Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2012 Oct 15;110(7):1439–1447. doi: 10.1093/aob/mcs211

Exogenous selection shapes germination behaviour and seedling traits of populations at different altitudes in a Senecio hybrid zone

Rebecca I C Ross 1,†, J Arvid Ågren 2,‡, John R Pannell 1,3,*
PMCID: PMC3489152  PMID: 23071216

Abstract

Background and Aims

The Senecio hybrid zone on Mt Etna, Sicily, is characterized by steep altitudinal clines in quantitative traits and genetic variation. Such clines are thought to be maintained by a combination of ‘endogenous’ selection arising from genetic incompatibilities and environment-dependent ‘exogenous’ selection leading to local adaptation. Here, the hypothesis was tested that local adaptation to the altitudinal temperature gradient contributes to maintaining divergence between the parental species, S. chrysanthemifolius and S. aethnensis.

Methods

Intra- and inter-population crosses were performed between five populations from across the hybrid zone and the germination and early seedling growth of the progeny were assessed.

Key Results

Seedlings from higher-altitude populations germinated better under low temperatures (9–13 °C) than those from lower altitude populations. Seedlings from higher-altitude populations had lower survival rates under warm conditions (25/15 °C) than those from lower altitude populations, but also attained greater biomass. There was no altitudinal variation in growth or survival under cold conditions (15/5 °C). Population-level plasticity increased with altitude. Germination, growth and survival of natural hybrids and experimentally generated F1s generally exceeded the worse-performing parent.

Conclusions

Limited evidence was found for endogenous selection against hybrids but relatively clear evidence was found for divergence in seed and seedling traits, which is probably adaptive. The combination of low-temperature germination and faster growth in warm conditions might enable high-altitude S. aethnensis to maximize its growth during a shorter growing season, while the slower growth of S. chrysanthemifolius may be an adaptation to drought stress at low altitudes. This study indicates that temperature gradients are likely to be an important environmental factor generating and maintaining adaptive divergence across the Senecio hybrid zone on Mt Etna.

Keywords: Senecio aethnensis, S. chrysanthemifolius, hybrid, germination, exogenous selection, early life history, temperature

INTRODUCTION

Most hybrid zones are thought to be maintained in a balance between natural selection and the diffusion of genes by dispersal across the species boundary (Moore, 1977; Barton and Hewitt, 1985). But what is the nature of such selection, and on what traits does it act? On the one hand, selection might be ‘exogenous’, whereby fitness depends on the ecological context in which genes are expressed (G × E interactions) (Endler, 1977; Moore, 1977; Harrison and Rand, 1989; Barton and Gale, 1993). On the other hand, selection might be ‘endogenous’, with the fitness of genes depending on the genomic context in which they are expressed (G × G interactions). Here, genes are selected against when they are expressed in the background of a different species' genome (Dobzhansky, 1936, 1937; Muller, 1942; Mayr, 1963; Barton and Hewitt, 1985; Barton, 2001). In practice, these two modes of selection are not mutually exclusive and they result in very similar clines in allele frequencies across a hybrid zone (Kruuk et al., 1999). A key question is therefore the relative extent to which endogenous vs. exogenous selection shapes the genetic and phenotypic clines we observe in a particular hybrid setting.

Recent studies in plants using reciprocal transplant experiments in hybrid zones across environmental gradients have found strong evidence for exogenous selection in terms of significant genotype by environment (G × E) interactions, but there is comparatively weak evidence for endogenous selection (including Wang et al., 1997; Miglia et al., 2005; Campbell et al., 2008; Kameyama et al., 2008; Kimball et al., 2008; Taylor et al., 2009). While it is accepted that environmental clines can contribute to the maintenance of hybrid zones, identifying the specific environmental variable responsible has been challenging (Campbell, 2004; Angert, 2006). In contrast to the number of studies considering the role of biotic variation, the extent to which abiotic gradients specifically play a role in shaping plant hybrid zones has been surprisingly little studied. Temperature is one of the primary determinants of species distributions on a global scale and so may also determine species distributions across a smaller spatial scale such as a narrow hybrid zone. Altitudinal gradients potentially provide a fruitful context in which to investigate this effect as ambient temperature decreases by approx. 5·5 °C per 1000 m increase in altitude (Körner, 2007), and this temperature change affects a wide range of life-history traits, including growth rates and germination behaviour (Körner and Larcher, 1988; Baskin and Baskin, 1998).

Mortality at the seed and seedling stages tends to be high (Harper, 1977; Kitajima and Fenner, 2000), meaning that selection acting on early life-history traits might have a disproportionate effect on gene flow across a hybrid zone. For logistical reasons, field experiments measuring exogenous selection often start with established seedlings, yet the direction of selection at these early stages of development need not match that acting at later stages (Johnston et al., 2003). Therefore field studies that start with seedlings may fail to find evidence for the role of exogenous selection in shaping a cline even when it is present (Gimenez-Benavides et al., 2008). Common garden experiments with seeds and seedlings are a useful way to explore which environmental factors affect fitness-related traits, indicating how exogenous selection might act.

Here, we investigate possible divergence of early life-history traits between the groundsel species Senecio aethnensis and S. chrysanthemifolius, which hybridize across a steep altitudinal gradient on the slopes of Mt Etna in Sicily, Italy. Both species are diploid (2n = 20), herbaceous, self-incompatible, short-lived perennials (Crisp, 1972). The main hybrid zone between them, confirmed by RAPD and chloroplast haplotype analysis (James and Abbott, 2005), lies along a road on the southern slope of the volcano, between about 1300 and 1800 m a.s.l. S. aethnensis occurs between 2000 and 3500 m a.s.l. and S. chrysanthemifolius occurs from sea level up to about 1000 m (Fig. 1). Numerous phenotypic traits vary between the two species across the hybrid zone, including leaf dissection (see Fig. 1), capitulum size and achene length, and patterns of gene expression (Ronsisvalle, 1968; Crisp, 1972; Abbott et al., 2000; Hegarty et al., 2009). Given the steep environmental gradient across which the hybrid zone occurs, it seems likely that exogenous selection plays a role in maintaining species integrity, but the nature of this selection remains unknown. Recent clinal analysis suggests that endogenous selection against hybrids has probably also been important in shaping the hybrid zone (Brennan et al., 2009), but there is as yet no direct evidence for reduced fitness in hybrid genotypes.

