Development of PCR‐Based Markers to Determine the Sex of Kelps

Sex discriminating genetic markers are commonly used to facilitate breeding programs in economically and ecologically important animal and plant species. However, despite their considerable economic and ecological value, the development of sex markers for kelp species has been very limited. In this study, we used the recently described sequence of the sex determining region (SDR) of the brown algal model Ectocarpus to develop novel DNA-based sex-markers for three commercially relevant kelps: Laminaria digitata, Undaria pinnatifida and Macrocystis pyrifera. Markers were designed within nine protein coding genes of Ectocarpus male and female (U/V) sex chromosomes and tested on gametophytes of the three kelp species. Seven primer pairs corresponding to three loci in the Ectocarpus SDR amplified sex-specific bands in the three kelp species, yielding at least one male and one female marker for each species. Our work has generated the first male sex-specific markers for L. digitata and U. pinnatifida, as well as the first sex markers developed for the genus Macrocystis. The markers and methodology presented here will not only facilitate seaweed breeding programs but also represent useful tools for population and demography studies and provide a means to investigate the evolution of sex determination across this largely understudied eukaryotic group.


Introduction
Brown algae (Phaeophyceae) are the dominant inhabitants of coastal environments and the major primary producers in these marine ecosystems. They are best known for forming kelp forests in shallow, rocky, cold water environments, especially in the northern Hemisphere. Subtidal kelp forests, together with their associated fauna and flora, are among the most diverse and productive ecosystems worldwide (reviewed in [1][2][3]). The importance of brown algae as a [31,32], a genetic map [33], gene mapping techniques, a large amount of transcriptomic data [34][35][36][37] and bioinformatics tools [38,39].
In this study, our objective was to use recently obtained knowledge about the sex-determination system in Ectocarpus to develop sex markers for commercially important kelp species. Genetic sex markers have broad applications in land plants such as jojoba [40], asparagus [41], papaya [42], ginkgo [43], hop [44], pistachio [45], and in animals such as chicken and fish produced by aquaculture [46][47][48][49]. For brown algae, genetic sex markers would represent an important tool for the efficient identification of male, female or bisexual individuals in breeding and production programs. Because gametophytes are small and isolation of single gametophyte clones is laborious, sex markers would allow rapid detection of contamination when only one haploid male or female clone is necessary. Moreover, sex markers are also useful to discriminate between diploid sporophytes derived from sexual reproduction and female or male partheno-sporophytes which can be diploid but fully homozygous. A method has recently been developed in which partheno-sporophytes are derived from female Undaria gametophytes. The zoospores produced by one of these partheno-sporophytes develop as gametophytes producing a large population of clonal, sexual individuals that can be used in crosses to produce large numbers of highly similar sporophytes [50]. Efforts are currently being made to develop reproductive male partheno-sporophytes for the same purpose. Mass crosses between sets of genetically identical gametophytes produced from zoospores would greatly improve breeding strategies for kelps and sex markers would represent a valuable resource for this approach, allowing verification of the unisexual genetic background of the parent partheno-sporophytes. The sexual cycle consists of an alternation between a macroscopic diploid sporophyte and microscopic, haploid, dioicous (male and female) gametophytes. The zoospores produced by the sporophyte are released and develop as male (carrying the V sex chromosome) or female (carrying the U sex chromosome) gametophytes. Egg and sperm produced by the gametophyte fuse to return to the diploid, asexual sporophyte generation, which carries both the U and the V sex chromosomes.
So far, very few sex markers have been developed for commercially grown seaweeds, rare exceptions including male and female markers for the red algae Gracilaria gracilis [51] and G. chilensis [52] and female microsatellite-based markers for the brown algae S. japonica [53,54] and U. pinnatifida [55]. Note, however, that when only female markers are available, it is not possible to distinguish male samples from female samples for which the PCR amplification has failed. Programs that use female markers alone therefore run the risk of a high level of falsely assigned males. Moreover, on a more general level, microsatellite-based markers have relatively high development costs, technical challenges with regards to the construction of libraries and primers, and the fact that flanking sequences may not be conserved across different groups of organisms. These features can make it difficult to transfer microsatellite-based markers across species. In addition, SDRs have not been precisely mapped in kelp species and it is possible that microsatellite loci identified as sex markers lie very close to, but not within, the SDR. If this is the case the markers may recombine with the SDR in rare instances resulting in false sex assignment. The best way to avoid this problem is to design sex markers based directly on the SDR region.
In Ectocarpus, sex is determined genetically during the haploid, gametophyte generation by a UV sex chromosome system [26]. The female and male haplotypes of the sex-determination region (SDR) each extend over about one Mbp and constitute about a fifth of the length of the sex chromosome (the remaining four fifths corresponding to the recombining pseudoautosomal region, the PAR, [56]). The male and female haplotypes of the SDR are thought to have stopped recombining at least 70 Mya ago [26] and are highly divergent at the sequence level. One consequence of this sequence divergence is that sex markers can easily be generated based on the amplification of either male or female sequences from within the SDR [26]. Moreover, because the divergence of the male and female SDR haplotype is relatively ancient, these regions can also be used to develop sex markers for other brown algal species, provided they share the same sex-determination system. Chromosome staining experiments and localization of the female specific SCAR marker strongly suggested that one of the chromosomes in Saccharina japonica carries a female-specific region [57] which was recently confirmed by a high density SNP linkage map [58]. However, a male-specific locus could not be visualized due to lack of male-specific markers.
Here we show that sequences from within the male or female SDR can be used to develop sex markers for diverse species of economically important kelps. We present the first report of male sex-specific markers for Laminaria digitata and Undaria pinnatifida, as well as the first sex markers developed for the genus Macrocystis. Our findings are expected not only to facilitate breeding programs of these seaweeds, but also to accelerate the study of kelps at the population and biogeographic level and to enable further research into the evolution of sex determination systems across different brown algal lineages.

