Construction of high-resolution genetic maps of Zoysia matrella (L.) Merrill and applications to comparative genomic analysis and QTL mapping of resistance to fall armyworm
- Xiaoen Huang†1,
- Fangfang Wang†1,
- Ratnesh Singh†1,
- James A. Reinert1,
- M. C. Engelke1,
- Anthony D. Genovesi1,
- Ambika Chandra1, 2 and
- Qingyi Yu1, 3Email author
© The Author(s). 2016
Received: 8 February 2016
Accepted: 26 July 2016
Published: 8 August 2016
Zoysia matrella, widely used in lawns and sports fields, is of great economic and ecological value. Z. matrella is an allotetraploid species (2n = 4x = 40) in the genus zoysia under the subfamily Chloridoideae. Despite its ecological impacts and economic importance, the subfamily Chloridoideae has received little attention in genomics studies. As a result, limited genetic and genomic information are available for this subfamily, which have impeded progress in understanding evolutionary history of grasses in this important lineage. The lack of a high-resolution genetic map has hampered efforts to improve zoysiagrass using molecular genetic tools.
We used restriction site-associated DNA sequencing (RADSeq) approach and a segregating population developed from the cross between Z. matrella cultivars ‘Diamond’ and ‘Cavalier’ to construct high-resolution genetic maps of Z. matrella. The genetic map of Diamond consists of 2,375 Single Nucleotide Polymorphism (SNP) markers mapped on 20 linkage groups (LGs) with a total length of 1754.48 cM and an average distance between adjacent markers at 0.74 cM. The genetic map of Cavalier contains 3,563 SNP markers on 20 LGs, covering 1824.92 cM, with an average distance between adjacent markers at 0.51 cM. A higher level of genome collinearity between Z. matrella and rice than that between Z. matrella and sorghum was revealed by comparative genomic analysis. Pairwise comparison revealed that two independent nested chromosome fusion events occurred after Z. matrella and sorghum split from a common ancestor. The high-resolution linkage maps were applied into mapping QTLs associated with fall armyworm (FAW) resistance and six loci located on LGs 8 and 20 were detected to be significantly associated with FAW resistance.
The high-resolution linkage maps provide anchor points for comparative genomics analysis between Z. matrella and other grass species. Our comparative genomic analysis suggested that the chromosome number reduction from 12 to 10 had occurred independently via a single-step in the subfamilies Chloridoideae and Panicoideae. The high-resolution genetic maps provide an essential framework for mapping QTLs associated with economically and agronomically important traits. The major QTLs mapped on LG8 of the Cavalier map provide a starting point for cloning FAW resistance genes and further studies for a better understanding of FAW resistance in zoysiagrass.
KeywordsZoysia Chloridoideae Genetic map Resistance to fall armyworm Restriction site-associated DNA sequencing (RADSeq)
Zoysia matrella (L.) Merrill (2n = 4x = 40), commonly known as Manila Grass, is naturally distributed in South-East Asian countries, along the coasts of Indian Ocean, and in southern Japan (Ryukyu Islands) and northern Australia . Z. matrella was first introduced into the United States from Manila in the early 20th century and since then it has been increasingly used as a turfgrass on athletic fields, golf courses, home lawns, and parks . Its fine leaf texture, high density, rhizomatous growth habit, and excellent tolerance to shade, heat, and salinity stresses has made it an ideal choice for high quality playing surfaces. Z. matrella has become an economically important warm-season turfgrass, widely growing in the southern United States, Japan, China, Southeast Asia, New Guinea, Australia and New Zealand.
The primary focus of the zoysiagrass (Zoysia spp.) breeding program is to improve resistance or tolerance to biotic and abiotic stresses in order to reduce replacement and management costs incurred by the end user. The fall armyworm (FAW), Spodoptera frugiperda (J.E. Smith), is a destructive pest in over 60 plant species, including the major cool-season and warm-season turfgrasses, corn, wheat, rice, sugarcane, and sorghum . Outbreaks of FAW are sporadic and unpredictable, making it very difficult to control. Utilization of host plant resistance is the most effective, efficient, and economical approach in pest control. In addition, the use of pest-resistant or tolerant varieties can help to reduce pesticide application and meet the increasing strict environmental quality and safety standards. For this reason, resistance to FAW was evaluated in major zoysiagrass cultivars [4–8]. Among the twelve cultivars or genotypes evaluated, ‘Cavalier’ (Z. matrella) exhibited the highest resistance to FAW, regardless of the development stage of the larvae . No FAW larvae were observed to be able to survive more than 17 days on Cavalier . While, cultivar ‘Diamond’ (Z. matrella) served as an excellent host and produced the largest larval weight in the test . An F1 segregating population was created from a cross between the FAW resistant cultivar Cavalier and the sensitive cultivar Diamond. In the present study, we developed high-resolution linkage maps of Z. matrella and identified major quantitative trait loci (QTLs) associated with FAW resistance utilizing this segregating population.
Effort has been made to map FAW resistance in zoysiagrass by Jessup et al. . However, a limited number of molecular markers were used to develop the linkage map of Z. matrella, which could lead to inaccurate genetic mapping of QTLs. The locus of FAW resistance was mapped on the linkage group (LG) 36 that only contains three markers , which significantly limited its application in marker-assisted breeding and cloning the genes controlling the resistance to FAW through a positional cloning approach.
The advent of next-generation sequencing (NGS) technologies provides us new opportunities to identify Single Nucleotide Polymorphism (SNP) with high resolution in a cost-effective manner at a much faster speed than ever before. Restriction site-associated DNA sequencing (RADSeq), a method that utilizes NGS and restriction enzyme digestion to reduce the complexity across the targeted genomes enabling discovery of genome-wide genetic markers [10, 11], has become a powerful tool for a wide range of genetic studies in both model and non-model organisms. RADSeq has been applied in genetic map construction as well as detection of QTLs that impact economically and agronomically important traits for a number of plant species, including eggplant , barley , ryegrass , grape , lotus , pear , and Z. japonica . In the present work, we developed high-resolution linkage maps of Z. matrella using RADSeq markers to serve as foundation for gene mapping, QTL studies, assigning genome sequence to chromosomes, and comparative genomics studies.