Fig. 1.

Fig. 1.

Open in a new tab

Sample populations of S. chrysanthemifolius, S. aethnensis and their hybrids on Mt Etna, Sicily. Senecio chrysanthemifolius is found up to 1000 m, the hybrid zone lies between 1000 and 2000 m, and S. aethnensis is found above 2000 m. The insert shows the variation in leaf morphology between the parental species and their hybrids. See Table 1 for an explanation of population codes.

Our study tests the hypothesis that germination behaviour and early seedling growth is sensitive to endogenous selection by comparing the performance of F1 hybrids and a natural hybrid population with that of the parental species across a range of temperatures. G × G interactions result in deviations from the expected result under an additive genetic architecture (Lynch and Walsh, 1998). We therefore expect that, if intergenomic or cytonuclear incompatibilities are weak or absent, the trait values of the F1 hybrid will equal the pooled average of the parents, while negative G × G interactions would be manifested by poorer F1 performance than predicted from the mid-parent average (Rhode and Cruzan, 2005). Alternatively, hybridization may lead to heterosis, whereby F1 trait values deviate in a positive direction from the mid-parental average (Falconer, 1981). The most common cause of heterosis is probably dominance or overdominance at loci where, as a result of inbreeding or genetic drift, inferior alleles have become fixed in the parental species (Whitlock et al., 2000; Burke and Arnold, 2001). Interestingly, microarray studies in the Senecio species complex studied here have also identified non-additive changes in gene expression in early-generation hybrids (Hegarty et al., 2009); such changes may also result in heterosis if they translate into observable changes in phenotype. As the level of heterozygosity decreases each generation after hybridization, we expect the natural hybrid population not to show heterosis (Rhode and Cruzan, 2005). Non-additive negative performance in the natural hybrid combined with non-additive positive performance in the F1 indicates both heterosis and hybrid breakdown. We also investigate whether populations of the two parental species differ in their response to different temperatures in terms of their seed germination behaviour, seedling growth and seedling survival, which gives an indication whether exogenous selection might be involved in structuring the hybrid zone.

MATERIALS AND METHODS

Sampling and seed material

We collected seeds of Senecio aethnensis and S. chrysanthemifolius by sampling at least 30 individuals from each of five populations on the southern slope of Mt Etna, including the parental species and three hybrid populations from intermediate altitudes and with intermediate morphology (Table 1 and Fig. 1). To minimize possible maternal effects on germination and seedling growth, we grew plants from each population to maturity in a common glasshouse at the Department of Plant Sciences, University of Oxford, and crossed them to produce seed for our experiment. Genetic background had previously been established at population level by genetic assays (James and Abbott, 2005; Brennan et al., 2009). Leaf morphology of glasshouse-grown plants supported the designation of populations as either hybrid or parental.

Table 1.

Locality information for the sample populations used to generate seed material for the experiments

Population Description Altitude (m) Latitude (N) Longitude (E)
CC S. chrysanthemifolius 749 37°36·861′ 15°00·961′
CH S. chrysanthemifolius-like hybrid 1532 37°40·783′ 14°58·901′
IH Intermediate hybrid 1624 37°41·128′ 14°59·215′
AH S. aethnensis-like hybrid 1820 37°41·669′ 14°59·505′
AA S. aethnensis 2097 37°42·409′ 14°59·830′
Open in a new tab

Genetic assays confirmed these populations to be either largely parental or hybrid genotypes (James and Abbott, 2005; Brennan et al., 2009). Populations were collected from the road side or on recent lava flows, growing in fine volcanic soil and generally unshaded. A map of the region is shown in Fig. 1.

Plants were established as single individuals from each of ten half-sib families per population in 5-inch pots containing a 3 : 1 mixture of peat-based compost and grit. Ten blocks were set up, each containing one replicate plant from each population. The glasshouse was maintained on a 16-h photoperiod with a constant temperature of approx. 22 °C, and watering was performed as necessary. Crosses were performed by rubbing ripe inflorescences together, as described in Hiscock (2000). The parental and natural hybrid populations were replicated through crosses among individuals within each population (intra-population crosses). The first generation seed produced from these crosses was used in the subsequent experiments to reduce maternal effects. Between seven and 15 intra-population crosses per natural population were performed. A similar number of inter-population crosses between the two parental populations were performed to generate the F1 hybrid cross-type. The F1 cross-type and the progeny of intra-population crosses are collectively referred to below as ‘genetic classes’.

Seed germination experiment

We established bulked seed samples of 180 seeds for each genetic class. Seeds were stratified by sowing on moist filter paper and placing in the dark at 4 °C for 1 week. A linear thermal gradient was set up between 7 °C and 35 °C using a one-metre temperature gradient bar and monitored at 17-cm intervals with a thermocouple array (Berti and Johnson, 2008). Temperatures were fixed without oscillation. Two lanes were randomly allocated to each genetic class and 90 seeds sown at equal spacing along each lane. Seeds were watered daily and lighting provided by fluorescent strip lights on a 12-h photoperiod. Germination was recorded daily as the time at which cotyledons were fully expanded. The experiment was run for 35 d, at which point almost all the seeds in the central part of the bar had germinated.

