa71 probe containing the 45S rDNA sequence (Gerlach and Bedbrook 1979) was labeled with biotin-14-dATP by nick translation. The pTa794 probe containing the 5S rDNA sequence (Gerlach and Dyer 1980) was labeled with digoxigenin-11-dUTP by nick translation. Each slide was treated with 34 μL of hybridization mixture containing 100 ng of labeled probe (2 μL of each probe), 100% formamide (15 μL), 50% polyethylene glycol (6 μL), 20 × SSC (3 μL), 100 ng of calf thymus DNA (1 μL) and 10% SDS (1 μL). The material was denatured at 90°C for 10 min and hybridization was performed overnight at 37°C in a humidified chamber. Post-hybridization washes were carried out in 2 × SSC, 20% formamide in 0.1 × SSC, 0.1 × SSC and 4 × SSC/0.2% Tween 20, all at 42°C. The probes were simultaneously detected with a solution composed of 5% BSA, avidin-FITC conjugated and anti-dig-rhodamine conjugated (100:1:1, v:v:v), and the post-detection baths were conducted in 4 × SSC/0.2% Tween 20 at room temperature. Slides were mounted with 25 μL of a medium composed of 23 μL of DABCO solution (1,4-diaza- bicyclo (2.2.2)-octane (2,3%), 20 mM Tris HCl, pH 8.0, (2%) and glycerol (90%), in distilled water), 1 μL of 2 μg/mL DAPI and 1 μL of 50 mM MgCl2.

All the images were acquired with a Leica DM 4500 B microscope equipped with a DFC 300FX camera and the Leica IM50 4.0 software. All the images were optimized for best constrast and brightness with iGrafx Image software.

Results

The conventional analysis showed that the species of Solanum, Lycianthes australe and Vassobia breviflora presented nonreticulated interphase nuclei, proximal condensation at prophase (Fig. 1e, q), 2n = 2x = 24, chromosomes measuring around 2 μm and predominance of meta- and submetacentrics (Figs. 1, 4; Table 1), with the exception of Solanum corymbiflorum (Sendtn.) Bohs which exhibited chromosomes with up to 6 μm (Figs. 1r, 4i; Table 1).

Fig. 1
figure 1

Conventional staining in a Solanum americanum, b S. gemellum, c S. atropurpureum, d prometaphase of S. viarum, e S. granuloso-leprosum, f S. palinacanthum, g S. pseudocapsicum, h prometaphase of S. scuticum, i prometaphase of S. variabile, j S. trachytrichium, k S. paniculatum, l S. mauritianum, m S. sisymbriifolium, n prometaphase of Vassobia breviflora, o S. guaraniticum, p S. amygdalifolium, q Lycianthes australe and r S. corymbiflorum. Arrows indicate satellites. Bar = 10 μm

Full size image

In the subgenera Leptostemonum Dunal (Bitter) and Solanum there was a predominance of metacentrics in relation to submetacentrics (Fig. 4; Table 1). However, the more frequent karyotype formulae was 6 m + 6 sm, observed in S. atropurpureum Schrank, S. sisymbriifolium Lam., S. trachytrichium Bitter, S. granuloso-leprosum Dunal, S. corymbiflorum and Lycianthes australe Barboza et Hunz. (Fig. 4c, d, l, m, i, q; Table 1). In spite of similar chromosome types in these species, the haploid set size was variable, from 13.08 to 27.9 μm, except to S. corymbiflorum that showed 77.88 μm (Table 1).

Karyotypes containing a larger number of metacentrics were observed in S. paniculatum L., S. pseudocapsicum L. and S. mauritianum Scop., with 7 m + 5 sm (Fig. 4e, k, n), S. guaraniticum St.-Hil., S. scuticum M. Nee and S. amygdalifolium Steud., with 8 m + 4 sm (Fig. 4f, g, p) and in Vassobia breviflora (Sendtn.) Hunz., with 9 m + 3 sm (Fig. 4r). In those species, the haploid set size varied from 13.08 to 25.87 μm (Table 1). Only three species were not grouped: Solanum viarum with 11 m + 1 sm and 18.14 μm of haploid set size, S. americanum Mill. with 10 m + 2 sm and 22.39 μm and S. variabile Mart. with only metacentrics and 20.50 μm (Fig. 4a, j, h, respectively; Table 1). Solanum palinacanthum Dunal and S. gemellum Sendtn. also stood out because they exhibited more submetacentrics (3 m + 9 sm, 5 m + 7 sm) and higher haploid set sizes with 27.90 and 25.74 μm (Fig. 4b, o, respectively; Table 1).

