Joshua R. Kohn

Species Selection Maintains Self-Incompatibility

Incompatible Self-Compatibility Macroevolutionary processes driving species differences in diversification rates are important in explaining the variation we see in nature, but the extent of this process and how much the traits within a single species can drive changes in the diversification rate are unknown. Goldberg et al. (p. 493; see the Perspective by Wright and Barrett) analyzed the phylogenetics of the plant family Solanaceae and found that rates of extinction are greater for self-pollinating species than outbreeding species. Species-level selection against the deleterious effects of inbreeding may explain why self-fertilization, despite its short-term evolutionary advantages, has not spread to become more common in the flowering plants. Self-incompatibility in a family of plants is associated with appreciably higher rates of speciation. Identifying traits that affect rates of speciation and extinction and, hence, explain differences in species diversity among clades is a major goal of evolutionary biology. Detecting such traits is especially difficult when they undergo frequent transitions between states. Self-incompatibility, the ability of hermaphrodites to enforce outcrossing, is frequently lost in flowering plants, enabling self-fertilization. We show, however, that in the nightshade plant family (Solanaceae), species with functional self-incompatibility diversify at a significantly higher rate than those without it. The apparent short-term advantages of potentially self-fertilizing individuals are therefore offset by strong species selection, which favors obligate outcrossing.

LOSS OF SELF-INCOMPATIBILITY AND ITS EVOLUTIONARY CONSEQUENCES

We review and analyze the available literature on the frequency and distribution of self‐incompatibility (SI) among angiosperms and find that SI is reported in more than 100 families and occurs in an estimated 39% of species. SI frequently has been lost but rarely has been gained during angiosperm diversification, and there is no evidence that any particular system of SI, once lost, has been regained. Irreversible loss of SI systems is thought to occur because transitions to self‐compatibility (SC) are accompanied by collapse of variation at the S‐locus and by accumulation of loss‐of‐function mutations at multiple loci involved in the incompatibility response. The asymmetry in transitions implies either that SI is declining in frequency or that it provides a macroevolutionary advantage. We present a model in which the loss of SI is irreversible and species can be SI, SC but outcrossing, or predominantly selfing. Increased diversification rates of SI relative to SC taxa are required to maintain SI at equilibrium, while transition rates between states, together with state‐specific diversification rates, govern the frequency distribution of breeding‐system states. We review empirical studies about the causes and consequences of the loss of SI, paying particular attention to the model systems Arabidopsis and Solanum sect. Lycopersicon. In both groups, losses of SI have been recent and were accompanied by loss of most or all of the functional variation at the S‐locus. Multiple loss‐of‐function mutations are commonly found. Some evidence indicates that mutations causing SC strongly increase the selfing rate and that SC species have lower genetic diversity than their SI relatives, perhaps causing an increase in the extinction rate.

Evolutionary relationships among self-incompatibility RNases.

T2-type RNases are responsible for self-pollen recognition and rejection in three distantly related families of flowering plants—the Solanaceae, Scrophulariaceae, and Rosaceae. We used phylogenetic analyses of 67 T2-type RNases together with information on intron number and position to determine whether the use of RNases for self-incompatibility in these families is homologous or convergent. All methods of phylogenetic reconstruction as well as patterns of variation in intron structure find that all self-incompatibility RNases along with non-S genes from only two taxa form a monophyletic clade. Several lines of evidence suggest that the best interpretation of this pattern is homology of self-incompatibility RNases from the Scrophulariaceae, Solanaceae, and Rosaceae. Because the most recent common ancestor of these three families is the ancestor of ≈75% of dicot families, our results indicate that RNase-based self-incompatibility was the ancestral state in the majority of dicots.