Seedling growth experiment

Seedlings were collected from the temperature bar once they had germinated, transplanted to 3-inch pots containing peat-based compost and randomly allocated to warm (25 °C/15 °C) or cold (15 °C/5 °C) growth cabinets (SANYO-Biomedical Europe MLR plant growth chambers) over a period of 16 d. Although it would have been preferable to use only individuals that germinated at the same temperature, the temperature at which individuals germinated will not have influenced the results because individuals were randomly allocated to the seedling growth treatments. Between 26 and 41 seedlings (mean = 35) were transplanted for each treatment × genetic class combination. The temperatures used for the warm and cold treatments were based on weather station data in the habitat of S. aethnensis (cold) and S. chrysanthemifolius (warm) during spring (available from Meteo Sicilia). All cabinets were set to 75 % humidity and a 12-h photoperiod. Pots were randomized within cabinets daily. As there may be other non-measured differences between cabinets, cabinets were allocated to different temperature regimes each week (leaving other conditions the same) and pots were moved between cabinets so that they maintained the same treatment. Pots were watered daily by irrigating the capillary matting beneath them.

Seedlings were harvested after 83 d (stratified by potting date). At harvest, the above-ground height was measured and chlorophyll content was estimated using a Minolta SPAD-502 meter (Kariya et al., 1982) on a living mid-cauline leaf, which was then air-dried and weighed using a Sartorius 1212 balance. The dry above-ground biomass was also weighed. Seedlings that died before harvesting were also recorded to determine the mortality rate. Mortality rate was only assessed on seedlings that had survived to the first true leaf stage, to exclude death due to damage incurred on transplantation as this was assumed to be independent of seed source.

Data analysis

All statistical analyses were implemented in the statistical software package, R, versions 2·12·1 to 2·14·1 (R Development Core Team, 2010–2011). The final germination proportion was compared between the two parental genetic classes (AA and CC) in a generalized linear model (GLM) with binomial errors. Linear orthogonal treatment contrasts were fitted to compare the F1 and intermediate altitude natural hybrid (IH) to the expected mid-parent value to test whether the traits being investigated conformed to an additive genetic structure (Campbell and Waser, 2001). Throughout the Results section, only the intermediate altitude hybrid is statistically compared with the F1 in this manner. Final germination proportion was analysed in a GLM with quasibinomial errors. Final biomass, height, leaf mass and SPAD-value were analysed using ANOVA. Final biomass, leaf mass and SPAD-value were transformed to help meet the assumptions of the Normal model, as suggested by the boxcox function in R. Seedling survival of the F1 and IH was compared against the combined survival of the parental genetic classes using a proportion test.

To investigate altitudinal variation in life-history traits, trait values were regressed on the collection altitude of the field-sampled populations that were used as the parents of the experimental generation (not including the F1 cross-type). Low temperature seed germination responses were investigated by stratifying the data into 2 °C temperature intervals from 7 to 19 °C and regressing the germination proportion for each interval on altitude of seed source. To account for non-independence of seeds sampled from the same population, a single value for the germination proportion was calculated for each population and temperature interval combination, logit transformed [log y/(1 – y)] (Warton and Hui, 2011) and regressed on altitude, rather than using a GLM with binomial errors. Pearson's product moment correlation was also calculated.

Seedling traits were analysed by ANCOVA, with altitude of seed source and temperature treatment as the independent variables. Where interaction terms were significant, the data for each treatment were independently regressed on altitude. To account for non-independence, we used a mixed-effects model, with population fitted as a random effect and altitude and temperature fitted as fixed effects. We used the lme package in R for linear mixed-model analyses and the MCMCglmm package to analyse seedling survival rates (Hadfield, 2010). Significance of fixed effects for lme models was tested by comparing the maximum likelihood fits of full and reduced models using a likelihood ratio test. In addition, the mean values for each population were regressed on the altitude of the source populations.

Phenotypic plasticity was quantified for each genetic class using the Relative Distances Plasticity Index (RDPI), which gives a value between 0 (no plasticity) and 1 (maximal plasticity) (Valladares et al., 2006):

graphic file with name mcs211eqnU1.jpg

where dij is the absolute value of the distance between trait values (x) for all pairs of individuals (j) within a genetic class grown under different temperature regimes (i) and n is the number of distances. Genetic-class-level RDPI was compared between traits using an ANOVA and compared between populations by linear regression.

RESULTS

Intrinsic genetic effects

Of the 42 inter-population crosses performed to generate the F1, 37 crosses were successful in at least one direction. Crosses seemed to be equally successful independent of the species of the maternal and paternal parent, i.e. the direction of crossing did not appear to affect success rates. S. chrysanthemifolius was the maternal parent for 11 of the 25 families used in the experiment. Thus there is no clear evidence for asymmetric pre-zygotic incompatibility.

For most seed and seedling traits, trait values of both natural and experimental F1 hybrid genotypes were intermediate between the parental values, with the exception of seedling growth and survival under cold conditions (Tables 2 and 3). Senecio chrysanthemifolius germinated significantly less well than S. aethnensis across the range of experimental temperatures (t = –3·26, P = 0·02; Fig. 2). Germination proportion of F1 and natural hybrid seeds was equal to the mid-parent value (F1: t = –0·65, P = 0·53; IH: t = 0·48, P = 0·64). Average time to germination was similar for all the genetic classes (ranging between 7 and 10 d) and the F1 and natural hybrid values were between the parental values. F1 and natural hybrid class means were not different from the expected mid-parent value for seedling biomass, height, leaf mass, chlorophyll content (SPAD value) and survival under warm conditions (Table 3). Under cold conditions, the mean biomass, height, and leaf mass of the F1 class exceeded the mid-parent average (P < 0·04 for all traits). The natural hybrid class mean was lower than the mid-parent average for height (P = 0·04) and survival rate (P < 0·01). F1 and natural hybrid class means were not different from the expected mid-parent value for chlorophyll content.

Table 2.

