| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Centre d'Ecologie Fonctionnelle et Evolutive (CNRS), 1919 Route de Mende, F-34293 Montpellier, France
Received for publication September 14, 2000. Accepted for publication January 26, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Asteraceae • Crepis sancta; • frequency and density dependence • inbreeding depression
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), the effects of inbreeding on fitness components, i.e., inbreeding depression, have been well studied, particularly in plants (Jarne and Charlesworth, 1993
). Strong inbreeding depression is expected in allogamous plant populations, whereas weaker inbreeding depression is expected in autogamous populations where inbreeding depression is caused by deleterious recessive alleles (Charlesworth and Charlesworth, 1987
). Yet, the efficiency of purging in plants could be limited (Byers and Waller, 1999
), and the magnitude of inbreeding depression could also be influenced by environmental conditions. Most studies concerning this subject have been performed in greenhouses or in experimental gardens with optimal environmental conditions. This may minimize the magnitude of inbreeding depression because recent empirical studies (Dudash, 1990
) have shown that the effect of inbreeding may be more severe in stressful conditions (i.e., competition, water stress, nutrient deficiency). Nevertheless, recent published results comparing the intensity of inbreeding depression in stressful and in normal conditions have not shown a clear trend (reviewed in Norman et al., 1995
). This may be due to the fact that environmental effects are complex. Hence, it is necessary to clearly define experimental conditions with regard to natural population characteristics in order to draw relevant conclusions.
Competition may strongly modify the expression of inbreeding depression (Antonovics, 1968
; Schmitt and Ehrhardt, 1990
; Cheptou et al., 2000b
). Uyenoyama, Holsinger, and Waller (1993)
hypothesized that the relative fitness of inbred plants could potentially be affected by density dependence or frequency dependence. First, inbreeding depression could be magnified at high density (density dependence) where space and resources become limiting, creating stressful conditions that may enhance inbreeding depression. Second, for a given density, inbred plants may be less competitive than outbred plants. Here the fitness of inbred plants could be influenced by whether or not competitors are inbred or outcrossed (frequency dependence). Previous studies including competition between outbred and inbred plants, however, have not shown any clear trend (Charlesworth, Lyons, and Litchfield, 1994
; Mayer, Charlesworth, and Meyers, 1996
). This could be because the experimental designs did not simulate the opportunity of selection in natural conditions and, in particular, the natural densities experienced by the species in the field. Indeed, in these experiments only two individuals were used (one inbred and one outbred) to simulate competitive conditions. Therefore, density-dependent processes were not considered and competition for light, a very important limiting resource (Tilman, 1988
), was minimized.
In this study, we examined the hypothesis that intraspecific competition can increase inbreeding depression via the density and type of competitors (selfed and outcrossed), in Crepis sancta, an annual colonizing plant. To test the hypothesis of Uyenoyama, Holsinger, and Waller (1993)
, the relative fitness of inbred plants was measured at two densities and inbred progeny were grown in competition with either inbred and outcrossed plants. The outcrossing rate for the population used in this experiment was also measured.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). In autumn, seeds germinate following the onset of rain. Rosettes grow during winter and flowering occurs from February to April. Pollination is entomophilous. The population used in this experiment occurs in a vineyard in Limagne in central France. In April 1998, this population was composed of a few thousand individuals at a density of 
50 plants/m2. In vineyards, Crepis sancta is often the dominant species (P.-O. Cheptou, personal observation) because of its early spring development (Imbert, Escarré, and Lepart, 1996
). Competitive interactions between plants are thus primarily intraspecific in vineyards. In 1998, 30 rosettes were sampled in the field and transplanted to a glasshouse at the CEFE-CNRS gardens in Montpellier in order to perform controlled pollinations. The mean distance between sampled plants was >10 m, to decrease the likelihood of sampling related individuals.
Pollination treatments
Three treatments were performed: (1) plants were nonpollinated in order to test the possibility of autonomous self-fertilization; (2) plants were self-pollinated by hand-pollination (with a brush); and (3) plants were hand-pollinated with another plant chosen at random. Hand-pollination was performed when all florets were receptive. The three treatments were randomized in time. When fruits were mature, we counted seed number and total number of florets per capitulum with a binocular microscope. Number of seeds compared to the total number of florets (seed : floret ratio) were analyzed. A sample of selfed seedling progeny were screened by enzymes electrophoresis (see below for technical details) to check that selfed progeny were not accidental outcrossing events.
