Background: The changing climate and the desire to use renewable oil sources necessitate the development of new oilseed crops. Field cress (Lepidium campestre) is a species in the Brassicaceae family that has been targeted for domestication not only as an oilseed crop that produces seeds with a desirable industrial oil quality but also as a cover/catch crop that provides valuable ecosystem services. Lepidium is closely related to Arabidopsis and display significant proportions of syntenic regions in their genomes. Arabidopsis genes are among the most characterized genes in the plant kingdom and, hence, comparative genomics of Lepidium-Arabidopsis would facilitate the identification of Lepidium candidate genes regulating various desirable traits. Results: Homologues of 30 genes known to regulate vernalization, flowering time, pod shattering, oil content and quality in Arabidopsis were identified and partially characterized in Lepidium. Alignments of sequences representing field cress and two of its closely related perennial relatives: L. heterophyllum and L. hirtum revealed 243 polymorphic sites across the partial sequences of the 30 genes, of which 95 were within the predicted coding regions and 40 led to a change in amino acids of the target proteins. Within field cress, 34 polymorphic sites including nine non- synonymous substitutions were identified. The phylogenetic analysis of the data revealed that field cress is more closely related to L. heterophyllum than to L. hirtum. Conclusions: There is significant variation within and among Lepidium species within partial sequences of the 30 genes known to regulate traits targeted in the present study. The variation within these genes are potentially useful to speed-up the process of domesticating field cress as future oil crop. The phylogenetic relationship between the Lepidium species revealed in this study does not only shed some light on Lepidium genome evolution but also provides important information to develop efficient schemes for interspecific hybridization between different Lepidium species as part of the domestication efforts. Keywords: Domestication, Field cress, Lepidium campestre, Pod shattering, Vernalization Background include reduced seed dormancy, seed dispersal resist- Domestication is the process of selecting genetic ance, free threshing and high number of seeds per plant polymorphisms for traits that suit human needs. The . Historically, domestication of plants has been a slow “domestication syndrome” describes the main differences and a labor-intense task as breeders have relied only on between domesticated and wild plants. Traits that nor- phenotypes to improve the crop. Although this approach mally are associated with the domestication syndrome has been very successful, it has its own drawbacks. It is time consuming, costly and can be influenced by the environment. The recent rapid development of genomics * Correspondence: Mulatu.Geleta.Dida@slu.se can make it possible to domesticate a new plant with Department of Plant Breeding, Swedish University of Agricultural Sciences, less labor and within a significantly shorter time frame. Box 101, SE-23053 Alnarp, Sweden Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Gustafsson et al. BMC Genetics (2018) 19:36 Page 2 of 15 The use of genomic tools and resources can speed up a approaches are being used for domestication of field cress. domestication process as breeders can select plants bear- In line with this, floral dip based-, and tissue culture based ing desirable traits before the traits have been expressed, transformation protocol have been successfully established using appropriate DNA markers and high-throughput for field cress [11, 12]. However, as the use of genetically precise phenotyping. modified plants is still very restricted within Europe, cross- While farming systems with annual crops have provided breeding techniques are still essential. us with unprecedented yield, they have also contributed to As the cost for sequencing technology has been ecosystem problems such as soil erosion and water runoff greatly reduced, comparative genomics has become an . Perennial crops normally have deeper root systems increasingly important tool in agriculture. Arabidopsis that can prevent soil erosion, reduce water runoff and thaliana (thale cress) is among the most investigated nutrient leakage. In addition, perennial crops may require plants in the plant kingdom and since the publication less herbicide treatment as its extended growing season of the genome in the year 2000 , a vast number of enables it to outcompete weeds. The few examples of genes have been identified and characterized. The successful breeding programs of perennial crops include publication of 1135 additional A. thaliana genomes perennial rice and intermediate wheatgrass [3, 4]. There provided information about the global pattern of poly- are also currently interesting attempts to develop peren- morphisms . Seed dispersal resistance, flowering nial oil crops such as sunflower . time, winter hardiness, perenniality, oil content and oil To meet the growing demand for a renewable oil produc- quality are among the major traits targeted for field tion and reduce the negative effects of annual crop systems, cress breeding at early stages of the domestication we are in the process of domesticating a new seed oil crop: process. A number of genes regulating these traits are field cress (Lepidium campestre (L.) R. Br., also called well known in Arabidopsis and Brassica, which are pepperworth), which is a wild species in the Brassicaceae in the same Brassicaceae family as field cress. The close family. It has many good agronomic characteristics, which evolutionary relationship between Arabidopsis and field makes it a promising candidate for domestication. It has a cress makes Arabidopsis an ideal species to identify relatively small diploid genome with a sporophytic orthologs of these genes in field cress through com- chromosome number of 2n =16 . Field cress is a parative genomics. Analyses of genes that are associated self-fertilized biennial plant with closely related perennials with vernalization and flowering time is important in (L. hirtum (L.) Sm. and L. heterophyllum Benth.), which can order to facilitate the development of a winter hardy be used as a source of perenniality genes for the develop- perennial field cress. MADS-box genes such as FLOW- ment of perennial field cress . Although polyploids are ERING LOCUS C (FLC) play a key role in regulating common among Lepidium species, both L. heterophyllum plant developmental responses to temperature [15, 16]. and L. hirtum are diploids with 2n =16 chromosomes and The gene FRIGIDA (FRI) enhances the expression of can be cross-hybridized with field cress to produce viable FLC and has been demonstrated to be of great import- hybrid plants . Moreover, field cress is extremely winter ance for vernalization and flowering time [17, 18]. hardy, resistant to the pollen beetle, have a high seed yield VERNALIZATION INSENSITIVE 3 (VIN3) is part of − 1 (> 4500 kg ha ) and a good seed size (half the size of that the polycomb repression complex PRC2 which estab- of large seeded rapeseed) [9, 10]. Field cress produces a lishes repression of FLC and is one of the earliest re- high-quality seed oil suitable for industrial use, with sponses to vernalization [19, 20]. Downregulation of linolenic and erucic acids being the main components. FLC promotes floral development and the low levels of Field cress has been targeted for domestication as an FLC are maintained after the cold treatment by RE- under sown cover-, catch- and oilseed crop to prevent nu- DUCED VERNALIZATION RESPONSE 1 and 2 (VRN1 trient leaching as well as being a high yielding crop . It and VRN2) which serves as a molecular memory of the can be under sown with a spring cereal and serve as a vernalization [20, 21]. SUPRESSOR OF OVEREXPRES- cover and catch crop, after the harvest of the cereal in the SION OF CO1 (SOC1, also AGAMOUS LIKE 20/AGL20) summer, and finally harvested as an oil crop the second and FRUITFULL (FUL) are controlling flowering as well year. Hence, field cress holds high agronomic potential and as determinacy of meristems but also seem to be in- eco-friendliness as it could simultaneously function as a volved in inhibiting longevity in annual plants [22–24]. high yielding oil crop and a cover/catch crop. Unlike rape- FLC, FRI, VIN3, VRN1 and VRN2 were among main tar- seed, which is not widely cultivated in areas with strong gets in the present study as their role in vernalization winter due to its poor winter-hardiness, field cress has and flowering time has