| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
First published online July 13, 2007; 10.1104/pp.107.102632 Plant Physiology 145:160-173 (2007) © 2007 American Society of Plant Biologists OPEN ACCESS ARTICLE
Differential Expression of Genes Important for Adaptation in Capsella bursa-pastoris (Brassicaceae)1,[W],[OA]Department of Evolution, Genomics and Systematics, Uppsala University, SE–752 36 Uppsala, Sweden (T.S., K.H., U.L., M.L.); and Department of Molecular Genetics and Microbiology, University of Florida, Gainesville, Florida 32610–0266 (L.M.M.)
Understanding the genetic basis of natural variation is of primary interest for evolutionary studies of adaptation. In Capsella bursa-pastoris, a close relative of Arabidopsis (Arabidopsis thaliana), variation in flowering time is correlated with latitude, suggestive of an adaptation to photoperiod. To identify pathways regulating natural flowering time variation in C. bursa-pastoris, we have studied gene expression differences between two pairs of early- and late-flowering C. bursa-pastoris accessions and compared their response to vernalization. Using Arabidopsis microarrays, we found a large number of significant differences in gene expression between flowering ecotypes. The key flowering time gene FLOWERING LOCUS C (FLC) was not differentially expressed prior to vernalization. This result is in contrast to those in Arabidopsis, where most natural flowering time variation acts through FLC. However, the gibberellin and photoperiodic flowering pathways were significantly enriched for gene expression differences between early- and late-flowering C. bursa-pastoris. Gibberellin biosynthesis genes were down-regulated in late-flowering accessions, whereas circadian core genes in the photoperiodic pathway were differentially expressed between early- and late-flowering accessions. Detailed time-series experiments clearly demonstrated that the diurnal rhythm of CIRCADIAN CLOCK-ASSOCIATED1 (CCA1) and TIMING OF CAB EXPRESSION1 (TOC1) expression differed between flowering ecotypes, both under constant light and long-day conditions. Differential expression of flowering time genes was biologically validated in an independent pair of flowering ecotypes, suggesting a shared genetic basis or parallel evolution of similar regulatory differences. We conclude that genes involved in regulation of the circadian clock, such as CCA1 and TOC1, are strong candidates for the evolution of adaptive flowering time variation in C. bursa-pastoris.
Flowering time is a major life-history trait contributing to reproduction and adaptation, especially in annual plants (Roux et al., 2006  ). The timing of flowering in relation to the environment is of crucial importance for seed production, and different flowering strategies may have evolved in response to local climatic conditions (Engelmann and Purugganan, 2006; Mitchell-Olds and Schmitt, 2006). The genetic basis of flowering time variation is well understood in Arabidopsis thaliana. Four main pathways, the photoperiod, vernalization, GA, and autonomous pathways, allow the plant to perceive and respond to changes in daylength, temperature, and hormonal status (Mouradov et al., 2002; Simpson and Dean, 2002; Koornneef et al., 2004). Floral pathway integrator genes integrate signals from these pathways and fine-tune the transition from vegetative to reproductive development, although recent studies also indicate that there is direct cross talk between pathways (Edwards et al., 2006; Gould et al., 2006; Salathia et al., 2007).
Understanding the genetic basis of natural variation is of primary interest for evolutionary studies of adaptation (Mitchell-Olds and Schmitt, 2006
Capsella bursa-pastoris L. Medik. is a predominantly selfing, disomic tetraploid crucifer with a nearly worldwide distribution (Hurka and Neuffer, 1997
Changes in the balance between flowering time pathways can result in dramatic differences in flowering time (Lempe et al., 2005
Flowering Time Variation in C. bursa-pastoris
Based on data from a survey of flowering time variation in a worldwide sample of C. bursa-pastoris (Ceplitis et al., 2005
Flowering Time Is Affected by Vernalization We assessed the flowering time of ecotypes PL and SE14, with and without vernalization, using survival analysis, an analysis method for time-dependent developmental traits (see "Materials and Methods") such as flowering time. We found that the survival function (i.e. the predicted probability of not flowering) was different across the four groups (P < 0.0001), and all pairwise comparisons, including that between vernalization treatments for the early-flowering accession PL, exhibited significantly different median flowering times (P < 0.001; Table I ; Fig. 2 ). Thus, vernalization had an effect on flowering time in both extreme flowering ecotypes, although the effect was greater for accession SE14 than accession PL (Table I).