Comparison of germination characteristics between genetic classes

Genetic class Germination proportion Average germination time (d) Germination range (°C)
CC 0·51 10·2 (0·6) 19·3
CH 0·76 8·4 (0·5) 23·8
IH 0·64 8·2 (0·6) 25·6
AH 0·73 8·1 (0·5) 25·3
AA 0·82 7·2 (0·5) 24·7
F1 0·64 9·2 (0·5) 23·8
Open in a new tab

Average germination time is given with standard errors in brackets.

Table 3.

Comparison of the performance of natural and F1 hybrid seedlings to seedlings of the parental species under cold (15/5 °C) and warm (25/15 °C) growing conditions

(A) Trait means
Trait means and standard error
Treatment Genetic class Biomass (mg) Height (mm) Leaf mass (mg) Chlorophyll content Survival (%)
Cold AA 23·2 (3·5) 53·4 (3·1) 3·6 (0·5) 34·4 (1·2) 100
CC 30·6 (8·1) 52·8 (5·9) 4·6 (0·8) 30·4 (1·7) 84
F1 38·1 (5·4) 60·5 (3·0) 4·9 (0·6) 34·1 (1·1) 78
IH 19·4 (2·7) 44·5 (2·7) 2·8 (0·3) 30·9 (1·4) 62
Warm AA 381·9 (51·1) 134·5 (7·4) 20·0 (2·2) 40·3 (2·1) 52
CC 271·1 (56·3) 104·3 (7·9) 15·5 (1·8) 36·8 (1·8) 96
F1 363·7 (62·7) 134·1 (7·2) 21·7 (3·0) 41·9 (2) 57
IH 413·1 (52·3) 122·7 (7·2) 17·7 (1·7) 41·4 (1·5) 50
(B) Linear treatment contrasts
Trait being compared
Treatment Genetic class Biomass Height Leaf mass Chlorophyll content Survival
Cold F1 Statistic 2·64 2·75 2·33 1·02 2·8
P <0·01 <0·01 0·02 0·31 0·09
IH Statistic 1·3 2·45 1·68 0·63 8·8
P 0·2 0·02 0·1 0·53 <0·01
Warm F1 Statistic –0·40 –1·56 1·30 1·25 1·40
P 0·69 0·12 0·20 0·22 0·23
IH Statistic –1·43 –0·02 0·44 –0·85 3·55
P 0·16 0·98 0·66 0·40 0·06
Open in a new tab

Part (A) gives the trait means for each genetic class (standard errors in brackets); part (B) presents the results of linear treatment contrasts between the natural and F1 hybrid genetic class to the mid-parent average.

The test statistic is t, except survival where statistic is χ2 (proportion test). Bold indicates P-values <0·05.

Fig. 2.

Fig. 2.

Open in a new tab

Germination proportions by the end of the experiment stratified by temperature for each genetic class. F1 hybrids (open symbols) are plotted at 1600 m a.s.l. for display purposes. Low temperatures = 7–14 °C; mid-temperatures = 14–26 °C; high temperatures = 26–35 °C.

Temperature effects on seed germination

Seeds from high-altitude populations were better able to germinate at low temperatures than seeds from low-altitude populations. Germination rate followed a negative quadratic relationship with temperature (χ2 = 358·85, d.f. = 2, P < 0·0001), and the shape of this relationship varied between populations (χ2 = 125·22, d.f. = 12, P < 0·0001). Germination occurred over a greater range of temperatures in higher-altitude populations (R2 = 0·68, P = 0·001; Table 2). Germination proportion at moderately low temperatures (9–11 °C interval and 11–13 °C interval) increased with altitude, whereas germination success was similar across all populations at very low temperatures (7–9 °C) and at higher temperatures (13–19 °C; Table 4 and Fig. 3). The natural hybrid and F1 genetic classes were intermediate between the parental species in their ability to germinate across the 7–19 °C range (F1: t = –0·94, P = 0·34; IH: t = 0·37, P = 0·72).

Table 4.

Comparison of germination responses to different temperatures in Senecio populations sampled from different altitudes

Temperature interval (°C) F d.f. P Pearson's r P
7–9 3·11 1,3 0·18 0·60 0·28
9–11 42·62 1,3 <0·01 0·98 <0·01
11–13 16·75 1,3 0·03 0·98 <0·001
13–15 3·55 1,3 0·16 0·59 0·30
15–17 3·50 1,3 0·16 0·27 0·66
17–19 4·53 1,3 0·12 0·74 0·15
Open in a new tab

For each 2 °C interval, germination proportion per genetic class was regressed on the sampling altitude (expressed in units of 100 m increase in altitude) and Pearson's product moment correlation also calculated. Bold indicates P-values <0·05.

Fig. 3.

Fig. 3.

Open in a new tab

Germination proportion at different temperatures. Each point represents germination proportion over a 2 °C window for one genetic class and the lines are the predictions of a GLM of germination proportion on altitude. F1 hybrids are plotted at 1600 m a.s.l. for display purposes and represented by open symbols; temperature ranges are indicated in the key (17–19 °C values are displaced slightly on the x-axis as points are overlapping). The 9–11 °C and 11–13 °C intervals exhibited a significant correlation with altitude (P < 0·01, P = 0·03); other intervals were uncorrelated (P > 0·1). The 13–15 °C and 15–17 °C intervals have not been plotted but were also not correlated (P > 0·1).

Temperature effects on seedling survival and growth

Seedlings from high-altitude populations grew better under warm conditions than those from low-altitudes but also suffered higher mortality rates; in contrast, there were no significant trends with altitude under cold conditions. A lower proportion of seedlings survived between the one leaf stage and harvest in the warm treatment compared with the cold treatment (MCMC P = 0·002; Fig. 4). In the warm treatment, the proportion of seedlings surviving decreased with increasing altitude of sampling site, but the difference fell short of significance (MCMC P = 0·088). In the cold treatment, there was no altitude effect (MCMC P = 0·394).