The true grasses, family Poaceae (also known as Gramineae), provide for the major source of carbohydrates and proteins need by humans as well as various herbivores and fowl. Panicoideae and Chloridoideae are the two major subfamilies in the family Poaceae, which include mostly C4 grasses that are adapted to warm climates. Z. matrella is a member of the subfamily Chloridoideae which has attracted less attention in genomics research due to its relatively smaller economic impact compared to its sister subfamily Panicoideae. The sequence-tagged high-resolution linkage map of Z. matrella provides a framework for comparative and evolutionary genomic studies thus allowing us to further investigate the evolutionary history of this underexplored grass family.
Plant materials and DNA isolation
The F1 segregation population of 95 individuals created from a cross between Z. matrella varieties Cavalier (resistant to FAW) and Diamond (susceptible to FAW) was used for construction of the high-resolution linkage maps of Z. matrella. The mapping population was planted in pots and maintained in a greenhouse at Texas A&M AgriLife Research Center at Dallas, Texas, USA.
Young and healthy leaf tissue was harvested from each individual of the mapping population and frozen in liquid nitrogen immediately. The leaf tissue was then stored in a −80 °C freezer until DNA isolation. Approximately 100 mg of leaf tissue of each sample was ground into fine powder using liquid nitrogen, and then used for genomic DNA isolation with DNeasy Plant Mini Kit (Qiagen, Valencia, CA). The extracted DNA samples were run on an agarose gel and measured with a NanoDrop 2000c (Thermo Fisher Scientific, Waltham, MA) to ensure the DNA quality for RADSeq library construction. The high-quality DNA was then quantified by a Qubit fluorometer with the Qubit double-stranded DNA HS Assay kit (Life Technologies, Carlsbad, CA).
RADSeq library construction and sequencing
The RADSeq libraries were constructed using a modified protocol described by Wang et al. . Approximately 1 μg of genomic DNA was digested with restriction enzymes NsiI (New England Biolabs, Ipswich, MA) and MseI (New England Biolabs, Ipswich, MA). The digested genomic DNA was ligated with P1 and P2 adapters containing unique identifying sequences (molecular identifier, MID, see Additional file 1: Figure S1). The ligation reaction was then deactivated at 65 °C for 20 min and purified with Axyprep Mag PCR clean-up beads (Axygen, Union City, CA) following the manufacturer’s instruction. Purified adaptor-ligated DNA was PCR amplified with12 cycles using Phusion High-Fidelity DNA polymerase (New England Biolabs, Ipswich, MA). Then, size selection of 300–500 bp DNA fragments were performed using Axyprep Mag PCR clean-up beads (Axygen, Union City, CA) following the manufacturer’s instruction. Quantification of recovered DNA was performed on a Qubit fluorometer (Life Technologies, Carlsbad, CA). The RADSeq libraries of the whole mapping population were pooled and sequenced on an Illumina HiSeq 2000 (Illumina, San Diego, CA).
RADSeq sequence analysis and SNP calling
Raw data processing and SNP identification were performed using modules implemented in the open-source STACKS pipeline v1.19 . Raw reads were de-multiplexed and trimmed to remove low-quality sequences and adapter contamination using the ‘PROCESS_RADTAGS’ module of the Stacks package. Sequences with base call accuracy lower than 99 % (Q20) in a sliding window of 15 % of the read-length were considered to be low quality and trimmed. Filtered reads were then subjected to build stacks using ‘USTACKS” module. A set of de novo loci were defined with no more than two nucleotide mismatches and at least six reads in each stack. A catalog of parental loci was created using ‘CSTACKS’ module. Progeny loci were compared to the parental loci in catalog and SNPs were called using maximum likelihood method implemented in ‘SSTACKS’ module. Finally, the haplotypes were determined and genotype data were exported as JoinMap compatible format using the ‘GENOTYPES’ module of STACKS pipeline.
Construction of genetic maps
Linkage maps were constructed using JoinMap 4.0  and outcross pollinated family (CP) was selected as the population type. Since both parents, Diamond and Cavalier, are tetraploids, we selected single-dose markers for linkage analysis to avoid statistical challenges caused by complicated segregation patterns for the reason as described by Wu et al. . Markers that are heterozygous in Diamond and homozygous Cavalier (‘lm x ll’ type) were selected to build Diamond linkage groups. Markers that are heterozygous in Cavalier and homozygous in Diamond (‘nn x np’ type) were selected to build Cavalier linkage groups. QC-filtered SNPs were further filtered by the following standards: 1) markers must be genotyped in at least 80 out of 95 individuals; 2) Individuals with over 10 % missing data were discarded; 3) Markers segregating at distorted Mendelian ratio (expected ratio for ‘lm x ll’ type and ‘nn x np’ type is 1: 1, χ2 test, P < 0.05) were discarded; 4) Redundant markers were removed by standard of similarity = 1.
The linkage groups were built using regression mapping algorithm, with a minimum logarithm of odds (LOD) value at 7, and a maximum recombination frequency at 0.35. Marker positioning calculation was performed with a goodness-of-fit jump at 5, followed by a “ripple” procedure (value =1). Kosambi mapping function was used to correct linkage distance. ‘N.N. fit’ function of JoinMap 4.0 was used to check the map quality. MAPCHART 2.2  was used to draw linkage maps.
Comparison with other grass genomes
We used BLASTN (BLAST 2.2.28+)  with default parameter settings and an e-value cutoff of 1 × 10−8 to blast the consensus sequences of mapped RADSeq markers against the genome sequences and gene models of rice and sorghum genomes. Markers that showed significant hits to the genome sequences and/or gene models of rice and sorghum genomes were extracted and used for comparative genomics study.
