Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2013 Nov 27;113(3):523–532. doi: 10.1093/aob/mct277

Clonal growth is enhanced in the absence of a mating morph: a comparative study of fertile stylar polymorphic and sterile monomorphic populations of Nymphoides montana (Menyanthaceae)

Azadeh Haddadchi 1,*, Mohammad Fatemi 1,†, C L Gross 1
PMCID: PMC3906970  PMID: 24287813

Abstract

Background and Aims

Many aquatic species with stylar polymorphisms have the capacity for clonal and sexual reproduction and are sensitive to the balance of the two reproductive modes when there are a limited number of mating morphs within a population. This study asked how the clonal and sexual reproductive modes perform in populations that contain only a single morph and where fitness gain through sexual reproduction is rare. In clonal aquatic Nymphoides montana, polymorphic populations normally contain two mating morphs in equal frequencies. Populations are sexually fertile and appear to be maintained by pollen transfer between the two partners. However, in a monomorphic population of N. montana where mating opportunities are unavailable, female and male function is impaired and clonality maintains the population. Here, the consequences of intraspecific variation in sexuality were explored between monomorphic and polymorphic N. montana populations in eastern Australia.

Methods

Comparative measurements of male and female fertility, total dry mass and genotypic diversity using ISSR markers were made between populations with variable sexuality.

Key Results and Conclusions

Very few seeds were produced in the monomorphic population under natural and glasshouse conditions due to dysfunctional pollen and ovules. Stigma–anther separation was minimal in the monomorphic population, which may be a consequence of the relaxed selective pressures that regulate the maintenance of sexual function. However, clonal reproduction was favoured at the expense of sexual reproduction in the monomorphic population; this may facilitate the establishment of sterility throughout the population via resource reallocation or pleiotropic effects. The ISSR results showed that the monomorphic population was one large, single genotype, unlike the multi-genotypic fertile polymorphic populations. Evolutionary loss of sex in a clonal population in which a mating morph is absent was evident; under these conditions clonal growth may assure reproduction and expand the population via spreading stolons.

Keywords: aquatic plant, biomass, genotypic diversity, mating morph, Menyanthaceae, Nymphoides montana, resource reallocation, stylar polymorphism, trait loss

INTRODUCTION

Many perennial flowering plants reproduce sexually and asexually (e.g. by clonal growth) (Richards, 1986), and the balance between clonal and sexual reproduction can vary between species or among populations within a species (Eckert, 2002; Shibayama and Kadono, 2007; Vallejo-Marín et al., 2010). Variation in reproductive modes may be influenced by ecological (e.g. unfavorable conditions for pollinator activity; Eckert and Barrett, 1993) or genetic factors (e.g. uneven ploidies; Lui et al., 2005) that in turn can limit the opportunity for sexual reproduction. Stochastic processes can also limit sexual reproduction where, for example, founder events result in the loss of mating partners (Ornduff, 1972). Under such ecological conditions, in which selection intensities may not be strong enough to maintain sexual function, sterility mutations may spread and establish within populations via clonal reproduction (Vallejo-Marín et al., 2010; Gross et al., 2012). Reduced sexual performance or complete sexual sterility may have important implications for the ecology and genetics of clonal plants (reviewed in Eckert, 2002). To date, the ecological and/or genetic consequences of intraspecific variation in reproductive modes have received little attention in clonal plants with stylar polymorphisms (but see Dorken and Eckert, 2001; Wang et al., 2005).

Species with stylar polymorphic populations have flowers that consistently vary in the length or orientation of style and anthers (Barrett et al., 2000; Pauw, 2005; Haddadchi, 2013). Within populations, the frequency of these morphologically distinct floral (or mating) morphs can deviate from parity (Eckert and Barrett, 1992; Brys et al., 2007; Meeus et al., 2012). Mating morphs often cannot set seed using self pollen and/or pollen from the same morph due to a polymorphic incompatibility system (Ganders, 1979; Lloyd and Webb, 1992b), i.e. each morph can only mate with the reciprocal morph(s) to set seed. Stylar polymorphic species usually have a great capacity for clonal as well as sexual reproduction (Eckert and Barrett, 1995; Dorken and Eckert, 2001; Wang et al., 2005), and they are particularly sensitive to the balance of sexual and clonal recruitment because the number of mating morphs is limited within a population (e.g. only two morphs in distylous populations). This raises the question of how the two reproductive modes (sexual versus clonal) perform in, for example, a population that contains a single mating morph (i.e. monomorphic) and where fitness gain is limited because sexual reproduction is limited. Sexual sterility due to a morph loss and a complete reliance on clonal reproduction is not uncommon among stylar polymorphic species, following, for example, a breakdown of the reproductive system in founding populations (e.g. Ornduff, 1972; Eckert and Barrett, 1992; Castro et al., 2007). Yet few researchers have experimentally compared the consequences of sterility for sexual functions in sterile monomorphic and fertile polymorphic populations (e.g. Barrett et al., 1993; Dorken and Eckert, 2001) or investigated the costs of sterility in populations by examining whether clonal reproduction should be favoured at the expense of sexual reproduction (e.g. Dorken et al., 2004) or not (e.g. Eckert et al., 1999).

In the clonal aquatic Nymphoides (Menyanthaceae) there are several reports of an association between prolonged clonal growth and failure of sexual reproduction, including N. peltata (Ornduff, 1966; Uesugi et al., 2004), N. cordata, N. aquatica (Ornduff, 1970) and N. indica (Barrett, 1980). Very few researchers, however, have examined the relative importances of clonal and sexual reproduction to the population persistence of Nymphoides in detail. For example, Ornduff (1966) proposed that extensive clonal propagation may cause evolutionary modifications in the breeding system of European N. peltata, whereas Wang et al. (2005) found no evidence of the breakdown of breeding system or evolution of selfing as the result of clonal growth in Chinese populations of N. peltata.

Here we investigated the cause and consequences of the extreme variation in reproductive mode in Nymphoides montana, a perennial aquatic herb endemic to Australia (Aston, 1982). In a previous study, the degree of sex-organ reciprocity was quantified in three eastern Australian populations where the stylar condition of N. montana was described as a stylar polymorphism, including distylous and stigma-height dimorphic populations (Haddadchi, 2013). These populations are sexually fertile, although variation in reciprocity and the extent of incompatibility between morphs has been observed among populations. We included a monomorphic population of N. montana (identified by H. I. Aston in 2001; NE 033671) and used comparative measurements of male and female fertility, clonal growth and genotypic diversity between the monomorphic and polymorphic populations.

