For SSCP analysis, PCR was performed using the 5S rDNA specific primers. PCR product (5µl) was denatured by adding the loading dye consisting of 95% formamide, 10% glycerol, 0.25% bromophenol blue, 0.25% xylene cyanol) followed by its immediate quenching on ice. The samples were electrophoresed on 6% polyacrylamide gel in 0.5X TBE. For staining polyacrylamide gels, silver staining protocol of Bassam et al. (1991) was followed.

SSAP analysis was performed as per the protocol of Pearce et al. (1999). The primers selected for the analysis were MKD-4, MKD-5, MKD-9, MDK-11, ASW-8, ASW-9 and ASW-10 (Kour et al. 2009). A total of twelve different combinations were made using these primers (Suppl. material 1: Table 1). The PCR amplified fragments were electrophoresed on 6% polyacrylamide denaturing gel in 0.5X TBE followed by silver staining. Only reproducible bands were considered for scoring.

For SSR amplification four primer pairs were used as given by Squirrell and Wolff (2001) for Plantago major, and ten other primers were tried (Suppl. material 1: Table 2). PCR amplification of SSR loci was performed in 20µl reaction mixture containing 20ng DNA, 1X PCR buffer, 2.5mM MgCl₂, 0.5µmol each primer, 200µM dNTPs, 1U Taq polymerase (Fermentas, USA). The thermal cycling conditions were as follows: initial denaturation at 94°C for 4min followed by 35 cycles of 94°C for 30sec, 55-65°C for 30sec, extension at 72°C for 1min and final extension for 10min at 72°C. The amplification products were resolved on 10% polyacrylamide denaturing gel in 0.5X TBE and subsequently visualized by silver staining. For each microsatellite locus, size of the alleles was estimated by comparison with standard size. After scoring the allelic bands, other bands were also scored.

The transmission of B chromosomes was ascertained through male and female sex tracks by attempting various reciprocal crosses. When 1B plant was used as a male, transmission rate was in accordance with the Mendelian expected ratio of 1:1 (p=0.05; χ²= 0.058) (Table 1). In the reciprocal crosses, when 1B plant was used as female parent, progeny based on a total of 187 plants was screened for chromosome number (Table 2); the ratio between the plants with and without B chromosome(s) was 3.1:1. Thus, it is clear that frequency of B chromosome bearing plants was higher than the Mendelian expected value (p < 0.05, χ²=50.30) when 1B chromosome plant was used as female parent. The deviation obtained from the Mendelian ratio was significant. These results clearly indicate that the B chromosome has preferential transmission through the female sex track.

To understand the meiotic behaviour of B chromosome in 1B plants, a total of 1450 pollen mother cells from 145 plants were scanned at anaphase I and II. At metaphase-I the B chromosome remained as a univalent and at anaphase-I, in 1160 (80%) cells the B chromosome was present at one of the poles. In majority of the cases, the B chromosome seemed to have reached the pole earlier than the A chromosomes. At anaphase-II, the B chromosome divided into chromatids, which segregated to the poles. Similarly, in 2B plant at metaphase-I of meiosis, B chromosomes existed either as a bivalent (Fig. 2c) or two univalents (Fig. 2d). During anaphase-I the two chromosomes moved to the opposite or to the same pole (Fig. 2e, f). In 2B plants, 670 pollen mother cells were scanned, of which 490 cells showed B chromosomes at the poles earlier than the A chromosomes.

Figure 2.

FISH on cells having 2 B chromosomes. Root tip metaphase cells at prometaphase (a) and metaphase (b) showing two B chromosomes. Segregation pattern of B chromosomes during meiosis (c–f). Note the two B chromosomes at the same pole (f). Scale bar: 10μm.

In some progenies of 1B plants, rarely, plants with very high number of chromosomes were observed, which included plants having 23, 26 and 28 chromosomes (Fig. 1g, h). FISH analysis of 23-chromosome plant with 5S rDNA probe revealed its chromosome constitution as 18A + 5B chromosomes. Similarly, plants with 26 and 28 chromosomes had 24A + 2B chromosomes and 24A + 4B chromosomes, respectively (figures not given). In other words, the B chromosomes were found in addition to triploid and tetraploid states.

In order to trace the origin of plants with 2n = 26 and 2n = 28 chromosomes, three molecular markers were tried. The details are presented below:

As shown in our earlier study (Dhar et al. 2002) and the present study, B chromosome is mainly composed of 5S rDNA sequences, therefore, SSCP analysis was carried out by targeting these sequences. For this purpose, DNA isolated from the selfed progeny plants of 1B and 2B parents (separately), including plants with 26 and 28 chromosomes, respectively, was used. The banding pattern of 1B (mother plant) and 26 chromosome progeny plant, 2B (mother plant) and 28 chromosome progeny plant showed 100% similarity. The remaining plants in the two progenies composed of 0B, 1B and 2B plants showed variable banding pattern (Suppl. material 3: Fig. S2).