In vitro insect feeding bioassays
The whole mapping population including the parents were subjected to in vitro insect feeding bioassays using the method described by Reinert and Engelke . The bioassay was a no-choice experiment and conducted in the laboratory with 9-cm diam. x 20-mm deep plastic petri dishes as larvae feeding chambers. Each dish was provided with 3 g of fresh leaf tissue on two layers of water-saturated filter paper. Water was added daily as needed to maintain filter paper saturation and grass-clipping turgor. Grass clippings were added daily as needed during the whole experiment to ensure turgid fresh grass was always available to the developing larvae. The FAW used in the bioassay was a corn strain requested from the USDA-ARS-IBPMRL at Tifton, GA. Each individual of the mapping population was phenotyped using 4-day-old larvae that had first fed on fresh tissue of a susceptible zoysiagrass genotype, DALZ 8516. DALZ 8516 is an excellent host of FAW with near 100 % survival. Three 4-day-old larvae were randomly selected and placed into the feeding chambers with clippings from the respective individuals making up the mapping population, in a randomized complete block design (RCB) replicated 8 times. Each replicate started with 3 larvae for the first 3–4 days. And then the 3 larvae were separated into separate dishes. Larvae survivorship was measured at 7-, 10-, and 17-day-old larvae. The fresh weight was measured at 12-day-old larvae. The number of days to pupation and to adult emergence, and the sex types of larvae were also recorded. The bioassay for the whole population was split into 8 experiments. The two parents and a susceptible zoysiagrass genotype, DALZ 8516, were included in each experiment and used as controls to normalize the data between experiments. Percentage of mortality was adjusted by Abbott’s Formula .
Phenotypic data which includes mortality rate of the 7-, 10-, and 17-day-old FAW larvae for each individual and the genetic map of Cavalier were used for QTL analysis with MapQTL 5 . The non-parametric Kruskal–Wallis (KW) K-test was performed for detection of the presence and locations of QTLs. QTL significance in the Kruskall-Wallis analysis was based on P value < 0.001 significance level as suggested by the software.
RAD sequencing and SNP discovery
We genotyped all 95 individuals and parents that make up the whole mapping population using RADSeq. A total of 270,460,752 raw sequence reads were obtained (Additional file 2: Table S1, Additional file 1: Figure S1). After filtering low-quality sequence reads, 267,772,333 high-quality reads remained. All the RADSeq sequences can be accessed through the NCBI SRA database under BioProject Accession PRJNA312939 (http://www.ncbi.nlm.nih.gov/bioproject/312939). To maximize the chances of detecting segregating SNPs in the parents, we RADSeq’ed the parents more extensively than the offspring. We obtained 10,652,144 high-quality reads for the parent Diamond and 13,307,302 high-quality reads for Cavalier. The number of sequence reads of the F1 progeny ranged from 0.91 to 4.32 million reads. The average sequence depth for the F1 progeny was 2,566,451 reads per progeny (Additional file 2: Table S1, Additional file 1: Figure S1).
Summary of SNP marker filtering procedure
Number of SNPs Remaining
Raw RADSeq processing
Number of single-dose alleles
Type of single-dose alleles
Number of single-dose alleles before filtering
After removal of identical markers
After removal of markers with more than 10 % missing data
After removal of significantly segregation distorted markers (P < 0.05)
Linkage map construction
A pseudo-testcross strategy was used to develop the genetic maps of Z. matrella. Because only single-dose markers were selected for genetic map construction, the segregation ratio of the selected markers should be 1:1 in the F1 progeny. The selected markers were then divided into two data sets based on their segregation patterns. Each data set contains the meiotic segregation information of one parent. Then the two segregation data sets were subjected to construction of two independent linkage maps of Z. matrella, one for each parent.
Using JoinMap 4 , the selected markers were grouped into distinct LGs. Among the 2,462 single-dose SNP markers selected for linkage analysis in Diamond, 2,455 (99.7 % of the input markers) of them were grouped into 20 major LGs with a LOD score ≥7, 1 marker remained as singleton, and 6 markers were assigned to 2 small LGs with 4 and 2 markers for each LG, respectively. Among the 3,793 SNPs selected for linkage analysis in Cavalier, 3,784 (99.8 % of the input markers) of them were grouped into 20 major LGs with a LOD score ≥7, 2 markers remained unlinked, and 7 markers were assigned to 2 small LGs with 4 and 3 markers for each LG, respectively.
The markers in each LG were positioned and ordered using regression mapping algorithm (see parameters in materials and methods). Markers of “suspect linkages” were inspected for accuracy in scoring and removed from the map calculation if they proved problematic or troublesome. A total of 148 markers from Diamond LGs and 221 markers from Cavalier LGs were identified as “suspect linkages” during mapping and thus were discarded from subsequent linkage analysis.
Summary statistics for Diamond and Cavalier linkage maps
Sorghum Chr. no.
Diamond inkage Groups
Cavalier Linkage Groups
No. of markers
Gaps <5 cM (%)
Max. gap size (cM)
No. of distorted markers
No. of markers
Gaps <5 cM (%)
Max. gap size (cM)
No. of distorted markers
The genetic map of Diamond consisted of 2,375 SNP markers distributed on 20 LGs, with an average interval of 0.74 cM (Fig. 1, Table 2). In the Diamond genetic map, the length of individual LG ranged from 56.8 cM (LG14) to 131.7 cM (LG19), and the number of markers mapped on each LG ranged from 67 (LG17) to 226 (LG7). LG2 showed the highest density of markers at 0.39 cM between adjacent markers, almost twice of the density at genome level. While, LG17 had the lowest density of markers at 1.85 cM between adjacent markers, 150 % reduction of the marker density compared with genome-wide average. No major interruption was observed on most LGs except three relatively large gaps (>10 cM), a 14.2 cM gap on LG3, a 10.7 cM gap on LG13, and a 15.9 cM gap on LG19. The detailed information of Diamond genetic map are given in Additional file 3: Figure S2 and Additional file 4: Table S2.