Biological material and DNA extraction
Diploid sporophytes of L. digitata, U. pinnatifida and M. pyrifera were collected in the North Atlantic, Mediterranean, and Southeast Pacific (Table 1). Gametophytes were obtained from released zoospores maintained in culture in Petri dishes at 13°C in autoclaved natural sea water (NSW) supplemented with half-strength Provasoli solution [59] with a light:dark cycle of 12h:12h (initially, to avoid gametogenesis, 1-2 μmol photons m −2 s −1 , in later cultures containing larger thalli 20 μmol photons m −2 s −1 ) using daylight-type fluorescent tubes. Individual gametophyte clones were sub-isolated and their phenotypic sexes were determined morphologically by inspecting cell size, female individuals having significantly larger cells than males [18].
Total genomic DNA was extracted from 10-20 mg of wet tissue of using the NucleoSpin Plant II DNA extraction kit from Machery-Nagel (http://www.mn-net.com/), following the manufacturer's instructions. DNA samples were diluted to 1 ng/μl final concentration and stored at -20°C.

Marker design and PCR amplification
Sex-specific markers were designed based on the sequence of the Ectocarpus sp. sex chromosomes [26,32] but also took into account, where possible, similarities between Ectocarpus gene sequences and publically available transcriptomic data for Saccharina japonica [60]. Several genes located either in the SDR or in the pseudoautosomal region (PAR) were targeted (for the list of primers and genes see Table 2). Primers were designed using Primer3 [61,62]. In total, we tested 22 primer pairs corresponding to nine Ectocarpus SDR genes and 15 primer pairs corresponding to eight Ectocarpus PAR genes. Test amplifications to detect sex markers were carried out on three to seven individuals of each sex for each of the three species. PCR was carried out in 12 μL total volume containing 1 ng of genomic DNA template, 2.5 mM MgCl2, 2 mM of each dNTP, 0.5 mM of forward and reverse primer and 0.3 U Taq polymerase (GoTaq, Promega). All loci were amplified using a touch-down PCR procedure with initial denaturation for 3min at 95°C followed by 10 cycles of denaturation at 95°C for 30s, 30s annealing at X°C decreasing 1°C per cycle and elongation at 72°C for 30-90s (see Table 2 for the exact elongation time and initial annealing temperature) and then 25 cycles of denaturation at 95°C for 30s, 30s annealing at (X-10)°C and elongation at 72°C for 30-90s, with a final elongation step of 10 min

