SEARCH

SEARCH BY CITATION

Keywords:

  • tomato;
  • L1 layer;
  • periclinal chimera;
  • gene expression;
  • layer-specific;
  • technical advance

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Plant organs are made from multiple cell types, and defining the expression level of a gene in any one cell or group of cells from a complex mixture is difficult. Dicotyledonous plants normally have three distinct layers of cells, L1, L2 and L3. Layer L1 is the single layer of cells making up the epidermis, layer L2 the single cell sub-epidermal layer and layer L3 constitutes the rest of the internal cells. Here we show how it is possible to harvest an organ and characterise the level of layer-specific expression by using a periclinal chimera that has its L1 layer from Solanum pennellii and its L2 and L3 layers from Solanum lycopersicum. This is possible by measuring the level of the frequency of species-specific transcripts. RNA-seq analysis enabled the genome-wide assessment of whether a gene is expressed in the L1 or L2/L3 layers. From 13 277 genes that are expressed in both the chimera and the parental lines and with at least one polymorphism between the parental alleles, we identified 382 genes that are preferentially expressed in L1 in contrast to 1159 genes in L2/L3. Gene ontology analysis shows that many genes preferentially expressed in L1 are involved in cutin and wax biosynthesis, whereas numerous genes that are preferentially expressed in L2/L3 tissue are associated with chloroplastic processes. These data indicate the use of such chimeras and provide detailed information on the level of layer-specific expression of genes.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Ever since Hooke's seminal observations that organisms are made from cells a fundamental question has existed as to what defines the components that make each cell unique. Plants are made from numerous different cell types that have different shapes, sizes and function. Each unique cell type has its own complement of mRNA and protein to perform its specialised function. Being able to define the components that make each cell unique is of fundamental importance for gaining a better understanding of all processes occurring in plants from vegetative growth to flowering, and from resistance to insect and pathogen attack to plant mineral nutrition. Many approaches have been used to dissect this complexity in which the differences in the RNA complement (transcriptome) and protein levels (proteome) are assessed between different tissue samples. The samples isolated for analysis are obtained from two major sources. The first source is from whole plants, seedlings, calli, etc., or from discrete organs, for example leaf, root, flower, petal, apex, etc., and such samples contain numerous different cell types (Rakwal and Agrawal, 2003; Ko and Han, 2004; Pischke et al., 2006; Ruffel et al., 2008; Zeller et al., 2009). The second source constitutes samples collected from specific cell types or groups of cells. These cells can be identified either by staining or cell-specific reporter gene expression and subsequent dissection and cell sorting, e.g. guard cells, by the isolation of specific cells or groups of cells, for example epidermal peels, or by the mechanical removal of trichomes (Birnbaum et al., 2005; Day et al., 2005; Bargmann and Birnbaum, 2010; Matas et al., 2010, 2011; Hu et al., 2011). In both scenarios the transcriptome can then be analysed using either arrays or high-throughput sequencing technologies directly on the cDNA or via the sequence analysis of PCR products that sample the transcriptome, i.e. SAGE (serial analysis of gene expression) (Bao et al., 2005; Brady et al., 2006).

When mixtures of cell types are used in transcript analysis the level of expression for any one gene is the summation of all the cell types and thus it is not possible to discern whether all cells respond in the same way. For example, it is conceivable that within the same organ certain genes may be up-regulated in one cell type and down-regulated in another. Layer-specific regulation of such genes will, therefore, not be observed. Similarly, when specific cell types are isolated usually only one type is taken for analysis and the remainder discarded, which therefore limits the holistic interpretation of results. In addition, the isolation of specific cell types by mechanical means, for example laser dissection, protoplasting and cell sorting, may lead to the generation of artefacts through the extraction process. The best scenario is to tag each cell type within an individual organ such that it can be discriminated from other cell types but yet carry out the analysis on the whole organ. Here we show that this is possible using a periclinal chimera.

Plants that have one layer of cells that is genetically distinct from another layer are called periclinal chimeras. In dicotyledonous plants there are three distinct cell layers, L1, L2 and L3 (Figure 1). The outermost layer, L1, is made up of epidermal cells, stomata and hairs and L2 comprises the sub-epidermal mesophyll cells (in a leaf). Layers L1 and L2 form the tunica in which cell division is normally anticlinal, i.e. divisions that are at right angles to the surface of the growing point and thus do not add new rows of cells. The innermost L3 cells are referred to as the corpus and they can divide both anticlinally and periclinally (periclinal cell divisions are those that are parallel to the surface) (Satina et al., 1940; Sussex, 1989; Ingram, 2004). A periclinal chimera may be between plants of the same species or between plants of different species (Marcotrigiano, 1986; Marcotrigiano and Bernetzky, 1995).

image

Figure 1. Schematic diagrams indicating the cell layers in an apical meristem.

Three different meristems are shown, two parental lines and a periclinal line in which the outermost L1 cell layer is derived from Solanum pennellii whereas the internal tissue is from Solanum lycopersicum.

Download figure to PowerPoint

Here we show the characterisation of layer-specific gene expression in a mixture of tomato cell types utilising high-throughput sequencing, detection of single nucleotide polymorphisms and quantification of parental-origin allele-specific expression. This was possible by generating a periclinal chimera that has the L1 epidermal layer of a wild species and the L2 and L3 layers from cultivated tomato. RNA-seq analysis and the availability of the tomato genome sequence and annotation enabled the genome-wide assessment of whether a gene is expressed in the L1 or L2/L3 layers.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Generation of the epidermal L1 chimera

To generate the periclinal chimera about 120 ‘V’ grafts were made between the tomato lines Heinz 1706 (Solanum lycopersicum) and LA716 (Solanum pennellii). Graft unions were cut and allowed to callus and generate adventitious shoots. From these shoots one was observed to be a potential periclinal chimera that has the L1 layer of S. pennellii and the internal tissue of Heinz 1706. This shoot was maintained as a cutting and called Periclinal 1 (Peri1).