Fig 4.

Fig 4.

Open in a new tab

Proportion of seedlings surviving to harvest in cold (15/5 °C) and warm (25/15 °C) growing conditions, as indicated in the key. F1 hybrids are plotted at 1600 m a.s.l. for display purposes and represented by open symbols. The lines are the predictions of a GLM of survival on altitude under warm (P = 0·09) and cold (P = 0·39) conditions.

Seedlings from all source populations reached greater biomass in the warm treatment than the cold treatment (P < 0·0001; Fig. 5). Seedlings from high-altitude populations had greater biomass than those from low-altitudes in the warm treatment (P = 0·01), while there was no altitude effect in the cold treatment. Height and leaf mass displayed similar patterns (Table 5). The SPAD-value of leaves, a measure of chlorophyll content, was significantly greater in warmer temperatures and was not affected by altitude (Table 5). There was a marginally significant trend for high-altitude populations to exhibit greater temperature-dependent plasticity of growth-related traits (traits = dry weight, height and leaf mass; P = 0·07, 0·09, and 0·05, respectively). Phenotypic plasticity varied significantly between traits (F4,25 = 179·5, P < 0·001). Biomass was the most plastic trait while chlorophyll content was quite unresponsive to growing temperature (Fig. 6).

Fig. 5.

Fig. 5.

Open in a new tab

Above-ground biomass of seedlings in cold (15/5 °C) and warm (25/15 °C) growing conditions. Squares and circles represent cold and warm, respectively. Open and closed symbols represent hybrid and natural populations, respectively.

Table 5.

Effect of temperature, seed source altitude and their interaction on seedling above-ground dry biomass; seedling height, dry mass of one mid-cauline leaf, and chlorophyll content of a mid-cauline leaf (estimated from the SPAD-value) under cold (15/5 °C) and warm (25/15 °C) growing conditions

Trait ANCOVA
 Regression on altitude
Temperature
 Interaction
 Warm
 Cold
LR P LR P LR P P* LR P P*
Biomass 220·31 <0·0001 8·40 <0·01 6·21 0·01 0·07 0·22 0·64 0·60
Height 212·29 <0·0001 10·22 <0·01 8·93 <0·01 <0·01 0·02 0·88 0·94
Leaf mass 211·86 <0·0001 10·25 <0·01 6·85 <0·01 0·02 1·42 0·23 0·29
Chlorophyll 46·52 <0·001 <0·01 0·98 2·10 0·15 0·60 2·71 0·10 0·93
Open in a new tab

P is the significance given by a likelihood ratio (LR) test with 1 d.f. and P* is the P-value obtained from an F-test using the mean values for each genetic class. Bold indicates P-values<0·05.

Fig. 6.

Fig. 6.

Open in a new tab

Population-level plasticity of growth traits compared between parental and natural hybrid populations (filled symbols) and experimental hybrids (open symbols) using the relative distances plasticity index, where 0 = no plasticity and 1 = maximal plasticity. Growth traits are above-ground biomass, height, mass of one mid-cauline leaf, and mid-cauline leaf chlorophyll content (as indicated in the key). Asterisks indicate significant relationships with altitude: **P = 0·05; *0·05 < P < 0·09.

DISCUSSION

Our study has revealed significant G × E interactions between temperature and sampling altitude for both seed germination and early seedling growth in experimental and natural hybrids from the Senecio hybrid zone on Mt Etna. In contrast, we found limited evidence for negative G × G interactions. Our results thus point to the likely role of exogenous selection and the evolution of local adaptation across a steep environmental gradient in structuring this plant hybrid zone.

Our study found no convincing evidence that hybrid genotypes (whether natural late-generation hybrids or experimentally generated F1 genotypes) have uniformly lower fitness than parental genotypes. Instead, seed germination rates and, for the majority of trait and treatment combinations, the seedling traits of natural hybrids were intermediate between the parental species. Heterosis can mask hybrid unfitness, particularly in early-generation hybrids (Fenster and Galloway, 2000; Johansen-Morris and Latta, 2006; Rhode and Cruzan, 2005). However, this is unlikely to be the case here, as trait values in experimentally generated F1 hybrids did not exceed the average of the two parental species for most treatment and trait combinations.

Instead, the impact of endogenous selection may be environmentally dependent. The seedling survival and height of intermediate altitude natural hybrids was lower than predicted from the parental performance under cold conditions but not under warm conditions. In contrast, we found evidence for F1 heterosis for several growth-related traits, also under cold conditions only. This suggests that the effects of deleterious recessive mutations on parental fitness (which give rise to heretosis in first-generation hybrids) might only be revealed under more stressful conditions (Parsons, 1971; Charlesworth and Charlesworth, 1987; Bijlsma et al., 2001). In any case these contrasting results suggest that plant performance across the Senecio hybrid zone on Mt Etna depends on interactions with the environment (G × E interactions) and is not solely determined by endogenous factors (G × G interactions).

We found that high- and low-altitude populations exhibited divergent seed germination responses at low temperatures. Specifically, we found that the minimum temperature at which germination occurred was lower for seeds derived from high altitude populations than seeds derived from low-altitude populations and, between 9 °C and 13 °C, seeds from higher-altitude populations germinated better than lower altitude populations. This pattern, also observed in the herb Reynoutria japonica on Mt Fuji, Japan (Mariko et al., 1993), probably reflects adaptation to the decrease in spring temperature with altitude.

Under the warm temperature regime in our experiment, seedlings from populations more likely to experience warm temperatures in the field displayed higher survivorship than populations sampled from colder sites (although the difference was only marginally significant). In contrast, we found no evidence that high-altitude genotypes were specially adapted to cold temperatures. Our results are similar to those found for a Mimulus altitudinal hybrid zone, where low-altitude M. cardinalis showed greater survival than high-altitude M. lewisii when grown experimentally under a warm temperature regime representative of the native site of M. cardinalis, and differences in survival were observed under warm conditions but not cold (Angert, 2006).