Results
Development of sex markers for kelps based on the sequence of the Ectocarpus sex chromosome The objective of this study was to explore the possibility of using information from the recently characterised sex chromosome of the filamentous brown alga Ectocarpus to develop sex markers for economically important kelp species. Previous work has shown that the marked divergence between the male and female haplotypes of the Ectocarpus SDR can be exploited to develop easily scorable, PCR-based sex markers for different species within the genus Ectocarpus [26]. The aim here was to determine whether this approach could be extended to more distantly related brown algal taxa. Current estimates indicate that the Ectocarpales and the Laminariales diverged about 100 Mya [63]. The search for markers therefore focused on protein-coding genes, because these sequences are more likely to show strong conservation across this evolutionary distance than other regions of the genome. Nine genes from within the Ectocarpus SDR were selected as candidates for marker development, but we also included eight genes from the two surrounding, recombining regions (PAR), as controls.
The Ectocarpus SDR includes both sex-specific (male-or female-specific) genes and genes which are present in both the U and V SDR haplotypes (so-called gametologues). Both types of sequence could potentially be exploited for the generation of sex markers because the male and female copies of gametologues are significantly different at the sequence level and markers designed from gametologue sequences have been successfully used to discriminate between sexes in different Ectocarpus species [26]. The nine SDR genes that were selected for marker development therefore included four sex-specific genes (two male-and two female-specific) and five gametologues (S1 Table).
Between one and five primer pairs were designed for each gene and all primer pairs were tested against at least three male and three female individuals of each of three target kelp species, with additional primer combinations being tested for some genes by combining primers from different pairs. The markers were named according to the Ectocarpus gene that had been used to design them, for example M_68_58_1 corresponded to marker 1 based on the male SDR gene Esi0068_0058.
For the test PCR amplifications, we focused on three economically important species of kelp, L. digitata, U. pinnatifida and M. pyrifera, which represent two major families (Laminariaceae for L. digitata and M. pyrifera and Alariaceae for U. pinnatifida). PCR amplifications were performed using genomic DNA extracted from gametophytes derived from individuals that had been collected from wild populations. Of the 22 SDR gene primer pairs tested, fifteen either failed to amplify a PCR product from DNA of any of the kelp species or gave spurious amplifications. Seven primer pairs allowed the successful amplification of products corresponding to three loci (one male-specific gene, Esi0068_0016, and two gametologue pairs, Esi0068_0058/FeV4scaf01_4 and Esi0285_0020/FeV4scaf08_1; Fig 2). Primers based on the Ectocarpus male-specific gene Esi0068_0016 specifically amplified products from L. digitata and U. pinnatifida males. In most cases a similar pattern was observed with primers based on gametologue pairs, with a product being amplified from only one of the sexes (L. digitata females for Esi0068_0058, U. pinnatifida females for Esi0285_0020). An interesting exception was marker M_68_58_1, which amplified products of different sizes from male and female M. pyrifera individuals (Fig 2). This marker alone is therefore sufficient to determine the sex of M. pyrifera individuals but the sex-specific markers also represent valuable sex-markers provided that male-and female-specific markers are combined to avoid the occurrence of false negatives due to non-amplification. Note that, by combining markers for more than one SDR gene, it was possible to define pairs of markers that positively detected both male and female individuals for each of the three test species (Fig 2).
Comparison of the sequences of the sex-specific PCR products amplified from the kelp genomic DNAs with the Ectocarpus genome using Blastn consistently detected the Ectocarpus reference gene used for the primer design as the best match (E value 10e-04 -5e-104).
The diploid sporophyte generation is expected to carry both the male and female haplotypes of the sex locus. To determine whether male and female markers could be employed as codominant markers in this diploid context, six L. digitata sporophytes collected from a wild population in Roscoff, France were tested for amplification. Fig 3A shows that both the male and female markers scored positive for all six individuals. In addition, we tested the ability of the markers to distinguish between diploid sporophytes (resulting from fertilization) and partheno-sporophytes (derived from unfertilized female eggs). Fig 3B demonstrates that female partheno-sporophytes score positive only for the female marker, whereas diploid sporophytes show amplification of both male and female markers.
Of the 15 PAR gene primer pairs, 12 failed to amplify a product from DNA of any of the kelp species. We therefore retained three markers corresponding to three PAR genes for further analysis (Esi0357_0003, Esi0248_0008 and Esi0285_0026). The products amplified from males and females were of identical size suggesting that these regions are not polymorphic and recombine between sexes, and therefore can be used as PCR controls in the kelp species analysed (Fig 2).

