An adventurous journey toward and away from fern apomixis: Insights from genome size and spore abortion patterns.

PREMISE: Apomixis in ferns is relatively common and obligatory. Sterile hybrids may restore fertility via apomixis at a cost of long-term genetic stagnation. In this study, we outlined apomixis as a possible temporary phase leading to sexuality and analyzed factors relating to transitioning to and away from apomixis, such as unreduced and reduced spore formation in apomict and apo-sex hybrid ferns. METHODS: We analyzed the genome size of 15 fern species or hybrids ("taxa") via flow cytometry. The number of reduced and unreduced gametophytes was established as a proxy for viable spore formation of either type. We also calculated the spore abortion ratio (sign of reduced spores) in several taxa, including the apo-sex hybrid Dryopteris × critica and its 16 apomictically formed offspring. RESULTS: Four of 15 sampled taxa yielded offspring variable in genome size. Specifically, each variable taxon formed one viable reduced plant among 12-451 sampled gametophytes per taxon. Thus, haploid spore formation in the studied apomicts was very rare but possible. Spore abortion analyses indicated gradually decreasing abortion (haploid spore formation) over time. In Dryopteris × critica, abortion decreased from 93.8% to mean 89.5% in one generation. CONCLUSIONS: Our results support apomixis as a transitionary phase toward sexuality. Newly formed apomicts hybridize with sexual relatives and continue to form haploid spores early on. Thus, they may get the genomic content necessary for regular meiosis and restore sexuality. If the missing relative goes extinct, the lineage gets locked into apomixis as may be the case with the Dryopteris affinis complex.

Apomictic fern fathers: an experimental approach to the reproductive characteristics of sexual, apomict, and hybrid fern gametophytes.

PREMISE: Apomixis and hybridization are two essential and complementary factors in the evolution of plants, including ferns. Hybridization combines characteristics from different species, while apomixis conserves features within a lineage. When combined, these two processes result in apo-sex hybrids. The conditions leading to the formation of these hybrids are poorly understood in ferns. METHODS: We cultivated spores from 66 fern samples (43 apomicts, 7 apo-sex hybrids, and 16 sexuals), and measured their development in vitro over 16 weeks. We evaluated germination, lateral meristem formation rates, sexual expression, and production of sporophytes and then compared ontogenetic patterns among the three groups. RESULTS: The three examined groups formed antheridia (male gametangia) but differed in overall gametophyte development. Sexual species created archegonia (female, 86% of viable samples), but no sporophytes. Apomicts rarely created nonfunctional archegonia (8%) but usually produced apogamous sporophytes (75%). Surprisingly, apomictic and sexual species showed similar development speed. The sexually reproducing parents of viable studied hybrids formed about twice as many meristic gametophytes as the apomictic parents (39% vs. 20%, respectively). CONCLUSIONS: We present the most thorough comparison of gametangial development of sexual and apomictic ferns, to date. Despite expectations, apomictic reproduction might not lead to earlier sporophyte formation. Apomicts produce functional sperm and thus can contribute this type of gamete to their hybrids. The development patterns found in the parents of hybrids indicate a possible increase of hybridization rates by antheridiogens. The apo-sex hybrids always inherit the apomictic reproductive strategy and are thus capable of self-perpetuation.

Insights into the evolutionary history and widespread occurrence of antheridiogen systems in ferns.

Sex expression of homosporous ferns is controlled by multiple factors, one being the antheridiogen system. Antheridiogens are pheromones released by sexually mature female fern gametophytes, turning nearby asexual gametophytes precociously male. Nevertheless, not all species respond. It is still unknown how many fern species use antheridiogens, how the antheridiogen system evolved, and whether it is affected by polyploidy and/or apomixis. We tested the response of 68 fern species to antheridiogens in cultivation. These results were combined with a comprehensive review of literature to form the largest dataset of antheridiogen interactions to date. Analyzed species also were coded as apomictic or sexual and diploid or polyploid. Our final dataset contains a total of 498 interactions involving 208 species (c. 2% of all ferns). About 65% of studied species respond to antheridiogen. Multiple antheridiogen types were delimited and their evolution is discussed. Antheridiogen responsiveness was not significantly affected by apomixis or polyploidy. Antheridiogens are widely used by ferns to direct sex expression. The antheridiogen system likely evolved multiple times and provides homosporous ferns with the benefits often associated with heterospory, such as increased rates of outcrossing. Despite expectations, antheridiogens may be beneficial to polyploids and apomicts.

Asymmetric hybridization in Central European populations of the Dryopteris carthusiana group.