The final genetic map of Cavalier consisted of 3,563 SNP markers mapped on 20 major LGs. Compared with the Diamond linkage map, the Cavalier map had a much higher density with an average interval of 0.51 cM (Fig. 1, Table 2). The number of markers mapped on each LG varied from 53 (LG3) to 351 (LG1). The average interval of individual LG ranged from 0.31 cM (LG12) to 1.09 cM (LG3), and the length of individual LG varied from 51.73 cM (LG14) to 130.62 cM (LG7). The longest LG was LG7, which contains 195 loci spanning 130.62 cM. The shortest was LG14, which contains 73 loci spanning 51.73 cM. Owing to the high density, only 1 gap >10 cM, a 12.2 cM gap on LG17, was found in Cavalier linkage map. The detailed information of Cavalier genetic map are given in Additional file 5: Figure S3 and Additional file 6: Table S3.
Conserved synteny relationships between Z. matrella and other grass genomes
Genetically mapped high-density RAD tags provide anchor points for comparative genomics analysis between Z. matrella and other grass species. One of the most important research topics in comparative and evolutionary genomics studies is karyotype evolution. Rice represents the most extensively studied and complete annotated genome among grasses. And it was suggested that the two major clades of grasses, Panicoideae-Aristidoideae-Centhothecoideae-Chloridoideae- Micrairoideae-Arundinoideae-Danthoideae (PACCMAD) and Bambusoideae-Ehrhartoidea-Pooideae (BEP), evolved from a common paleo-ancestor genome having a base chromosome number of n = 12 [26–30]. Comparative genomic analysis revealed the rice genome resembled the most of the ancestral form of the paleo-ancestor genome [26–30]. To address the question how Z. matrella genome obtained its reduced chromosome count from 12 to 10, we conducted comparative genomics analysis between Z. matrella and rice and identified the synteny relationships between these two genomes.
Z. matrella and Sorghum are closely related plant species and belong to sister subfamilies, Chloridoideae and Panicoideae, respectively, under PACCMAD clade. Like Z. matrella, sorghum has a base chromosome number of n = 10. Did the same pairs of ancestral chromosomes fuse in the karyotype evolution of the sorghum genome as fused in the Z. matrella lineage? To address this question, we conducted comparative genomics analysis between Z. matrella and sorghum genomes.
We identified anchor points for comparative analysis by searching the consensus sequences of the mapped RAD tags on Z. matrella genetic maps against sorghum genome sequences and gene models. Among the 3,563 RAD tags mapped on the Cavalier LGs, 801 (22.5 %) of them can be placed on sorghum chromosomes (Additional file 9: Table S6). Among the 2,375 RAD markers positioned on the Diamond LGs, 495 (20.8 %) of them can be found homologous positions on sorghum chromosomes (Additional file 10: Table S7). Dot-plot diagrams were drawn by plotting the genetic positions of anchored markers on Z. matrella maps against the physical positions of their homologs on sorghum chromosomes (Fig. 3c, d). The dot-plot diagrams revealed extensive collinearity along chromosome arms between genomes of Z. matrella and S. bicolor, but interruption of collinearity occurred at centromeric and pericentromeric regions of sorghum chromosomes for each pair of orthologous chromosomes. For each sorghum chromosome, we identified two orthologous LGs in Z. matrella, reinforcing the tetraploid genome nature of Z. matrella. We numbered the Z. matrella LGs according to their homology to sorghum chromosomes to simplify the synteny relationship for comparative genomics studies.
Syntenic relationships between Zoysia matrella LGs and rice and sorghum chromosomes. Numbers of markers that can be positioned on rice and sorghum chromosomes were present and the number of markers that hit rice or sorghum gene models were present in the bracket
Orthologous rice chromosome
Number of markers that can be positioned on rice chromosomes
Orthologous sorghum chromosome
Number of markers that can be positioned on sorghum chromosomes
LG 01, 02
LG 01, 02
LG 03, 04
LG 03, 04
LG 05, 06
LG 05, 06
LG 07, 08
LG 07, 08
Os02 + Os10
Sb04 + Sb01
LG 09, 10
LG 09, 10
LG 11, 12
LG 11, 12
LG 13, 14
LG 13, 14
LG 15, 16
LG 15, 16
LG 17, 18
LG 17, 18
LG 19, 20
LG 19, 20
Os06 + Os09
Sb10 + Sb02
Besides inter-chromosome rearrangements, dot-plot also revealed few intra-chromosomal rearrangements between Z. matrella and sorghum genomes. One inversion event was observed near one ends of LG5 and LG6 when the comparison was done between Z. matrella and sorghum genomes, but not between Z. matrella and rice genomes. This inversion event was also revealed by the comparison between sorghum and rice genomes. Similarly, another two inversion events were identified between Z. matrella LG7 and LG8 and their corresponding sorghum chromosome Sb4, and between Z. matrella LG13 and LG14 and their corresponding sorghum chromosome Sb7. These two inversion events were also observed when we compared sorghum chromosomes and their orthologous chromosomes in rice genome, but not between Z. matrella LGs and their rice homologous chromosomes.