The type of chimera generated was verified by phenotypic observations, crossing experiments and progeny testing. Figure 2 shows the phenotype of tissues from this periclinal line whose leaves have a lighter green colour compared with Heinz 1706 due to the numerous trichomes present on the leaf surface. When touched the leaves have the same oily feel as those of S. pennellii (LA716) due to the presence of glandular trichomes, as previously observed by Goffreda et al. (1990). The chimera also has altered leaf morphology in comparison with the parental types in that the leaflets are smaller and less serrated than the Heinz variety but more elongated than LA716 (Figure 2). The leaflets of the F1 cross between these species, in comparison, are much larger due to heterosis. An additional key observation is that the periclinal chimera does not set fruits by selfing, whereas the Heinz and LA716 lines are self-compatible. In crossing experiments when the chimera was used as the female parent it failed to set seed using pollen from Heinz plants but did set seed when crossed with LA716 pollen (Table 1). Pollen from the chimera was crossed to LA716 plants but no seed was produced; however, seed was set when pollen from the chimera was crossed to Heinz plants. It is known that Spennellii LA716 cannot be fertilised by other tomato species, and that this incongruous behaviour is associated with the L1 layer (Liedl et al., 1996).

Table 1. Compatibility of crosses between parental lines and layer 1 periclinal chimera 1
 Male
Heinz 1706LA716Periclinal 1
FemaleHeinz 1706YesYesYes
LA716NoYesNo
Periclinal 1NoYesNo
image

Figure 2. Morphology of L1 periclinal chimera.

(a) Flowers at anthesis.

(b) Leaves, eighth leaf from the apex.

(c) Mature green fruits. From left to right, Solanum lycopersicum (Heinz 1706), Solanum pennellii (LA716), F1 [S. lycopersicum (Heinz 1706) × S. pennellii (LA716)] and Periclinal 1.

Download figure to PowerPoint

These data therefore confirm the fact that the L1 surface of the stigma in the chimera is of S. pennellii that will only allow successful fertilisation by S. pennellii pollen. However, as pollen is generated from the L2 layer (Huala and Sussex, 1993), the pollen from the chimera is S. lycopersicum and it therefore cannot self-fertilise the S. pennellii-‘covered’ styles of the chimera. The pollen can, however, be used to fertilise the Heinz plants.

The size and weight of seeds from the parental lines and the crosses generated confirm these observations (Figure S1). The F1 seeds from a cross between Slycopersicum and Spennellii have a size and weight intermediate between those of the parental lines. While the size and weight of seeds from crosses between Slycopersicum and the chimera are very similar to Heinz seeds, the seeds from Peri1 and Spennellii crosses are similar or the same as those of F1 seeds (Figure S1). Progeny testing of these lines also confirmed these results; all seedlings from the Peri1 × LA716 cross had the F1 phenotype between these two species and all progeny from Heinz 1706 × Peri1 had a Heinz 1706 phenotype (Table 2).

Table 2. Phenotypes of seedlings from progeny test crosses
 Male
Heinz 1706LA716Periclinal 1
FemaleHeinz 1706

100% Heinz

= 12

100% F1

= 12

100% Heinz

= 12

LA716No seed

100% LA716

= 12

No seed
Periclinal 1No seed

100% F1

= 12

No seed

Polymorphism detection

To confirm that the periclinal line was a mixture of both species' genomes, genomic DNA was extracted from the chimera and the parental cultivars. Eleven genes were amplified by PCR (see Table 3) and sequence analysis of the PCR products revealed 1.3% sequence polymorphisms in coding regions [135 single nucleotide polymorphisms (SNPs)/10317 bp] and about 2% in non-coding regions (65 SNPs/3140 bp). Assuming that both alleles amplify equally, the relative frequency of each base at each polymorphic site will give an estimate of the proportion of template DNA from each species. On average the level of the S. pennellii allele was about 20% of the total, which provides an approximate level of S. pennellii tissue in the sample, i.e. the L1 tissue was about one-fifth of the sample.

Table 3. Analysis of layer 1 (L1) specificity of genes involved in processes that may exhibit layer-specific expression
GeneLocus nameSolyc IDArabidopsis thaliana protein homologue% L1 specificityReference
SlML1Meristem layer 1Solyc10g005330AAB49378.1100 ± 1Lu et al. (1996)
SlPDF1Protodermal factor 1Solyc07g055950AAD3386998 ± 3Abe et al. (1999)
SlSYS ProsysteminSolyc05g051750na8 ± 2Jacinto et al. (1997)
SlzFPS Z-isoprenyl pyrophosphate synthaseSolyc08g005680na98 ± 4Sallaud et al. (2009)
SlSBS Santalene and bergamotene synthaseSolyc08g005640NM_106594.391 ± 8Sallaud et al. (2009)
SlBRI1 Brassinosteroid insensitive 1Solyc04g051510NP_19565053 ± 5Montoya et al. (2002)
SlBIN2 Brassinosteroid-insensitive 2Solyc02g072300NP_19360656 ± 6Li and Nam (2002)
SlBAK1 Brassinosteroid insensitive1-associated receptor kinase 1Solyc01g104970NP_56792058 ± 4Li et al. (2002)
SlCYP85A1 Dwarf Solyc02g089160NP_97486231 ± 5Bishop et al. (1996, 1999)
SlTMM Too many mouthsSolyc12g042760NP_17812545 ± 4Yang and Sack (1995)
SlStomagen StomagenSolyc08g066610NM_117366.314 ± 4Sugano et al. (2010)

Layer L1 specificity

To show that it is possible to discriminate the relative layer-specific expression of a gene from a leaf sample containing a mixture of cell types, analysis of genes known to have layer-specific expression was carried out. First total RNA was extracted from leaf tissue from S. lycopersicum (lyc), S. pennellii (penn) and Peri1. These RNAs were used in RT-PCR experiments with PCR primers designed to amplify the tomato homologues of genes known to be L1 specific in Arabidopsis, namely Meristem Layer 1 (ML1) (Lu et al., 1996) and Protodermal Factor 1 (PDF1) (Abe et al., 1999). As shown in Figure 3, sequence analysis of the PCR products of the ML1 and PDF1 homologues shows that the products from the chimera have identical sequence to the penn sequences. This is predicted as the L1 layer in the chimera is S. pennellii and therefore any genes specifically expressed in L1 will only have the penn sequence. This analysis was extended further to the tomato prosystemin gene that is known to be expressed in layer L3 (Jacinto et al., 1997). RT-PCR and sequence analysis of the PCR product from the Peri1 chimera showed that the prosystemin sequence was identical to the S. lycopersicum allele, confirming L3 specificity. This analysis was also performed on products of various genes involved in different processes, and the percentage expression in layer L1 was calculated after normalising to the amount of L1 tissue (Table 3).

image

Figure 3. Sanger sequencing detecting polymorphisms and layer-specific expression.