been well established. SOC1 and been proven to be productive in a colder climate. Thus, FUL were also interesting for their reported effect on cultivation of field cress could be of great value to farmers perenniality. Analysis of field cress orthologs of AGA- in the northern parts of temperate regions, e.g. Nordic MOUS LIKE 6 (AGL6), AGAMOUS LIKE 16 (AGL16) Europe. Both crossbreeding and genetic engineering based and MADS AFFECTING FLOWERING 2–5 (MAF2–5) Gustafsson et al. BMC Genetics (2018) 19:36 Page 3 of 15 which are reported to have effect on these traits [25–28] among 18 Lepidium samples used for sequencing the are also of interest. restriction site associated DNA (RAD) . The remaining Shattering of pods (fruit dehiscence) can lead to sig- 19 individuals –comprising 14 field cress, three L. hirtum nificant yield losses and pod shattering resistance is and two L. heterophyllum– (Table 1)wereused for the therefore a highly advantageous trait in any domesti- sequencing of partial sequences of target genes after they cated crop. Hence, identification and analysis of tran- were identified based on comparative genomic analysis of scription factor genes, FUL and REPLUMLESS (RPL), Lepidium RAD-sequences and genomic sequences of together with the valve identity genes INDEHISCENT Arabidopsis and Brassica.These Lepidium accessions were (IND), ALCATRAZ (ALC), SHATTERPROOF 1 and 2 obtained from various genebanks and botanical gardens in (SHP1 and SHP2) that are responsible for the establish- Europe as well as after collecting populations from various ment of the valve margin in the seed-containing pod, regions in Sweden. Except the CHe hybrids, the samples and regulate thereby the release of seeds from the pod, were selfed at least for four generations in a greenhouse. [29–34] should be targeted as important steps towards The CHe hybrids are the result of interspecific the development of shatter proof field cress cultivars. In hybridization followed by two rounds of selfing. These addition, genes encoding ARABIDOPSIS DEHISCENCE samples were chosen for their geographical spread and ZONE POLYGALACTURONASE 1 and 2 (ADPG1 and their phenotypic variation. ADPG2) and the NAC DOMAIN CONTAINING PRO- TEIN 12 (NAC012/SND1) which are contributing to Comparative genomics/bioinformatics and primer design fruit dehiscence [35, 36] are also important targets. The RAD-sequencing conducted on 18 Lepidium samples Because field cress is targeted as an oilseed crop, genes comprising 10 field cress, two F CHe hybrids and three that code for both oil content and quality need to be progenies of each of the two CHe hybrids at Edinburgh analyzed as integral part of the domestication process so Genomics (School of Biological Sciences, University of that acceptable levels of oil content and quality can be Edinburgh, EH9 3FL, Edinburgh, UK) produced over achieved. Among the various genes involved in the 190,000 consensus RAD-sequences . A BLAST search biosynthesis of fatty acid ACYL-COA:DIACYLGLY- for sequences that match the RAD sequences in the CEROL ACYLTRANSFERASE 1 (TAG1) and WRINKLED National Center for Biotechnology Information (NCBI) 1 (WRI1) regulate oil production in many species, in- database identified sequences that have 75 to 90% cluding Arabidopsis and FATTY ACID DESATURASE 2 sequence identity with the Arabidopsis genome sequences, − 11 (FAD2), FATTY ACID ELONGATION 1 (FAE1) and with e-values ranging from 0.0 to 1e .Additionaltop 3-KETOACYL-COA-SYNTHASE 8 (KCS8) are control- hits included sequences of other Brassicaceae species such ling seed oil composition in most oil crops [37–42]. as Brassica rapa, Camelina sativa and Boechera stricta. Other important traits in field cress include host plant All DNA sequences used for comparative genomic ana- resistance to pathogens and seed dormancy. Hence, lysis except those of Lepidium were retrieved from the genes involved in regulating these traits are important NCBI database. targets. Homologous sequences in field cress for host DNA sequences of 30 genes (see Table 2)from A. thali- plant resistance gene FERONIA (FER) and plant defense ana and field cress known to be regulators of desirable gene AUTOPHAGY RELATED 5 (ATG5), genes that are traits for plant domestication were used as query se- involved in germination of seeds AGAMOUS-LIKE 11 quences to find any homologous sequences in the RAD (AGL11)and HIGHLY ABA-INDUCED PP2C PROTEIN sequence pool. However, since the RAD-sequences were 2 (HAI2/AIP1)[43–46] were also in focus. short sequences (117 nucleotides (nt) to 567 nt long), they The present study aimed at the identification of the only represent short segments of these genes. In total, orthologues of the aforementioned genes in field cress DNA sequences homologous to partial sequences of 24 through comparative analysis of its genome with that of different genes were identified through using this Arabidopsis and Brassica; characterization of the genetic approach (Additional file 1: Table S1). Primers were then diversity of these genes in field cress; and analyzing the designed using field cress RAD-sequences as a template effect of different mutations in each gene on the plant by mainly targeting coding regions in 18 of these 24 genes. phenotypes in field cress and related species. Although sequences homologous to partial sequences of the remaining six genes (AGL11, ALC, FUL, HAI2, IND, Methods and SOC1) were found among the RAD-sequences, con- Plant material served Arabidopsis sequences were used to design primers A total of 31 individual plants of field cress, L. hirtum, L. in order to expand the regions to be sequenced. Six of the heterophyllum and interspecific hybrids of field cress and 30 genes (ADPG1, AP2, FAE1, FLC, NAC012 and RPL) L. heterophyllum (CHe hybrids) was used for this study did not show any homology to the RAD sequences, and (Table 1). Ten field cress and two F CHe hybrids were hence conserved sequences in A. thaliana or previously 3 Gustafsson et al. BMC Genetics (2018) 19:36 Page 4 of 15 Table 1 Sample codes, source and country of origin of different genotypes of three Lepidium species and CHe hybrids analyzed for genetic variation within partial sequences of various genes regulating desirable traits. L. hirtum is represented by three subspecies. Genotypes 1–14 and 27–31 were amplified and sequenced using newly designed primers (Additional file 1: Table S1) while genotypes 15–26 were those used in the RAD-Sequencing project No Sample code Species Source accession/population Country of origin name obtained from 1 LcSma L. campestre Mörbylånga Newly collected Öland, Sweden 2 LcSstu L. campestre Stuvsta Newly collected Södermanland, Sweden 3 LcCze L. campestre PI 633248 USDA-ARS Czechoslovakia 4 LcGer1 L. campestre LEP 122 IPK, Germany Germany 5 LcSho L. campestre Höör2 Newly collected Skåne, Sweden 6 LcGer2 L. campestre LEP 93 IPK, Germany Germany 7 LcSlj L. campestre Ljugarn Newly collected Gotland, Sweden 8 LcGer3 L. campestre PI 633251 USDA-ARS Germany 9 LcSkr L. campestre Kristianstad Newly collected Skåne, Sweden 10 LcSar L. campestre Årsta1 Newly collected Södermanland, Sweden 11 LcSvi L. campestre Viken Newly collected Skåne, Sweden 12 LcSsk L. campestre Skövde Newly collected Västergötland, Sweden 13 LcSsp L. campestre Spjutstorp Newly collected Skåne, Sweden 14 LcSve L. campestre Ventlinge Newly collected Öland, Sweden 15 LcFra L. campestre PI 633252 USDA-ARS France 16 LcGre1 L. campestre LEP 89 IPK, Germany Greece 17 LcSst L. campestre 094–10 Newly collected Stjärnelund, Sweden 18 LcUK L. campestre 0018580 Royal Botanic Garden, UK ? 19 LcSga L. campestre Gävle Newly collected Gävleborg, Sweden 20 LcStr L. campestre Trelleborg Newly collected Skåne, Sweden 21 LcGer4 L. campestre LEP 94 IPK, Germany Germany 22 LcGre2 L. campestre LEP 92 IPK, Germany Greece 23 LcDen1 L. campestre 4932 4 Botanic Garden-Denmark ? 24 LcDen2 L. campestre NGB22634 NordGen, Sweden Denmark a a 25 CH1 L. campestre x heterophyllum LEP 89 & 597856 IPK, Germany & USDA-ARS Germany & Spain a a 26 CH2 L. campestre x heterophyllum Huddinge & 597856 Newly collected & USDA-ARS Sweden & Spain 27 LheGer L. heterophyllum 1988/690–148 Marburg Bot. garden, Germany ? 