We did not find fruit or seed in the monomorphic population during our preliminary field studies. Therefore, our first aim was to determine if plants from this population could set seed under a common glasshouse environment, and whether factors reducing seed set have a genetic or ecological component. A genetic basis to sexual sterility is revealed when there is no difference in sexual reproduction between the field and glasshouse environments (Eckert et al., 1999; Dorken and Eckert, 2001). Secondly, we compared within-flower traits associated with sexual reproduction in the monomorphic and fertile polymorphic populations to assess if sterility is associated with a degeneration of sexual traits. Thirdly, we determined chromosome counts and ploidy levels to understand the underlying factors that could disrupt seed production in the monomorphic population. A clonal growth experiment was then undertaken to investigate if reduced investment in sexual reproduction is associated with enhanced clonal reproduction, and how sterile plants increase in frequency in natural populations. Finally, since complete or near complete sterility has been shown to have various effects on the genotypic structure of clonal plant populations (Ellstrand and Roose, 1987; Rossetto et al., 2004), a genotypic study using ISSR molecular markers, a DNA fingerprinting technique, was undertaken. We conclude by discussing the genetic and ecological consequences of sterility in the absence of a mating morph.

MATERIALS AND METHODS

Study species and populations

Nymphoides montana (Menyanthaceae, marshwort) is an aquatic clonal plant that is endemic to eastern Australia, where it inhabits the shallow margins of lakes and swamps (Jacobs, 1992). Nymphoides montana has a stylar polymorphism and sexually fertile populations contain two mating morphs that differ in the reciprocal positioning of stigma and anthers (Haddadchi, 2013). The flowers are attractive to insects, with five yellow petals. Nymphoides montana possesses a dimorphic incompatibility system, although between-morph variation in the extent of incompatibility has been observed in one population (Haddadchi, 2013). Crosses between plants of the reciprocal morph produce 89–93 % seed set. Nymphoides montana combines sexual reproduction with clonal reproduction from spreading stolons. Clonal propagation occurs when the aboveground shoots die over winter and ramets become dismembered from the parent plant and are transported by water to favourable sites, where they re-establish.

Four populations of N. montana located on the Northern Tablelands of New South Wales were used in this study: one monomorphic [Cooney Creek (CC), 1000 m elevation asl, 30° 32′ 06″ S, 151° 50′ 20″ E] and three stylar polymorphic populations [Dumaresq Dam (DD), 1070 m elevation asl, 30° 25′ 48″ S, 151° 35′ 50″ E; Thomas Lagoon (TL), 1035 m elevation asl, 30° 32′ 43″ S, 151° 33′ 10″ E; Glencoe (GC), 1150 m elevation asl, 29° 55′ 30″ S, 151° 43′ 17″ E]. Observations and experiments on the CC population were carried out between February 2005 and March 2007. Approximately 600 flowers were scored in a 1200 m2 area throughout the whole population. Flowers of the CC population are similar to the N. montana short-style (hereafter S) flowers. No flower of the reciprocal and compatible morph (long-style, hereafter L) was sighted within the CC population during the study years. The CC population is situated in eastern Australia and separated from the three polymorphic (DD, TL and GC) populations by 26, 27 and 70 km, respectively.

Floral morphometrics

To determine the positions of stigmas and anthers on the monomorphic plants, we randomly collected 20 flowers from each of the monomorphic CC and polymorphic DD, TL and GC populations. Here and elsewhere, sampled flowers were at least 5 m apart; this minimized the chances of sampling genetically uniform plants (or ramets). We measured stigma height, stigma width, stigma length, anther height, anther length, stigma–anther separation, corolla tube length and corolla diameter. The height measurements were taken from the base of the superior ovary. To assess whether the variation in sexual systems is associated with changes in sex-organ heights and herkogamy, we compared stigma height, anther height and stigma–anther separation between the monomorphic plants and each of the S- and L- plants from the DD population. The DD population is representative of other polymorphic populations as the population is typically distylous and displays strong sex-organ reciprocity, in which anther heights perfectly overlap the corresponding stigma heights (see Fig. 2 in Haddadchi, 2013).

Fig. 2.

Fig. 2.

Open in a new tab

Comparison of proportion of seed set after within-population cross-pollination (M × M, open circles) and between-population cross-pollination with the fertile S (M × S, filled symbols) and L plants (M × L, open symbols). Polymorphic populations: DD (triangles) and TL (squares). No seed was set in the within-population crosses. In the between-population crosses, the monomorphic plants produced more seeds when crossed with the L than with the S morphs. Means ± s.e. of proportion of seed set per population are plotted.

Floral traits that contributed to the distinction between sexual morphs (monomorphism versus polymorphism) were explored by discriminant function analysis. To understand the structure of the data without making a priori assumptions about the sexual morphs and population, principal components analysis was used on all floral traits. A covariance matrix was used to extract eigenvalues. One-way ANOVAs were used to compare stigma height, anther height and stigma-anther separation between the sexual morphs.

All multivariate and univariate analyses were performed using the Minitab® (version 16) statistical package. Where necessary, data were log10-transformed to satisfy the assumptions of ANOVA.

Chromosome counts

To assess whether ploidy levels were associated with fertility, root-tip meristems were obtained from young plants of the monomorphic CC and the polymorphic DD and TL populations (number of plants: n = 22 at CC and 25 at each of DD and TL). Plants were left in water to produce roots, which were then harvested and placed in a metaphase-arresting agent (2 mm 8-hydroxyquinoline) for 3–5 h and then fixed in absolute ethanol and glacial acetic acid (3:1) for at least 24 h at 4°C. Samples were macerated in 1 m hydrochloric acid at 60°C for 15–20 min, washed and then suspended in 45 % acetic acid for 10 min (Li and Zhang, 1991). After maceration, the root-tip meristems were transferred to a drop of 0·5 % carbon–fuchsin solution on a microscope slide. We obtained ∼250 cells from each population. The large numbers of aggregated chromosomes and the small size of the chromosomes made an accurate chromosome count difficult; mean (± s.d.) chromosome numbers are therefore presented here.