Estimation of population outcrossing rate
In the same population as above, 15 individuals with mature seeds were harvested in May 1998 in order to infer the natural selfing rate. Twelve seeds per plant (hereafter family) were sown in the glasshouse in September 1998, and height of randomly chosen seedlings of each of the 15 families was analyzed using allozymes. Enzymes were extracted from fresh leaves using 0.05 mol/L Tris-HCl buffer (pH: 7.5), 5g/L sucrose, and 0.6% mercaptoethanol. The multilocus genotypes of progeny were scored on two starch gels for six polymorphic loci. These loci were coherent with Mendelian inheritance (data not shown). Gels with 12% starch were used for electrophoresis. PGI-1 (E.C. 5.3.1.9), PGD-1, PGD-2 (E.C. 1.1.1.44), and SKD (E.C. 1.1.1.25) resolved on Histidine gels at pH 6.5 (Soltis and Soltis, 1989
). PGM-1 (E.C. 5.4.2.2) were resolved on lithium borate gels at pH 8.3 (Soltis and Soltis, 1989
). Enzyme stain recipes are those described by Soltis and Soltis (1989)
.
Population outcrossing rate was obtained using MLTR program (available from K. Ritland). The MLTR program calculates maximum likelihood estimates for outcrossing rate based on the model described in Ritland and Jain (1981)
and Ritland (1990)
. Multilocus and average single-locus outcrossing rate were estimated. The inbreeding coefficient of maternal parents was also calculated. The Newton-Raphson method was used for the maximum likelihood estimates, and standard errors of estimates were calculated using 100 bootstraps.
Measurements of inbreeding depression in an intraspecific competition context
Seeds produced from controlled pollination were used to measure the performance of plants produced by selfing (F = 0.5) and outcrossing (F = 0). Two types of seeds (outcrossed and inbred) were used for 16 families. Masses of 20 seeds per family and per cross type were measured. The mass of selfed seeds was significantly lower than the mass of outbred seeds (mean ± SE: 2.337 ± 0.159 mg for inbred seeds and 2.725 ± 0.131 mg for outbred seeds, paired t test = 2.37, df = 15, P = 0.031). The experiment was constructed in a cross-factorial design. Seeds were sown in two types of culture: (1) pure culture of selfed or outcrossed seeds in separate pots and (2) mixed culture of 50% selfed seeds and 50% outcrossed seeds in the same pots. The substrate was composed of 50% sterile soil and 50% commercial compost. The design was conducted at two density levels with four replicates for a total of 512 individuals. Sixteen seeds per pot (one seed per family and per cross) were planted in pure culture arranged in a 4 x 4 square pattern. Thirty-two seeds per pot (one seed per family and per cross) were planted in pure culture arranged in a 4 x 8 checkerboard pattern. The position of families in pots was randomized. Each pot was surrounded by an outer border row to avoid edge effects on target plants. At high density, the distance between plants was 2.5 cm for a density of 1600 plants/m². At low density, the distance between plants was 5 cm for a density of 400 plants/m². The low density is found in many old fields that were abandoned >20 yr ago. The high density corresponds to an active or recently abandoned vineyard where it is possible to observe a uniform density of 2000 seedlings/m² at the end of September (P.-O. Cheptou, unpublished data). Because we used 32 individuals (16 inbred seeds and 16 outbred seeds) in mixtures, the size of pots was modified to keep the same density of the pure cultures. Therefore at high density, we used 16 x 16 x 16 cm pots (4 L) for pure culture (for a total of 16 plants plus 20 plants for the border) and 25 x 16 x 16 cm (6.5 L) for mixed culture (for a total of 32 plants plus 28 plants for the border). At a low density, we used 27 x 27 x 15 cm pots (11 L) for pure culture (for a total of 16 plants plus 20 plants for the border) and 50 x 25 x 15 cm (18.75 L) for mixed culture (for a total of 32 plants plus 28 plants for the border). So, although pot sizes vary, this experimental design allowed us to provide the same available resources per plant and the same soil depth, for a given density. The experimental design is very similar to the one used by Schmitt and Ehrhardt (1990)
.
Seeds were sown in a heated greenhouse with 12 h artificial light at 25°C and 12 h darkness at 12°C (Cheptou et al., 2000b
). Germination was monitored daily. Seeds germinated very rapidly (>80% in 2 d). Thus variation in germination time was not analyzed. Additional seeds for each family were sown at the same date in order to replace nongerminated seeds. We chose to replace seedlings at the seedling stage (i.e., two cotyledons) to avoid unbalanced competition at following stages, although replacement seedlings were not analyzed in the estimation of total fitness (see below). After 1 mo in the greenhouse, pots were placed outdoors on a table and watered daily. Free-pollination was ensured by bees. Pots were moved randomly every four 4 d to minimize position effects. We measured weekly survival to reproduction. At the end of the life cycle, we counted the total number of flowers that produced seeds (number of capitula). Because of the relatively high density in the whole experiment, no intermediate measures of growth were performed. The individual final biomass was not measured because all plants were harvested at the end of the life cycle when leaves could not be properly harvested. However, previous studies have shown that biomass is highly correlated to the number of capitula (Cheptou et al., 2000a
). Total fitness estimates were calculated as the numbers of capitula for each seed that was sown at the beginning of the experiment. If one seed did not germinate or if it died before flowering, its total fitness was considered zero.