28 LheSha L. heterophyllum Hästveda Newly collected Skåne, Sweden 29 LhiIta L. hirtum ssp. nebrodense PI633253 USDA-ARS Italy 30 LhiSpa L. hirtum ssp. calycotrichum PI597858 USDA-ARS Spain 31 LhiMor L. hirtum ssp. atlanticum Ames 21387 USDA-ARS Morocco hybrid of L. campestre and L. heterophyllum;? = no information published field cress mRNA sequences were used as a (0.1 M Tris, 20 mM EDTA, 1.4 M NaCl, 2% CTAB, template to design primers for amplification of ortholo- pH 7.5) for 1 h at 52 °C. The samples were then centri- gous regions in field cress. fuged for 15 min at 14.1 rpm in an Eppendorf miniSpin tabletop centrifuge. The supernatant was then trans- DNA extraction ferred to sample plate and DNA was extracted using the Genomic DNA was obtained by sampling leaf tissue Qiacube DNA extraction robotic workstation (Qiagen). from young plants, grown in a green house, which was The extracted DNA was finally ran on a 1% agarose gel flash-frozen in liquid nitrogen and homogenized by and checked for impurities on a Nanodrop. DNA from vigorously shaking in a Retsch MM400. The samples the 19 individuals (Table 1) was used for screening par- were incubated with 1 ml of pre-warmed CTAB buffer tial sequences of the 30 target genes (Table 3). Gustafsson et al. BMC Genetics (2018) 19:36 Page 5 of 15 Table 2 Sequence identity (%) between field cress and seven other Brassicaceae species within partial sequences of coding regions of 30 genes regulating desirable traits in crops Gene Trait/gene function A.lyrata A. thaliana B.rapa B.napus B.oleracea Camelina sativa Capsella rubella Mean AGL11 SD 95 95 91 92 92 96 96 93.9 SOC1 FT 94 94 93 93 93 94 93 93.4 TAG1 OC 94 93 93 93 93 93 95 93.4 AGL6 FT 95 98 90 91 91 92 92 92.7 VRN1 VRN 94 94 90 90 90 92 94 92.0 NAC012 PSH 92 92 91 91 90 93 92 91.6 FER DR 93 93 89 89 89 92 91 90.9 FUL FT, PSH 92 92 89 89 89 90 90 90.1 FLC FT, VRN 89 90 88 88 88 90 92 89.3 MAF2 VRN 93 95 87 84 85 91 89 89.1 MAF5 VRN 93 95 87 84 85 91 89 89.1 AP2 FT 89 89 89 89 89 90 88 89.0 ADPG1 PSH 89 91 88 86 87 91 90 88.9 SHP1 PSH 90 90 89 89 89 89 86 88.9 SHP2 PSH 91 89 86 87 87 89 92 88.7 FAD2 OQ 91 91 85 85 85 90 91 88.3 VRN2 VRN 92 90 83 84 84 89 88 87.1 GTR2 GTR 88 88 86 86 86 88 88 87.1 FAE1 OQ 89 88 85 85 85 89 88 87.0 RPL FT, PSH 89 89 85 85 84 88 88 86.9 VIN3 VRN 87 86 85 85 84 85 85 85.3 WRI1 OC 88 88 82 82 83 86 87 85.1 KCS8 OQ 88 87 82 83 82 87 86 85.0 ADPG2 PSH 85 84 82 82 82 89 90 84.9 IND PSH 86 85 84 86 76 85 84 83.7 ALC PSH 82 83 81 81 81 82 84 82.0 AGL16 FT 82 84 81 81 82 82 81 81.9 FRI VRN, FT 84 81 81 81 80 82 83 81.7 HAI2 SD 82 82 81 80 80 81 82 81.1 ATG5 PD 82 81 79 80 72 84 80 79.7 Mean 89.3 89.2 86.1 86.0 85.4 88.7 88.5 87.6 DR Disease resistance, FT Flowering time, OC Oil content, OQ Oil quality, PD Plant defense, PSH Pod shattering, SD Seed dormancy, GTR Glucosinolate Transport, VRN Vernalization. Note: The matching of the Lepidium sequences with the right Arabidopsis gene sequences is highly significant, with e-values ranging from 0.0 −11 to 1e PCR and DNA sequencing GenBank and the accession numbers of the sequences are PCR mix for a 25 μl reaction volume contained 1× PCR given in Additional file 2: Table S2. buffer, 1.5 mM MgCl , 0.3 mM dNTPs, 0.2 μMof each primer, 1 U Dream Taq polymerase (ThermoFisher Scien- Sequence analyses tific) and 1 ng/μl DNA template. The conditions of PCR The diversity analyses are based on the 12 individuals (10 cycling were: 95 °C for 5 min; 35–40 cycles of 95 °C for field cress and two CHe hybrids) from the 15 s, 52 to 60 °C, (depending on the primer-pair) for 15 s, RAD-sequencing and the 19 individuals (14 field cress, and 72 °C for 40 s; and a final step of 72 °C for 10 min. three L. hirtum,two L. heterophyllum)fromthe resequen- PCR product clean-up and sequencing was performed at cing work. From the RAD-sequencing data, variants corre- MWG Eurofins Genomics, Germany. DNA sequence data sponding to each individual were generated using GATK’s generated in this study have been deposited at the NCBI/ HaplotypeCaller command (GATK version 3.5; a toolkit Gustafsson et al. BMC Genetics (2018) 19:36 Page 6 of 15 Table 3 Variation (SNPs and indels) found in aligned sequences of L. campestre, L. heterophyllum and L. hirtum as well as in aligned sequences of different L. campestre genotypes. The analyzed sequence length, number of polymorphism (Polym), number of nonsynonymous mutations (Non-syn), the number of species specific polymorphisms (Species spec. polym) and percent polymorphism per nucleotide (Polym/nt) are listed according to gene Gene Trait/gene function Analyzed L. campestre + L. hirtum + L heterophyllum L. campestre only sequence a b a a c a b a Polym Non-syn Polym/nt Species spec. polym , Polym Non-syn Polym/nt length (%) (%) FRI vernalization, flowering time 606_498 13_8 6 2.1_1.6 0_0 (0) 3_3 2 0.5_0.6 MAF2 vernalization 532_181 7_3 2 1.3_1.7 0_0 (0) 7_3 2 1.3_1.7 MAF5 vernalization 365_206 4_4 1 1.1_1.9 0_1 (0) 1_1 0 0.3_0.5 VIN3 vernalization 1347_1121 7_5 4 0.5_0.4 0_1 (1) 0_0 0 0_0 VRN1 vernalization 1545_501 2_1 1 0.1_0.2 0_0 (0) 2_1 1 0.1_0.2 VRN2 vernalization 1405_536 3_1 0 0.2_0.2 0_1 (0) 0_0 0 0_0 FLC vernalization, flowering time 840_170 17_5 2 2_2.9 9_5 (3) 0_0 0 0_0 AGL6 flowering time 570_182 2_0 0 0.4_0 1_0 (0) 0_0 0 0_0 AGL16 flowering time 772_182 9_1 1 0.3_0 3_0 (1) 1_0 0 0.13_0 AP2 flowering time 750_447 9_4 2 1.2_0.9 0_4 (0) 0_0 0 0_0 SOC1 flowering time 797_290 6_1 1 0.8_0.3 0_0 (0) 0_0 0 0_0 FUL flowering time, pod shattering 942_295 3_0 0 0.3_0 1_0 (1) 0_0 0 0_0 RPL flowering time, pod shattering 560_524 1_1 1 0.2_0.2 0_0 (0) 1_1 1 0.2_0.2 ADPG1 pod shattering 650_138 15_0 0 2.3_0 11_0 (2) 0_0 0 0_0 ADPG2 pod shattering 465_212 4_3 0 0.9_1.4 0_0 (0) 0_0 0 0_0 ALC pod shattering 849_321 7_1 1 0.8_0.3 3_0 (2) 0_0 0 0_0 IND pod shattering 1025_440 11_5 3 1.1_1.1 0_3 (2) 0_0 0 0_0 NAC012 pod shattering 400_305 1_0 0 0.3_0 0_0 (0) 0_0 0 0_0 SHP1 pod shattering 1112_330 19_5 0 1.7_1.5 0_0 (0) 11_3 0 1_0.9 SHP2 pod shattering 962_411 15_6 2 1.6_1.5 1_1 (1) 1_1 1 0.1_0.2 FAD2 oil quality 710_416 5_4 2 0.7_1 0_0 (0) 1_1 1 0.1_0.2 FAE1 oil quality 490_478 2_2 0 0.4_0.4 0_2 (0) 0_0 0 0_0 KCS8 oil quality 1382_1362 12_9 5 0.9_0.7 0_0 (0) 0_0 0 0_0 TAG1 oil content 962_414 15_5 0 1.6_0 1_0 (0) 1_0 0 0.1_0 WRI1 oil content 1738_736 10_3 1 0.6_0.4 1_0 (1) 0_0 0 0_0 AGL11 seed dormancy 587_253 8_1 0 1.4_0.4 1_0 (1) 1_0 0 0.2_0 ATG5 plant defense 1672_392 11_0 0 0.7_0 5_0 (1) 2_0 0 0.1_0 FER disease resistance 1112_1112 7_7 1 + 2 0.6_0.6 0_0 (0) 2_2 1 0.2_0.2 GTR2 glucosinolate transport 1200_1016 7_6 1 0.6_0.6 1_3 (4) 0_0 0 0_0 HAI2 seed dormancy 1034_865 11_4 1 1.1_0.5 1_1 (1) 0_0 0 0_0 Total 27,381_14,334 243_95 40 39_22 (21) 34_16 9 Mean 0.9_0.8 0.14_0.16 a b Values to the left of the underscore are for full length of the sequenced regions whereas values to the right of the underscore are only for the coding regions; c d = applies only to the coding regions; = values within parentheses refer to the total number of polymorphisms unique to L. campestre; = polymorphism predicted as deleterious for protein function developed by Broad Institute). These variants were subse- sites to ensure that these sites were correctly scored. Then, quently used to generate individual specific consensus multiple alignment for each of the 24 gene sequences of the sequences using GATK’s FastaAlternativeReferenceMaker 31 samples (19 samples from resequencing and 12 samples command . All chromatograms generated through rese- from RAD-sequencing) was performed using ClustalX ver- quencing of the 19 samples were visually evaluated using sion2.1 software , and then the sequences were manu- BioEdit version 7.0.5 , especially at the polymorphic ally edited using BioEdit version 7.0.5. Finally, variable sites Gustafsson et al. BMC Genetics (2018) 19:36 Page 7 of 15 in the forms of single nucleotide polymorphism (SNPs) and the gene sequences had at least one polymorphism that indels were tabulated. The same procedure was followed lead to an altered amino acid sequence. Polymorphisms for the remaining six genes except that only sequences that were predicted to be deleterious for protein function from 19 genotypes were used. The online PROVEAN tool were recorded in six of the genes (AGL16, FER, GTR2, (J. Craig Venter Institute) was used to predict the impact of HAI2, MAF5 and SOC1)(Table 3). the polymorphisms leading to a change in the amino acid In total, 243 polymorphic sites were observed in 27,381 sequence . nucleotides long aligned sequences, which constitutes about 0.9% of the total analyzed nucleotides. The majority Phylogenetic analyses of the variations were single nucleotide substitutions but Two data sets were used for phylogenetic analyses of dif- indels of variable lengths were also observed. Ninety-five of ferent genotypes of three Lepidium species and their hy- the polymorphisms were located in the predicted coding brids. The first data set contains polymorphism only regions which constitute about 0.7% of the total length of within the coding regions of the 30 genes whereas the the coding regions analyzed (14,334 nt). None of the second data set contain polymorphism within both