Open pollination and pollinations within and between populations

To assess the reproductive success of the monomorphic plants in the natural environment, in February 2005–2006 we tagged 40–45 mature flowers and then scored the number of flowers that formed mature fruits 2 weeks later. In March 2006, we tested the level of sexual fertility in 16 monomorphic (M) plants in a pollinator-free glasshouse. On each M plant, one flower was allocated to each of the following hand pollination treatments: self-pollination, within-population cross-pollination, and between-population cross-pollination, including reciprocal crosses between the monomorphic plants and each of the S (M × S) and L plants (M × L) of the polymorphic DD and TL populations. We counted the number of flowers forming mature fruits with at least one seed and the number of ovules forming mature seeds in each fruit to calculate percentage fruit set and percentage seed set, respectively. Controlled cross-pollinations between the polymorphic populations were performed and percentage fruit set was calculated (plants of each morph: n = 10 from each of DD and TL).

We used a random block ANOVA to compare percentage seed set sired by S and L pollen from the two polymorphic populations with plant as a random factor and both morph and population as fixed factors. We considered population as a fixed factor, since the two polymorphic populations were the only fertile populations in this study.

Male sterility

To assess male sterility, we examined pollen morphology, the quantity of aborted pollen produced per flower, variation in pollen size, and viability by stainability and germinability experiments, as described below.

Anther and pollen morphology

To assess male sterility based on variation in anther and pollen morphology between the monomorphic and polymorphic plants, we used scanning electron microscopy (SEM). Anthers and pollen of three flowers were collected from each of the monomorphic and the polymorphic DD populations, and were air-dried and sputter-coated with gold for observation under various magnifications with a Jeol JSM-5800LV scanning electron microscope (Tokyo, Japan).

Proportion and size of aborted pollen

To assess male sterility based on the proportion of aborted pollen, pollen from flowers of ten randomly chosen monomorphic plants were collected and prepared semi-permanently on microscope slides (Beattie, 1971). We counted and compared the proportions of pollen that appeared healthy (swollen and triangular) and aborted (shrivelled with a convoluted surface) from a total of 100 pollen grains per plant. Mean (± s.e.) percentages are reported here.

To assess male sterility by variation in pollen size between the healthy and aborted pollen, we calculated pollen area of each type from each of the monomorphic flowers used to count pollen, using a Nikon digital sight DS-L1 attached to a light microscope (pollen of each type: n = 15 from each plant). Pollen grains are triangular and the height and length of pollen were measured to compare pollen size.

We used a random block ANOVA to compare pollen size among the plants and between the two types of pollen with plant and pollen type as random and fixed factors, respectively.

Pollen germinability and stainability

To assess pollen viability by germinability, pollen from each of the monomorphic CC and the polymorphic DD and TL plants was collected (number of plants: n = 10 for each population) and germinated in a series of media with increasing sucrose concentrations (5–25 %). Pollen showed the highest germinability on the 15 % medium. Viable pollen was able to grow a tube in vitro, whereas inviable pollen did not germinate. We then tested the stainability of pollen with 1 % acetocarmine from each of the plants used to germinate pollen. Viable pollen stained light red to red, whereas inviable pollen did not stain. Mean (± s.e.) percentages are reported here.

Clonal growth experiment

To examine if reduced investment in sexual reproduction is associated with enhanced clonal reproduction, in February 2006 we collected, weighed and planted 14 young ramets of similar sizes (containing apical buds and a pair of leaves on the last vegetative nodes on stolons) from each of the monomorphic CC and the polymorphic DD and TL populations in the glasshouse. We set up the experiment in a complete randomized block. A total of 70 plants were randomly assigned to seven blocks (ten plants, including two monomorphic plants and two plants of each morph from each of the polymorphic DD and TL populations, per block). The total number of leaves, flowers and vegetative nodes borne on stolons were counted every week for one flowering season. The total plant dry mass allocated to sexual and clonal reproduction was also calculated. Flowers were harvested throughout the season and placed in an oven at 70°C to dry. At the end of the flowering season, just as the aboveground structures began to senesce, all ramets were harvested, dried to a constant weight at 70°C for at least 72 h and weighed.

We used analysis of covariance (ANCOVA) to detect differences between the two sexual systems (clonality versus sexual reproduction) in the number of leaves, vegetative nodes and flowers, and total plant dry mass, while controlling for variation in the fresh mass (primary weight) character. An ANCOVA based on type III sum of squares was conducted with fresh mass as a covariate, block, sexual system, population nested within sexual system, and morph nested within sexual system and population. Block was considered as a random factor and population as a fixed factor.

Sampling, DNA isolation and ISSR analysis

To measure the impact of sterility on the genotypic diversity of the monomorphic population, in 2007 we used DNA-based molecular markers (inter-simple sequence repeats, ISSRs). During March–November, we collected fresh leaves from 21–43 plants from each of the monomorphic CC and the polymorphic DD and TL populations. Genomic DNA was extracted from ∼20 mg of dried leaf tissues using the DNeasy Plant Mini Kit (Qiagen). Forty-five ISSR primers (GeneWorks, Melbourne, Australia) were obtained to screen for polymorphisms. Nine primers (808, 818, 826, 827, 834, 841, 855, 880 and 900) that produced reproducible bands were selected for amplification for all of the DNA samples. PCR reactions, gel electrophoresis and scoring were prepared following Haddadchi et al. (2013).

All the amplified bands were treated as binary data and scored as 1 (present) or 0 (absent) to assemble a rectangular matrix. ISSR diversity was analysed using GenAlEx version 6·5 (Peakall and Smouse, 2012) to assess the proportion of polymorphic bands (P%), Shannon's diversity index (I) and the number of putative genotypes.

Genotypic diversity was estimated by calculating the following two parameters in each population. (1) The proportion of distinguishable genotypes (Ellstrand and Roose, 1987; Eckert and Barrett, 1993) was measured as G/N, where G is the number of genotypes detected and N is the total number of individuals (ramets) sampled. (2) Simpson's diversity index (D) modified for finite sample size by Pielou (1969) was calculated as D = 1− ∑{[ni(ni − l)]/[N(N − 1)]}, where ni is the number of ramets of the ith genotype and N is the sample size. The index D varies from 1 in a population where every individual sampled is a different genotype to 0 in a population composed of a single clone.

