Genome size , morphological and palynological variations , and heterostyly in some species of the genus Linum L . ( Linaceae ) in Iran

Heterostyly is the occurrence of flowers with different sexual organ arrangements in different plants of the same species. This floral polymorphism occurs in four sections of genus Linum. The present study compares the morphological, palynologycal and genome size (C-value content) characteristics in the long-styled and short-styled plants in three Linum species, that is, Linum austriacum L., Linum album Ky.ex Boiss. and L. glaucum Boiss. & Nöe . 15 qualitative and quantitative morphological characters from both vegetative and reproductive organs of these plants were studied. A higher mean value of the plant height, size of the basal leaves width, flower leaves width, calyx length, sepal length and petal length occurred in the long-styled plants, while the mean value of branch number, basal leaves length, flower leaves length, calyx width, pedicel length and sepal length was higher in the short-styled plant populations. T-test analysis of morphological characters showed significant difference (p<0.05) for some of the characters studied. Principal coordinate analysis (PCoA) plot of long-styled and shortstyled plant populations based on all morphological characters also separated these two kinds of plants in the three species studied. The pollen obtained from the mature buds was prepared for light microscopy (LM) and scanning electron microscopy (SEM). The polar and equatorial views of the pollen grains were similar in the long-styled (Ls) and short-styled (Ss) plants but the aperture shape differed in these populations. C-values obtained by flow cytometry, differed in the long-styled and shortstyled plants of the species studied. The analysis of variance (ANOVA) test performed among the three Linum species showed a significant difference in 2C-value content. Positive significant correlation was observed between 2C-value and northern distribution of the Linum species studied, while a negative significant correlation occurred with eastern distribution.


INTRODUCTION
The occurrence of flowers with different sexual organ arrangements in different plants of a single species is called heterostyly (Darwin, 1888).It is a sexual polymorphism in which populations are composed of two (distyly) or three (tristyly) floral morphs with reciprocal arrangements of anthers and stigmas (reciprocal herkogamy) (Ganders, 1979).Plants in distylous species produce either all long-styled (LS or Pin) or all shortstyled (SS or Thrum) flowers.Flowers with the LS morphology have stigma(s) positioned above anthers, whereas flowers with the SS morphology have anthers above stigma(s).Heterostyly has been documented in 28 angiosperm families (Barrett et al., 2000).
Family Linaceae is geographically widespread with about 300 species worldwide distributed (Hickey, 1988).Several species are shrubs and occur in tropical areas, while perennial and annual species are found in temperate areas.About 22 species, subspecies or varieties are reported from Iran (Sharifnia and Assadi, 2001) which are classified into five sections (Rechinger, 1974).Due to wide range of diversity within the genus Linum, it has received considerable attention from botanists (Diederichsen and Richards, 2003).Distyly is widespread and very common in Linum and about 40% of these species are distylous occurring in four out of five sections of the genus, namely Linum, Syllinum, Dasylinum and Linastrum (Rogers, 1979;Sharifnia and Assadi, 2001).
Various differences have been reported in the longstyled and short-styled plants in different species, for example, Armbruster et al. (2006) found variation in distyly of Linum suffructicosum L., with bent styles and stamens, achieving a three-dimensional arrangement.In heterostylous species, some morphological and micromorphological characters were different in long-styled and short-styled plants including the number and size of pollen grains, stamens shape, shape and color of stigma and its surface papillae (Richards and Barrett, 1992).In some heterostylous species of Linum such as; Linum perenne, Linum grandiflorum and Linum alpinum exine sculpturing structure differed in the long-styled and shortstyled plants (Dulberger, 1981).
C-value (genome size = size of the monoploid chromosome set) data is considered as useful characteristics for infrageneric classification, species delimitation or hybrid identification (Keller et al., 1996;Buitendijk et al., 1997;Bare et al., 1998Bare et al., , 2004)).Moreover, correlation between DNA content and plant life-histories, plant phenology, environmental factors, climatic variation and geographical plant distribution has been determined in various studies performed on plant species (Bennett, 1976;Poggio et al., 1998;Baranyi and Greilhuber, 1995;Bureš et al., 2004).Evans (1968) found that varieties of Linum usitatissimum which grew in different ecological condition, such as; high nitrogen concentrations and high temperatures increased 10% in DNA content.In the present study, difference in morphological and palynological characters as well as genome size (C-value content) was investigated in the long-styled and short-styled plants of three Linum species namely Linum austriacum L., Linum album Ky.ex Boiss, and Linum glaucum Boiss., for the first time.