the sequences analyzed had mutations that would lead to an coding and non-coding regions of these genes. MEGA7 obvious severe defect in the protein, such as premature  software was used for the phylogenetic analyses STOP codon or frame-shift indels. However, 40 of the through the application of methods that include Max- identified polymorphisms were nonsynonymous and would imum Likelihood, Neighbor-Joining, Minimum Evolu- result in an altered protein sequence. Seven of these poly- tion, and Maximum Parsimony. However, most of the morphisms were SNPs that were predicted to be deleteri- analyses produced highly similar results and hence trees ous for the protein function (Table 3). Moreover, 61 of the generated using the Neighbor-Joining method  based observed polymorphisms (39 in non-coding, and 22 in on the evolutionary distances calculated according to coding regions) were shown to be species specific, of which Nei and Kumar  are presented in this paper. The 21 were unique to field cress. bootstrap test  for the robustness of the tree Within the field cress group of 24 individuals, 34 poly- branches was conducted based on 10,000 replications. morphisms were observed in 13 different genes. Sixteen of the polymorphisms were located in the predicted cod- Results ing regions of the FAD2, FER, FRI, MAF2, MAF5, RPL, To identify polymorphisms that may contribute to domes- SHP1, SHP2 and VRN1 genes. Nine of these polymor- tication syndrome traits, perenniality and oil content and phisms are nonsynonymous but none were predicted to quality, partial sequences of 30 genes were analyzed in have a severe impact on the protein function, according Lepidium. The protein coding sequences of field cress to PROVEAN. The gene sequence of SHP1 stands out showed a high level of homology to A. thaliana with the by having most polymorphisms among the field cress in- sequence identity between A. thaliana and field cress ran- dividuals with six SNPs and two indels in the ging between 81 to 98% with a mean value of 89.2% (Table non-coding region and additionally three SNPs in the 2). A similar level of average sequence identity (89.3%) coding region. MAF2 has variable sites that accounted was observed when comparing field cress to A. lyrata,al- for 1.3% of the total nucleotides analyzed and 1.7% in though differences were observed in more than 50% of the coding region, which is significantly higher than the genes investigated. The comparative sequence analysis other genes in this study. with additional Brassicaceae species showed that an aver- Two different data sets were used for analysis of age sequence identity between field cress and Camelina phylogenetic relationship between the individuals repre- sativa, a minor oil crop grown in Europe and North senting the three Lepidium species and CHe hybrids: a America was 88.7%; Capsella rubella –a model plant used data set that contains only sequences of coding regions to study self-incompatibility in plant reproduction– and another one that contains sequences of both coding showed similar level of sequence identity with field cress and non-coding regions. Cladograms generated based on (88.5%). The commercially grown agricultural and horti- both data sets showed a formation of clearly separated cultural crops B. napus, B. oleracea and B. rapa shared a branches representing the different species and the slightly lower sequence identity with field cress with a hybrids with a strong bootstrap support (Figs. 1 and 2). mean of around 86% for the coding regions. In both cases, field cress is shown to be most closely re- The alignment of partial gene sequences of field cress, lated to CHe hybrids and then to L. heterophyllum.The CHe hybrids, L. heterophyllum,and L. hirtum revealed two cladograms slightly differ in the clade that com- polymorphisms in all of the analyzed gene sequences. The prised the field cress individuals although both showed a distribution of polymorphisms for each gene sequence are clear separation of genotypes LcGre2 and LcSma from listed in Table 3. Twenty-five of the gene sequences had the rest. Three genotypes (LcGer1, LcSlj and LcSsk), two SNPs or indels in the predicted coding regions and 19 of of which represent Sweden and one represents Germany, Gustafsson et al. BMC Genetics (2018) 19:36 Page 8 of 15 Fig. 1 Cladogram showing the clustering pattern of individual genotypes representing field cress, CHe hybrids, L. heterophyllum and L. hirtum based on polymorphisms in coding regions. Green diamonds denote field cress accessions collected in Sweden and grey diamonds field cress from other parts of Europe. Green triangles = CHe hybrids, red triangles = L. heterophyllum and blue circles = L. hirtum. Numbers at the base of branches are bootstrap values formed a separate branch with a moderate level of boot- species. The close evolutionary relationship between strap support in both data set. The use of combined data field cress and the extensively studied model plant A. of coding and non-coding sequences produced a better thaliana is advantageous and will be highly valuable for resolution in this clade although some of the branches the domestication and further breeding of field cress. were supported by low bootstrap values, which is mainly Alignments of the coding regions of the 30 genes because of low sequence divergence between the differ- confirm the close phylogenetic relationship between ent genotypes of field cress. these two species. Comparing the mean sequence iden- tity of the 30 genes suggests that Lepidium is more Discussion closely related to Arabidopsis than to Brassica (Table 2). Sequence homology between various genes of field cress The data also suggests closer evolutionary relationship and other Brassicaceae species of Lepidium with Camelina and Capsella than with Comparative genomics is a powerful tool that facilitates Brassica. This finding correlates well with the recently our understanding of the genetics of poorly studied plant released phylogeny of the Brassicaceae family using 113 species based on data from a closely related well-studied nuclear DNA markers . Gustafsson et al. BMC Genetics (2018) 19:36 Page 9 of 15 Fig. 2 Cladogram showing the clustering pattern of individual genotypes representing field cress, CHe hybrids, L. heterophyllum and L. hirtum based on polymorphisms in both coding- and noncoding regions. Green diamonds denote field cress accessions collected in Sweden and grey diamonds field cress from other parts of Europe. Green triangles = CHe hybrids, red triangles = L. heterophyllum and blue circles = L. hirtum. Numbers at the base of branches are bootstrap values Many of the genes that are regulating flowering time seems to have a less pronounced role in flowering time and vernalization have more than 90% sequence identity than in A. thaliana . to the Arabidopsis counterpart in this study (Table 2). Moreover, the top three genes with the highest level of Sequence variation in various genes between and within polymorphisms per nucleotide in the coding regions Lepidium species (FLC, FRI and MAF5) also belong to this group of traits. Our genomic data showed that field cress is more closely However, in this group of genes, FRI stands out as one related to L. heterophyllum than to L. hirtum (Figs. 1 and of the least conserved genes as it has a lower level of 2), which is in agreement with previous research based on sequence homology to the other species it was compared chloroplast DNA analysis . Field cress is a with and has a relatively high level of variation in the self-fertilizing plant, and as inbred populations they are coding region. Thus, this result suggests that FRI is expected to have low within-population variation. Hence, evolving faster than most of the other genes analyzed in the clustering pattern of the genotypes (Figs. 1 and 2) this study. Although FRI has been attributed to play a clearly show lack of population differentiation in field major role in response to vernalization and controlling cress according to geographical origin within Europe. flowering time in A. thaliana, the function of this gene Genotypes representing Sweden and other parts of Europe in other species such as A. lyrata seems to be of less im- are distributed across the sub-branches within the field portance . The role of FRI has not been completely cress branch. This is similar to what was previously elucidated for B. rapa, B. napus and B. oleracea, but it reported for A. thaliana; i.e., lacking a clear branching of Gustafsson et al. BMC Genetics (2018) 19:36 Page 10 of 15 different ecotypes, and only significant isolation by dis- explained by the fact that individual plants displaying tance –such as separate continents– divides populations clear phenotypic differences for these traits were into distinct clusters . However, the addition of con- intentionally included in this study. A number of poly- siderably more DNA markers and screening of additional morphic sites among the three Lepidium species in the individuals have altered this hypothesis and clear global genes regulating oil content and quality was observed, population structures seem to exist for A. thaliana . but only two polymorphisms were observed within the In this study, only 24 field cress genotypes and short seg- field cress for these genes, whereof one is an indel in ments of the 30 genes were analyzed. Hence to reach a FAD2 causing a deletion of a serine residue in two of the firm conclusion regarding the lack of population structure field cress genotypes. Field cress has generally a higher in European field cress populations, the use of additional oil content, with average of about 20%, than the other polymorphic DNA markers and genotyping more individ- two Lepidium species, with average about 15% (unpub- uals may be necessary. lished data). Hence further analysis of the variations The genotype LcGre2 differentiated from the main body within these genes followed by well-thought crossbreed- of the field cress genotypes (Figs. 1 and 2). This genotype ing may lead to an increase in oil content in field cress. is of particular interest as it is more resistant to pod shat- Flowering time is one of the most important agro- tering and has a higher oil content compared to other field nomic traits and is preceded by a vernalization period in cress genotypes (unpublished data). It carries two muta- perennial, biennial and winter type annual plants. Tim- tions in the non-coding part of the SHP1 gene which ing of flowering at the most favorable conditions is im- makes it different from all other field cress genotypes in- portant to optimize the seed production in a crop. cluded in the present study, and which may be linked to Vernalization ensures that the frost-sensitive transition mutation(s) in the coding region of this gene that confer from vegetative to reproductive growth to occur at shattering resistance. However, no mutations were found milder temperatures. In this study, 13 gene sequences in the genes that encode for oil content and quality, and related to flowering time and vernalization were ana- hence sequencing the full coding regions of these genes is lyzed for natural variation and compared with other needed to identify any useful mutations. One synonymous closely related species in terms of sequence homology. and one nonsynonymous mutation in the coding regions This group of genes show the highest level of variation that contributed to the separation of this genotype from in the sequenced coding regions, compared to the genes other field cress genotypes were found in the FRI gene, coding for other traits considered in this study. FLC, which regulates vernalization. No distinct vernalization MAF5, MAF2 and FRI come on top as per their poly- requirement or winter hardiness were observed for the morphisms per nucleotide in the protein coding regions accession from which this genotype was developed. (Table 3), suggesting that they are the fastest evolving Multi-environment field trials are needed to determine genes, in that order, when compared to other genes. the effect of these mutations. Similarly, LcSmar, a Lepidium FRI also has a lower sequence homology to A. relatively high seed yielding genotype, was also clearly sep- thaliana FRI (AtFRI), compared to other genes in this arated from the field cress cluster and have several muta- study except ATG5, which has the same level of hom- tions in the coding regions of FRI, VRN1 and MAF5, ology. In this gene, a high degree of variation between which are genes regulating vernalization. This genotype the biennial species (field cress) and the perennial spe- has a unique SNP mutation in FRI, which is of particular cies (L. heterophyllum and L. hirtum) was observed, thus interest. Hence, it will be very interesting to see how this the genetic variation between the two groups could be accession perform in upcoming field trials held in the linked to perenniality. northern part of Sweden. In our ongoing crossbreeding The transition between vegetative and reproductive experiments, hybridization of some field cress genotypes phase is mainly controlled by FLC and its positive regu- results in F hybrids with strong hybrid vigor, and some lator FRI. In fact, polymorphisms in these genes account recombinant inbred lines derived from such hybrids main- for most of the natural variation found in flowering time tain their vigor. On the other hand, some crosses result in in A. thaliana . Several naturally occurring weak F hybrids and subsequent generations. Hence, the non-functional AtFRI alleles have been reported and the clustering pattern of the 24 field cress genotypes in the majority of them derive from deletions in the protein present study facilitates the designing of an efficient cross- coding region [17, 62]. Six nonsynonymous SNPs were breeding scheme to identify best pairs of genotypes in observed in the FRI sequence among the three Lepidium terms of their combining ability. species but no polymorphisms resulting in a deletion of Genetic variation was found in all analyzed genes but amino acid sequence was found. Our data shows a high a higher level of polymorphisms was observed in the level of polymorphisms in the FRI gene, both in coding coding regions of genes regulating flowering time, and non-coding sequences. This finding may indicate vernalization and pod shattering. This result could be that FRI does not have a dominant role in the biennial Gustafsson et al. BMC Genetics (2018) 19:36 Page 11 of 15 field cress as it has been shown that FRI has a redundant region is less variable than other parts of the gene as function in A. lyrata, which has a perennial life cycle only 2% of the nucleotides are polymorphic in A. . As mentioned above, the sequence homology for thaliana. However, that is still ten times more variable the coding sequences of the FRI gene is higher between than in Lepidium. Hence, the VRN2 gene appears to be field cress and A. lyrata (84%) than between field cress more conserved within the Lepidium species than in A. and A. thaliana (81%). However, the homology is basic- thaliana. Moreover, very few polymorphisms were ally the same for both pairs at the protein level. observed in VRN1 among the Lepidium species in this Multiple studies in A. thaliana failed to find any non- study, but this is more in line with what have been synonymous polymorphisms in the AtFLC gene. Of the reported for the Arabidopsis ortholog. few polymorphisms found, the majority were located in SOC1 is an integrator of the floral pathway under dir- the first intron . One of these polymorphisms separ- ect repression of FLC. FUL is a MADS box gene that ate AtFLC alleles into two distinct haplogroups that operates downstream of SOC1 in the same pathway. Per- flowers at significantly different times in a null-FRI back- enniality is normally considered to be a complex trait. ground . The low abundance of polymorphisms in However, a double null mutant of FUL and SOC1 could the coding region of AtFLC is in line with our findings transform an annual Arabidopsis into something that is in the partial sequences of field cress FLC gene, as no reminiscent of a perennial plant . While the controls variation was detected among the field cress genotypes. senesced after flowering, the FUL/SOC1 mutant plant However, among the three Lepidium species FLC had returned to vegetative stage and eventually developed one of the highest number of polymorphisms per nucle- into a highly-branched shrub with woody stems, which otides, both in coding and non-coding regions. Hence, are traits associated with perennials. Moreover, expres- although this gene seems to be conserved within the sion levels of FUL and SOC1 are significantly reduced in field cress, a high level of diversity between closely the A. thaliana ecotype SY-0, which has a similar related Lepidium species is apparent. On the other hand, morphology as the FUL/SOC1 mutant . This indicate the sequence identity between field cress and several that SOC1 and FUL genes may be important in control- Brassicaceae species was overall high with a sequence ling longevity in a plant. Only a single nonsynonymous identity of 88 to 92% for the coding sequence of this polymorphisms have been reported for AtFUL and none gene (Table 2). The FLC gene in A. lyrata has been tan- for AtSOC1 . Such a result is comparable with