RESULTS

Floral morphometrics

Monomorphic flowers were more similar to the S than the L flowers in which the stigma was located below the anthers. In the principal components analysis, the monomorphic plants formed a distinct group that clustered in close proximity to the S-morph group along the PC1 axis (Fig. 1). The first two principal components explained ≥93·9 % of the variation in the traits (eigenvalues and total variance: PC1 = 14·30 and 83·9 %, respectively; PC2 = 1·60 and 10·0 %). In the discriminant function analysis, cross-validation verified this result and 89 % of the total variation among the individuals was correctly classified.

Fig. 1.

Fig. 1.

Open in a new tab

Plot of the first and second principal components analysis scores for the S (left, closed symbols), monomorphic (middle, +) and L plants (right, open symbols) of Nymphoides montana. Principal components PC1 and PC2 explained 83·9 and 10 % of the floral trait variation among the individuals, respectively. Symbols referring to the polymorphic populations are indicated in the key.

Stigma height (mean ± s.e.) of the monomorphic plants (6·20 ± 0·06) showed no significant difference from that of the S plants (6·17 ± 0·13; F1,28 = 0·54, P  = 0·468) but was significantly shorter than that of the L plants (10·70 ± 0·11; F1,28 = 1098·13, P  < 0·001). Anther height (mean ± s.e.) of the monomorphic plants (7·03 ± 0·08) was significantly shorter than that of the S plants (8·99 ± 0·15; F1,28 = 117·58, P  < 0·001) and longer than that of the L plants (5·91 ± 0·07; F1,28 = 92·85, P < 0·001). Consequently, the monomorphic plants showed significantly reduced stigma–anther separation (mean ± s.e., 0·74 ± 0·09) compared with the polymorphic plants (S morph, 2·81 ± 0·13; F1,28 = 159·97, P < 0·001; L morph, 4·78 ± 0·11; F1,28 = 739·94, P < 0·001).

Chromosome count

The basic chromosome number in Nymphoides is x = 9 (Ornduff, 1970). Somatic chromosome counts in the monomorphic CC population of N. montana showed average counts close to the hexaploid number 6x = 54 (mean ± s.d., 52·4 ± 2·9). The polymorphic populations showed more than one cytotype (4x and 6x). Hexaploid 6x = 54 (52·1 ± 2·2) and tetraploid 4x = 36 (34·58 ± 1·88) counts were obtained for the polymorphic DD and TL populations, respectively.

Open pollination and pollinations within and between populations

In the natural environment, no flower in the monomorphic population set fruit or seed during the 2 years of the study. Furthermore, in the glasshouse, following within-population hand-pollinations, from a total of 32 monomorphic flowers that received either self or outcross pollen, no fruit or seed was formed (Fig. 2). Following between-population pollinations, very few seeds were produced (Fig. 2; <4 % seed set). The monomorphic plants produced significantly more seeds when they were crossed with the L plants compared with the S plants (Fig. 2 and Table 1). The effect of monomorphic plant was not significant (Table 1). There were no significant differences in percentage seed set sired by pollen from the polymorphic DD and TL populations (Table 1). The population × morph interaction was also not significant (Table 1). In addition, the monomorphic pollen donors failed to produce any seeds in either of the two pollen-recipient polymorphic populations under the glasshouse conditions. In contrast, the control cross-pollination (DD × TL) resulted in 100 % fruit set for each morph. The viability and fertility of F1 seeds from crosses between the polymorphic populations with different ploidy levels were not tested.

Table 1.

Results of the random block ANOVA showed a significant effect of fertile morphs (as pollen donors) on the proportion of seed set in monomorphic of Nymphoides montana plants

Source d.f. Mean square F P
Monomorphic plant 15 0·0002 0·48 0·938
Fertile population 1 0·0009 2·10 0·154
Fertile morph 1 0·0023 5·12 0·029
Fertile population × fertile morph 1 0·0001 0·03 0·836
Error 45 0·0004
Open in a new tab

Male sterility

Most monomorphic anthers did not dehisce and their wall remained intact (Fig. 3A, B). This was in contrast with the fertile morphs from the polymorphic populations, where anthers dehisced longitudinally and released pollen (Fig. 3C). The dehisced anthers of the monomorphic flowers showed two distinct pollen shapes: swollen healthy-looking pollen and shrunken aborted-looking pollen with collapsed exine layers (Fig. 3D). On average, 17 % of pollen produced in the monomorphic population looked healthy, whereas aborted pollen accounted for 83 % of the pollen. Healthy pollen grains were significantly larger than aborted pollen grains (Fig. 4; F1,9 = 100·36, P < 0·001). There was no significant variation between monomorphic plants in the number of viable and inviable pollen grains produced (F9,9 = 0·99, P = 0·507). Approximately 1 % of pollen germinated and 7 % stained in the monomorphic population. In contrast, the polymorphic DD and TL populations showed high rates of pollen stainability and germinability (70–85 %).

Fig. 3.

Fig. 3.

Open in a new tab

SEM micrographs of air-dried anthers and pollen of Nymphoides montana, comparing an indehiscent anther (A) with intact anther walls (B) from the monomorphic population and a dehisced anther (C) from a polymorphic population at the time of anthesis. (D) Degenerating pollen grains with collapsed exine from the monomorphic population.

Fig. 4.

Fig. 4.

Open in a new tab

Comparison of the proportion of pollen exhibiting collapsed exine and pollen size in the monomorphic population of Nymphoides montana. Shrivelled aborted-looking pollen was smaller and produced in greater proportion than healthy-looking pollen.

Clonal growth experiment

Results of the ANCOVAs showed that, under controlled glasshouse conditions, plants from the monomorphic population exhibited higher vegetative and clonal performance than plants from the polymorphic DD and TL populations (Table 2, Fig. 5). The numbers of leaves, vegetative nodes and flowers and total plant dry mass of the monomorphic plants were significantly greater than those of the S and L plants (Table 2, Fig. 5). The effects of fresh mass and block were not significant (Table 2). There were no significant differences in the number of vegetative nodes, the number of flowers and final biomass between the S and L morphs (Table 2).

Table 2.