Data analysis
Data from the experimental design were analyzed with a generalized linear model using GLIM (Payne, 1985
), which allows the specification of error structure appropriate to this type of data. Binomial responses such as germination and survival before reproduction were analyzed with a logit link specifying binomial error. Count data (number of capitula, total fitness) were analyzed with a log link specifying Poisson error (Crawley, 1993
). The full linear model included four factors: family, cross (selfed or outcrossed), density (high or low), and type of culture (mixed or pure) and their interactions. The significance of the factors was tested using a backward suppression of nonsignificant effects (Crawley, 1993
). Factors were tested and included in the residual deviance if nonsignificant. Overdispersion of the residual deviance was corrected by scaling the residual deviance using the Pearson chi-square divided by the number of degrees of freedom (Crawley, 1993
). For the analysis of number of capitula, only living plants at the end of the life cycle were taken into account. In this model, the effect of a factor (density, culture) on inbreeding depression is shown by a significant interaction between cross and this factor.
Furthermore, we calculated inbreeding depression values for the different variables as 
= 1 – wself/wout where wself and wout are the mean fitness estimates for self progeny and outcross progeny, respectively.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Loci used for the analysis of outcrossing rate were highly polymorphic (Table 1). In particular, PGI-1 and PGI-2 loci had four alleles, which insures a good estimate of the selfing rate (Ritland, 1986
). The population outcrossing rate based on multilocus estimates showed a value close to one and nonsignificantly different from complete outcrossing (Table 1). Single-locus estimates showed a similar value, indicating an absence of biparental inbreeding (Sun and Ritland, 1998
). The inbreeding coefficient for the maternal parents (F) is nonsignificantly different from zero, which is coherent with a highly allogamous population.
|
|
|
Reproduction
The four main factors were all significant (Table 2). Outbred plants produced significantly more capitula than inbred plants. At high density there was a significant reduction of the number of capitula produced compared to low density (Fig. 1C). However, the numbers of capitula for inbred plants were affected by the mixed culture in both densities. This produced a significant cross x culture interaction (Table 2). As for germination and survival, the family x cross interaction was significant.
Total fitness
The four main factors were also highly significant (Table 2). As for the number of capitula, total fitness for inbred plants was clearly reduced in mixed culture. This is supported by the significant cross x culture interaction (Fig. 1D). The family x cross interaction was highly significant. Though families differed in inbreeding depression values, we checked that inbreeding depression by family for the different fitness estimates (germination, survival, and reproduction) were not correlated. Table 3 summarizes the inbreeding depression calculated for germination, survival, number of capitulum, and total fitness. Inbreeding depression increased during the life cycle in mixed culture at both densities, whereas it did not show a clear trend in pure culture.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). It provides evidence that sampled plants are not related, and consequently experimental crosses with individuals from the same population (outcrossed pollination) are not consanguineous matings (i.e., F = 0). Such an outcrossed mating strategy is expected to be maintained in populations that experience high inbreeding depression > 0.5). Some natural populations of Crepis sancta may contain dense patches where intraspecific competition is likely to influence the relative fitness. A clear result from this study is that intraspecific competition may play an important role in the expression of inbreeding depression, although, at the germination stage, no clear pattern emerges. Density and type of competitors have significant effects on the relative mortality and reproductive fitness of selfed and outcrossed plants. For mortality, inbreeding depression was highest in mixed culture at high density, hence the significant culture x density x cross interaction. As expected, this pattern is consistent with a cumulative effect of density and culture effect. High density and outbred competitors intensify competition and lead to a higher mortality of inbred plants.