The chromosome banding showed four heterochromatin distribution types. In Vassobia breviflora (Figs. 2q, 4r) and in the subgenera Leptostemonum and Solanum (S. pseudocapsicum, S. sisymbriifolium, S. americanum, S. guaraniticum and S. variabile), terminal C-CMA3 + bands in the largest chromosomes and major blocks associated to satellites were found (Figs. 2e, f, i, l, m, 4k, d, j, f, g, h). In other species, as S. atropurpureum (Figs. 2a, 4c), S. viarum Dunal (Figs. 2b, 4a), S. mauritianum (Figs. 2c, 4n), S. trachytrichium (Figs. 2d, 4l), S. granuloso-leprosum (Figs. 2h, 4m), S. amygdalifolium (Figs. 2k, 4p), S. gemellum (Figs. 2p, 4o) and Lycianthes australe (Figs. 2r, 4q), only C-CMA3 + bands associated with satellites were observed. Solanum paniculatum and S. palinacanthum, besides of C-CMA3 + associated with satellites, exhibited proximal, terminal and interstitial dots (Figs. 2n, o, 4b, e). Solanum corymbiflorum, of the section Pachyphylla Dunal (subgenus Bassovia (Aubl.) Bitter), with the largest chromosomes, showed only two pairs with terminal C-CMA3 + bands (Figs. 2t, 4i). Except Lycianthes australe (Figs. 2s, 4q) which showed centromeric C-DAPI+ bands, the remaining species did not present this heterochromatin type.

Fig. 2
figure 2

Chromosome banding in species of Solanum, Vassobia breviflora and Lycianthes australe. C-CMA3 banding in Solanum artropurpureum (a), S. viarum (b), S. mauritianum (c), S. trachytrichium (d), S. pseudocapsicum (e), S. sisymbriifolium (f), S. granuloso-leprosum (h), S. americanum (i), S. amygdalifolium (k), S. guaraniticum (l), S. variabile (m), S. paniculatum (n), S. palinacanthum (o), S. gemellum (p), Vassobia breviflora (q), Lycianthes australe (r), S. corymbiflorum (t). C-DAPI banding in S. sisymbriifolium (g), S. pseudocapsicum (j), S. corymbiflorum (u), see negative bands (arrows), Lycianthes australe (s), with centromeric C-DAPI positive bands. Bar = 10 μm

Full size image

The double FISH with 45S and 5S rDNA probes always showed two terminal sites of 45S rDNA and two terminal or pericentromeric hybridization sites of 5S rDNA (Fig. 3). The exceptions were Vassobia breviflora (Figs. 3p, 4r) and S. corymbiflorum (Figs. 3r, 4i), which exhibited two pairs with terminal 45S rDNA sites and two signals of 5S rDNA. In all cases, the hybridization sites with 45S rDNA probes were coincident with satellites detected by C-CMA3 + banding (Fig. 4). However, the pair bearer of 45S rDNA varied among species. These sites were located in the pair ten in Solanum sections Acanthophora, Pachyphylla and Brevantherum, in the pair 11 at section Torva and Vassobia breviflora and in different pairs at the remaining groups (Fig. 4). The hybridization signals obtained with 5S rDNA probe were always located in the pair nine in Solanum viarum, S. atropurpureum, S. palinacanthum, S. scuticum, S. guaraniticum, S. americanum and S. pseudocapsicum. However, the pair nine was not constant in shape, being meta or submetacentric, and the 5S rDNA location was terminal or interstitial, independent of the chromosome shape (Figs. 3a–e, g, i, 4a–c, f–g, j–k). The remaining species showed hybridization signals in different chromosome pairs, e.g., Vassobia breviflora in the pair six (Figs. 3p, 4r), Solanum amygdalifolium and Lycianthes australe in the pair seven (Figs. 3j, q, 4p, q), S. paniculatum and S. corymbiflorum in the pair eight (Figs. 3f, r, 4e, i), S. sisymbriifolium and S. variabile in the pair ten (Figs. 3h, k, 4d, h), S. granuloso-leprosum and S. mauritianum in the pair 12 (Figs. 3n, o, 4m, n). On the other hand, Solanum trachytrichium and S. gemellum exhibited four 5S rDNA hybridization signals. In the first one, in the interstitial position of the pair five and in the terminal position of the pair eight (Figs. 3l, 4l). In Solanum gemellum, in the interstitial position of the pair eight and in the terminal position of the pair ten (Figs. 3m, 4o).