Plant samples
Three heterostylous species including L. austriacum, L. album and Talebi et al. 16041 L. glaucum Boiss.& Nöe were studied.Plant specimens were collected from both short-styled and long-styled population of these species during the growing season in spring 2010 and 2011 (Table 1).From each species three populations were studied and in each population four plant specimens were used for detailed investigation.

Morphology
15 qualitative and quantitative morphological characters from both vegetative and reproductive organs of these plants were studied.Morphological characters used include: the length and diameter of stems, number and status of stem branches, size, shape and diameter of the basal leaves and inflorescence leaves, dimensions of calyx and corolla and pedicle length.T-test analysis was performed to show morphological differences between long-styled and short-styled plants, while unweighted paired group using average method (UPGMA) and neighbor joining (NJ) trees as well as ordination plot based on principal coordinate analyses (PCoA) were used for grouping the species and also long-styled and shortstyled plants in each species.
For multivariate analyses the mean of quantitative characters were used, while qualitative characters were coded as binary/ multistate characters.Standardized variables (mean = 0, variance = 1) were used for statistical analyses.The average taxonomic distance and Manhattan distance were used as dissimilarity coefficient in cluster analysis of morphological data (Podani, 2000).UPGMA and NJ trees as well as ordination plot based on PCoA were used for grouping the species and also long-styled and shortstyled plants in each species.

Palynology
The pollen was obtained from the mature buds and used for light microscopy (LM) and scanning electron microscopy (SEM) by the prolonged acetolysis method of Erdtman (1960).For LM, the pollen was mounted in glycerin jelly and sealed with paraffin.The polar (P), equatorial (E) and colpus lengths, and sizes of pollen grains from six populations in three species were measured under the light microscope and P/E ratios were calculated.For SEM, the pollen grains were transferred directly to double-sided tape affixed stubs and were vacuum-coated with gold in Biorad E5200 auto sputter coater and photographed with a Camscan MV2300 scanning electron microscope.The terminology in this paper is based on Moore et al. (1991) work.T-test analysis was performed to show pollen characteristic difference between long-styled and shortstyled plants.

Flow cytometery
Three species of the genus Linum were analyzed with flow cytometer.For each taxon, three to five populations were collected and their 2C-value DNA content was determined.The nuclei suspensions were prepared from small amount of mature fresh leaf tissue together with an equal weight of mature leaf tissue of the external standard.The external standard used in for L. austriacum was Parsley (Petroselinum crispum cv.Champion Moss Curled) which had a 2C DNA value of 4.46 pg (Yokoya et al., 2000), for L. album, Rosa wichurana Cre´p was used with 2C value of 1.13 pg (Yokoya et al., 2000) and for L. glaucum Boiss.& Nöe was Allium cepa which had a 2C DNA value of 33.5 pg (Greilhuber and Ebert, 1994).One-step protocol was used for preparation of the nuclear suspension.The leaves were chopped with a single-used sharp scalpel with adding 400 μl nuclei isolation buffer in the plastic Petri dish at room temperature.To staining the nuclei, 1600 μl DNA fluorochorome or 2, 6-diamidino-2-phenylindole (DAPI) was added and suspensions was filtered through a 50 μm nylon mesh into a labeled sample tube.Stained nuclei suspensions were analyzed with a Partec flow cytometer (Partec Germany).The flow cytometric statistics such as; coefficient of variation (CV), mode, mean and the no. of cells counted in each sample, are showed in the histograms obtained.
DNA amounts were measured in picograms (pg) and the status of nuclei described in terms of 'C' values (Doležel et al., 2007). 1 pg of DNA represents 978 mega base pairs (Mbp).The amount of nuclear DNA of each sample was calculated based on the values of the G1 peak means (Doležel and Bartoš, 2005) as follows: T-test analysis was performed to show C-value (genome size) difference between long-styled and short-styled plant populations, while analysis of variance (ANOVA) test was performed to showed difference in 2C-value content among the three Linum species studied.Pearson coefficient of correlation was determined between C-value and geographical distribution of plant populations studied.SPSS ver. 9 (1988) and NTSYS ver.2 (1988) softwares were used for statistical analyses.