Analyses of covariance of the effect of sexual system on the number of leaves, vegetative nodes and flowers and total plant dry mass in the monomorphic and polymorphic populations ofNymphoides montana

Source d.f. No. of leaves
 No. of vegetative nodes
 No. of flowers
 Total plant dry mass
Mean square F P Mean square F P Mean square F P Mean square F P
Fresh mass 1 0·03 2·25 0·140 0·33 2·59 0·114 0·97 3·16 0·081 0·02 1·10 0·299
Block 6 0·01 1·18 0·330 0·15 1·20 0·320 0·31 1·01 0·428 0·03 1·87 0·104
Sexual system 1 1·01 82·43 < 0·001 5·25 41·28 < 0·001 9·13 29·80 < 0·001 0·32 18·17 < 0·001
Population (sexual system) 1 0·06 4·69 0·035 0·07 0·51 0·477 0·67 2·19 0·145 0·13 7·62 0·008
Morph (sexual system, population) 2 0·08 6·19 0·004 0·28 2·17 0·124 0·52 1·70 0·193 0·01 0·33 0·723
Error 52 0·01 0·13 0·31 0·02
Open in a new tab

All parameters differed significantly between the two sexual systems (bold type).

‘Sexual system’ refers to clonality or sexual reproduction; ‘Morph’ refers to monomorphism or polymorphism. Fresh mass of plant at the start of the experiment was the covariate.

Fig. 5.

Fig. 5.

Open in a new tab

Variation in the capacity for clonal reproduction between plants from the polymorphic populations (as indicated; black, S; white, L) and the monomorphic population of Nymphoides montana grown in a common glasshouse environment. The monomorphic population exhibited higher vegetative performance and flower production than the polymorphic populations. Means (± s.e.) of vegetative and reproductive traits and plant total dry mass per morph per population are plotted.

ISSR and genotypic diversity

A total of 53 different ISSR band positions were scored, of which 44 (83·01 %) were polymorphic among populations. The analyses of ISSR diversity revealed no variation in the monomorphic CC population (Shannon's I = 0·00 and P% = 0·00), whereas in the polymorphic populations of DD and TL, I ranged between 0·292 and 0·399 and P% ranged between 88·68 and 90·57.

A total of 35 unique genotypes were identified among the 83 sampled ramets (all populations). Of these, 18 and 17 distinct genotypes were specific to the polymorphic DD and TL populations, respectively, whereas only one genotype was detected in the monomorphic population and it was unique to that population. The number of observed genotypes expressed as a proportion of the sample size (G/N) and Simpson's diversity index (D) were 0·782–0·809 and 0·980–0·990, respectively, in the polymorphic populations, whereas the monomorphic population showed no diversity (G/N = 0·026, D = 0·000).

DISCUSSION

Origin of sterility (genetics versus environment)

Several authors have commented on the origin of sexual sterility when associated with clonal reproduction in aquatic (Eckert et al., 2000; Dorken and Eckert, 2001; Eckert, 2002) and terrestrial plants (Gross and Caddy, 2006; Gross et al., 2012; Haddadchi et al., 2013) and whether genetic or environmental factors cause this sterility. The former is confirmed when a sterile population shows reduced, or a lack of, seed production following cross-pollination experiments involving fertile and sterile populations even under benign glasshouse conditions (Eckert et al., 1999; Dorken and Eckert, 2001; Eckert, 2002). The results of our study strongly support a genetic basis for the observed sterility. The N. montana monomorphic population was both female- and male-sterile and produced zero to very few seeds (<4 %) following cross-pollinations using pollen from the fertile polymorphic plants under glasshouse conditions. Furthermore, no open-pollinated fruit or seed was detected in the natural environment for the monomorphic population.

In the principal components analysis and floral morphometrics analyses, the monomorphic plants resembled only short-style (S) plants with much reduced stigma–anther separation. Furthermore, sexual fertility in the monomorphic population was partially restored in experimental glasshouse pollinations using viable compatible pollen from the polymorphic populations (i.e. higher seed set in crosses using the pollen of the rare L mating partner than in crosses using the S pollen). There are three possible scenarios for the loss of mating morph in the monomorphic population: (1) the population was founded by a single S genotype that had accumulated mutations sterilizing sexual functions; (2) the founder was of hybrid origin and sterile; or (3) in a source population for CC that had multiple mating genotypes, all but one S genotype was eliminated following a severe bottleneck; the self-incompatible plant would have experienced mate scarcity (long styles) and subsequently failed to produce seeds. Reduced pressure from the absence of a mating morph may have selectively favoured the loss of sexual function and enhanced clonal reproduction if resources made available by not producing viable pollen and ovules are reallocated to other fitness components, such as spreading stolons.

Indeed, low mating partner diversity has been shown to diminish the capacity for sexual reproduction in other species with stylar polymorphisms (e.g. Ornduff, 1970; Barrett et al., 1993). Another genetic factor responsible for sexual sterility in some species is an uneven ploidy level [e.g. triploidy in Butomus umbellatus (Eckert et al., 2000) or heptaploidy in Arundo donax (Haddadchi et al., 2013)]. However, our study of ploidy showed that a change in chromosome number could be ruled out as the possible cause of genetic sterility as a similar ploidy level was observed in the monomorphic and one of the polymorphic study populations. Sexual sterility may be attributed to chromosome aberrations, such as aneuploidy (Klekowski, 1988), which is often associated with abnormalities of different meiotic products in both male and female components, such as a lack of embryo sac and egg apparatus, or poorly developed microspores and male gametes (Fernando and Cass, 1997a, b). Data on pre-pollination and post-fertilization developmental stages are needed to evaluate these possibilities.

The possibility of a joint contribution of ecological and genetic factors to the sexual sterility of N. montana should not be overlooked. Unfavourable environmental conditions [e.g. short flowering season in geographically peripheral populations of Decodon verticillatus (Eckert and Barrett, 1993)] could first limit pollination success and seed production, leading to the evolution of genetic sterility. In an evaluation of ecological hypotheses about conditions that favour clonality versus sexual reproduction, Silvertown (2008) reported an association between clonal reproduction and geographically marginal populations. Study populations of N. montana are located at the northern periphery of the species' range (Australia's Virtual Herbarium), and therefore a widespread sampling across the geographical range is required to determine whether northern populations are more likely to be monomorphic and sterile than southern populations (e.g. see D. verticillatus in Dorken and Eckert, 2001).