Identification of QTLs associated with resistance to FAW
We evaluated the FAW resistance in the whole mapping population of 221 individuals using in vitro insect feeding bioassays. The bioassay for the whole population was split into 8 experiments and the two parents and a susceptible zoysiagrass genotype, DALZ 8516, were included in each experiment. The data between experiments were normalized using the susceptible genotype, DALZ 8516, by Abbott’s Formula . The mortality rates of larvae that were fed with leaf tissue of each individual of the mapping population were recorded at 7-, 10-, and 17-day. Significant difference in resistance to FAW between the two parents was observed during the whole procedure of in vitro insect feeding bioassay. Overall, the frequency distribution of the mortality rate of larvae fed leaf tissue from each individual of the mapping population exhibited non-normal distribution (Additional file 11: Figure S4). After 7 days of feeding, the mortality of larvae that were fed leaf tissue from the FAW resistant parent Cavalier was 57.7 % versus 1.7 % for the sensitive parent Diamond. The F1 population showed a skewed distribution toward FAW sensitive (Additional file 11: Figure S4A). A total of 88 F1 (39.8 % of the whole population) exhibited similar level of lethality rate (<10 %) as the FAW sensitive parent Diamond at 7-day. While, a total of 33 F1 individuals (14.9 % of the whole population) showed higher lethality rates than the FAW resistant parent Cavalier, indicating transgressive segregation of the mapping population for FAW resistance. After 10 days of feeding, the mortality of larvae that were fed leaf tissue from the FAW resistant parent Cavalier was at 87.47 %, a significant increase over the data collected at 7-day. For the FAW sensitive parent Diamond, the mortality of larvae remained low at 3.97 %. Among the 221 F1 individuals, almost equal number of individuals exhibited extreme tolerance and extreme sensitive to FAW (Additional file 11: Figure S4B). After 17 days of feeding, the mortality rate of larvae that were fed leaf tissue from Cavalier had increased to 100 %, while these that fed on Diamond tissue remained at similar levels as the one collected at 10-day (we want to clarify that the slightly lower mortality rate on Diamond at 17-day was caused by data normalization.). The frequency distribution of the mortality rate of larvae fed leaf tissue from the F1 individuals shifted toward FAW resistance at 17-day (Additional file 11: Figure S4C).
List of QTLs detected to be significantly associated with FAW resistance in zoysiagrass Cavalier
7-d-old larvae mortality rate
10-d-old larvae mortality rate
17-d-old larvae mortality rate
Grasses are the most important plant family to humans. They not only provide more than ¾ of our food, but also serve as a major producer of our oxygen due to their ecological dominance, wide geographic range, and enormous biomass. The Chloridoideae, consisting of more than 1,600 species, is one of the three largest subfamilies of the grasses, along with Pooideae and Panicoideae . Chloridoideae species share unusual features of leaf anatomy, and many of them exhibit extreme levels of tolerance to drought and high soil salinity. Furthermore, the transition from C3 to C4 photosynthesis that confers ecological success in many biomes firstly occurred in subfamily Chloridoideae, approximately 32.0–25.0 Mya . And the subfamily Chloridoideae has been identified as the largest wholly C4 clade in plants . Despite their ecological and economic importance, Chloridoideae species have received little attention on genomics studies, resulting in limited genetic and genomic information available for this important plant family. As a Chloridoideae species, Z. matrella has a base chromosome number of n = 10 which represents the predominant karyotypic number among the Chloridoideae. The sequence-tagged high-density linkage maps of Z. matrella developed in the present study provide a high-resolution framework for understanding evolutionary processes of this underexplored plant family, and serve as a foundation for future comparative genomic studies to expand our understanding of the origin of C4 photosynthesis in grasses.
Karyotype evolution in the subfamily Chloridoideae
Genome collinearity between zoysiagrass and rice, and between zoysiagrass and sorghum
Modern grass genomes have diverged at the level of genome size tremendously since they evolved from a common ancestor. The common ancestor genome had undergone a genome-wide duplication approximately 70 Mya [38–40]. Diversification of the two main grass clades, BEP and PACCMAD, occurred approximately 55 Mya following the genome-wide duplication [41–43]. Both zoysiagrass and sorghum are under the PACCMAD clade and rice belongs to the BEP clade. Although zoysiagrass is evolutionarily closer to sorghum than rice, our comparative analysis revealed a higher level of conservation between zoysiagrass and rice than that between zoysiagrass and sorghum. The genome size of the ancestral genome at the base of Poales was reconstructed as 1C = 1.7 pg , and the genome size of rice was estimated at 1C = 0.43–0.46 pg , indicating extensive DNA loss during the diploidization following the genome-wide duplication. The genome sizes of zoysiagrass and sorghum were estimated by flow cytometry at 0.86 pg/2C  and 1.56–1.74 pg/2C , respectively. Zoysiagrass is a tetraploid genome, and both rice and sorghum are diploid genomes. Therefore, the monoploid genome size of zoysiagrass would be the smallest among the three genomes, approximately 0.215 pg, only half of the size of the monoploid genome of rice and a quarter of the monoploid genome of sorghum. In general, Chloridoideae species have relatively smaller genome sizes and a narrower range of chromosome sizes compared to Panicoideae species , which may suggest the two subfamilies followed two distinctive paths toward genome size evolution after they diverged from a common ancestor. Genome size differences in plants are mainly caused by amplification and removal of repetitive DNAs [47, 48]. Therefore, we predicted that zoysiagrass genome contained much less repetitive sequences than sorghum genome. Mutations and genetic rearrangements are frequently associated with repetitive sequences. Large chromosomal rearrangements can be caused by ectopic recombination between repeat sequences in different genomic regions . Thus, the sorghum genome might have undergone more frequent genome rearrangements than did zoysiagrass due to its higher amount of repetitive sequences. On the other hand, the ancestor genome structure might have been better preserved in zoysiagrass than in sorghum.