Contrary to our expectation of local adaptation, seedlings from higher-altitude populations grew faster under warm conditions than seedlings from lower-altitude populations. This would suggest that plasticity of growth-related traits tends to increase with altitude in these species. The patterns observed in this study contrast with those typically found in temperate altitudinal gradients, where alpine plants tend to be less plastic than related low-altitude species (Körner and Larcher, 1988). We propose that this difference stems from the additional constraint of drought and light stress faced in Mediterranean ecosystems (Calloway et al., 2002; Gimenez-Benavides et al., 2005).

The combination of low temperature seed germination, rapid seedling growth and temperature-dependent growth plasticity observed in S. aethnensis may reflect an adaptation to high altitudes by enabling individuals to capitalize on short periods of favourable weather. In Mediterranean ecosystems, increases in altitude can coincide with a shift in the major environmental stress facing plants, from drought in the lowlands (Thompson, 2005; Padilla and Pugnaire, 2007; Gimenez-Benavides et al., 2008) to chilling and frost damage in montane environments (Mariko et al., 1993; Galen and Stanton, 1999). In montane habitats, germination at low temperatures and the ability to grow quickly allows plants to reach a size required to flower in their first year, or to survive the next winter (Harper, 1977; Mariko et al., 1993; Vera, 1997; Shimono and Kudo, 2003, Milla et al., 2009). For lowland Mediterranean plants, establishment success depends on an ability to avoid water shortage during the summer months, and slow above-ground growth may thus reflect increased allocation to roots as part of a stress-tolerance strategy (Volis et al., 2002; Padilla and Pugnaire, 2007; Wu et al., 2008).

Previous studies on the relative importance of endogenous versus exogenous selection in maintaining the Senecio hybrid zone on Mt Etna have reached conflicting conclusions (James and Abbott, 2005; Brennan et al., 2009). On the one hand, the steep environmental gradient and fertility of hybrids points to a key role for exogenous selection (James and Abbott, 2005); on the other hand, the genetic and phenotypic clines across the zone better match the predictions of an endogenous selection model (Brennan et al., 2009). While the present study does not resolve this issue, we have uncovered significant G × E interactions that affect seed and seedling traits and that are likely to translate into differences in the relative fitness of both parents and hybrids with variation in ambient temperature across the hybrid zone. The combination of early-season seed germination and rapid seedling growth might enable high-altitude genotypes to capitalize on periods of favourable weather, which is adaptive at high altitude as the growing season is shorter, but may reduce survival at low altitudes. In our study, putatively adaptive traits, such as germinability at low temperatures or temperature-dependent growth plasticity, varied in such a way as to approximately track the temperature gradient across the Mt Etna hybrid zone. Hybrids were therefore likely to exhibit more adaptive phenotypes at intermediate altitudes, i.e. the centre of the zone, than pure parental genotypes. These patterns are consistent with the suggestions of Anderson (1948) and Moore (1977) that hybrids might flourish in disturbed habitats in environments intermediate to those occupied by their parental species, referring to the role in hybrid maintenance played by ‘hybridization of the habitat’. Certainly, our results point to a potentially important role for exogenous selection acting on the early life stages in maintaining the hybrid zone.

ACKNOWLEDGEMENTS

We thank Richard Abbott and Adrian Brennan (St Andrews University, Scotland) for assistance with sourcing seed material and Richard Abbott for use of the temperature bar. This work was supported by a postgraduate research studentship from the Gatsby Charitable Foundation to R.I.C.R. and grants from NERC, UK, to J.R.P.