Genetic sex markers represent an important new tool for kelp breeding programs
In vitro production of zygotes and young sporophyte seedlings through hybridization of male and female gametophyte lines is an important process in commercial kelp cultivation [64]. The filamentous, microscopic gametophyte generation can be easily maintained in the laboratory and is used not only for controlled breeding but also for germplasm preservation and vegetative propagation [65][66][67][68][69]. Gametophytes can be sexed based on cell size and the presence of distinctive reproductive structures (oogonia and antheridia) but these structures appear only after several weeks of cultivation [70]. Attempts have been made to develop alternative approaches based on sexual dimorphisms to determine sex at an earlier stage of development. For example, measurement of filament width has been used, exploiting sex-dependent differences gametophyte cell size [71,72]. However, this approach cannot be applied in the species where gametophytes are monomorphic such as Phyllariopsis brevipes (C. Agardh) [73], which belongs to the Tilopteridales [12] but resembles a small kelp morphologically, and Laminaria longipes Bory de Saint-Vincent [74], for example. Another possible method is based on differences in fluorescence between the meiospores that will give rise to male and female gametophytes. Flow cytometry experiments detected lower levels of blue fluorescence in male meiospores of three Laminariales species (Alaria marginata, Laminaria saccharina and Cymathere triplicata) [72]. However, survival of meiospores after sorting was low (approximately 1.5%) making this an inefficient method of sex determination. Genetic sex markers represent a rapid, efficient and cost effective method to sex gametophytes and they do not suffer from the limitations encountered with the phenotypic markers described above. In particular, genetic sex markers can be reliably applied to material from any stage of the life cycle and the individuals tested do not need to be sexually mature.
Another life cycle feature that is of interest for kelp breeding is the production of sporophytes by parthenogenetic development of gametes. The homozygotic nature of these individuals allows direct selection of optimal genetic variations underlying desired phenotypic characteristics. Parthenogenesis has been observed under laboratory conditions in S. japonica, L. digitata, M. pyrifera, U. pinnatifida and Lessonia nigrescens when male and female gametophytes have been cultivated separately [50,[75][76][77]. In these studies, female partheno-sporophytes were able to produce normal gametophytes that could be crossed with male gametophytes to produce diploid sporophytes [76]. The co-dominant nature of the sex-markers developed in this study allows their use to discriminate between diploid sporophytes and haploid partheno-sporophytes, and therefore informs on the success of a breeding experiment.
Taken together, our experiments indicate that Ectocarpus SDR sequences can be used to rapidly develop effective and broadly applicable sex markers for kelp species and possibly for other brown algae, particularly species that are more closely related to Ectocarpus than kelps. The wide applicability of this approach is underlined by the fact that two of the markers (M_68_58_1 and M_68_58_2) ( Table 2) amplified sex-specific products from individuals corresponding to two different families within the Laminariales (Laminariaceae and Alariaceae) (Fig 1) [78,79].

Additional applications for sex markers in population and demographic studies
The sex markers developed in this study also have multiple applications for the analysis of the structure of wild kelp populations, allowing the detection and analysis of the life cycle variations in the field. For example, it will be possible to test the degree of parthenogenetic reproduction in marginal habitats at species range limits, a phenomenon that was recently described in populations of L. digitata in Brittany [80]. The study of selection against hybrids during speciation is another possible application, as in the case of the Lessonia nigrescens species complex for which changes in sex ratio was observed specifically in the populations where sister-species were in close contact [81]. The availability of sex markers could allow a better understanding of the meiotic drive leading to sex ratio changes and to hybrid avoidance, and ultimately to mating system evolution. Moreover, as most of the markers span intron regions and are applicable to a broad range of strains or species, they could also potentially be used to sample polymorphism within populations by coupling PCR amplification with sequencing of PCR products.

Sex chromosome evolution
The analyses carried out in this study provide the first evidence that sex chromosomes in two major brown algal orders, the Ectocarpales and the Laminariales, are derived from a common ancestral sex-determination system. All of the kelp sequences that were amplified using primers based on Ectocarpus SDR genes were sex-linked in the kelp species, whereas the markers based on PAR genes that were tested failed to detect evidence of sex-specific alleles in kelps. This suggests that the SDR of Laminariales and Ectocarpales have shared ancestry, i.e, the sexdetermination system originated more than 100 Mya, before the divergence of the two orders [63]. However, more detailed information about kelp genomes, including good quality whole genome assemblies, will be required to further investigate the kelp sex chromosome structure in detail. If sex chromosome systems are conserved across major groups within the brown algae, as this study suggests, this will provide a highly interesting system to investigate how sex chromosomes evolution is influenced by modifications to the sexual system. Brown algal sexual systems are particularly diverse with regard to a number of factors, including for example relative gamete size (isogamy, anisogamy and oogamy), sexual identity (dioicy and monoicy) and the extent to which sexual reproduction is part of the life cycle (asexual cycles, parthenogenesis) [82], and it will be of considerable interest to determine how this diversity has impacted on sex chromosome evolution and function.

Conclusion
In this study we have developed novel DNA-based sex-markers for the three economically important kelp species, L. digitata, U. pinnatifida and M. pyrifera. The markers are robust and easy to apply, because sex is determined simply by PCR-amplification of sex-specific DNA, visualized as bands on agarose gels. This is an advantage over alternatives such as microsatellite-based markers for example, which require the application of more sophisticated analyses to measure the exact size of the amplified product. Moreover, the three target kelp species were phylogenetically diverse, suggesting that the approach should be adaptable to a broad range of species within the Laminariales. The sex markers described here, and the general approach of developing sex markers based on Ectocarpus sex chromosome sequences, represent important new tools for seaweed breeding and biotechnology. They also have potential applications in natural population studies and in studies that aim at understanding the mechanisms and evolutionary principals underlying sex determination in brown algae.
Supporting Information S1 Table. Expression levels in Ectocarpus and the functions of the genes used in this study. (DOCX)