FAW resistance in zoysiagrass influenced by a major gene effect
High-resolution genetic map provides an essential tool for efficient detection of QTLs of economically and agronomically important traits. The accuracy of QTL mapping relies to some extent on the density of the genetic maps. In our present study, the genetic maps consisted of 2,375 and 3,563 SNP markers with average interval of 0.74 cM and 0.51 cM, respectively. Considering the genome size of zoysiagrass estimated at 421 Mbp , the average physical distance between adjacent markers is 177 kb for the Diamond map and 118 kb for the Cavalier map. The total lengths of the genetic maps are 1745.48 cM and 1824.94 cM for Diamond and Cavalier, respectively. Thus, the global ratio of physical distance (bp)/genetic distance (cM) is 241 kb/cM for the Diamond map and 231 kb/cM for the Cavalier map. The major QTL for FAW resistance was mapped on the 1.4 cM region (48.9 cM to 50.3 cM) of the LG8 of Cavalier map. Based on this estimation, the major target of FAW resistance encompasses approximately 320 kb.
Plant resistance to herbivores can be attributed to plant secondary metabolites and plant physical traits. Studies have shown that leaf tensile strength and lignin concentration were positively correlated with FAW resistance in varieties such as Cavalier and Emerald . In addition, the secondary metabolite luteolin-3 was found in zoysiagrass that has an inverse relationship with FAW mortality, while luteolin-9 was positively correlated with FAW mortality in zoysiagrass . We evaluated the FAW resistance in the whole mapping population of 221 individuals using in vitro no choice insect feeding bioassays. Our phenotype evaluation revealed a clearly non-normal distribution, implying the possibility that FAW resistance in zoysiagrass was influenced by specific genes with major effects. After 10 days of feeding, the ratio of extremely tolerant F1s to extremely sensitive F1s fitted 1:1 ratio, reinforcing our hypothesis that the FAW resistance in zoysiagrass is a quantitative trait with a major gene effect.
Our QTL analysis provides us a major target for cloning FAW resistance gene in zoysiagrass, which will serve as a basis for a better understanding of host resistance adaptations in zoysiagrass. The major QTLs associated with FAW resistance will be validated using different mapping populations and a large germplasm collection, which will help refine estimates of the genetic structure of FAW resistance in zoysiagrass. Molecular markers developed from the present study offer breeders a more efficient approach to select for targeted chromosome regions during zoysiagrass improvement efforts.
We constructed sequence-based high-resolution genetic maps of Z. matrella. These maps offer valuable resource for comparative genomic analysis between Z. matrella and other plant species. Our comparative analysis revealed that Z. matrella shared a higher level of genome collinearity with rice than that with sorghum. Our result suggested that the sister subfamilies Chloridoideae and Panicoideae followed separate karyotype evolutionary pathways reducing the chromosome number from 12 to 10. These high-resolution maps also provide us important tools for detection of QTLs associated with economically and agronomically important traits. Six loci located on LGs 6, 8, and 20 were detected to be significantly associated with FAW resistance, providing us a major target for cloning FAW resistance gene in zoysiagrass.
BEP, Bambusoideae-Ehrhartoidea-Pooideae; CF, chromosome fusion; cM, centiMorgan; FAW, fall armyworm; KW, Kruskal-Wallis; LG, linkage group; Mya, million years ago; NCF, nested chromosomal fusion; NGS, next-generation sequencing; PACCMAD, Panicoideae-Aristidoideae-Centhothecoideae-Chloridoideae-Micrairoideae-Arundinoideae-Danthoideae; QTL, quantitative trait loci; RADSeq, restriction site-associated DNA sequencing; RCB, randomized complete block; SDA, single-dose allele; SDR, segregation distortion region; SNP, single nucleotide polymorphism; WGD, whole genome duplication
We would like to acknowledge Sandy Wisdorf for maintenance of greenhouse plant materials.
This work was supported by National Institute of Food and Agriculture (NIFA) – Specialty Crop Research Initiative (SCRI) Grant 2010-51181-21064 to AC and QY and the USDA National Institute of Food and Agriculture Hatch Project TEX0-1-9374 to QY.
Availability of data and materials
The dataset supporting the conclusions of this article is available in the NCBI SRA database under BioProject Accession PRJNA312939, http://www.ncbi.nlm.nih.gov/bioproject/312939.
XH, FW, and RS carried out genetic map construction, mapping QTLs associated with the resistance to FAW, and comparative genomics analysis. JAR conducted in vitro insect feeding bioassays. MCE, ADG, and AC created and maintained the mapping population. XH, FW, RS, and QY wrote the manuscript. QY and AC coordinated and organized all research activities. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Open AccessThis 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.