our demly duplicated and one of the paralogs are having a findings in the three Lepidium species, in which we only more pronounced effect on vernalization than the other found one missense mutation that differentiates L. hir- [65, 66]. Two separate partial sequences of the field cress tum ssp. atlanticum from field cress, L. heterophyllum FLC gene were sequenced in this project, targeting the and the other two subspecies of L. hirtum in SOC1 and 5′ and 3′ regions of the coding sequence. A nucleotide none in FUL. Hence, this mutation cannot be attributed BLAST search shows that the 5′ region is more similar to perenniality trait. Lepidium FUL and SOC1 also share to the FLC2 gene in A. lyrata, whilst the 3′ region is a very high sequence homology with AtFUL and more homologous to the FLC1 gene. Hence it cannot be AtSOC1, thus indicating that these genes are highly con- ruled out that there are more than one copy of the FLC served. The difference in regulation of these genes, ra- gene in field cress, and that we have targeted both ther than polymorphism within the coding gene paralogs. sequence, is therefore likely to be the cause of the ob- Downregulation of FLC promotes floral development served phenotypic variation in the double mutant of while low levels of FLC are maintained after cold treat- FUL and SOC1, and the difference in longevity for field ment by VRN1 and VRN2 in Arabidopsis, which are cress versus its perennial relatives. serving as a molecular memory of the vernalization [21, For wild plants, it is generally advantageous to disperse 67]. Interestingly, compared to other genes included in all seeds to increase the survival chances of future gener- this study, an extremely high level of polymorphisms has ations. In contrast, domesticated plants are more resist- been reported for AtVRN2, over 140 mutations in the ant against shattering which ensures farmers’ maximum UTR and introns, and 55 in the coding region, resulting seed harvest. Resistance to pod shattering and seed dis- in 4% of the nucleotides in the coding region being poly- persal is an important agronomic trait as shattering can morphic . Even though 536 nt long coding sequences result in serious yield losses. In Arabidopsis, the of Lepidium VNR2 were analyzed, only one SNPs was seed-containing fruit is an ovary composed of three observed among the three Lepidium species. Thus, less main tissue types: the valve, the replum and the valve than 0.2% of the nucleotides in this region are variable. margin. The valve margin separates the valve from the When delimiting the observed polymorphisms of replum and also referred to as the dehiscence zone. AtVRN2 to the corresponding region to this 536 nt long Upon fruit dehiscence, the valve margin detaches from segment, and recalculating based on this, it seems as this the replum and the seeds are released. The two Gustafsson et al. BMC Genetics (2018) 19:36 Page 12 of 15 transcription factor genes (ALC and IND) are required the protein as an additional in-frame start codon is po- for valve margin formation and are promoted by MADS sitioned close by. Moreover, this shortening was not box genes SHP1 and SHP2 [31–33]. ALC, IND, SHP1 predicted to have a severe effect on the final protein ac- and SHP2 are collectively known as the valve identity cording to PROVEAN. In FAE1 there are two SNPs in genes. FUL and RPL genes are expressed in the valve the coding region differentiating field cress from the L. and replum respectively and negatively regulates the heterophyllum and L. hirtum. As field cress has a rela- valve identity genes to ensure that these genes are tively higher oil content than the other two Lepidium expressed in the proper tissue [30, 34]. Many aspects of species these polymorphism could be important fruit dehiscence have been studied in field cress. There markers for this trait. is a high degree of conservation of the fruit development The immediate challenges of breeding field cress into pathways between field cress and Arabidopsis and an economically viable new perennial oil crop includes expression levels of the field cress valve identity genes countering unstable longevity and weak shattering resist- resembles those of A. thaliana [12, 70] Many of the ance, and increasing the oil content [9, 73]. In this study, genes involved in seed shattering displayed a relatively we have characterized partial sequences of the genes high level of variation per nucleotides in this study. This regulating these traits. The continuation of the work in- mirrors the phenotypic variation in this particular trait cludes identifying the full gene sequences and among the field cress genotypes in the present study characterize them. In future research, it would be neces- (unpublished data). sary to include cis-regulatory elements as there are ex- The SHP1, ADPG2, IND pod shattering genes analyzed amples of important polymorphisms in these regions in Lepidium also had a higher level of variation in the responsible for the traits that are associated with the coding region per nucleotides (1.1–1.5%) than the aver- “domestication syndrome” . Linkage mapping, and age variation for all genes obtained in this study. The se- detailed quantitative trait loci maps, which are in pro- quences of ADPG2 and IND also are less similar to the gress at the moment, will also be needed to advance the A. thaliana counterparts (84 and 85%, respectively), thus breeding of field cress. Furthermore, the agronomic indicating that these genes are not as conserved as most practice for perennial crops is not fully developed, which other genes in this study. A nonsynonymous mutation is also the case for field cress. in RPL divided the field cress individuals into two With these findings, we have begun to unravel the groups, where the individuals having one of the alleles genomics of field cress and the results will be the display medium to high level of pod shattering while foundation for the future breeding strategy of this those having the other allele have a lower tendency to- potential oilseed crop. Moreover, the outcome of this wards shattering. RPL has proven to be an important study has contributed to the overall understanding of regulator of this trait in B. rapa, as well as in Oryza Lepidium genome evolution. sativa as a single point mutation in the regulatory se- quence of this gene seems to be responsible for seed Conclusion shattering resistance [71, 72]. The shattering type allele This study is the first performed on Lepidium genes with is present in A. thaliana as well as in field cress which is the purpose of improving field cress as a future oilseed naturally highly shattering . crop. It has revealed significant variation among the Lepi- Seeds with a high oil content and a desirable oil com- dium species within the partial sequences of 30 genes position is essential for developing an economically vi- known to regulate traits such as flowering, perenniality, able oilseed crop. Partial sequences of Lepidium that pod shattering and more. The phylogenetic relationship are orthologous to Arabidopsis FAD2, FAE1, WRI1, demonstrated between the three Lepidium species in this TAG1 and KCS8 which are all involved in fatty acid study can guide the development of interspecific biosynthesis were analyzed in this study. Much of the hybridization to advance the domestication process of natural variation in seed oil composition in Arabidopsis field cress. has been attributed to FAD2 and FAE1 . Indeed, downregulation of FAD2 and FAE1 in field cress also Additional files results in altered oil composition, drastically increasing the proportion of desirable oleic acid in the oil . All Additional file 1: Table S1. Genomic DNA sequences. (XLSX 23 kb) the five genes shared high homology to the Arabidopsis Additional file 2: Table S2. List of genes, accession numbers and primer sequences. (DOCX 36 kb) orthologs and nonsynonymous polymorphisms were observed in FAD2, KCS8 and WRI1. One of the nonsy- Abbreviations nonymous SNPs was located in the start codon of the ADPG1/2: ARABIDOPSIS DEHISCENCE ZONE POLYGALACTURONASE1/2 gene; predicted amino acid sequence of KCS8. This allelic dif- AGL11: AGAMOUS-LIKE 11 gene; AGL16: AGAMOUS LIKE 16 gene; ference only results in a five amino-acid shortening of AGL6: AGAMOUS LIKE 6 gene; ALC: ALCATRAZ gene; AP2: APETALA2 gene; Gustafsson et al. BMC Genetics (2018) 19:36 Page 13 of 15 At: Arabidopsis thaliana; ATG5: AUTOPHAGY RELATED 5 gene; BLAST: Basic local perennial grain. In: Batello C, Wade L, Cox S, Pogna N, Bozzini A, Choptiany J, alignment search tool; CHe hybrids: Lepidium campestre- Lepidium editors. Perennial crops for food security. Proceedings of