The SSAP exploits the insertional polymorphism of long terminal repeats (LTR retrotransposons) in the genome. In the present case SSAP analysis was used with twelve different primer combinations. It was observed that plant with 26 chromosomes showed 100% similarity with the 1B mother plant while other progeny plants showed polymorphism. Similarly, progeny plant with 28 chromosomes recovered from the selfed progeny of 2B plant showed 100% similarity with its mother plant as compared to other progeny plants.

For the present investigation we used 14 SSR primers. From SSR data generated demonstrated that 1B and 2B mother plants showed 100% similarity with the 26 (Suppl. material 4: Fig. S3a, b) and 28 (Suppl. material 4: Fig. S3c, d) progeny plants respectively, while differences were detected in other progeny plants.

The results obtained from the above three molecular markers clearly establish the maternal origin of 26 and 28 chromosome plants.

In Plantago lagopus Dhar et al. (2002) demonstrated that the B chromosome is mainly composed of 5S rDNA sequences. Using FISH and reverse GISH techniques, the entire chromosome was found to get painted with 5S rDNA probe, while 45S rDNA sequences were localized at the two ends, just below the telomeric sequences. There are several reports on identification of B chromosomes using FISH (reviewed in Jones et al. 2008a). Most of the B chromosomes reported in different organisms are heterochromatic mainly due to the presence of repetitive DNA sequences made of satellite DNA, ribosomal DNA and transposable elements (Martis et al. 2012). Since 0B, 1B and 2B plants are indistinguishable morphologically (Dhar et al. 2002), therefore, in order to confirm the chromosomal status of the parents and the progeny plants, FISH with 5S rDNA probe was used in the present investigation.

The B chromosome does not pair or recombine with any A chromosome. B chromosomes, in general, have been reported to follow non-Mendelian mode of inheritance (Jones and Rees 1982) attributed to their irregular mitotic and meiotic behavior (Jones 1991). The meiotic behavior of the B chromosome in the pollen mother cells was very interesting. During anaphase, the B chromosome generally reached the poles earlier than A chromosomes. Thus, the B chromosome showed meiotic drive, which ensures its segregation and subsequent inclusion in the microspore mother cells.

In the present case, transmission of the B chromosome was ascertained through male and female sex tracks by following the progenies of the reciprocal crosses. Interestingly, when 1B plant was used as a male, transmission rate was in accordance with the expected Mendelian ratio. The differences in segregation ratio observed among various cross combinations can be attributed to the heterozygous nature of P. lagopus - being a cross-pollinated species. On the other hand, when 1B plant was used as a female, there was significant deviation from 1:1 ratio; frequency of B chromosome bearing plants was higher than the Mendelian expected value. These results clearly indicate preferential transmission of B chromosome through the female sex track. According to Houben and Carchilan (2012), variation in transmission rate is a common feature of B inheritance, such that the Bs tend to distort Mendelian expectation in their favor. Their rate of transmission can be irregular with different levels of meiotic and post-meiotic drive or drag (Beukeboom 1994, Rusche et al. 1997). The drive through the female sex track can be explained on the basis of the fact that during female meiosis in plants, only one out of the four meiotic products survives. Therefore, in order to ensure inclusion in the surviving megaspore/egg, the B chromosome must be hosting a meiotic drive locus, as has been shown in Mimulus Linnaeus, 1753 - a phenomenon called female meiotic drive (Fishman and Saunders 2008). These authors have demonstrated that selfish chromosomal drive can be an important fitness determinant in natural populations. It can therefore be concluded that the B chromosome of P. lagopus is perpetuated preferentially due to drive, expressed during both male and female meiosis. The same is true of rye, where the drive happens in both male and female sex tracks (Jones and Houben 2003).

One of the important characteristics of B chromosomes is their accumulation in selfed or outcrossed progenies. In some species the accumulation is mainly due to non-disjunction of B chromosomes during pollen mitosis. The present case is perhaps the first in plants where the entire complement of the species gets duplicated in presence of a B chromosome. Earlier, Bidau (1987) reported the presence of macrospermatids (>diploid chromosome number) in B containing individuals of a grasshopper, Dichroplus pratensis Bruner, 1900. The formation of macrospermatids was attributed to nuclear fusion. In the present case, the plants with 23, 26 and 28 chromosomes were obtained from the selfed progenies of B chromosome bearing plants. The origin of these plants can be attributed to formation of unreduced gametes, followed by their fusion with other gametes or their endoreduplication and parthenogenetic development of the plant. Earlier, it has been postulated by Sharma et al. (1985) that in P. lagopus unreduced egg of the trisomic mother plant may have developed parthenogenetically giving rise to plants with 13 chromosomes. The authors reported formation of aneutriploid individuals with 19 chromosomes (Sharma et al. 1985) in the progeny of a cross between a trisomic and disomic plants. Similarly, Bhan et al. (1990) reported an autotetraploid in P. lagopus isolated from an experimental population, which could be the result of fusion of unreduced gametes.