PREMISE: Hybridization is a key process in plant speciation. Despite its importance, there is no detailed study of hybridization rates in fern populations. A proper estimate of hybridization rates is needed to understand factors regulating hybridization. METHODS: We studied hybridization in the European Dryopteris carthusiana group, represented by one diploid and two tetraploid species and their hybrids. We sampled ~100 individuals per population in 40 mixed populations of the D. carthusiana group across Europe. All plants were identified by measuring genome size (DAPI staining) using flow cytometry. To determine the maternal parentage of hybrids, we sequenced the chloroplast region trnL-trnF of all taxa involved. RESULTS: We found hybrids in 85% of populations. Triploid D. ×ambroseae occurred in every population that included both parent species and is most abundant when the parent species are equally abundant. By contrast, tetraploid D. ×deweveri was rare (15 individuals total) and triploid D. ×sarvelae was absent. The parentage of hybrid taxa is asymmetric. Despite expectations from previous studies, tetraploid D. dilatata is the predominant male parent of its triploid hybrid. CONCLUSIONS: This is a thorough investigation of hybridization rates in natural populations of ferns. Hybridization rates differ greatly even among closely related fern taxa. In contrast to angiosperms, our data suggest that hybridization rates are highest in balanced parent populations and support the notion that some ferns possess very weak barriers to hybridization. Our results from sequencing cpDNA challenge established notions about the correlation of ploidy level and mating tendencies.

Interploidy gene flow via a 'pentaploid bridge' and ploidy reduction in Cystopteris fragilis fern complex (Cystopteridaceae: Polypodiales).

KEY MESSAGE: Our results indicate the existence of interploidy gene flow in Cystopteris fragilis, resulting in sexual triploid and diploid gametophytes from pentaploid parents. Similar evolutionary dynamics might operate in other fern complexes and need further investigation. Polyploidization and hybridization are a key evolutionary processes in ferns. Here, we outline an interploidy gene flow pathway operating in the polyploid Cystopteris fragilis complex. The conditions necessary for the existence of this pathway were tested. A total of 365 C. fragilis individuals were collected covering representatives of all three predominant ploidy levels (tetraploid, pentaploid, and hexaploid), cultivated, had their ploidy level estimated by flow cytometry, and their spores collected. The spores, as well as gametophytes and sporophytes established from them, were analysed by flow cytometry. Spore abortion rate was also estimated. In tetraploids, we observed the formation of unreduced (tetraploid) spores (ca 2%). Collected pentaploid individuals indicate ongoing hybridization between ploidy levels. Pentaploids formed up to 52% viable spores, ca 79% of them reduced, i.e. diploid and triploid. Reduced spores formed viable gametophytes, and, in the case of triploids, filial hexaploid sporophytes, showing evidence of sexual reproduction. Some tetraploid sporophytes reproduce apomictically (based on uniform ploidy of their metagenesis up to filial sporophytes). Triploid and diploid gametophytes from pentaploid parents are able to mate among themselves, or with "normal" reduced gametophytes from the sexual tetraploid sporophytes (the dominant ploidy level in the sporophytes in this populations), to produce tetraploid, pentaploid, and hexaploid sporophytes, allowing for geneflow from the pentaploids to both the tetraploid and hexaploid populations. Similar evolutionary dynamics might operate in other fern complexes and need further investigation.

An experimental assessment of competitive interactions between sexual and apomictic fern gametophytes using Easy Leaf Area.

Premise: Few studies have explored competition in fern gametophyte populations. One limiting factor is the tedious measurement of gametophyte size as a proxy for biomass in these small plants. Here, an alternative approach of estimating the number of green pixels from photos was employed to measure the competitive interactions among apomictic and sexual Dryopteris gametophytes. Methods: We cultivated the gametophytes of two apomictic (diploid and triploid) and one sexual (tetraploid) Dryopteris species in monocultures and in two-species mixtures in the ratios 1 : 1 and 1 : 3. The total gametophyte cover of each population originating from 20 spores was assessed using Easy Leaf Area. Assessments were performed weekly between weeks 4 and 10 of cultivation. Additionally, during week 5, the cover of each species in each mixture was estimated separately. Results: We identified a positive correlation between gametophyte size and ploidy level as well as sexual reproduction. The performance of the tested species in mixtures was dependent on the competitor species identity, indicating the importance of competition between gametophytes. Discussion: The methods outlined can be used for a rapid assessment of fern gametophyte cover in large gametophyte populations. Ploidy level and reproduction type seem to play a major role in the competitive abilities of fern gametophytes, but more research is needed on this topic.