(a) Sanger sequencing traces of cDNA sequences of ML1, PDF1 and SYS tomato genes using template cDNA from parental (Sl = Solanum lycopersicum, Sp = Solanum pennellii) and Peri1 lines.

(b) Mean and SE percentage of expression in layer L1 after normalization to the amount of L1 tissue as indicated by the frequency of the pennellii allele (SlML1 n = 6; SlPDF1 n = 3; SlSYS n = 6). (Note: L1-specific genes ML1 and PDF1 have identical traces to S. pennellii whereas the L3-specific gene has an identical trace file to S. lycopersicum).

Download figure to PowerPoint

Genome-wide analysis of layer-specific gene expression

To gain a global understanding of layer-specific gene expression we sequenced transcripts from three major tissue samples from the parental lines and the chimera, namely leaves, leaves undergoing water stress that induced increased ABA levels (Table S1) and fruit. Water stress was used as S. pennellii is more drought tolerant than S. lycopersicum (Gong et al., 2010) (Figure S8) and water stress induces gene expression (Cameron et al., 2006). Such water stress samples would therefore enhance our ability to discriminate for layer-specific expression. Sequencing was carried out using the Applied Biosystems SOLiD3 platform, which generated around 162 million reads of 50 base pairs (bp) length for all nine samples. Best alignments of about 61.6 million reads to known coding sequencing regions in the S. lycopersicum genome yielded a total transcript coverage of 32×, 16×, 29× for the lyc, penn and peri1 libraries, respectively (Table S2).

Out of about 35 000 genes annotated in the tomato genome (TGC, 2012), 21 938 were found to be expressed in both parental lines and the chimera independent of tissue sample (Table S4). Mapped reads across tissue samples were pooled for the parental lines to identify polymorphisms between lyc and penn alleles in the coding sequences of the 22 000 expressed genes. Detection of SNPs by VARiD (Dalca et al., 2010) and FreeBayes (Garrison and Marth, 2012) commonly identified 13 277 genes with at least one SNP with a minimum 4 × coverage in both parental lines (Table S3). All methods commonly detected 63 669 SNPs, which correspond to about 0.5% of all examined coding sequence bases with at least 4 × depth in both lyc and penn pooled samples (Table S3). Results of the comparison of the three variant detection methods are presented in Figure S2.

Expression levels of genes were assessed separately for each tissue sample in all three plant types at identified SNP positions and with respect to the parental origin in the chimera. Our analysis has the structure of a 2 × 2 factorial experiment with one factor being the parental-origin (lyc or penn) allele-specific expression. The second factor relates to the genome examined, chimeric (peri) or parental wild type (lyc or penn). The lyc allele-specific expression in the parental genome (Lw) and in the chimera (Lc), as well as the pennellii ones (Pw and Pc) were quantified (Table S4). Ideally, Lc must be 0 for a gene that is expressed only in L1. Decreased or increased expression of genes due to the generation of the chimera was accounted for.

Differential expression analysis to classify genes as L1 or L2/L3 was conducted using a negative binomial generalized linear model (Robinson et al., 2010). Genes that exhibit differential expression in the chimera were identified by treatment-contrast parameterization (Smyth, 2005). L1 genes are those that are down-regulated in the chimera compared with the wild-type genome with respect to their lyc origin (Figure 4b) and not with respect to their penn origin (Figure 4a), and for which a statistically significant interaction effect is observed (Figure 4c). Those L1 genes which show upregulation in the chimera with respect to their penn allele-specific expression and in comparison with their lyc one (Figure 4d), are denoted as L1 specific, otherwise as L1 related. In a similar way, genes with a (highly) biased lyc allele-specific expression in the chimera are classified as L2/L3. After correcting for multiple testing (Benjamini and Hochberg, 1995; Benjamini and Yekutieli, 2001) and applying a cut-off value of 0.05 for the false discovery rate (FDR), 382 genes were classified as L1 and 1159 as L2/L3, with 254 of the 382 genes being called L1 specific and 811 of the 1159 being L2/L3 specific. The lists of classified genes are provided in Tables S5 and S6, while comparison of the classification resulting from the three variant detection programmes are shown in Figure S3.

image

Figure 4. Differential expression analysis on parental-origin allele-specific expression identifies layer-specific genes.

The relative abundance of each gene plotted versus the P-value, for all tissues and for the four contrasts examined.

(a) Difference of lyc allele-specific expression between wild type (Lw) and chimera (Lc).

(b) Difference of penn allele-specific expression between wild type (Pw) and chimera (Pc).

(c) Interaction effect, i.e. difference of the differences.

(d) Difference of penn and lyc allele-specific expressions in chimera. Grey circles, all genes examined; purple and blue circles, L1-specific and L1-related genes, respectively; brown and red circles, L2-specific and L2-related genes, respectively. Squares denote genes that are classified based on more than one tissue sample. Expression and the resulting fold changes and P-values are calculated based on polymorphisms detected by VARiD.

Download figure to PowerPoint

The tomato homologue of Arabidopsis Meristem Layer 1, a gene known to be expressed specifically in L1 (Lu et al., 1996), was identified in our data (Table S5). Prosystemin, a gene known to be expressed in L3 (Jacinto et al., 1997), was found in our results to be L2/L3 specific (Table S6). To validate our classifications, we randomly selected 14 predicted L1 genes, three predicted L2/L3 genes, as well as the two predicted known layer-specific genes. The parental-origin allele-specific expression for selected genes is plotted in Figure S5. Layer 1 genes show down-regulation of lyc allele expression in the chimera and L2/L3 genes show down-regulation of penn allele expression. Analysis of the Sanger trace files confirmed all our classifications (Tables 3 and S7) with the exception of one ambiguous case – a receptor-like kinase involved in brassinosteroid signalling was found to be expressed evenly between L1 and L2/L3 layers, while genome-wide analysis classified it as L2/L3 specific.

Gene Ontology (GO) enrichment analysis for genes classified as L1 (Figures 5a and S6) and L2/L3 (Figures 5B and S7) identified the function of genes and biological processes that are layer specific. Genes in L1, as opposed to L2/L3 ones, are involved in lipid and wax synthesis, and in cellulose synthesis in cell wall production. Five of the top 10 genes that are L1 specific are associated with lipid synthesis (Table S5). This fits with the epidermis producing cutin and wax (Javelle et al., 2011) that act as a barrier to prevent water loss. Other L1 genes identified are involved in cell wall synthesis and are likely to influence organ expansion (Savaldi-Goldstein et al., 2007). Genes preferentially expressed in L2/L3 include those associated with photosynthesis and the chloroplast. This is not unexpected, as the epidermal pavement cells lack chloroplasts and any nuclear gene whose protein is targeted to the chloroplast is expected to be expressed in the L2/L3 tissues rather than the L1.

image

Figure 5. Ranked list of the top 10 over-represented Gene Ontology (GO) terms in layer L1 (a) and layers L2 and L3 (b) identified genes.