- Manidool C. In: ‘t Mannetje L, Jones RM, editors. Zoysia matrella (L.) Merrill. in Plant Resources of South-East Asia. 4. Forages. Wageningen: Pudoc Scientific Publishers; 1992. p. 234–5.Google Scholar
- Cai H, Li M, Wang X, Yuyama N, Hirata M. In: Cai H, Yamada T, Kole C, editors. Zoysiagrass in genetics, genomics and breeding of forage crops. Boca Raton: CRC Press, Taylor & Francis Group; 2013. p. 168–86.Google Scholar
- Vickery RA. Studies on the fall army worm in the gulf coast district of Texas. Beltsville: U.S. Department of Agriculture, National Agricultural Library, Alternative Farming Systems Information Center; 2004. p. 1–63.Google Scholar
- Braman SK, Duncan RR, Engelke MC. Evaluation of turfgrass selections for resistance to fall armyworms (Lepidoptera : Noctuidae). Hort Sci. 2000;35(7):1268–70.Google Scholar
- Braman SK, Duncan RR, Hanna WW, Engelke MC. Turfgrass species and cultivar influences on survival and parasitism of fall armyworm. J Econ Entomol. 2004;97(6):1993–8.View ArticlePubMedGoogle Scholar
- Anderson WF, Snook ME, Johnson AW. Flavonoids of zoysiagrass (Zoysia spp.) cultivars varying in fall armyworm (Spodoptera frugiperda) resistance. J Agric Food Chem. 2007;55(5):1853–61.View ArticlePubMedGoogle Scholar
- Joseph SV, Braman SK. Predatory Potential of Geocoris spp. and Orius insidiosus on Fall Armyworm in Resistant and Susceptible Turf. J Econ Entomol. 2009;102(3):1151–6.View ArticlePubMedGoogle Scholar
- Reinert JA, Engelke MC. Resistance in Zoysiagrass (Zoysia spp.) To the fall armyworm (Spodoptera frugiperda) (Lepidoptera: noctuidae). Fla Entomol. 2010;93(2):254–9.View ArticleGoogle Scholar
- Jessup RW, Renganayaki K, Reinert JA, Genovesi AD, Engelke MC, Paterson AH, Kamps TL, Schulze S, Howard AN, Giliberto B, et al. Genetic mapping of fall armyworm resistance in zoysiagrass. Crop Sci. 2011;51(4):1774–83.View ArticleGoogle Scholar
- Miller MR, Dunham JP, Amores A, Cresko WA, Johnson EA. Rapid and cost-effective polymorphism identification and genotyping using restriction site associated DNA (RAD) markers. Genome Res. 2007;17(2):240–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Baird NA, Etter PD, Atwood TS, Currey MC, Shiver AL, Lewis ZA, Selker EU, Cresko WA, Johnson EA. Rapid SNP Discovery and Genetic Mapping Using Sequenced RAD Markers. Plos One. 2008;3(10):e3376.View ArticlePubMedPubMed CentralGoogle Scholar
- Barchi L, Lanteri S, Portis E, Acquadro A, Valè G, Toppino L, Rotino GL. Identification of SNP and SSR markers in eggplant using RAD tag sequencing. BMC Genomics. 2011;12:304.View ArticlePubMedPubMed CentralGoogle Scholar
- Chutimanitsakun Y, Nipper RW, Cuesta-Marcos A, Cistué L, Corey A, Filichkina T, Johnson EA, Hayes PM. Construction and application for QTL analysis of a Restriction Site Associated DNA (RAD) linkage map in barley. BMC Genomics. 2011;12:4.View ArticlePubMedPubMed CentralGoogle Scholar
- Pfender WF, Saha MC, Johnson EA, Slabaugh MB. Mapping with RAD (restriction-site associated DNA) markers to rapidly identify QTL for stem rust resistance in Lolium perenne. Theor Appl Genet. 2011;122(8):1467–80.View ArticlePubMedGoogle Scholar
- Wang N, Fang L, Xin H, Wang L, Li S. Construction of a high-density genetic map for grape using next generation restriction-site associated DNA sequencing. BMC Plant Biol. 2012;12:148.View ArticlePubMedPubMed CentralGoogle Scholar
- Zhang Q, Li L, VanBuren R, Liu Y, Yang M, Xu L, Bowers JE, Zhong C, Han Y, Li S, et al. Optimization of linkage mapping strategy and construction of a high-density American lotus linkage map. BMC Genomics. 2014;15:372.View ArticlePubMedPubMed CentralGoogle Scholar
- Wu J, Li L-T, Li M, Khan MA, Li X-G, Chen H, Yin H, Zhang S-L. High-density genetic linkage map construction and identification of fruit-related QTLs in pear using SNP and SSR markers. J Exp Bot. 2014;65(20):5771–81.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang FF, Singh R, Genovesi AD, Wai CM, Huang XE, Chandra A, Yu QY. Sequence-tagged high-density genetic maps of Zoysia japonica provide insights into genome evolution in Chloridoideae. Plant J. 2015;82(5):744–57.View ArticlePubMedGoogle Scholar
- Catchen JM, Amores A, Hohenlohe P, Cresko W, Postlethwait JH. Stacks: building and genotyping loci De novo from short-read sequences. G3-genes genomes. Genetics. 2011;1(3):171–82.Google Scholar
- Van Ooijen. JoinMap 4, Software for the calculation of genetic linkage maps in experimental populations. Kyazma B.V., Wageningen, Netherlands. 2006Google Scholar
- Wu KK, Burnquist W, Sorrells ME, Tew TL, Moore PH, Tanksley SD. The detection and estimation of linkage in polyploids using single-dose restriction fragments. Theor Appl Genet. 1992;83(3):294–300.View ArticlePubMedGoogle Scholar
- Voorrips RE. MapChart: Software for the graphical presentation of linkage maps and QTLs. J Hered. 2002;93(1):77–8.View ArticlePubMedGoogle Scholar
- Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421.View ArticlePubMedPubMed CentralGoogle Scholar
- Abbott WS. A method of computing the effectiveness of an insecticide. J Econ Entomol. 1925;18:265–7.View ArticleGoogle Scholar
- Van Ooijen. MapQTL 5, Software for the mapping of quantitative trait loci in experimental populations. Kyazma B.V., Wageningen, Netherlands. 2004.Google Scholar
- Luo MC, Deal KR, Akhunov ED, Akhunova AR, Anderson OD, Anderson JA, Blake N, Clegg MT, Coleman-Derr D, Conley EJ, et al. Genome comparisons reveal a dominant mechanism of chromosome number reduction in grasses and accelerated genome evolution in Triticeae. Proc Natl Acad Sci U S A. 2009;106(37):15780–5.View ArticlePubMedPubMed CentralGoogle Scholar