Fig. 3
figure 3

Double FISH with rDNA probes in species of Solanum, Vassobia breviflora and Lycianthes australe. Note two and four 45S rDNA sites (the largest signals) always located in terminal regions of the chromosomes. Arrows indicate always two 5S rDNA hybridization signals in almost species, excepted for S. trachytrichium (l) and S. gemellum (m). In S. gemellum only two sites of 5S rDNA are indicate. FISH in Solanum viarum (a), S. artropurpureum (b), S. palinacanthum (c), S. scuticum (d), S. guaraniticum (e), S. paniculatum (f), S. americanum (g), S. sisymbriifolium (h), S. pseudocapsicum (i), S. amygdalifolium (j), S. variabile (k), S. trachytrichium (l), S. gemellum (m), S. granuloso-leprosum (n), S. mauritianum (o), Vassobia breviflora (p), Lycianthes australe (q) and S. corymbiflorum (r). Bar = 10 μm

Full size image
Fig. 4
figure 4

Idiograms with the physical location of repetitive DNA segments in Solanum species, Lycianthes australe and Vassobia breviflora. The species of Solanum are organized according to Nee (1999). Subgenus Leptostemonum section Acanthophora: Solanum viarum (a), S. palinacanthum (b), S. artropurpureum (c). Section Melongena: S. sisymbriifolium (d). Section Torva: S. paniculatum (e), S. guaraniticum (f), S. scuticum (g), S. variabile (h). Subgenus Bassovia section Pachyphylla: S. corymbiflorum (i). Subgenus Solanum section Solanum: S. americanum (j). Section Geminata: S. pseudocapsicum (k), S. trachytrichium (l): Section Brevantherum: S. granuloso-leprosum (m), S. mauritianum (n). Section Extensum: S. gemellum (o). Section Dulcamara: S. amygdalifolium (p). Lycianthes australe (q) and Vassobia breviflora (r). Bar = 10 μm

Full size image

Discussion

The chromosome number 2n = 24 was found in all the studied species and is in accordance with those available at Index to Plant Chromosome Numbers (http://mobot.mobot.org/W3T/Search/ipcn.html), except to Solanum scuticum, S. gemellum, S. amygdalifolium and Lycianthes australe, which are new records. About 50% of the species of Solanaceae present chromosome numbers multiple of x = 12, which is considered the basic chromosome number (Hunziker 2001). This number is predominant in the subfamily Solanoideae, in the genera Triguera, Lycopersicon, Lycianthes and Vassobia (Hunziker 2001), as well as in Solanum (Olmstead et al. 1999; Hunziker 2001; Acosta et al. 2005). Only two species of the section Acanthophora showed 2n = 22, Solanum mammosum L. and S. platense Dieckman (Chiarini and Bernardello 2006). Numbers different from x = 12 have been reported in other groups, as in the subfamily Cestroideae (cf. Hunziker 2001).

The conventional analysis also showed invariably nonreticulated nuclei and proximal condensation at prophase. This is in accordance with Guerra (2000) that suggests that small chromosomes (<3 μm) are associated with these features. In relation to chromosome sizes, our results showed that except S. corymbiflorum, whose haploid set size was 77.8 μm, the remaining species presented small chromosomes with haploid set size varying from 13.08 to 27.9 μm, about three times smaller than in S. corymbiflorum. A wide variation in the haploid complement size of Solanum and Lycianthes (from 16.30 to 38.75 μm) has been also reported by Bernardello and Anderson (1990), Bernardello et al. (1994), Acosta et al. (2005) and Chiarini and Bernardello (2006). All the species studied here exhibited predominance of meta and submetacentric chromosomes, with the smallest being always submetacentric, except Solanum variabile and S. americanum that showed the smallest metacentric. The tendency of chromosome size reduction accompanied by a change in the centromere position was also observed in the idiograms reported by Bernardello and Anderson (1990), Pringle and Murray (1991), Bernardello et al. (1994), Acosta et al. (2005) and Chiarini and Bernardello (2006). However, our results showed that there is no correlation between karyotype features and groupings and section of Nee (1999).

Differences on the chromosome size have been useful to delimit some groups in Solanaceae. The section Pachyphylla of Solanum possess larger chromosomes in relation to other groups of the genus. Besides, some morphological features, such as anthers with thickened and glandular connectives, were also considered important to maintain this group as an independent genus, Cyphomandra (Bohs 1989, 1994; Hunziker 2001; Child and Lester 2001). However, phylogenetical analysis based in the sequence of conserved chloroplast genes, ndhF and rbcL, focusing Solanum and closely related genera, indicate that Cyphomandra can be included in Solanum (Bohs 1995; Olmstead and Palmer 1997; Bohs and Olmstead 1997, 1999; Olmstead et al. 1999; Bohs 2004, 2005). Solanum corymbiflorum, previously named Cyphomandra corymbiflora Sendtn. exhibits much bigger chromosomes and the largest haploid set size. The conventional cytogenetic analysis indicate that this species can be maintained in a genus separated of Solanum (Cyphomandra), as proposed by Child and Lester (2001), Moscone (1992) and Hunziker (2001). If Cyphomandra is included in Solanum, as pointed out by the above mentioned phylogenetic studies, therefore we will have to admit that the karyotypes of the section Pachyphylla become differentiated from the other species of Solanum by a substantial DNA accumulation, without important changes in the chromosome number and shape (see Fig. 4i).