Characterization of Gene Expression Differences between Flowering Ecotypes
To test whether genes involved in regulation of flowering time in A. thaliana were differentially expressed between flowering ecotypes of C. bursa-pastoris, we used A. thaliana CATMA 25k (Complete Arabidopsis Transcriptome Microarray; Allemeersch et al., 2005
We assembled a list of 214 genes that have been identified as involved in flowering time in A. thaliana, based on Gene Ontology (GO) annotation (see "Materials and Methods"; Supplemental Appendix S2). Of these, 112 probes were analyzed for differential expression, and 21 were significantly differentially expressed (false discovery rate [FDR]
The expression of several genes in the GA pathway differed between accessions (Table II; Fig. 2). Two genes involved in GA biosynthesis, GA4 encoding GA 3-  -dioxygenase/GA 3--hydroxylase (At1g15550; Talon et al., 1990; Chiang et al., 1995) and a gene encoding GA 20-oxidase (GA20OX2; At5g51810; Xu et al., 1995), were down-regulated in SE14 compared to PL, whereas genes encoding RGL1, a GA regulatory protein that represses GA signaling (Wen and Chang, 2002), and SLY1 (At4g24210), a gene involved in regulation of GA signaling (Dill et al., 2004), were up-regulated in the late-flowering accession SE14 compared to PL. In addition, six myb-family transcription factors whose expression is affected by GA (Chen et al., 2006) were differentially expressed across accessions. Three of these were up-regulated in accession SE14 (At5g65790, At5g02840, and At5g37260), and three were down-regulated (At3g46130, At5g67580, and At5g47390).
Other differentially expressed candidate genes for flowering included two genes in the vernalization pathway: VIP4 (At5g61150) and FRL1 (At5g16320), both involved in regulation of FLC expression (Zhang and van Nocker, 2002 Microarray data for an additional 10,859 probes were also analyzed for differential expression. The expression of a total of 1,642 differed significantly between groups at 10% FDR. The largest difference in gene expression was found between nonvernalized seedlings of accessions PL and SE14 (PLNV versus SENV, 1,493 genes). Fewer genes were differentially expressed between vernalized seedlings of the two accessions (PLV versus SEV, 874 genes), and very few gene expression differences were found between vernalized and nonvernalized seedlings (PLV versus PLNV, and SEV versus SENV, two genes). However, GO annotation of the 1,642 genes indicates that most of these genes function in various biological processes with no obvious relation to control of flowering time (Supplemental Appendix S3). Genes differentially expressed by vernalization encode a Gly-rich, endomembrane-located protein (At4g29030) and a microtubule-associated protein (MAP70-1) that have not been implicated previously in the vernalization response.
We used list enrichment analysis to assess whether there was an overrepresentation of differentially expressed genes in GO categories of relevance to flowering time (see "Materials and Methods"). We found a significant overrepresentation of significantly differentially expressed genes in the category "circadian rhythm" (20 genes in category, seven significant, two-sided P = 2.3 x 10–2, Fisher's exact test). There was also a significant overrepresentation of genes involved in GA metabolism and signaling (49 genes in category, 13 significant at FDR 0.1, two-sided P = 4.23 x 10–2, Fisher's exact test).
To determine whether the positions of differentially expressed genes were random or clustered, we examined the chromosomal position of each differentially transcribed probe, based on the A. thaliana genome annotation. We found that part of A. thaliana chromosome 2, corresponding to ancestral chromosome 4 (ak4) in Capsella (Schranz et al., 2006
We selected four genes for verification of the microarray results (SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 [SOC1], TOC1, CCA1, and FLC). Although FLC was not differentially expressed after correction for multiple testing, there was some evidence for differential expression (P = 0.03), and the literature on this gene as well as the vernalization response led us to include it in our panel. Real-time reverse transcription (RT)-PCR
Because the microarray data analysis indicated that circadian core genes were differentially expressed, we set up two experiments to assess differences in the expression of circadian genes over time. The rhythmic expression of the circadian core oscillator genes TOC1 and CCA1 differed between accessions PL and SE14 under both constant light and long-day conditions (Fig. 4 ).
