RESEARCH PAPER |
Transcript profiles at different growth stages and tap-root zones identify correlated developmental and metabolic pathways of sugar beet
1Max-Planck-Institute for Plant Breeding Research, Department of Plant Breeding and Yield Physiology, Carl-von-Linné Weg 10, D-50827 Köln, Germany
2KWS SAAT AG, Grimsehlstrasse 31, D-37555 Einbeck, Germany
3Institute of Genome Research, University of Bielefeld, D-33594 Bielefeld, Germany
4Fondazione Parco Tecnologico Padano, Via Albert Einstein c/o Polo Universitario, Cascina Codazza, I-26900 Lodi (Lo), Italy
5GSF – National Research Center for Environment and Health, Institute for Radiation Protection, Ingolstädter Landstrasse 1, D-85764 Neuherberg, Germany
* To whom correspondence should be addressed. E-mail: katharina.schneider{at}gsf.de
Received 4 May 2006; Revised 24 October 2006 Accepted 25 October 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Field-grown sugar beets were analysed for morphological characters, sucrose content, and reproducible transcript profiles by macroarray analyses with 11 520 unique sugar-beet cDNA targets in two different years. Seasonal differences were partly compensated by expressing sampling dates as thermal time. During early beet development the number of cambial rings, root length, and sucrose concentration had already achieved >40% of their final values. Sucrose levels rose from 10% to 17% over the thermal time of 1300–1400 °Cd with only small changes later when lower concentrations were restricted to the exterior zone at the minimum of the spatial sucrose gradient through the beet. The number of leaves and root diameter followed the same temporal growth pattern, but mass increased until beet maturity at around 2000 °Cd. Cluster analysis identified 543 transcripts with reproducible preferential expression between 1300–1400 °Cd, and 170 showing the highest transcript levels later. In maturing beets, 373 transcripts were over-represented in the inner zone and 148 in the outer zone. During early development, genes involved in cytoskeletal reorganization and transport processes showed the highest transcript levels. Cell wall biogenesis-, defence-, stress-, and degradation-related transcripts were identified in all samples, and associated with pathogen attack during late development and in the outer zone. Candidates with potential roles in carbohydrate metabolism appeared to serve anaplerotic functions by converting excess intermediates to sucrose production. Transcripts preferentially occurring in sucrose-accumulating young beet cells and newly generated peripheral cells of mature beets are discussed as potential breeding targets to improve sink strength and growth.
Key words: Candidate genes, macroarray analysis, sucrose accumulation, sugar beet, tap root formation
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
During the first year of its biennial life cycle, sugar beet (Beta vulgaris L.) develops a sucrose-rich tap root composed of 90% root tissue and 10% hypocotyl tissue (Elliott and Weston, 1993). The root contains up to 20% sucrose per fresh weight at maturity. Sugar beet breeding aims towards increasing the extractable sucrose and to lower the concentration of nitrate, potassium, and sodium which inhibit sucrose purification, as well as to increase resistance to biotic or abiotic stress factors.
After germination, dry biomass of sugar beet leaves exceeds that of roots (Elliott and Weston, 1993), but at the 8–10 leaf stage, leaves and roots start growing simultaneously, and, later on, roots take up the major part of dry weight. Sucrose accumulates during growth in the absence of a ripening phenomenon (Milford, 1973). The morphology of the tap root is conditioned by a central cambium produced between primary xylem and phloem, and by 12–15 supernumerary secondary cambium rings initiated centrifugally. During the first 8 weeks after emergence, eight cambium rings are laid down. Only four or five more rings are formed during later development when the focus is on rapid cell expansion and maximum cell division of cambial cells and their derivatives. Within a vascular ring, xylem and phloem are organized in bundles and separated by parenchymatous rays. The innermost six rings account for about 75% of the tap root; the rings outside contribute progressively less to the radial expansion. The most expanded sector of the beet is characterized by the highest cambial activity. The transport and distribution of sucrose follow the same pattern (Elliott and Weston, 1993). A correlation between cell size and sucrose concentration exists for cells up to 10–15x10–8 cm3, but in larger cells sucrose concentration decreases. There is evidence of apoplastic sucrose unloading from the phloem elements, followed by sucrose cleavage involving sucrose synthases and invertases, and by resynthesis of sucrose in storage cells (Fieuw and Willenbrink, 1990). Finally, sucrose is stored in the vacuoles of parenchymatous cells, and a proton–sucrose antiport system is involved in this process (Willenbrink et al., 1984).