Terms are ranked based on the marginal posterior probability calculated using the MGSA algorithm as implemented in Ontologizer (Bauer et al., 2008, 2010). Posterior probability is shown in yellow. GO terms having a posterior >0.5 are considered to be active, i.e. statistically significantly over-represented. Terms with posterior bars with shadowing lines are active when considering only the genes characterized as L1 specific (or L2/L3 specific) and thus excluding the layer-related ones. Error bars (95% confidence intervals) are obtained with 20 runs. The GO fraction is the fraction of all genes with the specific GO-term that is L1 or L2/L3 (coloured blue). The set fraction is the fraction of all L1 or L2/L3 genes assigned with the specific GO-term (coloured red). The numbers of L1 and L2/L3 genes assigned with a GO-term are denoted in the right y-axis.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Peri1, an L1 chimera

Periclinal chimeras have previously been generated in tomato and used to show how cell layers contribute to different processes (Heichel and Anagostakis, 1978; Goffreda et al., 1990; Liedl et al., 1996). One key observation from this earlier work was that layer L2 contributes to the gametes and that layer L1 contributes to the compatibility of viable crosses between S. pennellii and S. lycopersicum (Liedl et al., 1996). These preceding observations were used to confirm the structure of the Peri1 chimera; namely crossing experiments and progeny testing showed that Peri1 is an L1 chimera similar to that generated previously by others (Goffreda et al., 1990). This is an important observation as Peri1 was made using Heinz 1706, a wild-type variety that has been utilised in the tomato genome sequencing project (TGC, 2012) and does not harbour any visible mutations to confirm the genotype of the different layers.

Sequence analysis of PCR products generated using genomic DNA from Peri1 as the template indicated polymorphic sequences. These polymorphisms were confirmed when comparing the sequence of PCR products from the Heinz 1706 and LA716 parental lines. The level of polymorphism of 1.3% in the 13 457 bp analysed is similar to that detected by others (Yamamoto et al., 2005; Kamenetzky et al., 2010). The relative proportion of the species-specific alleles provides an estimate of the amount of pennellii/lycopersicum template DNA and thus a rough estimate of the proportion of pennellii/lycopersicum cells that contribute to the layers in the chimera. For the leaf sample used in our experiments the amount of S. pennellii was approximately 20% of the tissue.

Layer-specific expression

To show that Peri1 can be used to discriminate layer-specific expression, the tomato homologues of the Arabidopsis ML1 and PDF1 genes known to be L1 specific (Lu et al., 1996; Abe et al., 1999) were identified. When the RT-PCR products of SLML1 and SLPDF1 were sequenced from Peri1 template cDNA around 100% of the cDNA sequence had the S. pennellii alleles. This highlights that a conserved mechanism exists between Arabidopsis and tomato in which L1-specific transcription takes place. The opposite scenario was also tested. Prosystemin promoter GUS fusion lines exhibit GUS expression in vasculature (Jacinto et al., 1997) and the resulting cDNA sequencing of prosystemin from Peri1 only identified the S. lycopersicum allele. These data further corroborated the results that Peri1 was an L1 chimera and that sequencing cDNA from this plant can give an indication of the amount of expression of a gene in the L1 or L2/L3 layers.

Genes are classified as L1 or L2/L3 based on multiple tissues. All libraries generated from dehydrated tissues exhibit higher read coverage within each plant type, while across plant types dehydrated libraries have most similar coverage to each other (Table S2). Examining the tissue support (Figure S4), we observe that 70% of gene classifications are verified by analysis of dehydrated leaf samples, while 48% are solely based on these. This indicates that dehydrated samples are highly informative and help better identify layer-specific genes. In addition our conservative method of analysis minimises any false predictions at the expense of coverage of the layer-specific transcriptome. In the future, deeper sequencing using a range of technologies will enable a more detailed analysis of the level of layer-specific expression in various tissues under varying conditions.

The methodology and the data we have generated provide numerous leads for genes that are preferentially expressed in the L1 and L2/L3 layers of plants (Tables S5 and S6) and these data corroborate those generated by laser dissection and sequencing of the fruit epidermis (Matas et al., 2011). It is not surprising to see that many of the L1-specific genes being expressed are associated with cutin and wax formation. This includes Solyc11g006250 that encodes a GDSL1 lipase, so called due it belonging to the family of Gly-Asp-Ser-Leu esterases/acylhydrolases. The cutin deficient 1 (cd1) mutant is defective in this gene and cd1 tomato fruits show increased sensitivity to water loss as only 5–10% of the amount cutin as in the wild type is present (Yeats et al., 2012). When this gene is silenced it leads to a reduction of cuticle thickness and the number of cutin monomers present per unit area (Girard et al., 2012). Interestingly, immunogold labelling of this lipase shows that it is embedded in the cuticle (Girard et al., 2012; Yeats et al., 2012) and thus correlates with the observed L1 expression. Our analysis also revealed L1 expression of a further seven members of the GDSL1 lipases (Table S5). Other genes related to cutin or wax formation that were shown to be preferentially expressed in L1 tissue include the cytochrome P450 enzymes CYP86A69, CYP77A19 and CYP77A20. Mutants of CYP86A69 are defective in cutin biosynthesis (Isaacson et al., 2009; Shi et al., 2013) and this P450 encodes an end chain fatty acid hydroxylase (Shi et al., 2013).

Tomato homologues of the different types of eceriferum (cer) mutants of Arabidopsis are also found to be expressed in L1. These include homologues of cer1 and cer3 mutants that are defective in epicuticular wax (alkane) formation (Aarts et al., 1995). Further genes identified as being L1 specific and involved in cutin biosynthesis include those encoding glycerol-3-phosphate acyltransferase activity (Yang et al., 2012) and fatty acid elongase 3-ketoacyl-CoA synthases (Joubes et al., 2008).