Degeneration of sexual floral traits

The ‘use it or lose it’ hypothesis has been a likely explanation for the loss of sex in some sterile taxa, including D. verticillatus (Eckert et al., 1999), Townsendia hookeri (Thompson et al., 2008) and Grevillea rhizomatosa (Gross et al., 2012), where sexual structures are degraded and sexual reproductive output is impaired. Sexual traits that have no fitness effects on reproduction are not under strong selection and may be lost via, for example, the accumulation of mutations (Haldane, 1933; Eckert, 2002). The sterile population of N. montana also expresses the degeneration of sexual traits which have lost both male and female function. Our results showed that (1) most anthers had malfunctioned and could not rupture to release pollen, (2) pollen was mostly inviable, as indicated by pollen germinability and stainability, (3) about 85 % of pollen produced in the monomorphic population exhibited pollen with collapsed exine layers, and (4) <4 % of flowers could set seed when fertile pollen from the polymorphic populations was used. Within the population, relaxed selection on pollen and ovule fertility may result in accumulation of sterility mutations, and then allow the spread of sterility via extensive clonal reproduction (see below). In addition, the N. montana sterile plants exhibited reduced spatial stigma–anther separation (i.e. herkogamy) when compared with the fertile polymorphic plants. In hermaphrodites, herkogamy is known as an adaptive feature that promotes cross-pollination and reduces self-pollination and interference between pollen donors and recipients (Webb and Lloyd, 1986), and has been detected in the build-up of reciprocal herkogamy in stylar polymorphic plants (Lloyd and Webb, 1992a). Degeneration of herkogamy in N. montana is concomitant with loss of male and female function. Similarly, in sterile B. umbellatus a reduction in dichogamy (i.e. temporal stigma–anther separation) has been reported to occur within and among flowers (Eckert, 2002).

Establishment of the sterile monomorphic population (clonality versus sexual reproduction)

Sterile plants can persist in a population if they show fitness advantages that compensate for their loss of sexual reproduction (Eckert et al., 1999; Eckert, 2002). In the N. montana sterile plants, the severe reduction in seed production was associated with enhanced vegetative growth (root and leaves), clonal propagation (stolons) and also flower production. The observed increase in total plant mass of the sterile plants suggests that either the sterile plants use the resources which are saved from not producing viable pollen and ovules for clonal growth in a trade-off between the two reproductive modes, or the sterility mutations could have pleiotropic effects on clonal growth (Eckert, 2002; Vallejo-Marín et al., 2010). Our results do not reveal any association between sterility and a decrease in overall vigour (mutational meltdown hypothesis) or null expectations of the neutral mutation hypothesis, as explained by Eckert (2002). Experimental removal of flowers to intentionally prevent the plants from investing in fruit or seed production could distinguish whether the advantages of the sterile plants over the fertile plants are the result of resource reallocation or antagonistic pleiotropy (Eckert, 2002; Dorken et al., 2004). A question, however, remains as to why non-functional flowers are still produced in the sterile population of N. montana. Increasing flower production could mean that (1) only high-fitness genotypes are preserved through clonal propagation regardless of whether they are sterile (Dorken et al., 2004), or (2) flowers and clonal structures do not compete for either resources or meristems (Eckert et al., 2000), since in N. montana the stolons produce both vegetative and reproductive nodes.

Predominantly clonal reproduction and failure of sexual recruitment in the sterile monomorphic population

In clonal species, the relative importance of clonality versus sexual reproduction may lead to variation in the amount of genetic diversity among populations (Piquot et al., 1996; Dorken and Eckert, 2001; Eckert, 2002). Because most clonal plants are able to reproduce sexually, and because even low levels of sexual recruitment can maintain high levels of within-population genetic diversity, clonal plant populations may be as genetically diverse as sexual populations (Ellstrand and Roose, 1987; Hamrick and Godt, 1990; Kimpton et al., 2002; Rottenberg and Parker, 2004; Gross et al., 2012). However, in clonal populations where sexual reproduction is totally suppressed, prolonged clonal growth is expected to affect negatively the level of genotypic diversity (Honnay et al., 2006; but see Haddadchi et al., 2013). Our ISSR study also detected a strong link between severely reduced seed production and genotypic uniformity; the monomorphic population was dominated by one large single-ramet genotype that spanned 1·2 km2 and was maintained by clonal propagation. In contrast, the observed proportions of distinguishable genotypes (G/N) and Simpson's D in the polymorphic populations were higher than similar estimates from other sexually reproducing clonal species (G/N = 0·17 and Simpson's D = 0·62; Ellstrand and Roose, 1987) and indicate recruitment of sexually outcrossed progeny. Similar patterns of the extreme rarity of sexual reproduction and mating associated with a lack of genotypic diversity have also been reported in other Nymphoides species [e.g. distylous N. peltata (Uesugi et al., 2004) and N. indica (Shibayama and Kadono, 2007)] and in tristylous D. verticillatus (Eckert and Barrett, 1993; Dorken and Eckert, 2001). These examples provide strong evidence for the important role that clonality versus sexual reproduction can play in governing population structure in species with polymorphic reproductive systems.

In the stylar polymorphic populations of the clonal and aquatic N. montana, the presence of two compatible mating morphs is a necessity for sexual reproduction. In a monomorphic population of N. montana, mating opportunities are absent and clonal propagation is the only mode of reproduction. Here, we conclude that in the absence of a mating morph and fertility, the monomorphic plants that no longer ‘use’ sexual traits/mechanisms can ‘lose’ them because the contribution of sex to fitness is zero. Moreover, in a trade-off between sexual and clonal reproduction, selection may favour diversion of resources from the inviable pollen and ovule to clonal reproduction and vegetative growth (although it is not possible to completely rule out the alternative hypothesis that sterility mutations can have negative pleiotropic effects on clonal reproduction). Under these conditions where sex is dysfunctional, clonality may assure reproduction and expand the monomorphic population via spreading stolons (see an example of clonality and reproductive assurance in Solanum in Vallejo-Marín and O'Brien, 2007). Our findings provide evidence for the evolutionary loss of sex in a clonal population in which a mating morph is absent.

ACKNOWLEDGEMENTS

A.H. thanks G. Vaughton and Mike Ramsey for comments on an earlier draft when in thesis format, I. Simpson for field assistance, R. Willis for glasshouse assistance, P. Littlefield in the scanning electron microscopy unit and I. Telford at the NCW Beadle Herbarium. This study was supported by the financial assistance of IPRS (International Postgraduate Research Scholarship) and UNERA (UNE Research Assistantship) scholarships and the School of Environmental and Rural Science, UNE, to A.H.