LITERATURE CITED

  1. Abbott RJ, James JK, Irwin JA, Comes HP. Hybrid origin of the Oxford ragwort, Senecio squalidus L. Watsonia. 2000;23:123–138. [Google Scholar]
  2. Anderson E. Hybridization of the habitat. Evolution. 1948;2:1–9. [Google Scholar]
  3. Angert AL. Growth and leaf physiology of monkey flowers with different altitude ranges. Oecologia. 2006;148:183–194. doi: 10.1007/s00442-006-0361-z. [DOI] [PubMed] [Google Scholar]
  4. Barton NH. The role of hybridization in evolution. Molecular Ecology. 2001;10:551–568. doi: 10.1046/j.1365-294x.2001.01216.x. [DOI] [PubMed] [Google Scholar]
  5. Barton NH, Gale RS. Genetic analysis of hybrid zones. In: Harrison RG, editor. Hybrid zones and the evolutionary process. New York, NY: Oxford University Press; 1993. [Google Scholar]
  6. Barton NH, Hewitt GM. Analysis of hybrid zones. Annual Review of Ecology and Systematics. 1985;16:113–148. [Google Scholar]
  7. Baskin CC, Baskin JM. Seeds: ecology, biogeography, and evolution of dormancy and germination. San Diego, CA: Academic Press; 1998. [Google Scholar]
  8. Berti MT, Johnson BL. Seed germination response of cuphea to temperature. Industrial Crops and Products. 2008;27:17–21. [Google Scholar]
  9. Bijlsma R, Bundgaard J, Van Putten WF. Environmental dependence of inbreeding depression and purging in Drosophila melanogaster. Journal of Evolutionary Biology. 1999;12:1125–1137. [Google Scholar]
  10. Brennan AC, Bridle JR, Wang A, Hiscock SJ, Abbott RJ. Adaptation and selection in the Senecio (Asteraceae) hybrid zone on Mount Etna, Sicily. New Phytologist. 2009;183:702–717. doi: 10.1111/j.1469-8137.2009.02944.x. [DOI] [PubMed] [Google Scholar]
  11. Burke JM, Arnold ML. Genetics and the fitness of hybrids. Annual Review of Genetics. 2001;35:31–52. doi: 10.1146/annurev.genet.35.102401.085719. [DOI] [PubMed] [Google Scholar]
  12. Callaway RM, Brooker RW, Choler P, et al. Positive interactions among alpine plants increase with stress. Nature. 2002;417:844–848. doi: 10.1038/nature00812. [DOI] [PubMed] [Google Scholar]
  13. Campbell DR. Natural selection in Ipomopsis hybrid zones: implications for ecological speciation. New Phytologist. 2004;161:83–90. [Google Scholar]
  14. Campbell DR, Waser NM. Genotype-by-environment interaction and the fitness of plant hybrids in the wild. Evolution. 2001;55:669–676. doi: 10.1554/0014-3820(2001)055[0669:gbeiat]2.0.co;2. [DOI] [PubMed] [Google Scholar]
  15. Campbell DR, Waser NM, Aldridge G, Wu CA. Lifetime fitness in two generations of Ipomopsis hybrids. Evolution. 2008;62:2616–2627. doi: 10.1111/j.1558-5646.2008.00460.x. [DOI] [PubMed] [Google Scholar]
  16. Charlesworth B, Charlesworth D. Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics. 1987;18:237–268. [Google Scholar]
  17. Crisp PC. Cytotaxonomic studies in the Section Annui of Senecio. 1972. PhD Thesis, University of London. [Google Scholar]
  18. Dobzhansky T. Studies on hybrid sterility. II. Localisation of sterility factors in Drosophila pseudoobscura hybrids. Genetics. 1936;21:113–135. doi: 10.1093/genetics/21.2.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Dobzhansky T. Genetics and the origin of species. New York, NY: Columbia University Press; 1937. [Google Scholar]
  20. Endler JA. Geographic variation, speciation and clines. Princeton, NJ: Princeton University Press; 1977. [PubMed] [Google Scholar]
  21. Falconer DS. Introduction to quantitative genetics. London: Longman; 1981. [Google Scholar]
  22. Fenster CB, Galloway LF. Population differentiation in an annual legume: genetic architecture. Evolution. 2000;54:1157–1172. doi: 10.1111/j.0014-3820.2000.tb00551.x. [DOI] [PubMed] [Google Scholar]
  23. Galen C, Stanton ML. Seedling establishment in alpine buttercups under experimental manipulations of growing-season length. Ecology. 1999;80:2033–2044. [Google Scholar]
  24. Gimenez-Benavides L, Escudero A, Perez-Garcia F. Seed germination of high mountain Mediterranean species: altitudinal, interpopulation and interannual variability. Ecological Research. 2005;20:433–444. [Google Scholar]
  25. Gimenez-Benavides L, Escudero A, Iriondo JM. What shapes the altitudinal range of a high mountain Mediterranean plant? Recruitment probabilities from ovule to seedling stage. Ecography. 2008;31:731–740. [Google Scholar]
  26. Hadfield J. MCMC methods for Multi-response Generalised Linear Mixed Models: the MCMCglmm R Package. Journal of Statistical Software. 2010;33:1–22. [Google Scholar]
  27. Harper JL. Population biology of plants. New York, NY: Academic Press; 1977. [Google Scholar]
  28. Harrison RG, Rand DM. Mosaic hybrid zones and the nature of species boundaries. In: Otte D, Endler JA, editors. Speciation and its consequences. Sunderland, MA: Sinauer; 1989. pp. 111–133. [Google Scholar]
  29. Hegarty MJ, Barker GL, Brennan AC, Edwards KJ, Abbott RJ, Hiscock SJ. Extreme changes to gene expression associated with homoploid hybrid speciation. Molecular Ecology. 2009;18:877–889. doi: 10.1111/j.1365-294X.2008.04054.x. [DOI] [PubMed] [Google Scholar]
  30. Hiscock S. Genetic control of self-incompatibility in Senecio squalidus L. (Asteraceae): a successful colonizing species. Heredity. 2000;85:10–19. doi: 10.1046/j.1365-2540.2000.00692.x. [DOI] [PubMed] [Google Scholar]
  31. James JK, Abbott RJ. Recent, allopatric, homoploid hybrid speciation: the origin of Senecio squalidus (Asteraceae) in the British Isles from a hybrid zone on Mount Etna, Sicily. Evolution. 2005;59:2533–2547. [PubMed] [Google Scholar]
  32. Johansen-Morris AD, Latta RG. Fitness consequences of hybridization between ecotypes of Avena barbata: hybrid breakdown, hybrid vigor, and transgressive segregation. Evolution. 2006;60:1585–1595. [PubMed] [Google Scholar]
  33. Johnston JA, Arnold ML, Donovan LA. High hybrid fitness at seed and seedling life history stages in Louisiana irises. Journal of Ecology. 2003;91:438–446. [Google Scholar]