Plant morphology
Morphological comparison (Table 2) between long-styled and short-styled plant populations in L. glaucum, showed a higher mean value of the plant height, size of the basal leaves width, flower leaves width, calyx length, sepal length and petal length in the long-styled plants, while the mean value of branch number, basal leaves length, flower leaves length, calyx width, pedicel length and sepal length was higher in the short-styled plant populations.T-test analysis of morphological characters showed significant difference (p<0.05) for basal leaf length, calyx width and length.
Similar analysis in L. austriacum showed a higher mean value of the plant height and branch number in the long-styled plants, while the mean value of the other characters was higher in the short-styled plant populations.In L. album, a higher mean value of the basal leaves length, flower leaves length, flower leaves width, calyx width and sepal width occurred in the longstyled plants, while, the short-styled plant populations had a higher mean value for the other characters studied.T-test analysis of morpho-logical characters showed significant difference (p<0.05) for some of the characters studied.PCoA plot of long-styled and short-styled plant populations based on all morphological characters also separated these two kinds of plants in the three species studied (Figures 1 and 2).
The three species studied also differed in morphological characters studied and were separated in to distinct clusters/groups in UPGMA tree and PCoA plot, irrespective of long styled/short styled (Figure 3).

Palynology
In general, the pollen shape was circular in the three species studied, but its shape varied in the equatorial view in these species that is, circular in L. austriacum but elliptic in the other two species.Moreover, it was elliptic-obtuse and elliptictruncate in L. album (Table 3).The polar and equatorial views of the pollen grains were similar in Ls and Ss plants of the species studied, while the aperture shape varied in these populations.For example, it was polygonal gemmate in L. austriacum long-styled plants but gemmate shape in the short-styled plants of this species.The aperture shape was pilate and baculate in the long-styled plant populations of L. album while, it was pilate shape in the short-styled plant populations.Similarly in the long-styled plant populations of L. glaucum the aperture shape was gemmate and pilate to clavate but it was clavate in the short-styled plant populations.(Figure 4).
Comparison of pollen characteristics (Table 3) between long-styled and short-styled plant populations in L. glaucum, showed a higher mean  value of the aperture width and length, distance between apertures, polar length and colpi width as well as colpi length, in the long-styled plants, while the mean value of mesocolpi and equatorial length was higher in the shortstyled plant popula-tions.T-test analysis of morphological characters showed significant difference (P < 0.05) for distance between apertures, aperture width and length.
Similar analysis in L. austriacum showed a higher mean value of almost all pollen characteristics higher in the short-styled plant populations compared to those in the long-styled plant populations.In L. album, a higher mean value of the aperture length and distance between apertures were observed in the long-styled plants, while, the short-styled plant populations had a higher mean value for the other characters studied.T-test analysis of morphological characters showed significant difference (P < 0.05) for these characters.
PCoA plot of long-styled and short-styled plant populations based on all palynological characters also separated these two kinds of plants in all studied three species (Figures 5 and 6).

Flow cytometry
In order to compare C-value (genome size), difference C-values obtained differed between long-styled and short-styled plants of the species studied.For example, in L. austriacum, the mean 2C-value of long-styled plants was 2.65 ± 0.30 pg but in the short-styled plants it was 1.99 ± 0.12 pg.However, this was not a significant difference (P = 0.11).In case of L. album, long-styled plants had the mean 2C-value of 4.61 ± 0.71 while and short-styled plants had the mean 2C-value of 3.61 ± 0.07.Difference of 2C-values between long-styled and short-styled plants in this species was also not significant (P = 0.23).The same holds true (P= 0.17) for L. glaucum Boiss.& Nöe with 2C-value of 1.98 ±0.01 in Ls plants and 1.78 ± 0.11 in Ss plants.However, the ANOVA test performed among the three Linum species showed a significant diffe-rence in 2C-value content (F = 11.88,P<0.001).
A negative significant correlation (r = -0.85,P < 0.05), was observed between 2C-value and the stem length, but a positive significant correlation was observed between 2C-value and sepal length (r = 0.83, P < 0.05).Similarly, a positive significant correlation was observed between 2C-value and northern distribution of the Linum species studied (r = 0.54, P < 0.05), while a negative significant correlation (r = -0.57,P < 0.05) occurred with eastern distribution.No correlation was noticed between 2C-value and altitude of the species studied (r = 0.2., P >0.05).