LITERATURE CITED

  1. Aston HI. New Australian species of Nymphoides Séguier (Menyanthaceae) Muelleria. 1982;5:35–51. [Google Scholar]
  2. Australia's Virtual Herbarium nd. via Centre for Plant Biodiversity Research, Council of Heads of Australian Herbaria. http://www.cpbr.gov.au/cgi-bin/avh.cgi. (10 September 2007) [Google Scholar]
  3. Barrett SCH. Dimorphic incompatibility and gender in Nymphoides indica (Menyanthaceae) Canadian Journal of Botany. 1980;58:1938–1942. [Google Scholar]
  4. Barrett SCH, Eckert CG, Husband BC. Evolutionary processes in aquatic plant populations. Aquatic Botany. 1993;44:105–145. [Google Scholar]
  5. Barrett SCH, Jesson LK, Baker AM. The evolution and function of stylar polymorphisms in flowering plants. Annals of Botany. 2000;85:253–265. [Google Scholar]
  6. Beattie AJ. A technique for the study of insect-borne pollen. Pan-Pacific Entomologist. 1971;47:82. [Google Scholar]
  7. Brys R, Jacquemyn H, Hermy M. Impact of mate availability, population size, and spatial aggregation of morphs on sexual reproduction in a distylous, aquatic plant. American Journal of Botany. 2007;94:119–127. doi: 10.3732/ajb.94.1.119. [DOI] [PubMed] [Google Scholar]
  8. Castro S, Loureiro J, Santos C, Ater M, Ayensa G, Navarro L. Distribution of flower morphs, ploidy level and sexual reproduction of the invasive weed Oxalis pes-caprae in the western area of the Mediterranean region. Annals of Botany. 2007;99:507–517. doi: 10.1093/aob/mcl273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dorken ME, Eckert CG. Severely reduced sexual reproduction in northern populations of a clonal plant, Decodon verticillatus (Lythraceae) Journal of Ecology. 2001;89:339–350. [Google Scholar]
  10. Dorken ME, Neville KJ, Eckert CG. Evolutionary vestigialization of sex in a clonal plant: selection versus neutral mutation in geographically peripheral populations. Proceedings of the Royal Society B: Biological Sciences. 2004;271:2375–2380. doi: 10.1098/rspb.2004.2875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Eckert CG. The loss of sex in clonal plants. Evolutionary Ecology. 2002;15:501–520. [Google Scholar]
  12. Eckert CG, Barrett SCH. Stochastic loss of style morphs from populations of tristylous Lythrum salicaria and Decodon verticillatus (Lythraceae) Evolution. 1992;46:1014–1029. doi: 10.1111/j.1558-5646.1992.tb00616.x. [DOI] [PubMed] [Google Scholar]
  13. Eckert CG, Barrett SCH. Clonal reproduction and patterns of genotypic diversity in Decodon verticillatus (Lythraceae) American Journal of Botany. 1993;80:1175–1182. [Google Scholar]
  14. Eckert CG, Barrett SCH. Style morph ratios in tristylous Decodon verticillatus (Lythraceae): selection vs. historical contingency. Ecology. 1995;76:1051–1066. [Google Scholar]
  15. Eckert CG, Dorken ME, Mitchell SA. Loss of sex in clonal populations of a flowering plant, Decodon verticillatus (Lythraceae) Evolution. 1999;53:1079–1092. doi: 10.1111/j.1558-5646.1999.tb04523.x. [DOI] [PubMed] [Google Scholar]
  16. Eckert CG, Massonnet B, Thomas JJ. Variation in sexual and clonal reproduction among introduced populations of flowering rush, Butomus umbellatus (Butomaceae) Canadian Journal of Botany. 2000;78:437–446. [Google Scholar]
  17. Ellstrand NC, Roose ML. Patterns of genotypic diversity in clonal plant species. American Journal of Botany. 1987;74:123–131. [Google Scholar]
  18. Fernando DD, Cass DD. Developmental assessment of sexual reproduction in Butomus umbellatus (Butomaceae): male reproductive component. Annals of Botany. 1997a;80:449–456. [Google Scholar]
  19. Fernando DD, Cass DD. Developmental assessment of sexual reproductive capacity in Butomus umbellatus (Butomaceae): female reproductive component. Annals of Botany. 1997b;80:457–467. [Google Scholar]
  20. Ganders FR. The biology of heterostyly. New Zealand Journal of Botany. 1979;17:607–635. [Google Scholar]
  21. Gross CL, Caddy HAR. Are differences in breeding mechanisms and fertility among populations contributing to rarity in Grevillea rhizomatosa (Proteaceae)? American Journal of Botany. 2006;93:1791–1799. doi: 10.3732/ajb.93.12.1791. [DOI] [PubMed] [Google Scholar]
  22. Gross CL, Nelson PA, Haddadchi A, Fatemi M. Somatic mutations contribute to genotypic diversity in sterile and fertile populations of the threatened shrub, Grevillea rhizomatosa (Proteaceae) Annals of Botany. 2012;109:331–342. doi: 10.1093/aob/mcr283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Haddadchi A. Stylar polymorphism, reciprocity and incompatibility systems in Nymphoides montana (Menyanthaceae), endemic to southeastern Australia. Plant Systematics and Evolution. 2013;299:389–401. [Google Scholar]
  24. Haddadchi A, Fatemi M, Gross CL. The expansion of sterile Arundo donax (Poaceae) in southeastern Australia is accompanied by genotypic variation. Aquatic Botany. 2013;104:153–161. [Google Scholar]
  25. Haldane JBS. The part played by recurrent mutation in evolution. American Naturalist. 1933;67:5–19. [Google Scholar]
  26. Hamrick JL, Godt MJW. Allozyme diversity in plant species. In: Brown AHD, Clegg MT, Kahler AT, Weir BS, editors. Plant population genetics, breeding, and genetic resources. Sunderland, MA: Sinauer; 1990. [Google Scholar]
  27. Honnay O, Jacquemyn H, Roldán-Ruiz I, Hermy M. Consequences of prolonged clonal growth on local and regional genetic structure and fruiting success of the forest perennial Maianthemum bifolium. Oikos. 2006;112:21–30. [Google Scholar]