Independent Biological Validation of Differential Expression Independent biological validation of differentially expressed flowering time genes was obtained in a pair of less extreme flowering ecotypes, representative of the average part of the flowering time distribution (Fig. 1; accessions US721 and US740). Gene expression microarray analysis (see "Materials and Methods") indicated that a total of 97 probes for flowering time genes, including probes for 18 of the 21 differentially expressed flowering time genes in the PL-SE14 comparison, were in common between experiments (Supplemental Appendix S5). As an independent biological validation, we asked whether the set of 18 flowering time genes that were differentially expressed between the extreme flowering ecotypes also had evidence for differential expression in the US721-US740 comparison. Out of 33 significant contrasts between accessions for these genes in the PL-SE14 comparison, 12 contrasts corresponding to eight different genes were also significant in the US721-US740 comparison (Table III ). These genes included circadian core genes such as LHY and TOC1, as well genes involved in GA biosynthesis and response (e.g. GA4, RGL1, MYB48, and the myb-family transcription factor At5g02840); FRL1, a gene involved in the vernalization response; and SVP, a floral repressor (Table III). Overall, this constitutes good agreement between experiments and indicates that flowering ecotypes with intermediate differences in flowering time also differ in the expression of genes regulating circadian rhythm and GA biosynthesis and response.
In this study we have characterized differential gene expression between flowering ecotypes of C. bursa-pastoris, to test whether gene regulation differences in known flowering time genes in Arabidopsis are also responsible for natural variation in flowering time in C. bursa-pastoris. In the close relative A. thaliana, a major part of natural flowering time variation is due to multiple independent mutations in the FRI gene, the function of which is to induce FLC expression that in turn represses the transition to flowering (Johanson et al., 2000 ). In our experiment, quantitative RT-PCR analysis of FLC expression showed that FLC was indeed down-regulated in both early- and late-flowering accessions as a result of vernalization, but did not differ significantly in expression between accessions before vernalization. Thus, although it seems likely that the function of FLC as an important mediator of the vernalization response is conserved across A. thaliana and C. bursa-pastoris, our data shows that similar mutations as those found in A. thaliana FRI have not been important in generating natural flowering time variation in the C. bursa-pastoris accessions we have studied. Other pathways and genes are more likely responsible for natural variation in flowering time in this species. Microarray analysis of differential expression between early- and late-flowering C. bursa-pastoris offered some insight as to which pathways these may be. Indeed, we found a significant enrichment of differentially expressed genes in two of the main A. thaliana flowering time pathways, the GA pathway and the photoperiodic pathway, and, more specifically, circadian clock-related genes in the latter. The fact that different pathways seem responsible for natural flowering time variation in A. thaliana and the studied accessions of C. bursa-pastoris could suggest that these species have experienced different selective constraints on flowering time, or that genetic variation at flowering time genes differed between species, providing different avenues to variation in flowering time.
In A. thaliana, variation in circadian rhythm among natural accessions contributes to fitness (Dodd et al., 2005
The GA pathway was also enriched for differentially expressed genes among the early-flowering accession PL and the late-flowering accession SE14. In A. thaliana, the GA pathway is generally considered as a default pathway acting mainly when flowering is not induced by long days. Although gene expression differences for genes in the GA pathway might be important for flowering time variation in C. bursa-pastoris, an attractive alternative hypothesis is that these expression differences are a secondary effect of altered circadian clock function, and that this altered clock function affects flowering time mainly through other pathways (e.g. through CO and FT). In this study, two GA biosynthesis genes displayed a higher expression in early flowering accession PL as compared to SE14, which might well be an effect of altered clock function (Blázquez et al., 2002
Differential expression of flowering time genes was biologically validated in a pair of less extreme flowering ecotypes from North America. The good agreement of flowering time gene expression differences between both pairs of accessions could indicate that the genetic basis of expression differences is shared by common ancestry, or that similar regulatory differences have evolved in parallel. Although the two pairs of accessions were sampled in widely different geographical regions (the extreme flowering ecotypes PL and SE14 from Taiwan and Sweden, respectively, and the less extreme flowering accessions US721 and US740 from the United States), a shared genetic background is not unlikely, as the species has apparently attained its present distribution recently (Ceplitis et al., 2005