Inbreeding depression on the number of capitula and total fitness estimates was clearly enhanced by the presence of outbred competitors. Contrary to effects on mortality, density had no significant effect on the influence of inbreeding depression on the number of capitula and total fitness (i.e., a nonsignificant density x cross interaction). When competition occurred between plants from the same type of cross (inbred–inbred or outbred–outbred), performance of the two types was equivalent, whatever the density level. In contrast, when competition occurred between inbred and outbred plants, the fitness of inbred plants was significantly reduced in both densities. This result shows that inbreeding depression at this stage of the life cycle is a frequency-dependent phenomenon, the relative fitness of inbred plants being dependent on the frequency of outcrossed competitors. These results also suggest that frequency dependence has a greater effect than density dependence for reproduction, at least for these two experimental levels of density. A similar trend was found by Schmitt and Ehrhardt (1990)
in Impatiens capensis, in which the size hierarchy increased with time when competition occurred between inbred and outbred plants. Carr and Dudash (1995)
found a similar trend in Mimulus guttatus. However, in Collinsia heterophylla, Mayer, Charlesworth, and Meyers (1996)
did not find significant effect on inbreeding depression when a selfed plant grew in competition with an outbred plant. However, these two studies considered only pairwise competition, whereas our experiment dealt with competition with a higher number of neighbors for a target plant.
Our pattern of inbreeding depression, based on naturally occurring densities of seedlings, can be interpreted in the light of the dominance and suppression hypothesis (Weiner, 1985
), which postulates that a small initial advantage (for instance in seed mass) is enhanced when competition occurs. Indeed, many studies have demonstrated that slight differences in seed mass may translate into high fitness differences (El-Keblawy and Lovett-Doust, 1998
). In the present study, the significant difference in achene mass between inbred and outbred types may be the basis for the patterns observed. In the absence of competition, Cheptou et al. (2000a)
have shown that outbred plants of Crepis sancta grow faster than inbred plants in stressful conditions (drought) but with no effect on the relative fitness. In addition, with competitors of an equivalent inbred status, competition has no influence on fitness characters. In contrast, in mixture with outcrossed plants, inbred plants suffer more from competition because of a reduced growth rate. A higher growth rate allow a better aptitude to compete for light, for example, and so reduced growth rate could enhance the inbreeding depression. Wolfe (1993)
found similar trends in the relative growth rate of selfed and outcrossed Hydrophyllum appendiculatum. The results of Mayer, Charlesworth, and Meyers (1996)
, who did not find an effect of competition on inbreeding depression with only two plants per pot, could thus be explained by the low competitive conditions of their experimental design.
The presence of the family x cross interaction also suggests that inbreeding depression varies among family since some families show an absence of inbreeding depression whereas others families exhibit a high inbreeding depression. However, the pattern of variation among family greatly varies for the different life cycle components.
Although our experiment only manipulated two levels of mixture (50–50 and 0–100), it provides strong evidence that the relative fitness of inbred plants can be modified by the composition of the competing plants, leading to a frequency-dependent expression of inbreeding depression as suggested by Uyenoyama, Holsinger, and Waller (1993)
. In our experiment, we showed that inbreeding depression can be highly modified by the type of competitors: inbreeding depression was <0.5 in pure culture ( = 0.04–0.27) but >0.5 in mixed culture ( = 0.6). To our knowledge, this is the first study showing that inbreeding depression can be either above or below 0.5 because of intraspecific competition. This may have implications for the selection of selfing. We can thus suggest that inbreeding depression will be positively correlated with the frequency of competing outbred plants at the density level tested. For instance, if the local population is mainly composed of inbred competing plants, inbreeding depression is expected to be low, whereas in a population mainly composed of outbred plants inbreeding depression is expected to be high. Nevertheless, further empirical results based on the composition of natural populations are needed to explore this phenomenon. This ecological component of the expression of inbreeding depression has not been taken into account in theoretical models (Uyenoyama, Holsinger, and Waller, 1993
), but its evolutionary consequences may be important. Several studies suggest that inbreeding depression is higher in outcrossing populations (Byers and Waller, 1999
). In addition, we showed that a high outcrossing rate, i.e., a high frequency of competing outbred genotypes in a population, may also cause inbreeding depression to be high.
In an outcrossing population, the progeny of a rare inbreeder would suffer high inbreeding depression because competition will occur primarily with outbred plants. The maintenance of a self-incompatibility system and a highly allogamous strategy in the studied population are consistent with a large inbreeding depression, an important component of which may be explained by intraspecific competition in a mostly outcrossing population.