DISCUSSION
Heterostylous plants have been characterized by the presence of two or three discrete morphs that differ in their sex organ position within popu-lations.This polymorphism is widely distributed among the angiosperms, but detailed studies are limited to few taxonomic groups and it is suggested that, when precise measurements of the sexual whorls are reported, an evolutionary meaningful variations of the heterostylous syndrome will be understood (Sánchez et al., 2010).
The evolution of heterostyly appears to occur within some general constrains on floral morpho-logy (Ganders, 1979).Heterostylous flowers are generally moderate sized and have a floral tube, a limited number of stamens, and a syncarpous ovary with few carpels.For example, in the genus Cordia (Boraginaceae), smallestflowered species showed dioecy but heterostyly occurred in medium to large size flowers (Opler et al., 1975), while in the genus Melochia (Sterculiaceae), the smallestflowered plant lacks distyly (Martin, 1966).In Hypericum aegypticum and Cratoxylum formosum (Guttiferae) the heterostylous plants which have many stamens, are often grouped into three to five bundles with 30 to 125 stamens (Ornduff, 1975;Lewis, 1982).
Detailed comparison of flower characters between the long-styled and short-styled plants showed that in L. glaucum, the long-styled plants have significantly a Higher mean value of the flower leaves length, and calyx width compared to those of the shortstyled plants, while in L. album, the long-styled plants show a significant higher mean values in the calyx width, petal length and petal width.However, no significant difference in flower characters was observed in L. austriacum between the short-styled and long-styled plant populations.However these species showed distinct range of morphological variations as they were separated from each other in the UPGMA tree.The reproductive features such as; petal and sepal dimensions, pedicle length and some of vegetative characters such as basal and flower leaves dimensions were bigger in Ss samples rather than Ls samples.Study of different aspects of floral morphology in heterostylous species has been widely used as a tool to understand the pollination process, as it significantly influences pollen transfer and reproduction (Turketti, 2010).Research shows that many of heterostylous species are selfcompatible (Barrett and Cruzan, 1994) Linum species have distinct hermaphrodite flowers with attractive petals and nectars which absorbed different pollinator's insect.Heterostyly facilitated cross-pollination in Linum and increases the insect mediated pollination (Darwin, 1864).Rogers (1979) described Linum suffruticosum as heterostylous and intramorph-incompatible in eastern Spain showing that the pollen size is the same in the two morphs, but that the exine sculpturing differs.Three-dimensional heterostyly observed in L. suffruticosum is very effective in pollen distribution.When a pollinator insect visits this heterostylous species, the pollens of longstyled plants are placed on the underside of pollinator while, the pollens of short-styled plants are placed on the top of thorax and abdomen.The stigmas of long-styled plants contact the flies on the dorsum and pick up predominantly shortstyled plants pollens.Moreover, the stigmas of the short-styled plants contact the flies on the ventral surface, picking up predominantly long-styled pollens (Armbruster et al., 2006).Armbruster et al. (2006) also reported morphological variations in distylylous L. suffruticosum L., with bent styles and stamens, achieving a three-dimensional arrangement.Pin (L) and thrum (S) morphs were found to be of nearly identical appearance, except for the length and orientation of the sexual parts.Style and stamen lengths differed significantly; however, the differences were small compared with those in many other species of Linum and other heterostylous species.There was no detectable difference in pollen size.However, stigma width differed significantly between morphs.The stamens differed between pin and thrum morphs.Pin stamens spread from near the base, appraised to the corolla wall, and extending onethird of the way up the petals; the openings of the dehiscing anthers face inwards.The thrum stamens were erect, forming a column in the center of the flower, and the anthers are rotated so that the openings of the dehiscing anthers face outwards.Phylogenetic analysis and MACCLADE reconstruction of character states performed by Armbruster et al. (2006) also showed that heterostyly has originated several times, not only within Linum, but also within sect.Linastrum, including one to three reversions to monomorphism.They suggested that heterostyly has evolved at least twice in the two Linastrum clades, although equivocal character transitions and the limited sample of taxa preclude inference of the exact number of shifts.The least restrictive optimization of heterostyly is consistent with at least three independent origins of heterostyly in Linum.Sánchez et al. (2010) also reported stigma height dimorphism, as opposed to distyly in the genus Nivenia as the only genus within the Iridaceae containing heterostylous species.The presence of different types of polymorphism within the genus is consistent with hypotheses of the evolution of heterostyly.The role of the pollinators as the leading force of the transition seems to be apparent, since floral integration is related to reciprocity.Variation of sexual whorls was observed in different Nivenia species with Nivenia fruticosa as the only monomorphic species.