Overall, most genes differed in expression across accessions, and not as a result of the vernalization treatment, although vernalization had an effect on flowering time. This could indicate that vernalization affected the expression of very few genes, or that the effect on gene expression was generally small so that we had limited power to detect these differences. Similar results have been obtained in other species, for example, in Lolium perenne, where cDNA microarray analysis identified only a handful of genes differentially expressed as a result of vernalization treatment (Ciannamea et al., 2006
Most of the differentially expressed genes were scattered across different chromosomal regions. However, the proportion of significant genes (out of all detected genes) was higher than expected for ancestral chromosome 4, which corresponds to the lower part of A. thaliana chromosome 2 (Schranz et al., 2006
In this study we have characterized gene expression differences between early- and late-flowering accessions of C. bursa-pastoris. Flowering time variation may have evolved rapidly in this species and is probably of adaptive importance (Ceplitis et al., 2005
Flowering Time
We compared vernalized and nonvernalized plants for each of the two accessions (PL and SE14). Thus, for this experiment there were four groups: PL nonvernalized (PLNV), PL vernalized (PLV), SE14 nonvernalized (SENV), and SE14 vernalized (SEV). For each accession, a single mother plant grown from seed collected in the wild was selected and selfed. Two seeds from this plant were grown and selfed to produce two lines. For each of the four groups, seed from the two lines was used to set up eight plates as follows. Approximately 50 surface-sterilized seeds were sown on each 0.8% agar plate with Murashige and Skoog medium (Duchefa). For the vernalization treatment, four plates per line were set up and incubated at 2.6°C for 28 d. On day 25 of the vernalization treatment, four plates per line for the nonvernalized treatment were set up in the same manner and stratified at 2.6°C for 4 d in order to break seed dormancy. On the 29th day of the experiment, all 32 plates (two lines for both accessions and two treatments, four plates per line and treatment) were placed in a growth chamber under long-day conditions (16/8 h photoperiod, 22°C/18°C), in a randomized complete block design (Cochran and Cox, 1992 Two plates, representing the two lines, from each of the two blocks for each vernalization treatment and accession were used to select 15 seedlings, which were transferred to individual pots. Pots were placed in a growth chamber under long-day conditions as before (16/8 h photoperiod, 18°C/22°C, average light intensity 200 µmol m–2 s–1), again in a randomized block design consisting of five blocks where each block was a tray that contained three plants of each treatment-accession combination or a total of 12 plants. Flowering time was recorded as the time from germination to the opening of the first flower. In addition, the number of true leaves at the onset of flowering was recorded.
The time to flowering is a time-dependent developmental trait. Survival analysis was initially developed to model human lifetimes (Cox, 1972
Seven-day-old seedlings from the experiment described above were sampled from the plates in block 1. From each of the four independent plates, two plates for each of the two lines, 15 whole seedlings were sampled and immediately flash-frozen in liquid nitrogen, to give four independent biological replicates of each treatment accession combination. Sampling took place at midday, 7 h after dawn. Sampling occurred in the same order as the randomized block design and, therefore, the order of sampling was random with respect to vernalization-treatment and accession. We measured gene expression in seedlings because previous studies have shown that several key flowering time regulators are apparent at a very early stage in Arabidopsis thaliana (Kobayashi et al., 1999
Total RNA was extracted using the RNeasy plant mini kit (Qiagen), including DNase treatment using the RNase-free DNase set (Qiagen), according to the manufacturer's instructions. Protocols for RNA amplification, labeling, and hybridization were modified from those used by Wellmer et al. (2004)
Hybridization was conducted according to a loop design (Kerr and Churchill, 2001a
Microarray images were quantitated using the Spot 3.0 R-based package (CSIRO), using the GOGAC segmentation option, and signal median was background corrected using the morph.open.close background estimate. Previous work has demonstrated that this is a reliable quantification approach (Slotte and McIntyre, 2007 The spot quality was assessed as follows. For each microarray and dye, all spots were ranked and divided into quartiles. Quartiles were compared using the kappa coefficient and spots that differed in rank by more than one quartile between replicates were flagged. In addition, individual spots that were saturated were flagged.