Isolation and characterization of microsatellite loci in a Phylloscopus warbler

ferences between closely related species (Taylor et al. 1994), subspecies (Paetkau & Strobeck 1994) and populations of the same species (Bowcock et al. 1994). Roy et al. (1994) presented a thorough analysis at all three levels for wolves and their close relatives. Gotelli et al. (1994) were able to reveal hybridization between the Ethiopian wolf and domestic dogs, and others have used microsatellite variation to suggest recent population bottlenecks (Paetkau & Strobeck 1994). As more accurate models to analyse population structure data of microsatellite loci become available (Slatkin 1995) it appears likely that the study of microsatellite variation will contribute significantly to the fields of population genetics and conservation. In this paper we present information on the isolation and characterization of microsatellite loci in the warbler Phylloscopus occipitalis. Our research aims at investigating genetic structuring along an altitudinal gradient in its Himalayan breeding quarters and how this is related to morphological differentiation. Genomic DNA isolated from blood of an adult female large crowned leaf warbler Phylloscopus occipitalis was digested with MboI and electrophoresed in a 1.5% agarose gel. DNA fragments in the size range of 300–600 bp were excised from the gel and isolated by electroelution. A size selected library was then constructed by ligating the DNA into the vector M13mp18 (opened with BamHI), followed by transfection into E. coli DH5αF’ by electroporation (E. coli Pulser, Bio-Rad). According to the ratio of blue and white colonies 80% of the vectors contained inserted DNA. A total of P 40 000 colonies were screened using the oligonucleotides CA15 and GA15 end-labelled with γ32P-ATP. Nine positive clones were detected and sequenced (Table 1). The sequence of POCC3 happened to be identical to clone POCC2 although they were found on different plates. Primers were designed to have a Tm of 60 nC. The primers were used on putatively unrelated individuals of 16 Phylloscopus occipitalis sampled at Manali, Himachal Pradesh, India. In addition, we checked a family (male, female and four chicks) to look for inheritance patterns. Initial PCR amplifications were performed under the following conditions: 30 s at 94nC, 30 s at 55 nC, 30 s at 72 nC (28 cycles). Before the cyclic reactions the samples were incubated at 94nC for 3 min, and after completion at 72 nC for 5 min. Reactions of 10 μL included 50 ng of total genomic DNA, 0.2 mM of each nucleotide, 1.5 mM MgCl2, 0.4 μM of reverse primer, 0.2 μM of unlabelled forward primer, 0.02 μM of γ32P-ATP labelled forward primer and 0.5 units of Taq DNA polymerase. The PCR products were resolved on a 8% denaturing polyacrylamide gel (long ranger, Bio-Rad). Gels were dried and exposed for 3–36 h. The primers first used to amplify locus POCC1 failed to amplify one of the paternal alleles (see Callen et al. 1993). This was indicated by two of the chicks in the family not inheriting the father’s allele and that several of the screened individuals appeared to be homozygotes for rare alleles. When one of the primers was redesigned a new allele appeared in the father and the two chicks. Primers for locus POCC4 failed to amplify readily scorable products, perhaps because of the long A-repeat adjacent to the CA-repeat (Table 1). For locus POCC9 we failed to amplify altogether. This clone was very difficult to sequence (with many stops on both sides of the repeat) and one can speculate that the PCR failed for the same reason that caused stops to occur in the sequencing reaction. In total, primers designed for five of the clones reliably amplified polymorphic loci (1, 2, 5, 6 and 8). We also screened primers isolated from other species: four from loggerhead shrikes Lanius ludovicianus (N. Mundy MS), two from the pied flycatcher Ficedula hypoleuca, two from the barn swallow Hirundo rustica (Ellegren 1992) and six from reed buntings Emberiza schoeniclus (Hanotte et al. 1994). Of these, one loggerhead shrike primer pair (LS2), one pied flycatcher primer pair (PTC3) and one reed bunting primer pair (Escμ4) amplified variable microsatellite loci. All of the eight primer pairs found to be useful in P. occipitalis also showed to be variable in P. reguloides and for at least five of the loci in the willow warbler P. trochilus (Table 2). This supports earlier observations that microsatellite primers often work across species PRIMER NOTE