the FAO expert heterophyllum hybrids; FAD2: FATTY ACID DESATURASE 2 gene; FAE1: FATTY workshop. Rome, Italy; 28–30 august; 2013. p. 72–89. ACID ELONGATION 1 gene; FER: FERONIA gene; FLC: FLOWERING LOCUS C 4. Sacks E. Perennial rice: challenges and opportunities. In: Batello C, Wade L, gene; FRI: FRIGIDA gene; FUL: FRUITFULL gene; HAI2: HIGHLY ABA-INDUCED Cox S, Pogna N, Bozzini A, Choptiany J, editors. Perennial crops for food PP2C PROTEIN 2 gene; IND: INDEHISCENT gene; KCS8: 3-KETOACYL-COA- security. Proceedings of the FAO expert workshop. Rome, Italy; 28–30 SYNTHASE 8 gene; Lc: Lepidium campestre; MADS box: MINICHROMOSOME august; 2013. p. 16–26. MAINTENANCE1, AGAMOUS, DEFICIENS and SERUM RESPONSE FACTOR box; 5. Kantar MB, Betts K, Michno JM, Luby JJ, Morrell PL, Hulke BS, Stupar RM, MAF2–5: MADS AFFECTING FLOWERING 2–5 genes; NAC012: NAC DOMAIN Wyse DL. Evaluating an interspecific Helianthus annuus x Helianthus CONTAINING PROTEIN 12 gene; Nt: nucleotides; RAD: Restriction site tuberosus population for use in a perennial sunflower breeding program. associated DNA; RPL: REPLUMLESS gene; SHP1/2: SHATTERPROOF1/2 genes; Field Crop Res. 2014;155:254–64. SNP: Single nucleotide polymorphism; SOC1: SUPRESSOR OF OVEREXPRESSION 6. Rice A, Glick L, Abadi S, Einhorn M, Kopelman NM, Salman-Minkov A, Mayzel J, OF CO1 gene; TAG1: ACYL-COA: DIACYLGLYCEROL ACYLTRANSFERASE 1 gene; Chay O, Mayrose I. The chromosome counts database CCDB; − a community VIN3: VERNALIZATION INSENSITIVE 3 gene; VRN1/2: REDUCED VERNALIZATION resource of plant chromosome numbers. New Phytol. 2015;206:19–26. RESPONSE1/2 gene; WRI1: WRINKLED 1 gene 7. Geleta M, Zhu L-H, Stymne S, Lehrman A, Hansson S-O. Domestication of Lepidium campestre as part of Mistra biotech, a research programme Acknowledgements focused on agro biotechnology for sustainable food. In: Batello C, Wade L, We would like to thank PlantLink for bioinformatics support. The authors Cox S, Pogna N, Bozzini A, Choptiany J, editors. Perennial crops for food would also like to acknowledge Professor Dirk-Jan De Koning for his major security. Proceedings of the FAO expert workshop. Rome, Italy, 28–30 role in the RAD-sequencing project within Component Project-2 (CP2) of august; 2013. p. 141–7. Mistra Biotech research program. 8. Lee J, Mummenhoff K, Bowman J. Allopolyploidization and evolution of species with reduced floral structures in Lepidium L. (Brassicaceae). Proc Natl Funding Acad Sci U S A. 2002;99(26):16835–40. This work is financed by grants from the Swedish Foundation for Strategic 9. Andersson A, Merker A, Nilsson P, Sörensen H, Åman P. Chemical Research (SSF), the Swedish Foundation for Strategic Environmental Research composition of the potential new oilseed crops Barbarea vulgaris, Barbarea (MISTRA) and Swedish University of Agricultural Sciences (SLU). verna and Lepidium campestre. J Sci Food Agric. 1999;79:179–86. 10. Merker A, Eriksson D, Bertholdsson N-O. Barley yield increases with Availability of data and materials undersown Lepidium campestre. Acta Agric Scand Sect B. 2010;60:269–73. Lepidium campestre genomic sequences homologous to Arabidopsis genes that 11. Ivarson E, Ahlman A, Li X, Zhu LH. Development of an efficient regeneration were derived from RAD-sequencing and used in this study is provided in and transformation method for the new potential oilseed crop Lepidium Additional file 1: Table S1. Representative partial sequences of the 30 genes campestre. BMC Plant Biol. 2013;13:115. were submitted to GenBank and their accession number is provided in 12. Lenser T, Theissen G. Conservation of fruit dehiscence pathways between Additional file 2: Table S2. Aligned partial sequences of the genes for all samples Lepidium campestre and Arabidopsis thaliana sheds light on the regulation used in this study can be obtained by contacting the corresponding author. of INDEHISCENT. Plant J. 2013;76:545–56. 13. Arabidopsis Genome I. Analysis of the genome sequence of the flowering Authors’ contributions plant Arabidopsis thaliana. Nature. 2000;408:796–815. MG and RO secured the funding; CG, MG and RO conceived and designed 14. Genomes Consortium, Nordborg, M, et al. 1,135 genomes reveal the the study; CG and FLP obtained the data; CG, MG, FLP and JW analyzed the global pattern of polymorphism in Arabidopsis thaliana. Cell. 2016; data; CG and MG wrote the paper; and FLP, JW and RO revised the paper. All 166:481–91. authors read and approved the final manuscript. 15. Michaels SD, Amasino RM. FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell. 1999;11:949–56. Ethics approval and consent to participate 16. Sheldon CC, Rouse DT, Finnegan EJ, Peacock WJ, Dennis ES. The molecular The plant materials used in this study were derived from accessions basis of vernalization: the central role of FLOWERING LOCUS C FLC. Proc Natl obtained from Gene Banks and Botanical Gardens as well as from newly Acad Sci U S A. 2000;97:3753–8. collected populations. These materials are permitted to be used for this 17. Johanson U, West J, Lister C, Michaels S, Amasino R, Dean C. Molecular study. analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science. 2000;290:344–7. Competing interests 18. Koornneef M, Blankestijndevries H, Hanhart C, Soppe W, Peeters T. The The authors declare that they have no competing interests. phenotype of some late-flowering mutants is enhanced by a locus on chromosome-5 that is not effective in the Landsberg erecta wild-type. Plant J. 1994;6:911–9. Publisher’sNote 19. De Lucia F, Crevillen P, Jones AM, Greb T, Dean C. A PHD-polycomb Springer Nature remains neutral with regard to jurisdictional claims in repressive complex 2 triggers the epigenetic silencing of FLC during published maps and institutional affiliations. vernalization. Proc Natl Acad Sci U S A. 2008;105:16831–6. 20. Sung S, Amasino RM. Vernalization in Arabidopsis thaliana is mediated by Author details the PHD finger protein VIN3. Nature. 2004;427:159–64. Department of Plant Breeding, Swedish University of Agricultural Sciences, 21. Gendall AR, Levy YY, Wilson A, Dean C. The VERNALIZATION 2 gene Box 101, SE-23053 Alnarp, Sweden. Department of Plant Protection Biology, mediates the epigenetic regulation of vernalization in Arabidopsis. Cell. Swedish University of Agricultural Sciences, Box 102, SE-23053 Alnarp, 2001;107:525–35. Sweden. Department of Animal Breeding and Genetics, Swedish University 22. Hempel FD, Weigel D, Mandel MA, Ditta G, Zambryski PC, Feldman LJ, of Agricultural Sciences, Box 7023, SE-750 07 Uppsala, Sweden. Yanofsky MF. Floral determination and expression of floral regulatory genes in Arabidopsis. Development. 1997;124:3845–53. Received: 26 September 2017 Accepted: 18 May 2018 23. Lee H, Suh SS, Park E, Cho E, Ahn JH, Kim SG, Lee JS, Kwon YM, Lee I. The AGAMOUS-LIKE 20 MADS domain protein integrates floral inductive pathways in Arabidopsis. Genes Dev. 2000;14:2366–76. References 1. Doebley JF, Gaut BS, Smith BD. The molecular genetics of crop 24. Melzer S, Lens F, Gennen J, Vanneste S, Rohde A, Beeckman T. Flowering- domestication. Cell. 2006;127:1309–21. time genes modulate meristem determinacy and growth form in 2. Randall GW, Mulla DJ. Nitrate nitrogen in surface waters as influenced by Arabidopsis thaliana. Nat Genet. 2008;40:1489–92. climatic conditions and agricultural practices. J Environ Qual. 2001;30:337–44. 25. Hu JY, Zhou Y, He F, Dong X, Liu LY, Coupland G, Turck F, de Meaux J. 3. DeHaan LR, Wang S, Larson SR, Cattani DJ, Zhang X, Viinanen T. Current efforts miR824-regulated AGAMOUS-LIKE16 contributes to flowering time repression to develop perennial wheat and domesticate Thinopyrum intermedium as a in Arabidopsis. Plant Cell. 2014;26:2024–37. Gustafsson et al. BMC Genetics (2018) 19:36 Page 14 of 15 26. Koo SC, Bracko O, Park MS, Schwab R, Chun HJ, Park KM, Seo JS, 48. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Grbic V, Balasubramanian S, Schmid M, Godard F, Yun DJ, Lee SY, Cho MJ, Garimella K, Altshuler D, Gabriel S, Daly M, DePristo MA. The genome Weigel D, Kim MC. Control of lateral organ development and flowering analysis toolkit: a MapReduce framework for analyzing next-generation DNA time by the Arabidopsis thaliana MADS-box gene AGAMOUS-LIKE6.Plant J. sequencing data. Genome Res. 2010;20:1297–303. 2010;62:807–16. 49. Hall T. BioEdit 7.0.5. North Carolina State University. 2005. 27. Ratcliffe OJ, Kumimoto RW, Wong BJ, Riechmann JL. Analysis of the 50. Larkin MABG, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Arabidopsis MADS AFFECTING FLOWERING gene family: MAF2 prevents Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG. Clustal vernalization by short periods of cold. Plant Cell. 2003;15:1159–69. W and Clustal X version 2.0. Bioinformatics. 