To determine whether there was evidence for hybridization for a given probe, the distribution of negative controls was used. There are 16 negative controls on the CATMA slide distributed across the slide. Two of these negative controls have evidence of contamination (data not shown) and were excluded from consideration, leaving 14 spots per slide. To conclude that the sample has hybridized to a particular spot, the signal from the spot should be above the 90th percentile of the signal of negative control spots (Li et al., 2004
When comparing different genotypes directly on a microarray, there is always a possibility that differences in gene expression are confounded with sequence divergence (Gilad and Borevitz, 2006
Intensity values for each microarray (log2 background-corrected signal) were lowess-transformed (Cleveland, 1979
We downloaded A. thaliana locus tags and GO annotation corresponding to the probes on the CATMA array from The Arabidopsis Information Resource (www.arabidopsis.org). While the species are different and one cannot be certain of the similarity of annotation across species, the species are closely related (e.g. Galloway et al., 1998 We assembled a list of genes that have been identified as involved in flowering time. An overview of the current knowledge of A. thaliana flowering time pathways is found in Figure 3. For the development of the flowering time list, we included a total of 214 probes (which were also present on the CATMA array) whose GO biological process annotation contained the terms "circadian rhythm" (GO:0007623), "flower development" (GO:0009908), "vegetative to reproductive phase transition" (GO:0010228), "photoperiod" (GO:0009648), "vernalization response" (GO:0010048), or "gibberellic acid" (gibberellic acid biosynthetic process, GO:0009686; gibberellic acid metabolic process, GO:0009685; or gibberellic acid-mediated signaling, GO:0009740; gibberellic acid catabolic process, GO:0045487). The final list was manually curated to include additional flowering time genes that were not annotated using these terms (e.g. FRL1, CATMA5a14630). The resulting list represents a group for which we were a priori interested in their responses, and they are listed in Supplemental Appendix S2. We tested for statistical overrepresentation or underrepresentation of significantly differentially expressed genes in the six categories listed above, using Fisher's exact tests. List enrichment analyses, lowess and median normalization, ANOVA, and FDR correction of microarray data were performed using SAS 9.1 (SAS Institute) and JMP 6.0 microarray (SAS Institute).
Total RNA from the four biological replicates of each group was used as source for the real-time RT-PCR verification of specific transcript levels. For each replicate, 0.5 µg of total RNA was reverse transcribed to cDNA using random hexamer primers (Invitrogen) and SuperScript III reverse transcriptase (Invitrogen) following the manufacturer's instructions. cDNA samples were diluted 1:100 and amplified using the Platinum SYBR Green qPCR SuperMix (Invitrogen), on an ABI PRISM 7000 sequence detection system (Applied Biosystems). The two-step cycling program was as follows: 50°C for 3 min and 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 30 s. Melt curve analyses were performed after each amplification to confirm specificity of products. Each cDNA sample was run in technical triplicates. As a further data quality control, PCR efficiencies were calculated for each individual amplification with the software LinRegPCR (Ramakers et al., 2003
Primers were designed to amplify both homoeologous loci based on direct sequences for TOC1 and CCA1. In other instances, we tested and used primers originally designed for A. thaliana, SOC1 (Czechowski et al., 2004
Expression levels of TOC1 and CCA1 were monitored in two time-series experiments under two light regimes: constant light and long day (16 h light/8 h dark). For each time series, approximately 40 plants of each accession for each time point were germinated on two separate 0.8% agar plates with Murashige and Skoog medium (Duchefa). The two plates were randomly positioned in the growth chamber, yielding two environmental replicates of each accession at each time point. Seeds were stratified for 5 d at 2.6°C, followed by entrainment at 22°C under long-day conditions with a light intensity of 52 µmol m–2 s–1 for 7 d, before release into either constant light (52 µmol m–2 s–1) or continued long-day (52 µmol m–2 s–1) conditions. Two pools of 15 to 20 seedlings were sampled from each plate on 12 time points over 48 h, at 4-h intervals. Sampling of the constant light time series was initiated at 4 h after dawn, whereas sampling of the long-day time series was initiated at dawn. Total RNA was isolated in two separate extractions per accession and plate, using the RNeasy plant mini kit (Qiagen). cDNA synthesis and amplification were conducted as for the real-time RT-PCR verification (see above). Each accession for each time point was run in technical PCR duplicates, which enabled the comparison of both accessions on one RT-PCR plate. TOC1 and CCA1 were amplified with primer sets CbpTOC1_1043Fq/1240Rq and CCA1_5/6, respectively. TUB expression levels were used for normalization.
To obtain an independent biological validation of flowering time gene expression differences, we assessed gene expression differences between two North American accessions of C. bursa-pastoris (US721 and US740), which are less extreme in their differences in flowering time (Fig. 1). Gene expression was measured using CATMA microarrays, in a setup identical to that described above except that sampling took place at 9 h after dawn, 2 h later than for the experiment including accessions PL and SE14. Differential expression was analyzed as outlined above.
The following materials are available in the online version of this article.