The location of 45S ribosomal genes follows the common distribution pattern found in most plant species, always occurring in the terminal chromosome regions. This has been also reported in other genera of Solanaceae as Solanum bulbocastanum (Dong et al. 2000), Capsicum (Moscone et al. 1995), Nicotiana (Lim et al. 2000), and Cestrum (Fregonezi et al. 2006). In spite of the constancy of 45S rDNA location, the number of sites was variable, since in S. corymbiflorum and Vassobia breviflora four sites were observed. However, this is not an innovation in Solanaceae, as long as the number of hybridization sites in Cestrum using of the same probe also varied (Fregonezi et al. 2006). FISH with 5S rDNA probe showed that in almost all cases the signals appear in the short arm of the smallest chromosomes of the complement, except S. trachytrichium and S. scuticum where signals in the large arm were found. Solanum trachytrichium and S. gemellum exhibited four hybridization sites and the remaining species two signals of 5S rDNA. In all cases, the signals appeared in the smaller chromosomes of the set. On the other hand, Dong et al. (2000) showed paracentromeric 5S rDNA sites in the major chromosome of the set in S. bulbocastanum. In other genera, as Nicotiana (Lim et al. 2000; Kitamura et al. 2001), Lycopersicon (Lapitan et al. 1991) and Cestrum (Fregonezi et al. 2006) the 5S rDNA were also located in the paracentromeric region. These data suggest that the 45S and 5S rDNA segments do not present a conserved location and number of sites among species of Solanum and other Solanaceae. This feature seems to be common to other plant groups, e.g., Sapindaceae (Urdampilleta et al. 2006).

The C-banding followed by CMA3/DAPI staining, as described by Vanzela and Guerra (2000), was useful to determinate different heterochromatin type distribution (see also Fregonezi et al. 2006). The different bands can be grouped in: (1) terminal C-CMA3 + bands in all the chromosomes, besides a block associated to satellites, (2) only C-CMA3 + bands associated to satellites, (3) proximal, terminal and interstitial C-CMA3 + dots, besides those C-CMA3 + associated to satellites, and (4) centromeric C-DAPI+ bands. The presence of C-CMA3 + terminal dots can approximate species of distinct groups, as S. sisymbriifolium, S. paniculatum, S. guaraniticum, S. scuticum, S. variabile (sections Melongena and Torva), S. americanum (section Solanum), S. pseudocapsicum (section Geminata) and Vassobia breviflora. The occurrence of interstitial dots in S. palinacanthum indicates that the section Acanthophora is apparently more diverse. Fregonezi et al. (2006) studied species of Cestrum and detected different repetitive DNA families (dots), occupying different equilocal/equidistant regions, as well as observed in Solanum and Vassobia studied here. A similar behavior of bands distribution was reported to Solanum section Petota (Pijnaker and Ferwerda 1984), Capsicum (Moscone et al. 1993) and Lycopersicon (Peterson et al. 1996).

To conclude, the results of chromosome banding and FISH with rDNA probes show that they can be useful to characterize species and possibly populations and that the cytogenetical tools can complement the phylogenetical studies. We considered here the karyotypes of Vassobia breviflora and Lycianthes australe as external members for comparison with Solanum. Vassobia breviflora presents anthers with longitudinal dehiscence and ovary with a nectary, and possess karyotype features similar to species of subgenera Melongena, Torva and Solanum. Lycianthes and Solanum are considered separate genera, in spite of possessing similar morphologic features as poricidal anthers and ovary without a nectary (Hunziker 2001). They also share similar cytogenetic features, as 2n = 24, nonreticulated nuclei, proximal condensation at prophase, small chromosomes, similar karyotype structure and occurrence of two terminal 45S rDNA sites. If the same comparisons are made between Cyphomandra and Solanum, where there are more morphological differences and cytogenetical contrasts that in Solanum and Lycianthes (in spite of the occurrence of proximal C-DAPI bands in Lycianthes), therefore Cyphomandra should be also maintained separated of Solanum, as Lycianthes is. In spite of the great importance and contributions of the molecular phylogeny studies, we have doubts about the sustentation of the recent arrangement in Solanum, which is based on only a few chloroplast genes. We consider that molecular data should be applied along with other tools, including morphological, anatomical, chemical and cytogenetical information.