  28. Jacobs SWL. Menyanthaceae. In: Harden GJ, editor. Flora of New South Wales. Volume 3. Sydney: Royal Botanic Gardens; 1992. [Google Scholar]
  29. Kimpton SK, James EA, Drinnan AN. Reproductive biology and genetic marker diversity in Grevillea infecunda (Proteaceae), a rare plant with no known seed production. Australian Systematic Botany. 2002;15:485–492. [Google Scholar]
  30. Klekowski EJJ. Mutation, developmental selection, and plant evolution. New York: Columbia University Press; 1988. [Google Scholar]
  31. Li M, Zhang M. Technology for plant chromosome research. Shenyiang, China: Northwest Forest University Press; 1991. [Google Scholar]
  32. Lloyd DG, Webb CJ. The evolution of heterostyly. In: Barrett SCH, editor. Evolution and function of heterostyly. Berlin: Springer; 1992a. [Google Scholar]
  33. Lloyd DG, Webb CJ. The selection of heterostyly. In: Barret SCH, editor. Evolution and function of heterostyly. Berlin: Springer; 1992b. [Google Scholar]
  34. Lui K, Thompson FL, Eckert CG. Causes and consequences of extreme variation in reproductive strategy and vegetative growth among invasive populations of a clonal aquatic plant, Butomus umbellatus L. (Butomaceae) Biological Invasions. 2005;7:427–444. [Google Scholar]
  35. Meeus S, Honnay O, Brys R, Jacquemyn H. Biased morph ratios and skewed mating success contribute to loss of genetic diversity in the distylous Pulmonaria officinalis. Annals of Botany. 2012;109:227–235. doi: 10.1093/aob/mcr272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Ornduff R. The origin of dioecism from heterostyly in Nymphoides (Menyanthaceae) Evolution. 1966;20:309–314. doi: 10.1111/j.1558-5646.1966.tb03368.x. [DOI] [PubMed] [Google Scholar]
  37. Ornduff R. Cytogeography of Nymphoides (Menyanthaceae) Taxon. 1970;19:715–719. [Google Scholar]
  38. Ornduff R. The breakdown of trimorphic incompatibility in Oxalis Section Corniculatae. Evolution. 1972;26:52–65. doi: 10.1111/j.1558-5646.1972.tb00173.x. [DOI] [PubMed] [Google Scholar]
  39. Pauw A. Inversostyly: a new stylar polymorphism in an oil-secreting plant, Hemimeris racemosa (Scrophulariaceae) American Journal of Botany. 2005;92:1878–1886. doi: 10.3732/ajb.92.11.1878. [DOI] [PubMed] [Google Scholar]
  40. Peakall R, Smouse P. GenAlEx 6·5: Genetic analysis in Excel. Population genetic software for teaching and research – an update. Bioinformatics. 2012;28:2537–2539. doi: 10.1093/bioinformatics/bts460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pielou EC. An introduction to mathematical ecology. New York: Wiley-Interscience; 1969. [Google Scholar]
  42. Piquot Y, Saumitou-Laprade P, Petit D, Vernet P, Epplen JT. Genotypic diversity revealed by allozymes and oligonucleotide DNA fingerprinting in French populations of the aquatic macrophyte Sparganium erectum. Molecular Ecology. 1996;5:251–258. doi: 10.1111/j.1365-294x.1996.tb00312.x. [DOI] [PubMed] [Google Scholar]
  43. Richards AJ. Plant breeding systems. London: Chapman & Hall; 1986. [Google Scholar]
  44. Rossetto M, Gross CL, Jonesc R, Hunterd J. The impact of clonality on an endangered tree (Elaeocarpus williamsianus) in a fragmented rainforest. Biological Conservation. 2004;117:33–39. [Google Scholar]
  45. Rottenberg A, Parker JS. Asexual populations of the invasive weed Oxalis pes-caprae are genetically variable. Proceedings of the Royal Society B: Biological Sciences. 2004;271:206–208. doi: 10.1098/rsbl.2003.0135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Shibayama Y, Kadono Y. Reproductive success and genetic structure of populations of the heterostylous aquatic plant Nymphoides indica (L.) Kuntze (Menyanthaceae) Aquatic Botany. 2007;86:1–8. [Google Scholar]
  47. Silvertown J. The evolutionary maintenance of sexual reproduction: evidence from the ecological distribution of asexual reproduction in clonal plants. International Journal of Plant Sciences. 2008;169:157–168. [Google Scholar]
  48. Swofford DL. PAUP*. Phylogenetic Analysis Using Parsimony (*and other methods) Sunderland, MA: Sinauer; 2002. [Google Scholar]
  49. Thompson SL, Choe G, Ritland K, Whitton J. Cryptic sex within male-sterile polyploid populations of the Easter daisy, Townsendia hookeri. International Journal of Plant Sciences. 2008;169:183–193. [Google Scholar]
  50. Uesugi R, Gokab K, Nishihiro J, Washitani I. Allozyme polymorphism and conservation of the Lake Kasumigaura population of Nymphoides peltata. Aquatic Botany. 2004;79:203–210. [Google Scholar]
  51. Vallejo-Marín M, O'Brien HE. Correlated evolution of self-incompatibility and clonal reproduction in Solanum (Solanaceae) New Phytologist. 2007;173:415–421. doi: 10.1111/j.1469-8137.2006.01924.x. [DOI] [PubMed] [Google Scholar]
  52. Vallejo-Marín M, Dorken ME, Barrett SCH. The ecological and evolutionary consequences of clonality for plant mating. Annual Review of Ecology, Evolution, and Systematics. 2010;41:193–213. [Google Scholar]
  53. Wang Y, Wang Q-F, Guo Y-H, Barrett SCH. Reproductive consequences of interactions between clonal growth and sexual reproduction in Nymphoides peltata: a distylous aquatic plant. New Phytologist. 2005;165:329–336. doi: 10.1111/j.1469-8137.2004.01234.x. [DOI] [PubMed] [Google Scholar]
  54. Webb CJ, Lloyd DG. The avoidance of interference between the presentation of pollen and stigmas in angiosperms II. Herkogamy. New Zealand Journal of Botany. 1986;24:163–178. [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES