Mecardonia Ruiz & Pavon, 1798 belongs to the tribe Gratiolae (Plantaginaceae) and is distributed across the America, reaching its southernmost distribution in Argentina. The species are erect or creeping herbs, annual or perennial, much branched, mostly glabrous, sometimes blackening on drying, gland dotted, and yellow and white flowers (D’Arcy 1979). Since the description of the genus (Ruiz and Pavon 1794), there have been a few problems in establishing its generic and infrageneric circumscription. Rossow (1987) considered only 10 species, which had been previoulsy placed in Bacopa Aublet, 1775 under subgenus Mecardonia (Descole and Borsini 1954). More recently, Souza (1997) and Souza and Giuletti (2009) carried out some taxonomic modifications to Rossow’s clasiffication. The demarcation of the genus is variable depending on the author consulted. Following Rossow’s classification, the genus includes five species growing in Argentina: M. flagellaris (Chamisso & Schlechtendal, 1827), M. grandiflora (Bentham) Pennell, 1946, M. procumbens Small, 1903, M. serpylloides (Chamisso & Schlechtendal, 1891) and M. tenella (Chamisso & Schlechtendal, 1891). Recently, M. kamogawae Greppi & Hagiwara, 2011 was described by as an endemic species of Corrientes Province (Argentina).

Mecardonia has ornamental value because some cultivars developed from native species from Northern Argentina were recently introduced in the trade of ornamental plants (Greppi 2012). Therefore, researches on genetic improvement are carried out in this genus.

Cytological and cytogenetic studies have proved useful data for taxonomic and evolutionary analyses, which are widely used in processes of conventional or biotechnological genetic improvement (Poggio et al. 2004). Characters such as chromosome number, morphology, and meiotic behavior, as well as nuclear DNA content, have been used as taxonomic markers helping to circumscribe taxa and infer their relationships (Kron et al. 2007, Guerra 2008, Loureiro et al. 2010, Castro et al. 2012). At present, only two species of Mecardonia have been evaluated cytologically. Lewis et al. (1962) reported 2n = 42±2 for a Northamerican species M. acuminata (Walter, 1891) Small, 1903. Kaul (1969) determined 2n = 2x = 22 for M. procumbens Small, 1903 (as M. dianthera (Swartz 1900, Pennell 1946). Therefore, to increase the knowledge of Mecardonia, other species were cytologically analyzed in this study.

Nuclear DNA content, understood as genome size, is very variable across angiosperm, and has been revealed as an important character in biodiversity. In Mecardonia species there are no reported measurements of DNA content, but genome size variation has been explored in some genera of Plantaginaceae. DNA C-values are currently available for 204 species belonging to 18 genera of this family, including Callitriche Linnaeus, 1753, Penstemon Schmidel, 1762, Plantago Linnaeus, 1753 and Veronica Linnaeus, 1753 with a range of variation of 0.32–4,.63 pg (Albach and Greilhuber 2004, Broderick et al. 2011, Bennett and Leitch 2012, Wong and Murray 2012, Prančl et al. 2014, Meudt et al. 2015). Herein we used flow cytometry to estimate the nuclear DNA content in five species of Mecardonia for the first time. Additionally, we report original chromosome numbers of some of them. The results are discussed in relation to the taxonomy and evolution of the genus.

We examined six populations from five species of Mecardonia collected in Argentina. Information about the studied material and the voucher specimens is provided in Table 1. Vouchers are deposited at the herbarium BAB of the Instituto Nacional de Tecnología Agropecuaria (INTA), Buenos Aires, Argentina.

The chromosome numbers of six populations belonging to five species of Mecardonia were determined. The analyzed species and their chromosome numbers are given in Table 1. Four species were analyzed for the first time. The chromosome number observed in the remaining taxon is in agreement with previous studies. All species analysed shared the same base chromosome number (x = 11), and chromosome numbers ranged between 2n = 22 to 2n = 66. Of these, only three species were diploids: M. flagellaris (Fig. 1 A), M. grandiflora (Fig. 1 B) and M. procumbens. The remaining species were polyploids, Mecardonia sp. (Fig. 1 C) was tetraploid with 2n = 44 and M. kamogawae was hexaploid with 2n = 66 (Fig. 1 D).

The chromosome number 2n = 22 found in M. procumbens, is consistent with the chromosome counts recorded in a previous cytological study (Kaul 1969). Sinha (1987) reported B chromosomes for this taxon; however, the populations here analyzed did not show these accessory chromosomes. Chromosome counts for M. flagellaris (2n = 22), M. grandiflora (2n = 22), M. kamogawae (2n = 66) and Mecardonia sp. (2n = 44) are described here for the first time. Our results showed diploid and polyploid species for the genus. Polyploidization has long been recognized as an important process in plant evolution (Otto and Whitton 2000, Soltis et al. 2004). In Plantaginaceae, polyploidy is a common phenomenon occuring in many genera, such as Antirrhinum Linnaeus, 1753, Chaenorhinum (DC.) Reichenbach, 1828, Cymbalaria Hill, 1756, Chelone Linnaeus, 1753, Digitalis Linnaeus, 1753, Linaria Miller, 1754, Plantago, Nuttallanthus D.A. Sutton, 1988, Stemodia Linnaeus, 1759, Veronica (Hair 1966, Subramanian and Pondmudi 1987, Sosa and Seijo 2002, Sosa et al. 2009, 2011, Wolfe et al. 2002, Vargas et al. 2004, Murray et al. 2010, Castro et al. 2012, Wong and Murray 2012, Ranjbar and Nouri 2015). Our results evidenced the presence of multiple cytotypes in Mecardonia, hence suggesting polyploidy as a key driver of the evolution of the genus. The present analysis, in addition to previous chromosome number reports, revealed that the genus have exclusively the basic chromosome number of x = 11.

The interest on the study of genome size increased in the last decade. These studies focused on the use of genome size as a taxonomic marker (Castro et al. 2012, Angulo and Dematteis 2013, Galdeano et al. 2016) and on finding correlations between ecological and environmental variables and this character (Chalup et al. 2014, Vega and Dematteis 2015). However, there are still many families and genera being neglected, including Mecardonia, for which the present study is the first analysis of genome size for the genus. The estimates 2C- and 1C-values calculated for the species in this study are within ranges of variation found in Angiosperms and Plantaginaceae (Leitch and Bennett 2004, Meudt et al. 2015). Based on the available genome size data, Mecardonia falls into the categories “very small” genomes (2C = <2.8 pg) to “small” genomes (2C = <7.0 pg) according to values reported by Leitch et al. (1998) and Soltis et al. (2004).

The 2C-values of Mecardonia species revealed a positive and significant correlation with chromosome number (r = 0.99, P = < 0.0001). Therefore, in the genus there is a trend for increasing 2C–value with increasing ploidy level. On the other hand, the variation of 1Cx-values is negative and significantly (r = -0.86; P = <0.0001) correlated with chromosome number. Consequently, the values of 1Cx of the species decrease in inverse proportional to the ploidy level. Our data reflect that both polyploids (tetraploid 1Cx = 0.92 pg and hexaploid 1Cx = 0.88 pg) have lesser values of monoploid genome size than diploid species (mean of Cx = 1.00 pg). Many polyploid angiosperms undergo genome downsizing and so have smaller average genome sizes than their diploid relatives (Leitch and Bennett 2004, Leitch et al. 2008) and Mecardonia seems not to be an exception. Several studies have indicated that during polyploidization different balancing processes at genomic level occur which may promote variation in nuclear DNA content. These changes point towards a possible need for harmonization of genome and removal of some unnecessary genomic redundancies (Petrov 2001; Bennetzen et al. 2005, Pellicer et al. 2010). Mechanisms leading to a decrease in genome size in polyploids may include non-random elimination of chromosome- and genome-specific sequences (Ozkan et al. 2003, Shaked et al. 2001), illegitimate crossing over (Devos et al. 2002) or unbalanced deletion–insertion rates (Petrov 2001, 2002). Counterbalancing mechanisms are probably also involved to reduce the genetic and structural instabilities that accompany DNA loss (Pellicer et al. 2010).

Recently, Meudt et al. (2015) established a relationship between the genome downsizing with diversification in polyploid lineages of Veronica (Plantaginaceae), but they do not know how general this pattern might be or what causes it. Several hypotheses have been proposed to explain this relationship. Kraaijeveld (2010) suggested that organisms with small genomes have more stably inherited mutations, or a nucleotypic effect, in which organisms with small genomes and shorter genes have a general advantage as a result of faster replication and transcription. The genome size changes in Mecardonia are probably related with the diversification of the species. Further studies comparing this genus with the closest extant relative to determine what aspect of genome downsizing facilitate diversification are needed.

The genus Mecardonia is currently under revision and some closely related species with intermediate morphological characteristics were found. It has been well documented in many plants that chromosome numbers and genome size can be used as extra taxonomic characters for discriminating between closely related taxa, helping to clarify the taxonomy of some species in problematic genus (Guerra 2008, Castro et al. 2012, Sosa and Dematteis 2014). For example, Mercardonia sp. is closely related to M. flagellaris. A detailed morphological analysis along with chromosome number here reported showed that it should be considered as different species. Mecardonia sp. is tetraploid with 2n = 44, while M. flagellaris is diploid with 2n = 22. Thus, both species differ in chromosome number and morphological features, such as aspect of plant, leaf shape, and trichome types of corolla. In addition, the new species has more restricted distribution to North of Argentina. However, M. flagellaris is expanding from Mato Grosso do Sul (Brazil) to Chubut (Argentina), arrived to Chile, Paraguay and Uruguay.

Another case is M. kamogawae that is morphologically related to M. procumbens from which it differs in the life-form, root types, leaf texture, and size of bracteoles and pedicels. Regarding chromosome number, M. kamogawae is hexaploid with 2n = 66, while M. procumbens is diploid with 2n = 22. Therefore, both species can be distinguished by morphological features, as well as by the chromosome number.

Mecardonia procumbens and M. flagellaris were diploids with 2n = 22. Although the chromosome number does not distinguish both species, differences in 2C-values were observed. Mecardonia flagellaris had higher value (2C = 2.06 pg.) than Mecardonia procumbens (2C = 1.91 pg.). D’Arcy (1979) and Souza (1997) placed M. flagellaris under the synonymy of M. procumbens by having similar morphological characteristics. We considered, however, that both species are morphologically distinct. Mecardonia procumbens differs by aspect of the plant, shape and length of leaf, and calyx pieces. Also, M. procumbens has a wider distribution as it extends from the South of the United States of America to the Argentine Northwest. However, M. flagellaris grow in Paraguay, South of Brasil, Uruguay and Northeast of Argentina.

The results of this study suggest that chromosome number is useful in distinguishing species of Mecardonia. The different ploidy levels of the taxa showed the importance of polyploidy in the evolution of the genus. The results here obtained combined with those reported previously confirm that the Mecardonia genus has basic number x = 11.

Regarding to the variation of genome size, decreases in DNA content have occurred during the evolution of genome size in the Mecardonia species.

Our results showed that differences in morphological features along with chromosome numbers and DNA content values support Rossow’s criterion (1987).