2007;23:2947–8. 28. Yoo SK, Wu X, Lee JS, Ahn JH. AGAMOUS-LIKE 6 is a floral promoter that 51. Choi Y, Sims GE, Murphy S, Miller JR, Chan AP. Predicting the functional negatively regulates the FLC/MAF clade genes and positively regulates FT in effect of amino acid substitutions and indels. PLoS One. 2012;7:e46688. Arabidopsis. Plant J. 2011;65:62–76. 52. Kumar S, Stecher G, Koichiro Tamura K. MEGA7: molecular evolutionary 29. Ferrandiz C, Liljegren SJ, Yanofsky MF. Negative regulation of the genetics analysis version 7.0 for bigger datasets. Mol Biol Evol. 2016;33:1870–4. SHATTERPROOF genes by FRUITFULL during Arabidopsis fruit development. 53. Saitou N, Nei M. The neighbor-joining method: a new method for Science. 2000;289:436–8. reconstructing phylogenetic trees. Mol Biol Evol. 1987;4:406–25. 30. Gu Q, Ferrandiz C, Yanofsky MF, Martienssen R. The FRUITFULL MADS-box 54. Nei M, Kumar S. Molecular evolution and phylogenetics. New York: Oxford gene mediates cell differentiation during Arabidopsis fruit development. University Press; 2000. Development. 1998;125:1509–17. 55. Felsenstein J. Confidence limits on phylogenies: an approach using the 31. Liljegren SJ, Ditta GS, Eshed Y, Savidge B, Bowman JL, Yanofsky MF. bootstrap. Evolution. 1985;39:783–91. SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. 56. Huang CH, Sun R, Hu Y, Zeng L, Zhang N, Cai L, Zhang Q, Koch MA, Al- Nature. 2000;404:766–70. Shehbaz I, Edger PP, Pires JC, Tan DY, Zhong Y, Ma H. Resolution of 32. Liljegren SJ, Roeder AH, Kempin SA, Gremski K, Ostergaard L, Guimil S, Reyes Brassicaceae phylogeny using nuclear genes uncovers nested radiations and DK, Yanofsky MF. Control of fruit patterning in Arabidopsis by INDEHISCENT. supports convergent morphological evolution. Mol Biol Evol. 2016;33:394–412. Cell. 2004;116:843–53. 57. Kuittinen H, Niittyvuopio A, Rinne P, Savolainen O. Natural variation in 33. Rajani S, Sundaresan V. The Arabidopsis myc/bHLH gene ALCATRAZ enables Arabidopsis lyrata vernalization requirement conferred by a FRIGIDA indel cell separation in fruit dehiscence. Curr Biol. 2001;11:1914–22. polymorphism. Mol Biol Evol. 2008;25:319–29. 34. Roeder AH, Ferrandiz C, Yanofsky MF. The role of the REPLUMLESS homeodomain 58. Fadina OA, Khavkin EE. The vernalization gene FRIGIDA in cultivated Brassica proteininpatterningthe Arabidopsis fruit. Curr Biol. 2003;13:1630–5. species. Russ J Plant Physiol. 2014;61:309–17. 35. Ogawa M, Kay P, Wilson S, Swain SM. ARABIDOPSIS DEHISCENCE ZONE 59. Mummenhoff K, Bruggemann H, Bowman JL. Chloroplast DNA phylogeny and POLYGALACTURONASE1 ADPG1, ADPG2, and QUARTET2 are biogeography of Lepidium Brassicaceae. American J Bot. 2001;88:2051–63. polygalacturonases required for cell separation during reproductive 60. Sharbel TF, Haubold B, Mitchell-Olds T. Genetic isolation by distance in development in Arabidopsis. Plant Cell. 2009;21:216–33. Arabidopsis thaliana: biogeography and postglacial colonization of Europe. 36. Zhong R, Richardson EA, Ye ZH. Two NAC domain transcription factors, Mol Ecol. 2000;9:2109–18. SND1 and NST1, function redundantly in regulation of secondary wall 61. Nordborg M, Hu TT, Ishino Y, Jhaveri J, Toomajian C, Zheng H, Bakker E, synthesis in fibers of Arabidopsis. Planta. 2007;225:1603–11. Calabrese P, Gladstone J, Goyal R, Jakobsson M, Kim S, Morozov Y, 37. Focks N, Benning C. wrinkled1: a novel, low-seed-oil mutant of Arabidopsis Padhukasahasram B, Plagnol V, Rosenberg NA, Shah C, Wall JD, Wang J, with a deficiency in the seed-specific regulation of carbohydrate Zhao K, Kalbfleisch T, Schulz V, Kreitman M, Bergelson J. The pattern of metabolism. Plant Physiol. 1998;118:91–101. polymorphism in Arabidopsis thaliana. PLoS Biol. 2005;3:e196. 38. Jako C, Kumar A, Wei Y, Zou J, Barton DL, Giblin EM, Covello PS, Taylor DC. 62. Corre VL. Variation at two flowering time genes within and among Seed-specific over-expression of an Arabidopsis cDNA encoding a populations of Arabidopsis thaliana: comparison with markers and traits. Mol diacylglycerol acyltransferase enhances seed oil content and seed weight. Ecol. 2005;14:4181–92. Plant Physiol. 2001;126:861–74. 63. Mendez-Vigo B, Pico FX, Ramiro M, Martinez-Zapater JM, Alonso-Blanco C. 39. Routaboul JM, Benning C, Bechtold N, Caboche M, Lepiniec L. The TAG1 Altitudinal and climatic adaptation is mediated by flowering traits and FRI, locus of Arabidopsis encodes for a diacylglycerol acyltransferase. Plant FLC, and PHYC genes in Arabidopsis. Plant Physiol. 2011;157:1942–55. Physiol Biochem. 1999;37:831–40. 64. Caicedo AL, Stinchcombe JR, Olsen KM, Schmitt J, Purugganan MD. Epistatic 40. Costaglioli P,JoubesJ, GarciaC, StefM,ArveilerB,Lessire R, interaction between Arabidopsis FRI and FLC flowering time genes Garbay B. Profiling candidate genes involved in wax biosynthesis in generates a latitudinal cline in a life history trait. Proc Natl Acad Sci U S A. Arabidopsis thaliana by microarray analysis. Biochim Biophys Acta. 2004;101:15670–5. 2005;1734:247–58. 65. Kemi U, Niittyvuopio A, Toivainen T, Pasanen A, Quilot-Turion B, Holm K, 41. James DW Jr, Lim E, Keller J, Plooy I, Ralston E, Dooner HK. Directed tagging Lagercrantz U, Savolainen O, Kuittinen H. Role of vernalization and of of the Arabidopsis FATTY ACID ELONGATION1 FAE1; gene with the maize duplicated FLOWERING LOCUS C in the perennial Arabidopsis lyrata. New transposon activator. Plant Cell. 1995;7:309–19. Phytol. 2013;197:323–35. 66. Nah G, Chen ZJ. Tandem duplication of the FLC locus and the origin of a 42. Okuley J, Lightner J, Feldmann K, Yadav N, Lark E, Browse J. Arabidopsis new gene in Arabidopsis related species and their functional implications in FAD2 gene encodes the enzyme that is essential for polyunsaturated lipid allopolyploids. New Phytol. 2010;186:228–38. synthesis. Plant Cell. 1994;6:147–58. 43. Kessler SA, Shimosato-Asano H, Keinath NF, Wuest SE, Ingram G, Panstruga 67. Levy YY, Mesnage S, Mylne JS, Gendall AR, Dean C. Multiple roles of R, Grossniklaus U. Conserved molecular components for pollen tube Arabidopsis VRN1 in vernalization and flowering time control. Science. 2002; reception and fungal invasion. Science. 2010;330:968–71. 297:243–6. 44. Kim W, Lee Y, Park J, Lee N, Choi G. HONSU, a protein phosphatase 2C, 68. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, Sirotkin K. regulates seed dormancy by inhibiting ABA signaling in Arabidopsis. Plant dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 2001;29:308–11. Cell Physiol. 2013;54:555–72. 69. Poduska B, Humphrey T, Redweik A, Grbic V. The synergistic activation of 45. Lenz HD, Haller E, Melzer E, Gust AA, Nurnberger T. Autophagy controls FLOWERING LOCUS C by FRIGIDA and a new flowering gene AERIAL ROSETTE plant basal immunity in a pathogenic lifestyle-dependent manner. 1 underlies a novel morphology in Arabidopsis. Genetics. 2003;163:1457–65. Autophagy. 2011;7:773–4. 70. Muhlhausen A, Lenser T, Mummenhoff K, Theißen G. Evidence that an evolutionary transition from dehiscent to indehiscent fruits in Lepidium, 46. Mizzotti C, Ezquer I, Paolo D, Rueda-Romero P, Guerra RF, Battaglia R, Brassicaceae; was caused by a change in the control of valve margin Rogachev I, Aharoni A, Kater MM, Caporali E, Colombo L. SEEDSTICK is a identity genes. Plant J. 2013;73:824–35. master regulator of development and metabolism in the Arabidopsis seed coat. PLoS Genet. 2014;10:e1004856. 71. Arnaud N, Lawrenson T, Ostergaard L, Sablowski R. The same regulatory 47. Lopes-Pinto F, Vanhala T, Geleta M, Risse J, Nichols J, Karim Gharbi K, de point mutation changed seed-dispersal structures in evolution and Koning DJ. RAD sequencing of diverse accessions of Lepidium campestre,a domestication. Curr Biol. 2011;21:1215–9. target species for domestication as a novel oil crop. San Diego, CA, USA: 72. Konishi S, Izawa T, Lin SY, Ebana K, Fukuta Y, Sasaki T, Yano M. An SNP caused Plant and animal genome conference XXIV; 2016. p. 09–13. loss of seed shattering during rice domestication. Science. 2006;312:1392–6. Gustafsson et al. BMC Genetics (2018) 19:36 Page 15 of 15 73. Mummenhoff K, Polster A, Muhlhausen A, Theissen G. Lepidium as a model system for studying the evolution of fruit development in Brassicaceae. J Exp Bot. 2009;60:1503–13. 74. O’Neill CM, Gill S, Hobbs D, Morgan C, Bancroft I. Natural variation for seed oil composition in Arabidopsis thaliana. Phytochemistry. 2003;64:1077–90. 75. Ivarson E, Ahlman A, Lager I, Zhu LH. Significant increase of oleic acid level in the wild species Lepidium campestre through direct gene silencing. Plant Cell Rep. 2016;35:2055–63.
BMC Genetics – Springer Journals
Published: May 29, 2018