Aerial plant surfaces are covered by epicuticular waxes that among other purposes serve to control water loss. Maize glossy mutants originally identified by their “glossy” phenotypes exhibit alterations in the accumulation of epicuticular waxes. By combining data from a BSR-Seq experiment and the newly developed Seq-Walking technology, GRMZM2G118243 was identified as a strong candidate for being the glossy13 gene. The finding that multiple EMS-induced alleles contain premature stop codons in GRMZM2G118243, and the one knockout allele of gl13, validates the hypothesis that gene GRMZM2G118243 is gl13. Consistent with this, GRMZM2G118243 is an ortholog of AtABCG32 (Arabidopsis thaliana), HvABCG31 (barley) and OsABCG31 (rice), which encode ABCG subfamily transporters involved in the trans-membrane transport of various secondary metabolites. We therefore hypothesize that gl13 is involved in the transport of epicuticular waxes onto the surfaces of seedling leaves.
Citation: Li L, Li D, Liu S, Ma X, Dietrich CR, Hu H-C, et al. (2013) The Maize glossy13 Gene, Cloned via BSR-Seq and Seq-Walking Encodes a Putative ABC Transporter Required for the Normal Accumulation of Epicuticular Waxes. PLoS ONE8(12): e82333. https://doi.org/10.1371/journal.pone.0082333
Editor: Mingliang Xu, China Agricultural University, China
Received: September 6, 2013; Accepted: October 31, 2013; Published: December 6, 2013
Copyright: © 2013 Li et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This material is based in part upon work supported by the National Science Foundation under Grant numbers DBI-0344852 and IOS-1027527. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Although co-author Delin Li is currently employed by DATA Biotechnology Beijing limited company all of his contributions to the current manuscript were completed while he was a graduate student at Chinese Agricultural University (CUA). A similar situation affects author Chuck Dietrich (i.e., although he is currently employed by Monsanto, the research reported in this manuscript was conducted while he was a student at Iowa State University). Thus, the authors do not have any conflicts of interest to report here.
Plant cuticles provide the first line of defense between a plant and its environment, providing protection from, for example, water loss [1-3] and UV irradiation . The plant cuticle also plays critical roles in plant interactions with the biotic environment, including bacterial, fungi and insects . As such, the physical and chemical properties of cuticle are of particular interest in a world that is experiencing global climate change. The cuticle consists of a thin layer of secreted compounds on the surface of the epidermis, and is synthesized exclusively by the epidermal cells [6,7]. The cuticle proper is covered by microcrystalline epicuticular waxes .
Epicuticular waxes are composed of several classes of compounds, including very-long-chain fatty acids (VLCFA; >18C), esters, primary and secondary alcohols, fatty aldehydes and ketones . Different species have distinct epicuticular waxes compositions and epicuticular wax composition can also differ among tissues within the same species . In Arabidopsis and barley (Hordeum vulgare), ~20 and ~80 independent eceriferum (cer) loci that affect wax accumulation have been defined genetically, respectively [9,10,11]. In maize (Zea mays), more than 30 glossy (or gl) loci have been identified that affect the quantity and/or the composition of eipcuticular waxes on the surfaces of seedling leaves , (and Schnable lab, unpublished data).
Seven of the maize glossy genes have been cloned [13-18]. gl1 encodes a protein that is similar to an Arabidopsis WAX2-related protein hydroxylase displaying transmembrane domains . The gl2 gene encodes a protein with transferase activity whose expression is restricted to epidermal cells and is involved in epidermal wax formation only in the green portions of seedlings . gl3 is a cell autonomous , putative myb transcription factor that affects the expression of other glossy genes . gl4 is a homolog of the Arabidopsis CUT1 gene which encodes a condensing enzyme involved in the synthesis of very-long-chain fatty acids . gl8a encodes a β-ketoacyl reductase of the fatty acid elongase complex involved in wax production . gl15 encodes an APETALA2-like transcription factor involved in the transition from juvenile to adult leaves .
As a first step in cloning the gl13 gene a collection of EMS-induced  and Mu-induced  mutants was isolated. BSR-Seq was used to map the gl13 locus to an 8 Mb interval of chromosome 3. Subsequently, a newly developed technology, Seq-Walking was used to identify Mu insertions within the 8 Mb mapping interval of a Mu-induced allele of gl13, thereby defining a gl13 candidate gene. Sequence analyses of multiple EMS-induced alleles confirmed that the candidate gene is gl13. The gl13 gene is a homolog of an ATP-Binding cassette G (ABCG32) family Arabidopsis gene, suggesting that gl13 is involved in the transport of epicuticular waxes onto the surfaces of epidermal cells.
Materials and Methods
The gl13-ref allele (AC520/gl13-N169; Maize Genetics Stock Center ID U440B; Schnable Lab #Ac 1200-1205) was induced by EMS from an uncertain of progenitor by Gerry Neuffer. A single Mu-induced allele of gl13 (gl13-Mu 90-3134B1, ID# 91g-6608) was isolated from the Q60 genetic background by the Schnable lab in 1990. A plant heterozygous for this allele was crossed to Q60 and a resulting heterozygous progeny was self-pollinated to create an F2 population. Wild-type progeny (Gl13/Gl13) from non-segregating F3 families and mutant plants (gl13-Mu/gl13-Mu) from a segregating F3 family were used for Seq-Walking (Figure 1C). Eight EMS-induced alleles (gl13-94-1001-1481, gl13-2:207-44, gl13-N211B, gl13-2:15-33, gl13-2:207-44, gl13-2:225-43, gl13-N1478B and gl13-N169 or gl13-ref) and two spontaneous gl13 alleles (gl13-Nec8495, gl13-U440B/PI251938b) derived from different labs as described by Schnable et al.  were used in this study.
(A) Numbers of informative SNPs used in BSR-Seq analyses of 3 EMS-induced gl13 alleles. (B) Results of BSR-Seq analysis for three alleles of gl13. (C) Crossing strategy used to generate seedlings homozygous for wild-type and gl13-Mu 90-3134B1 alleles for Seq-Walking. (D) Identification of shared, wild-type specific, and mutant-specific Mu-flanking sequence (MFS) sites and identification of mutant-specific MFS located with the gl13 mapping interval.
BSR-Seq and Differential Gene Expression Analysis
Three EMS-induced alleles (gl13-94-1001-1481, gl13-2:207-44 and gl13-N211B) were used for BSR-Seq mapping and expression analysis. Approximately 35 mutant and 35 wild-type seedlings per segregating F2 population were grown in a greenhouse sand bench under 27°C days and 24°C nights with 15 hours of daylight. Above ground tissues were collected 8 DAP (Days After Planting) for the isolation of RNA. Total RNAs were quantified with a Bioanalyzer and then subjected to RNA-Sequencing using an Illumina HiSeq2000 paired-end sequencing instrument. Sequencing data were trimmed and aligned to the B73 reference genome (AGPv2) and subjected to SNP calling as well as reads count normalization . SNPs detected between the B73 reference genome and the 25 NAM founders (plus Mo17) were used for SNP filtering. Final filtered SNPs from both mutants and wild-type pools were used for BSR-Seq analysis . Genes with an average of at least five uniquely mapped read across samples were tested for differential expression between the gl13 mutant and sibling wild-type seedlings using the R package QuasiSeq (http://cran.r-project.org/web/packages/QuasiSeq). The negative binomial QLSpline method implemented in the QuasiSeq package was used to compute a p-value for each gene. The 0.75 quantile of reads from each sample was used as the normalization factor . A multiple test controlling approach  was used to convert p-values to q-value. To obtain approximate control of the false discovery rate at 5%, genes with q-values no larger than 0.05 were declared to be differentially expressed. Gene Ontology (GO) terms were determined using tools and resources available at: http://geneontology.org/
To identify the causal Mu insertion in gl13-Mu 90-3134B1, homozygous F2 wild-type and mutant plants derived from same F2 family were used for Seq-Walking (Figure 1C). Genomic DNAs from mutant and wild-type pools were fragmented via sonication (BioRuptor UCD-200, Diagenode, USA) for 8-10 cycles at intervals of 30s /30s on/stop. Purified DNA fragments were used for library preparation using the NEBNext® DNA Library Prep Master Mix Set for Illumina® kit (NEW ENGLAND BioLabs® inc.). The Seq-Walking library was prepared according to Ion Torrent library preparation pipeline (Figure S1). Equal quantities of DNA (based on results from the Bioanalyzer’s 2100 expert High Sensitive DNA Assay) from the mutant and wild-type pools (each of which had distinct DNA barcodes) were pooled. Final libraries were sequenced on an Ion Torrent PGM instrument using a 318 Chip.
Raw PGM reads were decoded based on library-specific barcodes with an in-house Perl script. After trimming barcodes, Mu sequences from the decoded reads were trimmed using the software package, SeqClean (sourceforge.net/projects/seqclean). Those reads that were longer than 90 bp after trimming were aligned to the B73 reference genome (AGPv2) using the aligner GSNAP . Uniquely mapped reads were used to define a set of non-redundant Mu-flanking sequence (MFS) sites for each library. The read count for each MFS site was determined for each library.
DNA and total RNA samples isolated from the 10 gl13 alleles that were not derived from Mu stocks were used as templates for PCR amplification with a set of gene-specific primers that cover the transcribed region of the gl13 gene (Table S11).
An MFS site from the Mu-insertion allele, gl13-Mu 90-3134B1 was amplified using a Mu primer (MuTIR: 5’AGAGAAGCCAACGCCA(AT)CGCCTC(CT)ATTTCGTC3’)  and gene-specific primer (gl13C1L11: 5’GGAGCAGATTCTTGGAGTGG3’ or gl13C1R11: 5’CCGAACTGAGAGGTCAGGAG3’ (also see Table S11). The resulting PCR products were sequenced using Sanger technology.
Quantitative RT-PCR analysis for expression pattern in 11 gl13 alleles
Seedlings from an F2 family segregating for gl13-ref were grown in a growth chamber under 18 hours of light at 35°C and 6 hours of dark at 20°C and then harvested at 6 DAP, 10 DAP and 15 DAP. Similarly, 11 F2 families each of which was segregating for one of the 11 alleles (Table 1) were grown in a growth chamber under 18 hours of light at 35°C and 6 hours of dark at 20°C. At 7~8 DAP seedlings were screened for phenotype and mutant and wild-type seedlings were harvested at 10 DAP. Total RNA was isolated from tissue samples using the RNeasy Mini Kit (Cat.No.74134, Qiagen, USA). cDNA was prepared using the iScript cDNA Synthesis Kit (Bio-Rad). Real-Time PCR was performed using iQ SYBR Green Supermix (Bio-Rad) on the Light Cycler® 480II instrument (Roche, USA). Relative quantification values (RQ) were calculated by the 2−ΔCt method (RQ = 2−ΔCt) with two biological replicates and two technical replicates using actin mRNA as an endogenous control. Oligonucleotides used for real-time PCR: forward primer: gl13C1L13.2, 5’ACCATTGCGCCTATTATTGC3’; reverse primer: gl13C1R11, 5’CCGAACTGAGAGGTCAGGAG3’ (also in Table S11); Actin408F: 5’CCAGGCTGTTCTTTCGTTGT3’; Actin520R: 5’GCAGTCTCCAGCTCCTGTTC3’
|ID||Allele||Progenitor||Mutagen||Pos. (bp) On Chr3||Mutation||Ref. (B73)||A632||Mo17||Lesion||AA change|
|1||gl13-94-1001-746a||A632||EMS||10282947-10282948||deld||AC||AC||AC||2 bp deletion in 47 bp downstream of stop codon||-|
|2||gl13-94-1001-1481a||A632||EMS||10273165||G->A||G||G||G||28 bp upstream of start codon||-|
|8||gl13-Nec 8495a||ND||ND||10282712||C->A||C||C||C||synonymous||NO CHANGE|
|11||gl13-Mu 90-3134B1c||Q60||Mu||10281062-10281070||-||-||-||-||Mu insertion (Mu1)||-|
Fresh tissues from leaf blades and stems of 10 DAP seedlings were subject to FAA fixation followed by dehydration with gradient alcohol solutions (50%, 70%, 95%, 100%), and then infiltrated with 50% xylene solution (1:1 of xylene:alcohol) and subsequently 100% Xylene. Tissues were embedded in a mixture of paraffin and xylene at 40°C overnight, followed with 60°C paraplast infiltration using pure paraffin. Imbedded tissues were sectioned on a Leica RM2135 rotary microtome (Thermo Scientific, USA), stained with Fast Green and imaged on Nikon Eclipse E1300 Microscope (Nikon, USA).
Isolation of glossy13 mutants
Allelism tests of a large collection of glossy mutants identified 11 alleles of gl13 (Table 1). One of these alleles (gl13-Mu 90-3134B1) was isolated via random mutagenesis using the Mu transposon . Eight of these alleles were generated via EMS-mutagenesis. The remaining 2 alleles were obtained from other sources or isolated by the Schnable lab .
Phenotypic characterization of glossy13 mutants
Ten day-old seedlings homozygous for the glossy13 reference allele (gl13-N169) exhibit a glossy phenotype (Figure 2A). Somewhat later in development, these mutant seedlings develop rolled leaves that exhibit necrotic lesions and homozygotes for some mutant alleles are ultimately lethal. Scanning electron microscopic analysis of juvenile leaves from seedlings homozygous for the gl13-ref allele revealed significant disruption of the epicuticular wax crystals as compared to wild-type. Paraffin sections of seedling leaves reveal significant differences in cell morphology between sibling wild-type and gl13-ref mutant leaves; as compared to wild-type leaves epidermal cells on mutant leaves are unordered (Figure 2B). Because epicuticular waxes serve as a water barrier, we tested the ability of the gl13 mutant to retain water post-harvest. Detached leaf sections were exposed to the open air at room temperature for two hours and then weighed. Leaf sections from the gl13-N169 (gl13-ref) mutant showed more water loss than sibling wild-type controls (Figure 2C, D).
(A) Comparisons of gross morphology and epicuticular wax accumulation and morphology in mutant (upper) and wild-type (lower) seedlings; (B) Comparisons of paraffin sections (10x) of leaves from mutant (left) and wild-type (right) seedlings stained with Fast Green; (C) Detached leaves from mutant (upper) and wild-type (lower) immediately after harvest and after 2 hours at room temperature; (D) Water loss from detached mutant and wild-type seedling leaves at room temperature. Error bars = SE.
Mapping gl13 to an 8 Mb interval on chromosome 3 via BSR-Seq
BSR-Seq technology  is a modification of Bulked Segregant Analysis , because it is based on RNA-Seq data, BSR-Seq provides both genetic mapping data and global expression data . BSR-Seq was conducted using three independent EMS-induced mutant alleles of gl13 (Methods). Three F2 populations, each of which was segregating for a different EMS-induced allele were subjected to RNA-Seq. SNPs were identified between mutants and sibling wild-type controls via comparisons of RNA-Seq reads to the B73 reference genome. Identified SNPs were subjected to a statistical test for linkage to gl13, based on the numbers of RNA-Seq read counts associated with each SNP in each of the two pools (Methods, Figure 1A). The probability of linkage for each SNP from each of the three RNA-Seq experiments was plotted. Clusters of significant SNPs were detected in an 8 Mb region on the long arm of chromosome 3 (chr3:5-13 Mb) in the two F2 families segregating for gl13-94-1001-1481 and gl13-2:207-44. The mapping interval detected in the F2 family segregating for gl13-N211B was larger, but overlapped with this 8 Mb region (Figure 1B).
Identification of the gl13 gene via Seq-Walking
Here, we report a novel technology, Seq-Walking that amplifies and sequences DNA fragments flanking Mu transposon insertions throughout the genome. A set of F3 families derived from homozygous mutant and non-mutant individuals from an F2 population segregating for the Mu-induced allele, gl13-Mu 90-3134B1 (Figure 1C; Methods) was used for Seq-Walking (Figure S1). Genomic DNA samples were extracted from F3 mutant plants (gl13/gl13) and wild-type plants from non-segregating F2 families (Gl13/Gl13). Pooled DNAs were used for NGS library preparation (Figure 1C) and sequenced using a Life Technologies PGM instrument (Methods).
561,749 raw reads were obtained from the PGM; 411,892 were from the gl13 mutant library and 149,857 from the wild-type library. After filtering for length (>90 bp) 69,578 and 25,645 reads from the mutant and wild-type samples, respectively could be uniquely mapped to the B73 reference genome . These reads are derived from 471 and 251 non-redundant MFS sites in the gl13 mutant and wild-type pools, respectively (Figure 1D). Among these sites, 82 were detected in both mutant and wild-type pools and thus removed. The remaining non-redundant 389 MFS sites in the mutant pool and 169 MFS sites in the wild-type pool were defined as “mutant specific” and “wild-type specific”, respectively. Four of the 389 “mutant specific” insertion with highest read counts were located in the 8 Mb mapping interval defined by the BSR-Seq mapping experiment (Table 2). Among these four genes, only GRMZM2G118243 was differentially expressed between mutant and wild-type based on RNA-Seq read counts from the BSR-Seq experiment; the corresponding MFS sites also had the highest reads count in the Seq-Walking analysis (Table 2). Using PCR primers specific to the Mu transposon and GRMZM2G118243 (Methods) we detected a Mu insertion (Mu1 ) in gl13-Mu 90-3134B1 at the site predicted based on the results from the Seq-Walking experiment (Figure 1C, D). GRMZM2G118243 was therefore identified as a gl13 gene candidate.
|Chr||Start of MFS (bp)a||End of MFS (bp)a||Read Count||Target gene||Insertion site||DEGb|
Validation of the gl13 candidate gene
The sequences of RNA-seq reads from the BSR-Seq mapping experiment that align to GRMZM2G118243 were compared to the progenitor alleles of the three gl13 EMS-induced alleles used in the BSR-Seq experiment to determine whether the EMS-induced alleles contained typical EMS-induced polymorphisms (G/C -> A/T transitions ) relative to their progenitor alleles. The gl13-2:207-44 and gl13-94-1001-1481 alleles were derived from Mo17 and A632, respectively. The sequences of the Mo17 and A632 progenitors were imputed by projecting known Mo17 vs. B73 [30,31] and A632 vs. B73 (Schnable Lab, unpublished results) SNPs discovered from RNA-Seq data onto the B73 allele of GRMZM2G118243. The progenitor of gl13-N211B is not known. Hence, we compared the sequences of the RNA-Seq reads from this allele to the sequences of all GRMZM2G118243 alleles present in the 26 founders of the NAM population  that had been determined by RNA-Seq analyses (Schnable Lab, unpublished data) and excluded from consideration any SNP present in any of the NAM founders. Based on these analyses two of the three EMS-induced alleles contain typical EMS-induced mutations that result in premature termination codons (PTCs). These PTCs are located in exon 9 (gl13-N211B) and exon 11 (gl13-2:207-44) of GRMZM2G118243 (Figure 3, red arrows 4, #6, Table 1, S1). A typical EMS-induced mutation (G/C -> A/T) was detected in the 5’ gl13-94-1001-1481 as compared with its progenitor A632 (Figure 3, 2). Because RNA-Seq data were obtained from only the transcribed region, it is also possible that a mutation outside of the transcribed region is responsible for the mutant phenotype of gl13-94-1001-1481.
The positions of detected lesions are indicated by arrows within the gl13 gene. Green boxes indicate gene coding region (CDS) and yellow boxes indicate un-translated region (UTR). Alleles are numbered according to Table 1. Four EMS-induced premature terminal codon (PTC) mutations are shown in red; two EMS-induced non-synonymous substitutions are shown in green; an EMS-induced G->A transition is shown in blue; the location of a Mu insertion (belong to Mu1 family ) is shown in pink; the location of a C->A transversion found in two spontaneous alleles are shown in yellow; and a 2bp deletion detected in one EMS-induced allele is shown in black.
To confirm the existence of the PTCs identified in 2/3 of the EMS-induced alleles via inspection of RNA-seq reads and discover further evidence in support of the gl13 candidate gene, the three EMS-induced alleles used in the BSR-Seq experiment plus five additional EMS-induced alleles and two spontaneous alleles (i.e., a total of 10 alleles) were sequenced via Sanger sequencing. In addition to confirming the two PTC alleles first identified in the BSR-Seq experiment, we detected PTCs in two additional EMS-induced alleles: gl13-2:225-43 (exon 18), gl13-N1478B (exon 14) (Figure 3, red arrows 5, #7; Table 1, S1).
G to A transitions were detected in three other alleles. gl13-2:15-33 and the reference allele (gl13-N169) contain non-synonymous mutations in exons 17 and 23, respectively (Figure 3, 3, 10 Table 1, S1). The gl13-94-1001-1481 allele, which was derived from the same genetic background as gl13-94-1001-746, contains a G to A transition 24 bp upstream of the start codon (Figure 3, #2; Table 1, S1).
Polymorphisms were also detected in other alleles, although it is not possible to establish whether these are causative. One EMS-induced gl13 allele (gl13-94-1001-746), which was derived from A632, contains a 2 bp deletion (of AC) 47 bp downstream of the stop codon (Figure 3, #1; Table 1, S1). The two spontaneous glossy alleles, gl13-Nec8495 and gl13-U440B/PI251938b both contain a synonymous substitution (A to C transversion) at the same position in exon24 (Figure 3, 8, 9; Table 1, S1).
It is notable that even though an ~1 kb section of the gl13 gene between exon 6 and exon 7 could not be cleanly amplified and sequenced, we were still able to identify PTC or non-synonymous mutations typical of EMS-mutagenesis in 6/8 of the EMS-induced alleles, thereby confirming that GRMZM2G118243 is gl13. This gl13 gene is 11,328 bp in length (located between positions 10,272,500 to 10,283,828 bp on chromosome 3) and contains 24 exons.
gl13 is predicted to encode a putative G family ABC transporter
The amino acid sequence of gl13 was compared with orthologs from other species in the NCBI database using Blastp. The GL13 protein (encoded by GRMZM2G118243) exhibits 97% similarity to the protein encoded by Sb03g0040410.1 (Sorghum), 94% similarity to OsABCG31 (rice), 92% similarity to HvABCG31 (barley) and 70% similarity to AtABCG32 (Arabidopsis). A phylogenetic tree was constructed based on the amino acid sequences of these proteins (Table S2, Figure 4). Domain analysis of the protein encoded by GRMZM2G118243 predicted that gl13 encodes an ATP-binding cassette subfamily G full transporter, a type of transporter that includes both white-brown complex (WBC) half transporters and pleiotropic drug resistance (PDR) full transporters . Consistent with our phenotyping data, AtABCG32 is predicted to encode an ABCG subfamily transporter involved in the export of epicuticular components from epidermal cells (Table S3) [33,34]. The homologs of gl13 in rice (OsABCG31) and barley (HvABCG31) both contribute to the accumulation of cutin [34,35].
The amino acid sequence of the 2 GL13 isoforms (GRMZM2G118243_P01 and GRMZM2G118243_P02) were aligned to 9 homologs downloaded from GenBank (Table S2). Neighbor-joining analysis was used to generate a phylogenetic tree, which was bootstrapped over 1,000 cycles, using MEGA5.0.
Expression and regulation of gl13
A Q-Teller analysis (http://www.qteller.com) (Figure S2) showed that the gl13 gene is highly expressed in tassel , young leaves [37,38] and seedling shoots , while being expressed at lower levels in seedling roots, seedling leaves (V2 stage), pollen and mesophyll cells, as well as bundle sheath cells [36-39].
Of the eight cloned glossy genes (gl1, gl2, gl3, gl4a, gl8a, gl8b and gl15) only gl2 was differentially expressed between gl13 mutant and sibling wild-type seedlings in the BSR-Seq experiment using a FDR=0.05. Surprisingly, the gl2 and gl13 genes were both significantly up-regulated in the gl13-ref (gl13-N169) mutant as compared with wild-type (q-values=0.032 and 0.042, respectively) (Table S4).
Based on a published eQTL analysis , gl13 is regulated by a trans eQTL located in the 853.4-861.4 cM interval of chromosome 3 that contains five genes. None of the 16 other genes that are also regulated by this eQTL interval are known to be associated with the accumulation of epicuticular waxes (data not shown).
RNA-Seq analysis of the gl13 mutant
In our BSR-Seq analysis a total of 30,536 genes were expressed (i.e., accumulated an average of at least five uniquely mapped RNA-Seq reads across samples) in wild-type and/or mutant seedlings. These genes were tested for differential expression between the gl13 mutant and wild-type (Methods). Controlling the false discovery rate (FDR) at 5%, 8.3% (2,522/30,536) of the expressed genes were differential expressed (DEGs), including 1,533 that were up-regulated (Table S5) and 989 that were down-regulated in the mutant as compared with the wild-type (Table S6). Genes involved in cell wall precursor synthesis, cell wall degradation and secondary metabolism of phenylpropanoids (which is involved in lignin biosynthesis), that encode peroxidases, lipid transfer proteins (LTP) or belong to the APETALA2/Ethylene-responsive element binding protein family which is involved in epidermal wax synthesis  are enriched (Methods) among the DEGS (Figure S3, Table S7, S8). Interestingly, most of the enriched pathway annotations are from the up-regulated DEGs rather of down-regulated DEGs (Figure S3, Table S8). Among these DEGs, 541 (35%) of the up-regulated and 49 (4.9%) of the down-regulated genes exhibited a log2FC>4 in gl13 mutants (Table S9, S10). The 541 strongly up-regulated genes are enriched in transcription factors (AP2/EREBP, APETALA2/Ethylene-responsive element binding protein family), biotic stress pathway genes, and cell wall degradation pathway genes (e.g., pectate lyases and polygalacturonases) (Figure S3, Table S7), consistent with roles in the biosynthesis of epicuticular waxes or that are induced by biotic stress.
gl13 is up-regulated in EMS-induced alleles
Surprisingly, based on the BSR-Seq data, the gl13 gene was up-regulated in the gl13 mutant. Consistently, the gl13 genes was up-regulated in all ten (8 EMS-induced and 2 spontaneous) of the non-Mu-induced mutant alleles as compared to wild-type based on RNA-Seq and qRT-PCR analyses of 10 DAP seedlings (Table 3, Figure 5). Only seedlings homozygous for the Mu-induced gl13 allele exhibited decreased expression of gl13 as compared to wild-type (Figure 5A). In seedlings homozygous for the gl13-ref (gl13-N169) allele, the gl13 gene was expressed at higher levels than a wild-type control at both 6 DAP and 10 DAP; in contrast, at 15 DAP, the gl13 gene was more strongly expressed in wild-type than mutant seedlings (Figure 5B).
(A) Transcript accumulation from the gl13 gene was measured via qRT-PCR in 8 EMS-induced alleles and 2 spontaneous alleles and compared to a wild-type control (left box) and the Mu-induced allele and its appropriate wild-type control (right box). (B) The relative expression of gl13 in shoots (6 DAP) and seedlings (10 DAP and 15 DAP) from wild-type and gl13-ref mutant.
During BSR-Seq  SNPs are discovered by comparing RNA-Seq reads derived from pools of mutant and wild-type siblings. We used BSR-Seq to map the gl13 gene to an 8Mb interval of the maize genome. As in any BSR-Seq analysis we were able to define an interval on the genetic map that contains the gene of interest. The SNPs discovered within this mapping interval are available for subsequent experiments, including fine mapping. Moreover, BSR-Seq enables the identification of genes that are differentially expressed between the mutant and wild-type pools. The candidate gene may be one of the differentially expressed genes located within the mapping interval. Differentially expressed genes genome-wide may also provide clues as to the molecular function(s) of the gene of interest. As described above, using the newly developed Seq-Walking technology, we identified a Mu-insertion in the coding region of gene GRMZM2G118243 within the gl13 mapping interval. GRMZM2G118243 was ultimately demonstrated to be gl13 via sequence analyses of a series of EMS-induced alleles.
Transposons, including the Mu transposons of maize  are widely used for gene cloning and functional analyses in model organisms [43-48]. After transposon insertion mutants have been isolated, they can be used to clone the underlying genes by identifying the sequences flanking the causative transposon insertions (MFS). Multiple methods have been reported to isolate MFS [49-54]. It is, however, still technically challenging to clone MFS due to the high copy number of Mu elements and their continued transposition in “Mu-active” lines . In this study, we report a new technology, Seq-Walking to obtain MFS using Life Technologies’ sequencing platform. In combination with the mapping interval defined by the BSR-seq  experiment a gl13 candidate gene was readily defined via Seq-Walking. We began by conducting Seq-Walking on DNA extracted from pools of homozygous mutant and wild-type relatives. This identified large numbers of MFS from each pool, which were then mapped to the B73 reference genome. Only those MFS that mapped to the gl13 mapping interval defined by the BSR-Seq experiment were used in subsequent analyses. Those MFS that were present in both pools (“universal insertions”) could be discarded. High copy “wild-type specific” MFS present only in the wild-type pool are likely derived from insertions that arose following the divergence of the wild-type and mutant families in this Mu-active genetic background. In contrast, “wild-type specific” MFS having low read counts are probably derived from somatic insertions, each of which would be present only in a single seedling in the wild-type pool. We therefore focused on the “mutant-specific” MFS. Because somatic insertions would also be expected to arise in the mutant pools we ranked the mutant-specific MFS by read counts to identify the gl13 candidate gene. These results demonstrate the ease with which it is possible to map and clone a gene via a combination of BSR-Seq and Seq-Walking if transposon-induced alleles are available. It should be straightforward to adapt Seq-Walking to enable the efficient cloning of sequences flanking other transposons.
Transposon insertion alleles are, however, not always available. This is the first report of conducting BSR-Seq on EMS-induced alleles. Using EMS alleles in a BSR-Seq experiment has the significant advantage that the mutant alleles would be expected to contain transition mutations (G:C→A:T) in the gl13 locus. In our search for a gl13 gene candidate within the 8Mb mapping interval we exploited this fact in parallel with the Seq-Walking experiment described above. Analysis of the three BSR-Seq experiments identified 6, 10 and 5 genes in the gl13 mapping interval that harbored non-synonymous typical EMS-induced transition SNPs relative to the B73 reference genome that were also not present in any of the 26 (non-glossy) inbred founders of the NAM population . Nine of the genes in the 8 Mb gl13 mapping interval were DEG at log2FC>=2 (6 down-regulated and 3 up-regulated relative to the wild-type allele; Table S12). We initially made the assumption that the mutant allele of the gl13 gene would be among the down-regulated genes. Two of the 6 down-regulated genes (GRMZM2G051753 and GRMZM2G352891) were at least potentially involved in wax biosynthesis (Table S9, S10). Neither of these genes, however, contained a typical EMS mutation in any of the three EMS-induced alleles. We next analyzed the 3 up-regulated genes contained within the mapping interval. Two of the EMS-induced mutants contained non-synonymous SNPs that encoded PTC codons in the same gene, GRMZM2G118243, which we subsequently demonstrated was the gl13 locus. These results demonstrate that by conducting BSR-Seq on EMS-induced mutant alleles it will be possible to at least sometimes directly identify the causative gene from the RNA-Seq data.
In the case of gl13 this discovery process was enabled by the fact that all 8 of the EMS-induced and 2 spontaneous alleles accumulated abundant levels of transcripts, indeed these levels exceeded those in the wild-type. This is surprising given that 4/8 of these up-regulated EMS-induced alleles contain PTC mutations (Table 3, Figure 5 red box) and that PTC-containing transcripts are typically degraded via nonsense-mediated decay (NMD) . There are, however, several reports of PTC alleles that are up-regulated as compared to wild-type controls [58-65]. Based on these results we are conducting a global analysis of the expression of PTC alleles among the NAM founders.
Full-sized ATP-binding cassette (ABC) transporters of the G subfamily (ABCG) are considered to be essential components of the plant immune system . This is consistent with the observation that epicuticular waxes play roles in plant/insect interactions . The fact that excised seedling leaves from gl13 mutants have lower water retention capacity than leaves from wild-type seedlings indicates that the gl13 genes, and by extension perhaps, epicuticular waxes in general may play important roles in water relations, as is the case for the gl13 homologs of barley and rice . These findings suggest a strategy for enhancing drought tolerance in crops.
Seq-Walking. Process by which genomic DNA is prepared for Seq-Walking.
Q-Teller analysis of the gl13 gene. Accumulation of transcripts from the gl13 gene (GRMZM2G118243) in multiple tissues and at multiple stages of development as measured via RNA-Seq.
Mapman analysis of the results of an RNA-Seq experiment comparing transcript accumulation in wild-type and gl13 mutant seedlings. Genes marked in blue and red are up- and down-regulated in mutant relative to wild-type, respectively. Darker shades designate larger fold changes.
Putative causal mutations in 10 gl13 alleles and the corresponding SNPtypes of the 25 NAM founders based on RNA-Seq reads. SNP data were discovered from RNA-Seq data of NAM (Nested associated mapping) founders in Schnable lab. The first 8/10 gl13 alleles shown in column 1 were derived from EMS-mutagenesis. The second and third columns indicate the physical position and nature of the putative causal mutation associated with each allele, if identified. In column 3 the B73 nucleotide is listed first. The SNPs at the putative causal positions are listed in columns 4-30 for each of the 25 NAM founders plus B73 and Mo17). The first number in parenthesis is the number of reads supporting the indicated SNP. The second number is the total reads that align in that particular SNP site. "--" indicates missing data or deletions.
Expression of 8 different glossy genes in gl13 mutant and wild-type seedlings.
Up-regulated DEGs in gl13 (mutant vs. wild-type). The F2 segregating populations of three gl13 EMS-induced alleles were used for the RNA-Seq experiment (Methods). Maize reference genome version 2 (Refgen2) was used as the reference. Gene orientation and position on chromosomes are listed in the table. RPM represents reads per million of mapped reads. log2FC designates log2 of fold changes. p-value was from the statistical test of differential expression. q-value is an adjusted p value. sig. indicates whether a gene is significantly differentially expressed.
Down-regulated DEGs in gl13 ( mutant vs. wild-type). The F2 segregating populations of three gl13 EMS-induced alleles were used for the RNA-Seq experiment (Methods). Maize reference genome version 2 (Refgen2) was used as the reference. Gene orientation and position on chromosomes are listed in the table. RPM represents reads per million of mapped reads. log2FC designates log2 of fold changes. p-value was from the statistical test of differential expression. q-value is an adjusted p value. sig. indicates whether a gene is significantly differentially expressed.
DEGs MapMan enrichment analysis.The enrichment analysis of gl13 differential expressed genes (DEGs) using MapMan annotation (http://mapman.gabipd.org). Four layers of annotation were defined by MapMan. The enrichment was tested for DEGs, up-regulated DEGs and down-regulated DEGs in each category of each annotation layer. Fisher's Exact test was used for the enrichment test and the resulting p-values were adjusted to account for multiple statistical tests (q-values).
Enrichment analysis for DEGs based on Gene Ontology. Most up-regulated DEGs in gl13 mutant or wild-type have enrichment information in Gene Ontology database.
541 significantly up-regulated DEGs with at least 16 fold changes in gl13 mutants relative to wild-type siblings. Maize reference genome version 2 (Refgen2) was used as reference. Gene orientation and position in chromosome were listed in table. Among 1,533 significantly up-regulated genes in the mutant as compared to the wild-type (Table S5), 541 exhibit at least 16-fold change. (Log2FC<=-4).
49 significantly down-regulated DEGs in gl13 mutants relative to wild-type siblings. Maize reference genome version 2 (Refgen2) was used as reference. Gene orientation and position in chromosome were listed in table. Among 989 significantly down-regulated genes in the mutant as compared to the wild-type (Table S6), 49 exhibit at least 16-fold change. (Log2FC<=-4). .
Primers used for gl13 sequence analysis and Seq-Walking library prepare.
We thank Life Technologies for providing sequencing reagents, Dr. Wei Wu, Cheng-Ting “Eddy” Yeh and Dr. An-Ping Hsia for technical support and/or helpful discussions, Mitzi Wilkening of the ISU Genomic Technologies Facility for sequencing services, and former graduate student, the late Joel Hansen, former research associate, Philip Stinard, and current nursery manager, Lisa Coffey for the generation and maintenance of the genetic stocks used in this study.
Conceived and designed the experiments: SL PS. Performed the experiments: LL DL SL CD XM. Analyzed the data: SL DL. Contributed reagents/materials/analysis tools: JZ GW. Wrote the manuscript: LL HH PS. Mentored graduate student authors: GZ ZL PS.
- 1. Kerstiens G (1996) Cuticular water permeability and its physiological significance. Journal of Experimental Botany 47: 1813–1832. doi:10.1093/jxb/47.12.1813.
- 2. Riederer M, Schreiber L (2001) Protecting against water loss: analysis of the barrier properties of plant cuticles. J Exp Bot 52: 2023–2032. doi:10.1093/jexbot/52.363.2023. PubMed: 11559738.
- 3. Jenks MA, Eigenbrode SD, Lemieux B (2002) Cuticular waxes of Arabidopsis. The Arabidopsis Book/American Society of Plant Biologists. p. 1.
- 4. Sturaro M, Hartings H, Schmelzer E, Velasco R, Salamini F et al. (2005) Cloning and characterization of GLOSSY1, a maize gene involved in cuticle membrane and wax production. Plant Physiol 138: 478–489. doi:10.1104/pp.104.058164. PubMed: 15849306.
- 5. Post-Beittenmiller D (1996) Biochemistry and molecular biology of wax production in plants. Annu Rev Plant Physiol Plant Mol Biol 47: 405–430. doi:10.1146/annurev.arplant.47.1.405. PubMed: 15012295.
- 6. Kolattukudy PE (1996) Biosynthetic pathways of cutin and waxes, and their sensitivity to environmental stresses. Plant cuticles: an integrated functional approach. BIOS Scientific Publishers Ltd: Oxford, UK. pp. 83–108.
- 7. Müller C, Riederer M (2005) Plant surface properties in chemical ecology. J Chem Ecol 31: 2621–2651. doi:10.1007/s10886-005-7617-7. PubMed: 16273432.
- 8. Kunst L, Samuels AL (2003) Biosynthesis and secretion of plant cuticular wax. Prog Lipid Res 42: 51–80. doi:10.1016/S0163-7827(02)00045-0. PubMed: 12467640.
- 9. Mariani M, Wolters-Arts M (2000) Complex waxes. Plant Cell Available: 12. doi:10: 1795–1798. PubMed: 11041876.
- 10. Lundqvist U, Lundqvist A (1988) Mutagen specificity in barley for 1580 eceriferum mutants localized to 79 loci. Hereditas 108: 1–12.
- 11. Lundqvist U (1985) Stock list for the eceriferum mutants VII. Barley Genetic Newsletter 15: 89–93.
- 12. Schnable PS, Stinard PS, Wen TJ, Heinen S, Weber D et al. (1994) The genetics of cuticular wax biosynthesis. Maydica 39: 279–287.
- 13. Moose SP, Sisco PH (1994) Glossy15 controls the epidermal juvenile-to-adult phase transition in maize. Plant Cell 6: 1343–1355. doi:10.2307/3869973. PubMed: 12244224.
- 14. Tacke E, Korfhage C, Michel D, Maddaloni M, Motto M et al. (1995) Transposon tagging of the maize Glossy2 locus with the transposable element En/Spm. Plant J 8: 907–917. doi:10.1046/j.1365-313X.1995.8060907.x. PubMed: 8580961.
- 15. Hansen JD, Pyee J, Xia Y, Wen T-J, Robertson DS et al. (1997) The glossy1 locus of maize and an epidermis-specific cDNA from Klenia odora define a class of receptor-like proteins required for the normal accumulation of cuticular waxes. Plant Physiol 113: 1091–1100. doi:10.1104/pp.113.4.1091. PubMed: 9112770.
- 16. Xu X, Dietrich CR, Delledonne M, Xia Y, Wen T-J et al. (1997) Sequence analysis of the cloned glossy8 gene of maize suggests that it may code for a β-ketoacyl reductase required for the biosynthesis of cuticular waxes. Plant Physiol 115: 501–510. doi:10.1104/pp.115.2.501. PubMed: 9342868.
- 17. Liu S, Yeh CT, Tang HM, Nettleton D, Schnable PS (2012) Gene mapping via bulked segregant RNA-Seq (BSR-Seq). PLOS ONE 7: e36406. doi:10.1371/journal.pone.0036406. PubMed: 22586469.
- 18. Liu S, Dietrich CR, Schnable PS (2009) DLA-based strategies for cloning insertion mutants: cloning the gl4 locus of maize using Mu transposon tagged alleles. Genetics 183: 1215–1225. doi:10.1534/genetics.109.108936. PubMed: 19805815.
- 19. Moose SP, Sisco PH (1996) Glossy15, an APETALA2-like gene from maize that regulates leaf epidermal cell identity. Genes & development 10: 3018–3027.
- 20. Amano E, Smith HH (1965) Mutations induced by ethyl methanesulfonate in maize. Mutat Res 2: 344–351. doi:10.1016/0027-5107(65)90070-9. PubMed: 5878310.
- 21. Lisch D (2002) Mutator transposons. Trends Plant Sci 7: 498–504. doi:10.1016/S1360-1385(02)02347-6. PubMed: 12417150.
- 22. Bullard JH, Purdom E, Hansen KD, Dudoit S (2010) Evaluation of statistical methods for normalization and differential expression in mRNA-Seq experiments. BMC Bioinformatics 11: 94. doi:10.1186/1471-2105-11-94. PubMed: 20167110.
- 23. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B Statistical Methodology): 289–300.
- 24. Wu TD, Nacu S (2010) Fast and SNP-tolerant detection of complex variants and splicing in short reads. Bioinformatics 26: 873–881. doi:10.1093/bioinformatics/btq057. PubMed: 20147302.
- 25. Liu S, Yeh CT, Ji T, Ying K, Wu H et al. (2009) Mu transposon insertion sites and meiotic recombination events co-localize with epigenetic marks for open chromatin across the maize genome. PLOS Genetics 5: e1000733.
- 26. Michelmore RW, Paran I, Kesseli RV (1991) Identification of markers linked to disease-resistance genes by bulked segregant analysis: a rapid method to detect markers in specific genomic regions by using segregating populations. Proceedings of the National Academy of Sciences of the USA 88: 9828–9832. doi:10.1073/pnas.88.21.9828. PubMed: 1682921.
- 27. Schnable PS, Ware D, Fulton RS, Stein JC, Wei F et al. (2009) The B73 maize genome: complexity, diversity, and dynamics. Science 326: 1112–1115. doi:10.1126/science.1178534. PubMed: 19965430.
- 28. Dietrich CR, Cui F, Packila ML, Li J, Ashlock DA et al. (2002) Maize Mu transposons are targeted to the 5′ untranslated region of the gl8 gene and sequences flanking Mu target-site duplications exhibit nonrandom nucleotide composition throughout the genome. Genetics 160: 697–716. PubMed: 11861572.
- 29. Greene EA, Codomo CA, Taylor NE, Henikoff JG, Till BJ et al. (2003) Spectrum of chemically induced mutations from a large-scale reverse-genetic screen in Arabidopsis. Genetics 164: 731–740. PubMed: 12807792.
- 30. Paschold A, Jia Y, Marcon C, Lund S, Larson NB et al. (2012) Complementation contributes to transcriptome complexity in maize (Zea mays L.) hybrids relative to their inbred parents. Genome Res 22: 2445–2454. doi:10.1101/gr.138461.112. PubMed: 23086286.
- 31. Li X, Zhu C, Yeh CT, Wu W, Takacs EM et al. (2012) Genic and non-genic contributions to natural variation of quantitative traits in maize. Genome Res 22: 2436–2444. doi:10.1101/gr.140277.112. PubMed: 22701078.
- 32. McMullen MD, Kresovich S, Villeda HS, Bradbury P, Li H et al. (2009) Genetic properties of the maize nested association mapping population. Science 325: 737–740. doi:10.1126/science.1174320. PubMed: 19661427.
- 33. Bessire M, Borel S, Fabre G, Carraça L, Efremova N, et al. (2011) A member of the PLEIOTROPIC DRUG RESISTANCE family of ATP binding cassette transporters is required for the formation of a functional cuticle in Arabidopsis. The Plant Cell Online 23: 1958–1970.
- 34. Chen G, Komatsuda T, Ma JF, Li C, Yamaji N et al. (2011) A functional cutin matrix is required for plant protection against water loss. Plant Signaling and Behavior 6: 1297–1299. doi:10.4161/psb.6.9.17507.
- 35. Chen G, Komatsuda T, Ma JF, Nawrath C, Pourkheirandish M et al. (2011) An ATP-binding cassette subfamily G full transporter is essential for the retention of leaf water in both wild barley and rice. Proceedings of the National Academy of Sciences of the USA 108: 12354–12359. doi:10.1073/pnas.1108444108.
- 36. Davidson RM, Hansey CN, Gowda M, Childs KL, Lin H et al. (2011) Utility of RNA sequencing for analysis of maize reproductive transcriptomes. Plant Genome 4: 191–203. doi:10.3835/plantgenome2011.05.0015.
- 37. Li P, Ponnala L, Gandotra N, Wang L, Si Y et al. (2010) The developmental dynamics of the maize leaf transcriptome. Nat Genet 42: 1060–1067. doi:10.1038/ng.703. PubMed: 21037569.
- 38. Wang X, Elling AA, Li X, Li N, Peng Z et al. (2009) Genome-wide and organ-specific landscapes of epigenetic modifications and their relationships to mRNA and small RNA transcriptomes in maize. Plant Cell 21: 1053–1069. doi:10.1105/tpc.109.065714. PubMed: 19376930.
- 39. Bolduc C, Larose M, Lafond N, Yoshioka M, Rodrigue MA et al. (2012) Adipose tissue transcriptome by serial analysis of gene expression. Obes Res 12: 750–757. PubMed: 15166294.
- 40. Li L, Petsch K, Shimizu R, Liu S, Xu WW et al. (2013) Mendelian and Non-Mendelian Regulation of Gene Expression in Maize. PLoS Genet 9: e1003202. PubMed: 23341782.
- 41. Lee SB, Suh MC (2013) Recent advances in cuticular wax biosynthesis and its regulation in Arabidopsis. Mol Plant 6: 246–249. doi:10.1093/mp/sss159. PubMed: 23253604.
- 42. Alleman M, Freeling M (1986) The Mu transposable elements of maize: evidence for transposition and copy number regulation during development. Genetics 112: 107–119. PubMed: 3002907.
- 43. Benito MI, Walbot V (1997) Characterization of the maize Mutator transposable element MURA transposase as a DNA-binding protein. Mol Cell Biol 17: 5165–5175. PubMed: 9271394.
- 44. Fedoroff NV, Furtek DB, Nelson OE (1984) Cloning of the bronze locus in maize by a simple and generalizable procedure using the transposable controlling element Activator (Ac). Proc Natl Acad Sci U S A 81: 3825–3829. doi:10.1073/pnas.81.12.3825. PubMed: 16593478.
- 45. Ballinger DG, Benzer S (1989) Targeted gene mutations in Drosophila. Proc Natl Acad Sci U S A 86: 9402–9406. doi:10.1073/pnas.86.23.9402. PubMed: 2556711.
- 46. Fadool JM, Hartl DL, Dowling JE (1998) Transposition of the mariner element from Drosophila mauritiana in zebrafish. Proc Natl Acad Sci U S A 95: 5182–5186. doi:10.1073/pnas.95.9.5182. PubMed: 9560250.
- 47. Martin C, MacKay S, Carpenter R (1988) Large-Scale Chromosomal Restructuring Is Induced by the Transposable Element Tam3 at the Nivea Locus of Antirrhinum Majus. Genetics 119: 171–184. PubMed: 17246424.
- 48. Balciunas D, Ekker SC (2005) Trapping fish genes with transposons. Zebrafish 1: 335–341. doi:10.1089/zeb.2005.1.335. PubMed: 18248211.
- 49. Shyamala V, Ames GFL (1989) Genome walking by single-specific-primer polymerase chain reaction: SSP-PCR. Gene 84: 1–8. doi:10.1016/0378-1119(89)90132-7. PubMed: 2691331.
- 50. Kilstrup M, Kristiansen KN (2000) Rapid genome walking: a simplified oligo-cassette mediated polymerase chain reaction using a single genome-specific primer. Nucleic Acids Res 28: e55–e55. doi:10.1093/nar/28.11.e55. PubMed: 10871354.
- 51. Edwards D, Coghill J, Batley J, Holdsworth M, Edwards KJ (2002) Amplification and detection of transposon insertion flanking sequences using fluorescent MuAFLP. BioTechniques 32: 1090–1097. PubMed: 12019782.
- 52. O'Malley RC, Alonso JM, Kim CJ, Leisse TJ, Ecker JR (2007) An adapter ligation-mediated PCR method for high-throughput mapping of T-DNA inserts in the Arabidopsis genome. Nat Protoc 2: 2910–2917. doi:10.1038/nprot.2007.425. PubMed: 18007627.
- 53. Vandenbussche M, Janssen A, Zethof J, Van Orsouw N, Peters J et al. (2008) Generation of a 3D indexed Petunia insertion database for reverse genetics. Plant J 54: 1105–1114. doi:10.1111/j.1365-313X.2008.03482.x. PubMed: 18346192.
- 54. Uren AG, Mikkers H, Kool J, van der Weyden L, Lund AH et al. (2009) A high-throughput splinkerette-PCR method for the isolation and sequencing of retroviral insertion sites. Nat Protoc 4: 789–798. doi:10.1038/nprot.2009.64. PubMed: 19528954.
- 55. Robertson DS (1978) Characterization of a Mutator system in maize. Mutation Research 51: 21–28. doi:10.1016/0027-5107(78)90004-0.
- 56. Till BJ, Reynolds SH, Weil C, Springer N, Burtner C et al. (2004) Discovery of induced point mutations in maize genes by TILLING. BMC Plant Biol 4: 12. doi:10.1186/1471-2229-4-12. PubMed: 15282033.
- 57. Frischmeyer PA, Dietz HC (1999) Nonsense-mediated mRNA decay in health and disease. Hum Mol Genet 8: 1893–1900. doi:10.1093/hmg/8.10.1893. PubMed: 10469842.
- 58. Nicholson P, Yepiskoposyan H, Metze S, Orozco RZ, Kleinschmidt N et al. (2010) Nonsense-mediated mRNA decay in human cells: mechanistic insights, functions beyond quality control and the double-life of NMD factors. Cellular and Molecular Life Sciences 67: 677–700. doi:10.1007/s00018-009-0177-1. PubMed: 19859661.
- 59. Brogna S, Wen J (2009) Nonsense-mediated mRNA decay (NMD) mechanisms. Nat Struct Mol Biol 16: 107–113. doi:10.1038/nsmb.1550. PubMed: 19190664.
- 60. Imam JS, Gudikote JP, Chan WK, Wilkinson MF (2010) Frame-disrupting mutations elicit pre-mRNA accumulation independently of frame disruption. Nucleic Acids Res 38: 1559–1574. doi:10.1093/nar/gkp1115. PubMed: 20007599.
- 61. David CJ, Manley JL (2010) Alternative pre-mRNA splicing regulation in cancer: pathways and programs unhinged. Genes Dev 24: 2343–2364. doi:10.1101/gad.1973010. PubMed: 21041405.
- 62. Mendell JT, Sharifi NA, Meyers JL, Martinez-Murillo F, Dietz HC (2004) Nonsense surveillance regulates expression of diverse classes of mammalian transcripts and mutes genomic noise. Nat Genet 36: 1073–1078. doi:10.1038/ng1429. PubMed: 15448691.
- 63. Nagy E, Maquat LE (1998) A rule for termination-codon position within intron-containing genes: when nonsense affects RNA abundance. Trends Biochem Sci 23: 198–199. doi:10.1016/S0968-0004(98)01208-0. PubMed: 9644970.
- 64. Lewis BP, Green RE, Brenner SE (2003) Evidence for the widespread coupling of alternative splicing and nonsense-mediated mRNA decay in humans. Proc Natl Acad Sci U S A 100: 189–192. doi:10.1073/pnas.0136770100. PubMed: 12502788.
- 65. Rayson S, Arciga-Reyes L, Wootton L, De Torres Zabala M, Truman W et al. (2012) A role for nonsense-mediated mRNA decay in plants: pathogen responses are induced in Arabidopsis thaliana NMD mutants. PLOS ONE 7: e31917. doi:10.1371/journal.pone.0031917. PubMed: 22384098.
- 66. Banasiak J, Biala W, Staszków A, Swarcewicz B, Kępczyńska E et al. (2013) A Medicago truncatula ABC transporter belonging to subfamily G modulates the level of isoflavonoids. J Exp Bot 64: 1005–1015. doi:10.1093/jxb/ers380. PubMed: 23314816.
- 67. Chen G, Komatsuda T, Ma JF, Li C, Yamaji N et al. (2011) A functional cutin matrix is required for plant protection against water loss. Plant Signaling and Behavior 6: 1297–1299. doi:10.4161/psb.6.9.17507.