Many of the biosynthesis genes involved in cutin/wax formation were found to be L1 specific; however, the SLShine3 transcription factor that regulates cuticle formation in tomato (Shi et al., 2013) was not identified as L1-specific in our analysis. The lack of identification of SLShine3 is a consequence of a lack of sequence coverage preventing SNP detection. If a SNP had been observed, the number of sequence reads corresponding to this gene obtained from the Periclinal libraries was minimal, thus preventing us from assigning the gene as L1-specific. It is also conceivable that there will be genes for which sequence coverage is good but no SNP exists between lycopersicum and pennellii, making the detection of layer-specific expression impossible. These observations highlight the need for deeper transcriptome sequencing of peri1 tissues. In addition, the future availability of the S. pennellii genome sequence will provide detailed and comprehensive SNP data between it and S. lycopersicum and thus greatly increase the power of layer-specific gene analysis.

The identification of the L1-expressed sequences provided the opportunity to determine if an L1-box motif was present in the 5′ region of these genes. In Arabidopsis an L1-box motif has been observed (TAAATGYA) that regulates the expression of Protodermal Factor 1 in L1 cells (Abe et al., 2001). Analysis of 1000 bp regions upstream of the translation start sites of the genes expressed and with at least one SNP identified by VARiD, showed an enrichment of the L1 box sequence (P-value = 0.005) in L1-related genes, based on Fisher's exact test. The L1 box was present in 18% of the L1 genes as opposed to 12% of L2/L3 genes and 13% of other genes examined. This enrichment is in agreement with our classification and also indicates the possibility of a conserved mechanism for L1-specific expression between Arabidopsis and tomato.

Future uses

The periclinal line that has been developed can be utilised to assess the relative level of L1 or L2 and L3 expression in different tissues and under different conditions. Most notably this has the potential to provide information on layer-specific expression of genes involved in epidermal-specific processes such as trichome and guard cell formation. The level of L1 specificity of genes involved in cutin formation and the production of metabolites involved in insect and pathogen resistance can also be discerned. Such chimeras also have the potential to unravel layer-specific gene expression in meristems, especially in the formation of leaf and flower primordia. Further different types of chimeras can be made to increase the power of analysis. It is conceivable to have a chimera in which the three layers are made from three different species. In practice, however, layer invasion between the L2 and L3 layers occurs at a rate that causes problems in maintaining such chimeras. L1 chimeras are, however, very stable and lack significant layer invasion (Schmulling and Schell, 1993; Jenik and Irish, 2000). Chimeras also need not be restricted to those between tomato species and can include other economically important solanaceous species, for example potato and aubergine. Chimeras can be made for shrubs and trees, for example +LaburnocytisusAdamii’ which is a chimera between a laburnum, Laburnum anagyroides, and a broom, Chamaecytisus purpureus. Grape and other species in which grafting is easily carried out and explants maintained as cuttings can also be used to generate chimeras. Thus this method of analysis can be extended widely.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

Plant growth

Tomato Heinz (S. lycopersicum 1706), S. pennellii LA716 and the periclinal line (Peri1) were grown in a glasshouse supplemented with artificial light for a 16-h/8-h photoperiod at 120 μmol m−2 sec−1 photosynthetically active radiation (PAR) irradiance.

Generation of chimeras

The grafting was carried out essentially as described previously (Montoya et al., 2005). Three rounds of grafting were carried out, with between 30 and 50 grafts being made at each attempt. For the successful generation of a single periclinal chimeric plant about 50 grafts between various plant genotypes were made using plants at the four true leaf stage of development. Graft unions were made and kept in place with appropriate size plastic tubing and either Vaseline or lanolin was used to seal the union. Grafted plants were kept on a misting bench for about 1–2 weeks. Around 90% of grafts were successful. After approximately 1 month the graft junction was cut and the cut tissue was then covered with lanolin paste or Vaseline to prevent desiccation. Callusing tissue and adventitious shoots were then allowed to grow from the graft union. Plants that lacked adventitious shoot formation were discarded. From the 50 lines one striking periclinal chimera was observed and maintained as a cutting.

DNA and RNA isolation

Genomic DNA and RNA were extracted from tissue of the eighth leaf from the apex of S. lycopersicum, Spennellii and Peri1 plants. All samples were frozen immediately in liquid nitrogen, ground and kept at −80°C. For DNA extraction the tissue was resuspended in nuclear extraction buffer [120 mm 2-amino-2-(hydroxymethyl)-1,3-propanediol (TRIS)-HCl, 30 mm EDTA, 1.2 m NaCl, 1.2% cetyl trimethylammonium bromide (CTAB), 40 mm sodium bisulphite, pH 7.5) and 1% sarkosyl. After incubation at 65°C the DNA was extracted twice with phenol:chloroform:isoamyl alcohol (25:24:1). The DNA was recovered by isopropanol precipitation. For total RNA isolation the tissue was homogenized in RNA extraction buffer (8 m guanidinium hydrochloride, 20 mm MES, pH 7) and extracted with phenol:chloroform: isoamyl alcohol. After centrifugation at 11 000 g the aqueous phase was re-extracted with phenol:chloroform:isoamyl alcohol and then subjected to three successive precipitations with ethanol/sodium acetate, 4 m LiCl and then ethanol/sodium acetate again. The final pellet was washed in 70% ethanol and resuspended in diethyl pyrocarbonated treated water. The total RNA concentration was estimated in a NanoDrop 1000 Spectrophotometer (Thermo Scientific, http://www.thermofisher.com) and analysed on a formaldehyde/3-(N-morpholino)propanesulphonic acid (MOPS) gel.

Expression analysis by RT-PCR

The RT-PCR was performed using Moloney Murine Leukaemia Virus (M-MLV) reverse transcriptase (Promega, http://www.promega.com/) according to the manufacturer's instructions. The single-strand DNA was synthesized in a 25-μl reaction mixture containing 2 μg of total RNA denatured by heating at 65°C for 5 min, 1 μg oligo(dT) anchor primer, 10 nm of deoxynucleotide triphosphate and 200 units M-MLV reverse transcriptase (Promega). The RT reactions were performed at 42°C for 45 min and stopped by incubation at 70°C for 15 min. Single-stranded DNA was amplified by PCR in mixtures containing 1 μl cDNA, 2.5 mm MgCl2, 10 mm TRIS–HCl, 50 mm KCl, 0.25 unit of Taq DNA polymerase (New England Biolabs, https://www.neb.com/), 2 mm of each deoxynucleotide triphosphate and 0.5 μm of each gene-specific primer as shown in Table S8.

The PCRs were performed in a GeneAmp PCR system 9700 thermocycler (Applied Biosystems, http://www.lifetechnologies.com) using the following conditions: 94°C for 3 min followed by 35 cycles of 94°C for 20 sec, 55°C for 30 sec, 72°C for 60 sec. The PCR products were analysed on a 1% (w/v) agarose gel containing 0.05 μl ml−1 of SYBR Safe (Invitrogen, http://www.invitrogen.com/).

Sequence analysis and polymorphism detection

The RT-PCR products were purified using QIAquick gel extraction kits (Qiagen, http://www.qiagen.com/) following the manufacturer's instructions. Sanger sequencing of PCR products was carried out at GATC Biotech, http://www.gatc-biotech.com (London, UK). Sequence analysis to identify the polymorphisms (SNPs and indels) between species was carried out by aligning sequences using the ContigExpress package in the Vector NTI sequence analysis software (Invitrogen).

The proportion of S. pennelli (L1 expression) of each polymorphic base in the periclinal line (Peri1) was calculated using the chromatogram's peak height and quality score (Ewing et al., 1998) for each base. The peak height was used to calculate the percentage level of the Spennellii or S. lycopersicum allele at the base. These percentages were used to obtain an arithmetic mean of all the polymorphic sites assessed and this value provides the proportion of L1 expression for an allele in Peri1. We performed a similar analysis using genomic DNA to estimate the proportion of tissue that corresponds to layer L1. Approximately 20% of the tissue in a sample corresponds to layer L1 (S. pennellii) and therefore the expression levels were adjusted to account for this level (see supporting data and example calculation in Methods S1).

RNA-seq

Total RNA was extracted from: (i) eighth leaf from the apex of 8-week-old plants, (ii) eighth leaf with 30% loss of fresh weight (D), (iii) mature green fruits (F) of S. lycopersicum, S. pennelli and Peri1 plants (see Figure 2b,c, for photographs of samples used). PolyA RNA was isolated as described in the Supporting Information. Whole transcriptome libraries were prepared from each sample, following the procedures of the SOLiD Whole Transcriptome Analysis Kit (Life Technologies, http://www.lifetechnologies.com/). Libraries were barcoded and pooled prior to ePCR amplification. One flowcell was loaded and 50-bp fragments were sequenced with a SOLiD version 3 instrument (Life Technologies).

Layer-specific transcriptome analysis

Whole transcriptome reads were aligned against the S. lycopersicum reference genome (TGC, 2012) using Applied Biosystems SOLiD BioScope Whole Transcriptome Analysis Alignment Pipeline (Life Technologies, 2010). Valid reads mapped to coding sequence regions were extracted and pooled across tissue samples for both parental lines to identify polymorphisms between lyc and penn alleles. Bases with coverage of less than four reads in either lyc or penn pooled samples were discarded. Genes were analysed for species-specific SNP expression to determine layer-specific expression. Details of data analysis are provided in Supporting Information.

The transcriptome sequence reads have been deposited with the EBI-SRA under study ERP002648.

A full description of the methods used is provided in Methods S1 in the Supporting Information.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

We thank Mark Bennett for ABA analysis, and Maike Paramor, Yasmeen Ghani and Amanda Jackson for the SOLiD3 sequencing. We also thank Geraint Barton for discussions on differential expression analysis and Michael Brudno's lab (University of Toronto, Canada) and in particular Misko Dzamba for their exceptional support on VARiD usage and implementation of requested features. The Bishop Lab has been funded by the EU and BBSRC, especially BBSRC grant BB/I023941.

Conflict of interest

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information

We, the authors, have no conflict of interest to declare regarding the publication of this paper.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information
  • Aarts, M.G., Keijzer, C.J., Stiekema, W.J. and Pereira, A. (1995) Molecular characterization of the CER1 gene of arabidopsis involved in epicuticular wax biosynthesis and pollen fertility. Plant Cell, 7, 21152127.
  • Abe, M., Takahashi, T. and Komeda, Y. (1999) Cloning and characterization of an L1 layer-specific gene in Arabidopsis thaliana. Plant Cell Physiol. 40, 571580.
  • Abe, M., Takahashi, T. and Komeda, Y. (2001) Identification of a cis-regulatory element for L1 layer-specific gene expression, which is targeted by an L1-specific homeodomain protein. Plant J. 26, 487494.
  • Bao, J.Y., Lee, S., Chen, C. et al. (2005) Serial analysis of gene expression study of a hybrid rice strain (LYP9) and its parental cultivars. Plant Physiol. 138, 12161231.
  • Bargmann, B. and Birnbaum, K. (2010) Fluorescence activated cell sorting of plant protoplasts. J. Vis. Exp. pii: 1673. doi: 10.3791/1673.
  • Bauer, S., Grossmann, S., Vingron, M. and Robinson, P.N. (2008) Ontologizer 2.0–a multifunctional tool for GO term enrichment analysis and data exploration. Bioinformatics, 24, 16501651.
  • Bauer, S., Gagneur, J. and Robinson, P.N. (2010) GOing Bayesian: model-based gene set analysis of genome-scale data. Nucleic Acids Res. 38, 35233532.
  • Benjamini, Y. and Hochberg, Y. (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Stat. Methodol. 57, 289300.
  • Benjamini, Y. and Yekutieli, D. (2001) The control of the false discovery rate in multiple testing under dependency. Ann. Stat. 29, 11651188.
  • Birnbaum, K., Jung, J.W., Wang, J.Y., Lambert, G.M., Hirst, J.A., Galbraith, D.W. and Benfey, P.N. (2005) Cell type-specific expression profiling in plants via cell sorting of protoplasts from fluorescent reporter lines. Nat. Methods, 2, 615619.
  • Bishop, G.J., Harrison, K. and Jones, J.D.G. (1996) The tomato dwarf gene isolated by heterologous transposon tagging encodes the first member of a new cytochrome P450 family. Plant Cell, 8, 959969.
  • Bishop, G.J., Nomura, T., Yokota, T., Harrison, K., Noguchi, T., Fujioka, S., Takatsuto, S., Jones, J.D.G. and Kamiya, Y. (1999) The tomato DWARF enzyme catalyses C-6 oxidation in brassinosteroid biosynthesis. Proc. Natl Acad. Sci. USA, 96, 17611766.
  • Brady, S.M., Long, T.A. and Benfey, P.N. (2006) Unraveling the dynamic transcriptome. Plant Cell, 18, 21012111.
  • Cameron, K.D., Teece, M.A. and Smart, L.B. (2006) Increased accumulation of cuticular wax and expression of lipid transfer protein in response to periodic drying events in leaves of tree tobacco. Plant Physiol. 140, 176183.
  • Dalca, A.V., Rumble, S.M., Levy, S. and Brudno, M. (2010) VARiD: a variation detection framework for color-space and letter-space platforms. Bioinformatics, 26, i343i349.
  • Day, R.C., Grossniklaus, U. and Macknight, R.C. (2005) Be more specific! Laser-assisted microdissection of plant cells. Trends Plant Sci. 10, 397406.
  • Ewing, B., Hillier, L., Wendl, M.C. and Green, P. (1998) Base-calling of automated sequencer traces using Phred. I. Accuracy assessment. Genome Res. 8, 175185.
  • Garrison, E. and Marth, G. (2012) Haplotype-based variant detection from short-read sequencing. http://arxiv.org/abs/1207.3907.
  • Girard, A.L., Mounet, F., Lemaire-Chamley, M. et al. (2012) Tomato GDSL1 is required for cutin deposition in the fruit cuticle. Plant Cell, 24, 31193134.
  • Goffreda, J.C., Szymkowiak, E.J., Sussex, I.M. and Mutschler, M.A. (1990) Chimeric tomato plants show that aphid resistance and triacylglucose production are epidermal autonomous characters. Plant Cell, 2, 643649.
  • Gong, P., Zhang, J., Li, H. et al. (2010) Transcriptional profiles of drought-responsive genes in modulating transcription signal transduction, and biochemical pathways in tomato. J. Exp. Bot. 61, 35633575.
  • Heichel, G.H. and Anagostakis, S.L. (1978) Stomatal response to light of Solanum pennellii, Lycopersicon esculentum, and a graft-induced chimera. Plant Physiol. 62, 387390.
  • Hu, T.-X., Miao, Y.U. and Zhao, J. (2011) Techniques of cell type-specific transcriptome analysis and application in researches of sexual plant reproduction. Front. Biol. 6, 3139.
  • Huala, E. and Sussex, I.M. (1993) Determination and cell interactions in reproductive meristems. Plant Cell, 5, 11571165.
  • Ingram, G.C. (2004) Between the sheets: inter-cell-layer communication in plant development. Philos. Trans. R. Soc. Lond. B Biol. Sci. 359, 891906.
  • Isaacson, T., Kosma, D.K., Matas, A.J. et al. (2009) Cutin deficiency in the tomato fruit cuticle consistently affects resistance to microbial infection and biomechanical properties, but not transpirational water loss. Plant J. 60, 363377.
  • Jacinto, T., McGurl, B., Franceschi, V., Delano-Freier, J. and Ryan, C.A. (1997) Tomato prosystemin promoter confers wound-inducible, vascular bundle-specific expression of the b-glucuronidase gene in transgenic tomato plants. Planta, 203, 406412.
  • Javelle, M., Vernoud, V., Rogowsky, P.M. and Ingram, G.C. (2011) Epidermis: the formation and functions of a fundamental plant tissue. New Phytol. 189, 1739.
  • Jenik, P.D. and Irish, V.F. (2000) Regulation of cell proliferation patterns by homeotic genes during Arabidopsis floral development. Development, 127, 12671276.
  • Joubes, J., Raffaele, S., Bourdenx, B., Garcia, C., Laroche-Traineau, J., Moreau, P., Domergue, F. and Lessire, R. (2008) The VLCFA elongase gene family in Arabidopsis thaliana: phylogenetic analysis, 3D modelling and expression profiling. Plant Mol. Biol. 67, 547566.
  • Kamenetzky, L., Asís, R., Bassi, S. et al. (2010) Genomic analysis of wild tomato introgressions determining metabolism- and yield-associated traits. Plant Physiol. 152 (4), 17721786.
  • Ko, J. and Han, K. (2004) Arabidopsis whole-transcriptome profiling defines the features of coordinated regulations that occur during secondary growth. Plant Mol. Biol. 55, 433453.
  • Li, J. and Nam, K.H. (2002) Regulation of brassinosteroid signaling by a GSK3/SHAGGY-like kinase. Science, 295, 12991301.
  • Li, J., Wen, J., Lease, K.A., Doke, J.T., Tax, F.E. and Walker, J.C. (2002) BAK1, an Arabidopsis LRR Receptor-like Protein Kinase, Interacts with BRI1 and Modulates Brassinosteroid Signaling. Cell, 110, 213222.
  • Liedl, B.E., McCormick, S. and Mutschler, M.A. (1996) Unilateral incongruity in crosses involving Lycopersicon pennellii and L. esculentum is distinct from self-incompatibility in expression, timing and location. Sex. Plant Reprod. 5, 299308.
  • Life Technologies. (2010). Applied Biosystems SOLiDTM System: BioScopeTM Software for Scientists Guide: Data Analysis Methods and Interpretation (version 1.2.1).
  • Lu, P., Porat, R., Nadeau, J.A. and O'Neill, S.D. (1996) Identification of a meristem L1 layer-specific gene in Arabidopsis that is expressed during embryonic pattern formation and defines a new class of homeobox genes. Plant Cell, 8, 21552168.
  • Marcotrigiano, M. (1986) Experimentally synthesized plant chimeras. 3. Qualitative and quantitative characteristics of the flowers of interspecific Nicotiana chimeras. Ann. Bot. 57, 435442.
  • Marcotrigiano, M. and Bernetzky, R. (1995) Arrangement of cell layers in the shoot apical meristems of periclinal chimeras influences cell fate. Plant J. 7, 193202.
  • Matas, A.J., Agusti, J., Tadeo, F.R., Talon, M and Rose, J.K.C. (2010) Tissue-specific transcriptome profiling of the citrus fruit epidermis and subepidermis using laser capture microdissection. J. Exp. Bot. 61, 33213330.
  • Matas, A.J., Yeats, T.H., Buda, G.J. et al. (2011) Tissue- and cell-type specific transcriptome profiling of expanding tomato fruit provides insights into metabolic and regulatory specialization and cuticle formation. Plant Cell, 23, 38933910.
  • Montoya, T., Nomura, T., Farrar, K., Kaneta, T., Yokota, T. and Bishop, G.J. (2002) Cloning the tomato Curl3 gene highlights the putative dual role of the leucine-rich repeat receptor kinase tBRI1/SR160 in plant steroid hormone and peptide hormone signaling. Plant Cell, 14, 31633176.
  • Montoya, T., Nomura, T., Yokota, T., Farrar, K., Harrison, K., Jones, J.G.D., Kaneta, T., Kamiya, Y., Szekeres, M. and Bishop, G.J. (2005) Patterns of Dwarf expression and brassinosteroid accumulation in tomato reveal the importance of brassinosteroid synthesis during fruit development. Plant J. 42, 262269.
  • Pischke, M.S., Huttlin, E.L., Hegeman, A.D. and Sussman, M.R. (2006) A transcriptome-based characterization of habituation in plant tissue culture. Plant Physiol. 140, 12551278.
  • Rakwal, R. and Agrawal, G.K. (2003) Rice proteomics: current status and future perspectives. Electrophoresis, 24, 33783389.
  • Robinson, M.D., McCarthy, D.J. and Smyth, G.K. (2010) edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics, 26, 139140.
  • Ruffel, S., Freixes, S., Balzergue, S. et al. (2008) Systemic signaling of the plant nitrogen status triggers specific transcriptome responses depending on the nitrogen source in Medicago truncatula. Plant Physiol. 146, 20202035.
  • Sallaud, C., Rontein, D., Onillon, S. et al. (2009) A novel pathway for sesquiterpene biosynthesis from Z,Z-farnesyl pyrophosphate in the wild tomato Solanum habrochaites. Plant Cell, 21, 301317.
  • Satina, S., Blakeslee, A.F. and Avery, A.G. (1940) Demonstrations of the three germ layers in the shoot apex of Datura by means of induced polyploidy in periclinal chimeras. Am. J. Bot. 27, 895905.
  • Savaldi-Goldstein, S., Peto, C. and Chory, J. (2007) The epidermis both drives and restricts plant shoot growth. Nature, 446, 199202.
  • Schmulling, T. and Schell, J. (1993) Transgenic tobacco plants regenerated from leaf disks can be periclinal chimeras. Plant Mol. Biol. 21, 705708.
  • Shi, J.X., Adato, A., Alkan, N. et al. (2013) The tomato SlSHINE3 transcription factor regulates fruit cuticle formation and epidermal patterning. New Phytol. 197, 468480.
  • Smyth, G.K. (2005) Limma: linear models for microarray data. In Bioinformatics and Computational Biology Solutions using R and Bioconductor (Gentleman, R., Carey, V., Dudoit, S., Irizarry, R. and Huber, W., eds). New York, NY: Springer, pp. 397420.
  • Sugano, S.S., Shimada, T., Imai, Y., Okawa, K., Tamai, A., Mori, M. and Hara-Nishimura, I. (2010) Stomagen positively regulates stomatal density in Arabidopsis. Nature, 463, 241244.
  • Sussex, I.M. (1989) Developmental programming of the shoot meristem. Cell, 56, 225229.
  • TGC - Tomato Genome Consortium (2012) The tomato genome sequence provides insights into fleshy fruit evolution. Nature, 485, 635641.
  • Yamamoto, N., Tsugane, T., Watanabe, M., Yano, K., Maeda, F., Kuwata, C., Torki, M., Ban, Y., Nishimura, S. and Shibata, D. (2005) Expressed sequence tags from the laboratory-grown miniature tomato (Lycopersicon esculentum) cultivar Micro-Tom and mining for single nucleotide polymorphisms and insertions/deletions in tomato cultivars. Gene, 356, 127134.
  • Yang, M. and Sack, F.D. (1995) The too many mouths and four lips mutations affect stomatal production in Arabidopsis. Plant Cell, 7, 22272239.
  • Yang, W., Simpson, J.P., Li-Beisson, Y., Beisson, F., Pollard, M. and Ohlrogge, J.B. (2012) A land-plant-specific glycerol-3-phosphate acyltransferase family in Arabidopsis: substrate specificity, sn-2 preference, and evolution. Plant Physiol. 160, 638652.
  • Yeats, T.H., Martin, L.B., Viart, H.M. et al. (2012) The identification of cutin synthase: formation of the plant polyester cutin. Nat. Chem. Biol. 8, 609611.
  • Zeller, G., Henz, S.R., Widmer, C.K., Sachsenberg, T., Rätsch, G., Weigel, D. and Laubinger, S. (2009) Stress-induced changes in the Arabidopsis thaliana transcriptome analyzed using whole-genome tiling arrays. Plant J. 58, 10681082.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Conflict of interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
tpj12250-sup-0001-FigS1.TIFFTIFF image47KFigure S1. Phenotype of seeds.
tpj12250-sup-0002-FigS2.TIFFTIFF image711KFigure S2. Comparison of variant detection methods.
tpj12250-sup-0003-FigS3.TIFFTIFF image555KFigure S3. Comparison of gene classifications resulting from all three variant detection methods.
tpj12250-sup-0004-FigS4.tiffTIFF image2839KFigure S4. Comparison of tissue support for all genes classified as from layer L1 or layers L2/L3.
tpj12250-sup-0005-FigS5.tiffTIFF image35703KFigure S5. Parental-origin allele-specific expression values for all Sanger-sequenced genes.
tpj12250-sup-0006-FigS6.tiffTIFF image2572KFigure S6. Gene Ontology graph of over-represented terms for layer L1 genes.
tpj12250-sup-0007-FigS7.tiffTIFF image1508KFigure S7. Gene Ontology graph of over-represented terms for layers L2/L3 genes.
tpj12250-sup-0008-FigS8.TIFFTIFF image48KFigure S8. Water loss.
tpj12250-sup-0013-FigureLegends.docxWord document17K 
tpj12250-sup-0009-TableS1-S3-TableS7-S8.docxWord document28K

Table S1. Levels of ABA in water-stressed leaves.

Table S2. Results from mapping of SOLiD reads for all nine libraries.

Table S3. Results from variant detection.

Table S4. Parental-origin allele-specific expression values in all samples.

Table S5. List of genes classified as layer L1, either specific or related.

Table S6. List of genes classified as layers L2/L3, either specific or related.

Table S7. Comparison of layer L1 specificity from Sanger sequencing with their Next Generation Sequencing based classification.

Table S8. Sequences of primers used.

tpj12250-sup-0010-TableS4.pdfapplication/PDF2532K 
tpj12250-sup-0011-TableS5.pdfapplication/PDF77K 
tpj12250-sup-0012-TableS6.pdfapplication/PDF192K 
tpj12250-sup-0015-SupportingMethods.docxWord document42KMethods S1. Supporting methods.
tpj12250-sup-0014-SupplementReference.docxWord document15K 

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.