We thank Mattias Myrenås and Myriam Heuertz for experimental assistance. Received May 23, 2007; accepted July 10, 2007; published July 13, 2007.
1 This work was supported by grants from the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (to M.L. and U.L.); a grant from the Swedish Research Council (to U.L.); and grants from the Nilsson-Ehle, Wallenberg, Sederholms, and Tullberg foundations (to T.S.).  The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Martin Lascoux (martin.lascoux{at}ebc.uu.se).
[W] The online version of this article contains Web-only data.
[OA] Open Access articles can be viewed online without a subscription. www.plantphysiol.org/cgi/doi/10.1104/pp.107.102632 * Corresponding author; e-mail tanja.slotte{at}ebc.uu.se.
Acarkan A, Rossberg M, Koch M, Schmidt R (2000) Comparative genome analysis reveals extensive conservation of genome organisation for Arabidopsis thaliana and Capsella rubella. Plant J 23: 55–62[CrossRef][ISI][Medline] Alabadi D, Oyama T, Yanovsky MJ, Harmon FG, Mas P, Kay SA (2001) Reciprocal regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock. Science 293: 880–883 Allemeersch J, Durinck S, Vanderhaeghen R, Alard P, Maes R, Seeuws K, Bogaert T, Coddens K, Deschouwer K, Van Hummelen P, et al (2005) Benchmarking the CATMA microarray: a novel tool for Arabidopsis transcriptome analysis. Plant Physiol 137: 588–601 Balasubramanian S, Sureshkumar S, Agrawal M, Michael TP, Wessinger C, Maloof JN, Clark R, Warthmann N, Chory J, Weigel D (2006) The PHYTOCHROME C photoreceptor gene mediates natural variation in flowering and growth responses of Arabidopsis thaliana. Nat Genet 38: 711–715[CrossRef][ISI][Medline] Bastow R, Mylne JS, Lister C, Lippman Z, Martienssen RA, Dean C (2004) Vernalization requires epigenetic silencing of FLC by histone methylation. Nature 427: 164–167[CrossRef][Medline] Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate—a practical and powerful approach to multiple testing. J R Stat Soc B 57: 289–300 Blázquez MA, Trenor M, Weigel D (2002) Independent control of gibberellin biosynthesis and flowering time by the circadian clock in Arabidopsis. Plant Physiol 130: 1770–1775 Boivin K, Acarkan A, Mbulu RS, Clarenz O, Schmidt R (2004) The Arabidopsis genome sequence as a tool for genome analysis in Brassicaceae. A comparison of the Arabidopsis and Capsella rubella genomes. Plant Physiol 135: 735–744 Brem RB, Yvert G, Clinton R, Kruglyak L (2002) Genetic dissection of transcriptional regulation in budding yeast. Science 296: 752–755 Browse J, Xin ZG (2001) Temperature sensing and cold acclimation. Curr Opin Plant Biol 4: 241–246[CrossRef][ISI][Medline] Bunning E (1936) Die endogene Tagesrhythmik als Grundlage der photoperiodischen Reaktion. Ber Dtsch Bot Ges 54: 590–607 Caicedo AL, Stinchcombe JR, Olsen KM, Schmitt J, Purugganan MD (2004) Epistatic interaction between Arabidopsis FRI and FLC flowering time genes generates a latitudinal cline in a life history trait. Proc Natl Acad Sci USA 101: 15670–15675 Ceplitis A, Yingtao S, Lascoux M (2005) Bayesian inference of evolutionary history from chloroplast microsatellites in the cosmopolitan weed Capsella bursa-pastoris (Brassicaceae). Mol Ecol 14: 4221–4233[Medline] Chen Y, Yang X, He K, Liu M, Li J, Gao Z, Lin Z, Zhang Y, Wang X, Qiu X, et al (2006) The MYB transcription factor superfamily of Arabidopsis: expression analysis and phylogenetic comparison with the rice MYB family. Plant Mol Biol 60: 107–124[CrossRef][ISI][Medline] Chiang HH, Hwang I, Goodman HM (1995) Isolation of the Arabidopsis Ga4 locus. Plant Cell 7: 195–201[Abstract] Churchill GA (2002) Fundamentals of experimental design for cDNA microarrays. Nat Genet 32: 490–495[CrossRef][ISI][Medline] Ciannamea S, Busscher-Lange J, de Folter S, Angenent GC, Immink RGH (2006) Characterization of the vernalization response in Lolium perenne by a cDNA microarray approach. Plant Cell Physiol 47: 481–492 