For the species with two floral morphs, stigma height was significantly different in all of them and the length of the stamens was statistically different between morphs in all cases, except for the population of Nivenia argentea at Aasvoëlkrans.Thus, those were considered as distylous, and N. argentea at Aasvoëlkrans as stigma-height dimorphic since two style-length morphs were present but anther heights remained indistinguishable between morphs.
Our palynological study revealed that the shape of pollen grains in the equatorial and polar views are uniform between Ss and Ls plant populations in the Linum species studied, but other pollen characteristics like the aperture shape and dimensions, polar and equatorial axis length differed between long-styled and short-styled plant populations.Previous reports indicate that in some heterostyled Linum species, the pollen surface in short-styled plants is composed of homo-sized apertures but the exine surface in long-styled plants contains small and big aperture with variable papillae shape and size (Dulberger, 1981).Nicholls (1986) in L. perenne found that short-styled plant was a better pollendonor and the long-styled plant was a better pollenreceiver.The short-styled samples produced more pollen, but the long-styled samples matured more seeds.These observations suggest a degree of sexual dimorphism in L. perenne, which the short-styled plants behaving functionally more male and the long-style plant functionally more female.Richards and Barrett (1992) reported differences in the number and size of pollen grains (for example Pontederia cordata), dimorphic in stamens position (Primula vulgaris), (Darwin, 1877), pollen grain color (Linum pubescens), (Wolfe, 2001), stylar bending (in Linum grandiflorum), shape and colors of stigma and its surface papillae (Linum hirsutum, Unal and Yildorim-Fazla, 2007).In some heterostylous species of Linum such as; L. perenne, L. grandiflorum and L. alpinum, exine sculpturing structure were differed between long-styled Heterostyly is usually associated with polymorphisms of pollen between samples (Barrett, 1992).Exine pattern of pollen interacts with biotic and abiotic pollination vectors and affected the surface area of the stigma interface and mediated stigma adhesions.
Aperture size, number and its complexity affected environmental vulnerability to desiccation, fungal invasion and mechanical stress, and serve as portals for pollen tube exit during germination (Edlund et al., 2004).Wang et al. (2009) found that in Pedicularis (Orobancacheae) there was a significant association between pollen aperture types and corolla types, as well as between pollination syndromes and corolla.There was a distinct correlation between exine ornamentation, floral morphology and pollination in Bauhinia.(Ferguson and Pearce, 1986).
Nuclear DNA C-value varies in different taxonomic taxa (Yokoya et al., 2000) and is considered as a mean for adaptation, and is affected key parameters of plant growth such as; the duration of the cell size, cell cycle, rate of cell division, sensitivity to radiation, ecological behavior in plant communities and life forms (Bennett et al., 2000).The 2C-values obtained here indicate genomic content difference between short-styled and long-styled plant populations and the possible role of quantitative genetic changes in heterostyly.Significant difference in 2C-values of the three Linum species studied indicates the possible role of DNA content change during species diversification.This may also suggest the use of 2C-value as supplementary data for species grouping in the genus Linum.Although not significant, negative relation between 2C-values and some morphological and palynologycal characters occurred in L. austriacum and L. album.The short-styled plants of L. austriacum and L. album have smaller genome size and nuclear DNA 2Cvalues compared to the long-styled plants.In the shortstyled plants, the petal and sepal dimensions, pedicle length, the mean size of basal and flower leaves and polar and equatorial axis of pollen are relatively bigger than the long-styled plants.
Significant correlation obtained between 2C-values and some of the quantitative morphological characters may indicate the effect of genome size on this character and the possible adaptive nature of these characters (although not apparent to us by now).Such correlation has been reported in other plant species including Cirsium species (Nouroozi et al., 2011).Significant positive correlation occurred between 2C-values and ratio of pappus length/seed size and involucre length.
Positive significant correlation was observed between 2C-value and northern distribution of the Linum species studied, while a negative significant correlation occurred with eastern distribution.No correlation was noticed between 2C-value and altitude of the species studied.This may indicate some relation with ecological and population differentiations which should be investigated to be confirmed.Studies performed on Cirsium species (Nouroozi et al., 2011) showed significant negative correlation between 2C-value and latitude, and positive significant correlation between 2C-value and the mean annual rain fall, while Bureš et al. (2004) reported that the species with larger genomes prefer dry habitats in areas with more oceanic climates, and species with smaller genomes grow in more humid habitats (wetlands) in continental areas.

Table 2 .
Morphology characteristics in long styled and short styled plant populations studied.
All values are in cm.

Table 3 .
Pollen characteristics in the Linum species studied.

Table 4 .
Nuclear DNA amount and genome size in studied taxa of the genus Linum.