Formation of unreduced gametes has been reported in many plants and has been proposed as an important mechanism for origin of polyploids (Mason et al. 2011, Mason and Pires 2015).

To substantiate the proposed mode of origin of plants with 26 and 28 chromosomes, recovered from selfed progeny of B chromosome plants, recourse was taken to molecular markers namely SSCP, SSAP and SSR. For SSCP analysis we targeted those sequences, which are present on the B chromosome. The technique was used to check whether Single Nucleotide Polymorphism exists among 1B and 2B mother plants and their progeny plants, including 26 and 28 chromosome plants. The SSCP pattern of 1B (mother) and 26 (progeny) plant showed 100% similarity, as compared to rest of the progeny plants. Similarly 2B (mother) and 28 chromosome progeny plants showed monomorphic pattern of bands, while in other 14 chromosome plants polymorphic pattern was observed. Thus SSCP analysis of 5S rDNA showed that the mother plants and the higher chromosome plants are 100% similar, thereby suggesting that they must have originated from the maternal genome. The experiments were repeated 3-4 times, however, similar SSCP pattern was observed in the mother plant and the offsprings with 26 and 28 chromosomes. SSCP technique is known to detect variation due to SNPs. Recombination is known to affect the SNPs; the SNP variation is less in regions of low recombination while it is more in high recombination regions (Brumfield et al. 2003). Recombination frequencies vary due to several genetic and non-genetic factors such as sex, the genetic background, genes and structures involved in meiotic recombination, age, irradiation, chemicals, nutrient salts and antibiotics (Barth et al. 2000). Occurrence of such processes more frequently during sexual reproduction in comparison to asexual reproduction, will lead to generation of SNPs in the former. This gets exemplified in the present case, by detection of large number of SNPs among progeny plants bearing 1B and 2B chromosomes than in 26 or 28 chromosome plants.

SSAP has also been used for the recombination studies in the selfed progeny plants of Pisum Linnaeus, 1753 (Jing et al. 2007). The marker has been used to detect the variation at genic level due to recombination and the combination of male and female genome mutation. Similarly, the inheritance of the B chromosome at the genome level has been analyzed using the SSR markers in Brassica sp. Linnaeus, 1753 (Navabi et al. 2011). SSAP and SSR used in the present case have also shown the same pattern in case of mother and the progeny plant having higher chromosome number. The molecular data does not support occurrence of recombination events in the origin of 26 and 28 chromosome plants, therefore, it can be presumed that the latter have originated through a two step process; formation of unreduced gametes in the parent, followed by parthenogenetic development of unreduced gametes, as has been proposed by Sharma et al. (1985).

Apomixis, or clonal propagation by seed, has been reported in many genera of higher plants following the gametophytic apomixis (Carman 1997). Plants arising from apomixis retain the maternal genotype. The main components of apomixis include unreduced gamete formation and parthenogenetic development (Koltunow and Grossniklaus 2003), which are also exemplified by the plants studied in present investigation. The genes and pathways involved in gametophytic apomixis have not been discovered as yet and are the subject of intense research (Hand and Koltunow 2014). Polyploidy is the common feature of almost all apomicts (Roche et al. 2001). B chromosomes have been documented in apomictic species (Jones and Rees 1982), however, there is no report on apomixis gene(s) necessarily residing on a B-chromosome (Roche et al. 2001). In an animal species, B-chromosomes were found in polyploid individuals reproducing by pseudogamous parthenogenesis but were conspicuously absent in the diploid sexual individuals (Beukeboom et al. 1998).

Preferential transmission of B chromosomes and occurrence of plants with high chromosome numbers (2n =23, 26, 28) can have serious implications in the evolution of Plantago lagopus genome and the speciation. Recently, in Arabidopsis thaliana (Linneaus) Heynh., 1842, unreduced gamete-producing mutants, on account of defects in the meiotic cell cycle machinery, have been identified which has further advanced our understanding of the mechanisms behind unreduced gamete formation (Brownfield and Kohler, 2011). In the present case, plants with 23, 26 and 28 chromosomes were isolated in the progenies of the plants carrying B chromosome(s), therefore, it can be presumed that some DNA element located on B chromosome is activating the gene(s) promoting unreduced gamete formation, which is (are) located on A chromosome. Similar observations have been made in rye (Carchilan et al. 2009). However, more intense molecular studies need to be conducted to identify the gene/genes responsible in P. lagopus.