|
FOOTNOTES |
|---|
2 Author for reprint requests (Tel: 33 4 67 61 33 05; cheptou{at}cefe.cnrs-mop.fr
; Fax: 33 4 67 41 21 38).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Byers D. L. D. M. Waller 1999 Do plant populations purge their genetic load? Effects of population size and mating system history on inbreeding depression. Annual Review of Ecology and Systematics 30: 479-513
Carr D. E. M. R. Dudash 1995 Inbreeding depression under a competitive regime in Mimulus guttatus: consequences for potential male and female function. Heredity 75: 437-445[ISI]
Charlesworth D. B. Charlesworth 1987 Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237-268[CrossRef][ISI]
Charlesworth D. E. E. Lyons L. B. Litchfield 1994 Inbreeding depression in two highly inbreeding populations of Leavenworthia. Proceeding of the Royal Society of London Series B 258: 209-214[CrossRef]
Cheptou P.-O. A. Berger A. Blanchard C. Collin J. Escarré 2000a The effect of drought stress on inbreeding depression in four populations of the Mediterranean outcrossing plant Crepis sancta (Asteraceae). Heredity 85: 294-302
———, E. Imbert J. Lepart J. Escarre 2000b Effects of competition on lifetime estimates of inbreeding depression in the outcrossing plant Crepis sancta (Asteraceae). Journal of Evolutionnary Biology 13: 522-531
Crawley J. M. 1993 GLIM for ecologists. Blackwell Scientific, Oxford, UK
Dimitrova D. J. Greilhuber 2000 Karyotype and DNA-content evolution in ten species of Crepis (Asteraceae) distributed in Bulgaria. Botanical Journal of the Linnean Society 132: 281-297[CrossRef]
Dudash M. R. 1990 Relative fitness of selfed and outcrossed progeny in a self-incompatible, protandrous species, Sabatia angularis L. (Gentianaceae): a comparison in three environments. Evolution 44: 1129-1139[CrossRef][ISI]
El-Keblawy A. J. Lovett-Doust 1998 Persistent, non-seed size maternal effects on life history traits in the progeny generation in squash, Cucurbita pepo. New Phytologist 140: 655-665[CrossRef][ISI]
Imbert E. J. Escarre J. Lepart 1996 Achene dimorphism and among-population variation in Crepis sancta (Asteraceae). International Journal of Plant Science 157: 309-315[CrossRef]
Jarne P. D. Charlesworth 1993 The evolution of the selfing rate in functionally hermaphrodite plants and animals. Annual Review of Ecology and Systematics 24: 441-466[CrossRef][ISI]
Mayer S. D. Charlesworth B. Meyers 1996 Inbreeding depression in four populations of Collinsia heterophylla Nutt (Scrophulariaceae). Evolution 50: 879-891[CrossRef][ISI]
Norman J. K. A. K. Sakai S. G. Weller T. E. Dawson 1995 Inbreeding depression in morphological and physiological traits of Schiedea lydgatei (Caryophyllaceae) in two environments. Evolution 49: 297-306[CrossRef][ISI]
Payne C. D. 1985 The GLIM system, release 3.77, manual. Oxford and Royal Statistical Society, London, UK
Ritland K. 1986 Joint maximum likelihood estimation of genetic and mating structure using open pollinated progenies. Biometrics 42: 25-43[CrossRef][ISI]
———. 1990 A series of FORTRAN computer programs for estimating plant mating systems. Journal of Heredity 81: 235-237[ISI]
———, and S. Jain 1981 A model for the estimation of outcrossing rate and gene frequencies using n independent loci. Heredity 47: 35-52[ISI]
Schmitt J. D. W. Ehrhardt 1990 Enhancement of inbreeding depression by dominance and suppression in Impatiens capensis. Evolution 44: 269-278[CrossRef][ISI]
Soltis D. E. P. S. Soltis 1989 Isozymes in plant biology. Dioscorides Press, Portland, Oregon, USA
Sun M. K. Ritland 1998 Mating system of yellow starthistle (Centaurea solstitialis), a successfull colonizer in North America. Heredity 80: 225-232[CrossRef][ISI]
Tilman D. 1988 Plant strategies and the dynamics and structure of plant communities. Princeton University Press, Princeton, New Jersey, USA
Uyenoyama M. K. K. E. Holsinger D. M. Waller 1993 Ecological and genetic factors directing the evolution of self fertilization. Oxford Surveys of Evolutionary Biology 9: 327-381
Waller D. M. 1993 The statics and dynamics of mating system evolution. In N. W. Thornhill [eds.], The natural history of inbreeding and outbreeding, University of Chicago Press, Chicago, Illinois, USA
Weiner J. 1985 Size hierarchies in experimental populations of annual plants. Ecology 66: 743-752[CrossRef][ISI]
Wolfe L. M. 1993 Inbreeding depression in Hydrophyllum appendiculatum: role of maternal effects, crowding, and parental mating history. Evolution 47: 374-386[CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  J. Friedman and S. C. H. Barrett High Outcrossing in the Annual Colonizing Species Ambrosia artemisiifolia (Asteraceae) Ann. Bot., June 1, 2008; 101(9): 1303 - 1309. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