  34. Kameyama Y, Kasagi T, Kudo G. A hybrid zone dominated by fertile F1s of two alpine shrub species, Phyllodoce caerulea and Phyllodoce aleutica, along a snowmelt gradient. Journal of Evolutionary Biology. 2008;21:588–597. doi: 10.1111/j.1420-9101.2007.01476.x. [DOI] [PubMed] [Google Scholar]
  35. Kariya K, Matsuzaki A, Machida H, Tsunoda K. Distribution of chlorophyll content in leaf blade of rice plant. Japanese Journal of Crop Science. 1982;51:134–135. [Google Scholar]
  36. Kimball S, Campbell DR, Lessin C. Differential performance of reciprocal hybrids in multiple environments. Journal of Ecology. 2008;96:1306–1318. [Google Scholar]
  37. Kitajima K, Fenner M. Ecology of seedling regeneration. In: Fenner M., editor. Seeds: the ecology of regeneration in plant communities. Wallingford, UK: CABI Publishing; 2000. pp. 331–359. [Google Scholar]
  38. Körner C. The use of ‘altitude’ in ecological research. Trends in Ecology and Evolution. 2007;22:569–574. doi: 10.1016/j.tree.2007.09.006. [DOI] [PubMed] [Google Scholar]
  39. Körner C, Larcher W. Plant life in cold climates. In: Long SP, Woodward FI, editors. Plants and temperature. Cambridge: The Company of Biologists; 1988. pp. 25–58. [Google Scholar]
  40. Kruuk LEB, Baird SJE, Gale KS, Barton NH. A comparison of multilocus clines maintained by environmental adaptation or by selection against hybrids. Genetics. 1999;153:1959–1971. doi: 10.1093/genetics/153.4.1959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Lynch M, Walsh B. Genetics and analysis of quantitative traits. Sunderland, MA: Sinauer Associates; 1998. [Google Scholar]
  42. Mariko S, Koizumi H, Suzuki J, Furukawa A. Altitudinal variations in germination and growth responses of Reynoutria iaponica populations on Mt Fuji to a controlled thermal environment. Ecological Research. 1993;8:27–34. [Google Scholar]
  43. Mayr E. Animal species and evolution. Cambridge, MA: Harvard University Press; 1963. [Google Scholar]
  44. Miglia KJ, Mcarthur ED, Moore WS, Wang H, Graham JH, Freeman DC. Nine-year reciprocal transplant experiment in the gardens of the basin and mountain big sagebrush (Artemisia tridentata: Asteraceae) hybrid zone of Salt Creek Canyon: the importance of multiple-year tracking of fitness. Biological Journal of the Linnean Society. 2005;86:213–225. [Google Scholar]
  45. Milla R, Gimenez-Benavides L, Escudero A, Reich PB. Intra- and interspecific performance in growth and reproduction increase with altitude: a case study with two Saxifraga species from northern Spain. Functional Ecology. 2009;23:111–118. [Google Scholar]
  46. Moore WS. Evaluation of narrow hybrid zones in vertebrates. Quarterly Review of Biology. 1977;52:263–277. [Google Scholar]
  47. Muller HJ. Isolating mechanisms, evolution and temperature. In: Dobzhansky T, editor. Temperature, evolution, development. Lancaster, PA: Jaques Cattell Press; 1942. pp. 71–125. [Google Scholar]
  48. Padilla FM, Pugnaire FI. Rooting depth and soil moisture control Mediterranean woody seedling survival during drought. Functional Ecology. 2007;21:489–495. [Google Scholar]
  49. Parsons PA. Extreme-environment heterosis and genetic loads. Heredity. 1971;26:479–482. doi: 10.1038/hdy.1971.59. [DOI] [PubMed] [Google Scholar]
  50. R Core Team. R: A Language and Environment for Statistical Computing. 2012 Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org . [Google Scholar]
  51. Rhode JM, Cruzan MB. Contributions of heterosis and epistasis to hybrid fitness. American Naturalist. 2005;166:E124–E139. doi: 10.1086/491798. [DOI] [PubMed] [Google Scholar]
  52. Ronsisvalle GA. Osservazioni biometriche su alcune specie di Senecio dell'Etna. 1. Variabilità di alcuni caratteri fogliari in popolazioni spontanee. Bolletim Accademia Gioenia Catania. 1968;9:332–345. [Google Scholar]
  53. Shimono Y, Kudo G. Intraspecific variations in seedling emergence and survival of Potentilla matsumurae (Rosaceae) between alpine fellfield and snowbed habitats. Annals of Botany. 2003;91:21–29. doi: 10.1093/aob/mcg002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Taylor SJ, Arnold M, Martin NH. The genetic architecture of reproductive isolation in Louisiana irises: hybrid fitness in nature. Evolution. 2009;63:2581–2594. doi: 10.1111/j.1558-5646.2009.00742.x. [DOI] [PubMed] [Google Scholar]
  55. Thompson JD. Plant evolution in the Mediterranean. New York, NY: Oxford University Press; 2005. [Google Scholar]
  56. Valladares F, Sanchez-Gomez D, Zavala MA. Quantitative estimation of phenotypic plasticity: bridging the gap between the evolutionary concept and its ecological applications. Journal of Ecology. 2006;94:1103–1116. [Google Scholar]
  57. Vera ML. Effects of altitude and seed size on germination and seedling survival of heathland plants in North Spain. Plant Ecology. 1997;133:101–106. [Google Scholar]
  58. Volis S, Mendlinger S, Ward D. Differentiation in populations of Hordeum spontaneum along a gradient of environmental productivity and predictability: life history and local adaptation. Biological Journal of the Linnean Society. 2002;77:479–490. [Google Scholar]
  59. Wang H, McArthur ED, Sanderson SC, Graham JH, Freeman DC. Narrow hybrid zone between two subspecies of big sagebrush (Artemisia tridentata: Asteraceae). IV. Reciprocal transplant experiments. Evolution. 1997;51:95–102. doi: 10.1111/j.1558-5646.1997.tb02391.x. [DOI] [PubMed] [Google Scholar]
  60. Warton DI, Hui FCK. The arcsine is asinine: the analysis of proportions in ecology. Ecology. 2011;92:3–10. doi: 10.1890/10-0340.1. [DOI] [PubMed] [Google Scholar]
  61. Whitlock MC, Ingvarsson PK, Hatfield T. Local drift load and the heterosis of interconnected populations. Heredity. 2000;84:452–457. doi: 10.1046/j.1365-2540.2000.00693.x. [DOI] [PubMed] [Google Scholar]
  62. Wu F, Bao W, Li F, Wu N. Effects of drought stress and N supply on the growth, biomass partitioning and water-use efficiency of Sophora davidii seedlings. Environmental and Experimental Botany. 2008;63:248–255. [Google Scholar]

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

RESOURCES