Experimental studies on the functional significance of heterostyly

Heterostyly has been viewed as both an antiselfing device and a mechanism that increases the proficiency of pollen transfer between plants. We used experimental manipulation of the morph structure of garden populations of self‐compatible, tristylous Eichhornia paniculata to investigate the function of floral polymorphism. Outcrossing rates (t), levels of intermorph mating (d), and morph‐specific male and female reproductive success were compared in replicate trimorphic and monomorphic populations. In trimorphic populations, t and d averaged 0.81 (2 SE = 0.03) and 0.77 (2 SE = 0.03) respectively, with no difference in either parameter among morphs. Ninety‐five percent of outcrossed seeds were therefore the result of intermorph fertilizations. Male reproductive success of the long‐styled morph was low, especially in comparison with plants of the short‐styled morph. Outcrossing rates for each morph were higher in trimorphic than monomorphic populations where t averaged 0.71 (2 SE = 0.01), 0.30 (2 SE = 0.04) and 0.43 (2 SE = 0.1) for the long‐, mid‐, and short‐styled morphs, respectively. Seed set was lower in monomorphic populations, particularly those composed of the L morph, reflecting reduced pollen deposition. Floral polymorphism therefore increased both outcrossing rate and fecundity but the magnitude of the differences varied among morphs. If the ancestral condition in heterostylous groups resembled the L morph, as has been suggested, data from this study suggests that the selective basis for the establishment of floral polymorphism could have been increased pollen transfer rather than higher levels of outcrossing.

Ancient polymorphism reveals unidirectional breeding system shifts

Loss of complex characters is thought to be irreversible (Dollos law). However, hypotheses of irreversible evolution are remarkably difficult to test, especially when character transitions are frequent. In such cases, inference of ancestral states, in the absence of fossil evidence, is uncertain and represents the single greatest constraint for reconstructing the evolutionary history of characters. Breeding system character transitions are of particular interest because they affect the amount and distribution of genetic variation within species. Transitions from obligate outcrossing to partial or predominant self-fertilization are thought to represent one of the most common trends in flowering plants. We use the unique molecular genetic properties (manifested as deep persistent polymorphisms) of the locus that enforces outcrossing to demonstrate that its loss is irreversible in the plant family Solanaceae. We argue that current phylogenetic methods of reconstruction are potentially inadequate in cases where ancestral state information is inferred by using only the phylogeny and the distribution of character states in extant taxa. This study shows in a statistical framework that a particular character transition is irreversible, consistent with Dollos law.

Primary structural features of the S haplotype-specific F-box protein, SFB, in Prunus

The gene SFB encodes an F-box protein that has appropriate S-haplotype-specific variation to be the pollen determinant in the S-RNase-based gametophytic self-incompatibility (GSI) reaction in Prunus (Rosaceae). To further characterize Prunus SFB, we cloned and sequenced four additional alleles from sweet cherry (P. avium), SFB1, SFB2, SFB4, and SFB5. These four alleles showed haplotype-specific sequence diversity similar to the other nine SFB alleles that have been cloned. In an amino acid alignment of Prunus SFBs, including the four newly cloned alleles, 121 out of the 384 sites were conserved and an additional 65 sites had only conservative replacements. Amino acid identity among the SFBs ranged from 66.0% to 82.5%. Based on normed variability indices (NVI), 34 of the non-conserved sites were considered to be highly variable. Most of the variable sites were located at the C-terminal region. A window-averaged plot of NVI indicated that there were two variable and two hypervariable regions. These variable and hypervariable regions appeared to be hydrophilic or at least not strongly hydrophobic, which suggests that these regions may be exposed on the surface and function in the allele specificity of the GSI reaction. Evidence of positive selection was detected using maximum likelihood methods with sites under positive selection concentrated in the variable and hypervariable regions.