Cleveland WS (1979) Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc 74: 829–836[CrossRef][ISI] Cochran WG, Cox GM (1992) Experimental Designs, Ed 2. Wiley InterScience, New York Corbesier L, Vincent C, Jang S, Fornara F, Fan Q, Searle I, Giakountis A, Farrona S, Gissot L, Turnbull C, et al (2007) FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis. Science 316: 1030–1033 Cox DR (1972) Regression models and life tables. J R Stat Soc B 34: 187–220 Czechowski T, Bari RP, Stitt M, Scheible WR, Udvardi MK (2004) Real-time RT-PCR profiling of over 1400 Arabidopsis transcription factors: unprecedented sensitivity reveals novel root- and shoot-specific genes. Plant J 38: 366–379[CrossRef][ISI][Medline] Darrah C, Taylor BL, Edwards KD, Brown PE, Hall A, McWatters HG (2006) Analysis of phase of LUCIFERASE expression reveals novel circadian quantitative trait loci in Arabidopsis. Plant Physiol 140: 1464–1474 de Meaux J, Pop A, Mitchell-Olds T (2006) Cis-regulatory evolution of chalcone-synthase expression in the genus Arabidopsis. Genetics 174: 2181–2202 Dill A, Thomas SG, Hu JH, Steber CM, Sun TP (2004) The Arabidopsis F-box protein SLEEPY1 targets gibberellin signaling repressors for gibberellin-induced degradation. Plant Cell 16: 1392–1405 Dodd AN, Salathia N, Hall A, Kevei E, Toth R, Nagy F, Hibberd JM, Millar AJ, Webb AAR (2005) Plant circadian clocks increase photosynthesis, growth, survival, and competitive advantage. Science 309: 630–633 Dudoit S, Yang YH, Callow MJ, Speed TP (2002) Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments. Statistica Sinica 12: 111–139[ISI] Edwards KD, Anderson PE, Hall A, Salathia NS, Locke JCW, Lynn JR, Straume M, Smith JQ, Millar AJ (2006) FLOWERING LOCUS C mediates natural variation in the high-temperature response of the Arabidopsis circadian clock. Plant Cell 18: 639–650 El-Assal SED, Alonso-Blanco C, Peeters AJM, Raz V, Koornneef M (2001) A QTL for flowering time in Arabidopsis reveals a novel allele of CRY2. Nat Genet 29: 435–440[CrossRef][ISI][Medline] Engelmann K, Purugganan M (2006) The molecular evolutionary ecology of plant development: flowering time in Arabidopsis thaliana. In DE Soltis, JH Leebens-Mack, PS Soltis, eds, Advances in Botanical Research: Incorporating Advances in Plant Pathology, Vol 44. Academic Press, San Diego, pp 507–526 Fleiss J (1981) Statistical Methods for Rates and Proportions. Wiley and Sons, New York Galloway GL, Malmberg RL, Price RA (1998) Phylogenetic utility of the nuclear gene arginine decarboxylase: an example from Brassicaceae. Mol Biol Evol 15: 1312–1320[Abstract] Gilad Y, Borevitz J (2006) Using DNA microarrays to study natural variation. Curr Opin Genet Dev 16: 553–558 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
RESULTS
0.1; 
2 = 9.00, d.f. = 1, P = 2.7 x 10–3). This region of A. thaliana chromosome 2 constitutes an entire, separate chromosome in both A. lyrata and Capsella rubella. In A. thaliana, it corresponds to approximately 10 Mb of the lower part of chromosome 2 (delimited by the loci At2g21160 and At2g47730) containing a total of 2,867 annotated loci. In this study, 1,235 of these were labeled "present" and 228 were differentially expressed. In the US721-US740 comparison, we found no overrepresentation of differentially expressed genes for ak4. 
CT values for differentially expressed candidate genes (SOC1, TOC1, CCA1) were consistent with array results (Supplemental Appendix S4). Thus, we did not identify any false positives among the genes assessed. Analysis of real-time RT-PCR gene expression measurements indicated that FLC expression did not differ between accessions prior to vernalization and was diminished after vernalization in both accessions, but to a greater extent in SE14. 
kl + 
ijkl was fit, where Y is a function of the fixed effects of dye (d), g is the effect of group where there are four groups (PLNV, PLV, SENV, SEV), and the random effect of slide