- Salse J, Abrouk M, Bolot S, Guilhot N, Courcelle E, Faraut T, Waugh R, Close TJ, Messing J, Feuillet C. Reconstruction of monocotelydoneous proto-chromosomes reveals faster evolution in plants than in animals. Proc Natl Acad Sci U S A. 2009;106(35):14908–13.View ArticlePubMedPubMed CentralGoogle Scholar
- Abrouk M, Murat F, Pont C, Messing J, Jackson S, Faraut T, Tannier E, Plomion C, Cooke R, Feuillet C, et al. Palaeogenomics of plants: synteny-based modelling of extinct ancestors. Trends Plant Sci. 2010;15(9):479–87.View ArticlePubMedGoogle Scholar
- Murat F, Xu JH, Tannier E, Abrouk M, Guilhot N, Pont C, Messing J, Salse J. Ancestral grass karyotype reconstruction unravels new mechanisms of genome shuffling as a source of plant evolution. Genome Res. 2010;20(11):1545–57.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang X, Jin D, Wang Z, Guo H, Zhang L, Wang L, Li J, Paterson AH. Telomere-centric genome repatterning determines recurring chromosome number reductions during the evolution of eukaryotes. New Phytol. 2015;205(1):378–89.View ArticlePubMedGoogle Scholar
- Aliscioni S, Bell HL, Besnard G, Christin P-A, Columbus JT, Duvall MR, Edwards EJ, Giussani L, Hasenstab-Lehman K, Hilu KW, et al. New grass phylogeny resolves deep evolutionary relationships and discovers C4 origins. New Phytol. 2012;193(2):304–12.View ArticleGoogle Scholar
- Christin P-A, Besnard G, Samaritani E, Duvall MR, Hodkinson TR, Savolainen V, Salamin N. Oligocene CO2 decline promoted C4 photosynthesis in grasses. Curr Biol. 2008;18(1):37–43.View ArticlePubMedGoogle Scholar
- Ingram AL, Christin P-A, Osborne CP. Molecular phylogenies disprove a hypothesized C4 reversion in Eragrostis walteri (Poaceae). Ann Bot. 2011;107(2):321–5.View ArticlePubMedGoogle Scholar
- Prasad V, Stromberg CAE, Alimohammadian H, Sahni A. Dinosaur coprolites and the early evolution of grasses and grazers. Science. 2005;310(5751):1177–80.View ArticlePubMedGoogle Scholar
- Edwards EJ, Smith SA. Phylogenetic analyses reveal the shady history of C-4 grasses. Proc Natl Acad Sci U S A. 2010;107(6):2532–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Murat F, Zhang R, Guizard S, Flores R, Armero A, Pont C, Steinbach D, Quesneville H, Cooke R, Salse J. Shared subgenome dominance following polyploidization explains grass genome evolutionary plasticity from a seven protochromosome ancestor with 16 K protogenes. Genome Biol Evol. 2014;6(1):12–33.View ArticlePubMedGoogle Scholar
- Gaut BS. Evolutionary dynamics of grass genomes. New Phytol. 2002;154(1):15–28.View ArticleGoogle Scholar
- Vandepoele K, Simillion C, Van de Peer Y. Evidence that rice and other cereals are ancient aneuploids. Plant Cell. 2003;15(9):2192–202.View ArticlePubMedPubMed CentralGoogle Scholar
- Yu J, Wang J, Lin W, Li SG, Li H, Zhou J, Ni PX, Dong W, Hu SN, Zeng CQ, et al. The genomes of Oryza sativa: a history of duplications. Plos Biology. 2005;3(2):266–81.View ArticleGoogle Scholar
- Paterson AH, Bowers JE, Bruggmann R, Dubchak I, Grimwood J, Gundlach H, Haberer G, Hellsten U, Mitros T, Poliakov A, et al. The Sorghum bicolor genome and the diversification of grasses. Nature. 2009;457(7229):551–6.View ArticlePubMedGoogle Scholar
- Bremer K. Gondwanan evolution of the grass alliance of families (Poales). Evolution. 2002;56(7):1374–87.View ArticlePubMedGoogle Scholar
- Christin P-A, Spriggs E, Osborne CP, Stroemberg CAE, Salamin N, Edwards EJ. Molecular dating, evolutionary rates, and the Age of the grasses. Syst Biol. 2014;63(2):153–65.View ArticlePubMedGoogle Scholar
- Ming R, VanBuren R, Wai CM, Tang H, Schatz MC, Bowers JE, Lyons E, Wang M-L, Chen J, Biggers E, et al. The pineapple genome and the evolution of CAM photosynthesis. Nat Genet. 2015;47(12):1435–46.View ArticlePubMedPubMed CentralGoogle Scholar
- Leitch IJ, Beaulieu JM, Chase MW, Leitch AR, Fay MF. Genome size dynamics and evolution in monocots. J Bot. 2010;2010(862516):18.Google Scholar
- Arumuganathan K, Earle ED. Nuclear DNA content of some important plant species. Plant Mol Biol Rep. 1991;9:208–18.View ArticleGoogle Scholar
- Arumuganathan K, Tallury SP, Fraser ML, Bruneau AH, Qu R. Nuclear DNA content of thirteen turfgrass species by flow cytometry. Crop Sci. 1999;39(5):1518–21.Google Scholar
- Bennetzen JL, Ma JX, Devos K. Mechanisms of recent genome size variation in flowering plants. Ann Bot. 2005;95(1):127–32.View ArticlePubMedPubMed CentralGoogle Scholar
- Vitte C, Bennetzen JL. Analysis of retrotransposon structural diversity uncovers properties and propensities in angiosperm genome evolution. Proc Natl Acad Sci U S A. 2006;103(47):17638–43.View ArticlePubMedPubMed CentralGoogle Scholar
- Gaut BS, Wright SI, Rizzon C, Dvorak J, Anderson LK. Opinion - Recombination: an underappreciated factor in the evolution of plant genomes. Nat Rev Genet. 2007;8(1):77–84.View ArticlePubMedGoogle Scholar
- Hale TC, Reinert JA, White RH. Resistance of zoysiagrasses (Zoysia spp.) to fall armyworm (Lepidoptera: Noctuidae): I. Leaf tensile strength and cell wall components. Int Turfgrass Soc Res J. 2009;11:639–48.Google Scholar
- Hale TC, White RH, Reinert JA, Snook ME. Zoysiagrass (Zoysia spp.) resistance to fall armyworm (Spodoptera frugiperda): II. Polyphenols and flavonoids-components of resistance. Acta Hort., ISHS Conf on Turfgrass Sci. Manage. Sports Field. 2008;783:507–17.Google Scholar