Reconstruction of the evolution of reproductive characters in Pontederiaceae using phylogenetic evidence from chloroplast DNA restriction-site variation

We reconstructed the phylogenetic history of Pontederiaceae using chloroplast DNA restriction‐site variation from approximately two‐thirds of the species in this family of aquatic monocotyledons. The molecular phylogeny was used to evaluate hypotheses concerning the evolution of reproductive characters associated with the breeding system. The family has four main genera, two of which (Eichhornia and Pontederia) have tristylous, predominantly outcrossing species, while two (Monochoria and Heteranthera) have enantiostylous taxa. Self‐incompatibility is restricted to some but not all tristylous species. In Eichhornia and Pontederia, predominantly selfing species with small monomorphic flowers (homostyly) have been hypothesized to result from the multiple breakdown of tristyly. Restriction‐site variation provided a well supported phylogeny of ingroup taxa, enabling the mapping of reproductive characters onto trees. Two contrasting optimization schemes were assessed, differing in the relative weights assigned to shifts in character states. The reconstructed sequence of floral character‐state change was used to assess competing hypotheses concerning the origin and breakdown of tristyly, and the relationships between tristylous and enantiostylous syndromes. Our results indicate that the class of optimization scheme used was the most critical factor in reconstructing character evolution. Despite some topological uncertainties and difficulty in reconstructing the primitive floral form in the family, several broad conclusions were possible when an unordered, unequally‐weighted optimization scheme was used: (1) tristyly originated either once or twice, while the occurrence of enantiostyly in Monochoria and Heteranthera was always found to have independent origins; (2) tristyly has repeatedly broken down leading to selfing, homostylous taxa; and (3) self‐incompatibility probably arose after the origin of floral trimorphism, a sequence of events that conflicts with some evolutionary models.

Phylogenetic Congruence and Discordance Among One Morphological and Three Molecular Data Sets from Pontederiaceae

A morphological data set and three sources of data from the chloroplast genome (two genes and a restriction site survey) were used to reconstruct the phylogenetic history of the pickerelweed family Pontederiaceae. The chloroplast data converged towards a single tree, presumably the true chloroplast phylogeny of the family. Unrooted trees estimated from each of the three chloroplast data sets were identical or extremely similar in shape to each other and mostly robustly supported. There was no evidence of significant heterogeneity among the data sets, and the few topological differences seen among unrooted trees from each chloroplast data set are probably artifacts of sampling error on short branches. Despite well-documented differences in rates of evolution for different characters in individual data sets, equally weighted parsimony permits accurate reconstructions of chloroplast relationships in Pontederiaceae. A separate morphology-based data set yielded trees that were very different from the chloroplast trees. Although there was substantial support from the morphological evidence for several major clades supported by chloroplast trees, most of the conflicting phylogenetic structure on the morphology trees was not robust. Nonetheless, several statistical tests of incongruence indicate significant heterogeneity between molecules and morphology. The source of this apparent incongruence appears to be a low ratio of phylogenetic signal to noise in the morphological data.

THE DISTRIBUTION OF PLANT MATING SYSTEMS: STUDY BIAS AGAINST OBLIGATELY OUTCROSSING SPECIES

Abstract Early models of plant mating-system evolution argued that predominant outcrossing and selfing are alternative stable states. At least for animal-pollinated species, recent summaries of empirical studies have suggested the opposite—that outcrossing rates do not show the expected bimodal distribution. However, it is generally accepted that several potential biases can affect conclusions from surveys of published outcrossing rates. Here, we examine one potential bias and find that published studies of outcrossing rates contain far fewer obligate outcrossers than expected. We approximate the magnitude of this study bias and present the distribution of outcrossing rates after compensating for it. Because this study examines only one potential bias, and finds it to be large, conclusions regarding either the frequency of mixed mating or the shape of the distribution of outcrossing rates in nature are premature.