Concerning all phenomena observed, the underlying molecular pathways and the genes involved have not yet been described. To enable a molecular genetic approach targeted to candidate genes with functions in sugar beet development and sucrose accumulation, a comprehensive unigene cDNA collection covering at least 40% of the sugar beet transcriptome is now available (Herwig et al., 2002). In addition, protocols for macroarray analysis of field-grown sugar beets, which allow transcripts to be detected in quantities as small as 0.01% of the poly(A)+ RNA used for probe synthesis (Bellin et al., 2002), have been established. With this technique, signal intensities from spots of the same cDNA on different filters varied <2-fold for about 90% of all cDNAs tested, and the variability within one filter was just about 1% (Bellin et al., 2002). Here, data sets on morphology and sucrose accumulation are presented for the same field-grown sugar beet hybrid samples for which transcript profiling by macroarray analysis has been conducted. In this paper the focus is (i) on beets of six to seven developmental stages from a time-course study carried out in two consecutive years, and (ii) on two separate regions of maturing beets which are characterized by different sucrose concentrations. For beet development, two expression profile gene clusters with increased expression at the early or the late stage, respectively, were identified. Correlations between the data sets depicting the temporal and spatial sucrose gradients in beet identify molecular pathways and the underlying genes which may play a role in sugar beet development and which are potential targets for sugar-beet breeding.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Plant material, cultivation, and harvesting of samples
Seeds of the sugar beet hybrid KWS86203 (KWS SAAT AG, Einbeck, Germany) were sown in field plots at the Max Planck Institute for Plant Breeding Research in Cologne (Germany) on 2 May 2001 and 4 April 2002. Plants within one row were 21 cm apart, and the distance between rows was 21 cm. Cultivation followed local agricultural practice.
Plants were harvested at seven time points each in two consecutive years. To account for seasonal differences the thermal time was calculated in growing degree days (°Cd) according to the formula thermal time=[(high+low temperature)/2]–T0 (Thornley and Johnson, 1990) using temperatures measured at Cologne airport and T0=6 °C. In the first year, 44 d after sowing (DAS) corresponded to 452 °Cd, 60 DAS to 657 °Cd, 76 DAS to 877 °Cd, 97 DAS to 1201 °Cd, 111 DAS to 1420 °Cd, 131 DAS to 1668 °Cd, and 168 DAS to 2019 °Cd. In the second year, sugar beet plants were harvested 62 DAS (514 °Cd), 82 DAS (807 °Cd), 99 DAS (996 °Cd), 120 DAS (1293 °Cd), 141 DAS (1599 °Cd), 166 DAS (1916 °Cd), and 188 DAS (2072 °Cd). At 1916 °Cd in the second year, additional beet material was harvested from the central 5-cm-wide region containing approximately the four innermost cambial rings and the most exterior non-green zone of 1 cm thickness excluding root tissue of at least three intermediate cambial rings.
Determination of morphological characters and sucrose content
At each time point of the kinetic study, 3–10 representative plants were harvested to measure beet diameter and length, to count the number of leaves and beet cambium rings in the region of the largest diameter, and to determine their mass. To estimate sucrose content, freeze-dried beet tissue was extracted, and sucrose was enzymatically determined (Boehringer Mannheim/R-Biopharm, Darmstadt, Germany). For each sample, sucrose measurements were performed in triplicate.
Sequence analysis
A total of 11 520 sugar beet cDNA clones selected by oligo fingerprinting (Herwig et al., 2002) were sequenced. EST sequences are accessible at EMBL/GenBank under the accession numbers BQ605167, BQ582276–BQ595857, and CF542669–CF543782. Trace files were reprocessed by the base calling program Phred (Ewing and Green, 1998; Ewing et al., 1998). Vector sequences were clipped from the raw EST sequences with Crossmatch (http://bozeman.mbt.washington.edu/phrap.docs/phrap.html). High quality sequences were obtained by using a sliding window approach using at least 50 bp with a minimal average Phred score of 20 and filtering out slippage (Telles and da Silva, 2001). The trimmed sequences were compared with cDNA sequences from Arabidopsis thaliana available from TIGR (ftp://ftp.tigr.org/pub/data/a_thaliana/ath1/SEQUENCES/ATH1.cdna, version 5.0) using NCBI-TBLASTX and to the non-redundant protein database from NCBI (ftp://ftp.ncbi.nlm.nih.gov/blast/db/nr, version 2005-01-13) using NCBI-BLASTX.
Macroarray generation and hybridization experiments
For macroarray generation, 11 520 unique sugar beet cDNAs were PCR-amplified and printed by the robotic system MicroGridII (Biorobotics, Carsen group, Ontario, Canada) onto 22 cmx22 cm nylon filters (Hybond N+; Amersham Biosciences, Heidelberg, Germany). Based on the observation that PCR products had an average concentration of 50 ng µl–1, c. 8 ng of DNA was immobilized per spot using pins 400 µm in size in eight repeated spotting steps. The printing pattern consisted of blocks comprising 4x4 spots with eight pairs of two identical spots each, and 384 blocks maximally containing eight different 384-well plates in duplicate constituted one sector. One 22 cmx22 cm filter accommodated six sectors with 23 800 spots including controls and additional empty positions in each of the six sectors for local adjustment of expression values (details available upon request). As controls for unspecific hybridization, vector sequences as well as the insert of pAW109 (derived from GeneAmplimer® pAW 109 RNA; Applied Biosystems, Weiterstadt, Germany) were spotted (Bellin et al., 2002). For sensitivity assessment, 0.8–32 ng of non-plant sequences such as a part of the human nebulin gene (Desprez et al., 1998) were immobilized, and the nebulin fragment was also used at a concentration of 50 ng µl–1 for local normalization of signal intensities.
For probe preparation, RNA was extracted from pooled tissue of three representative beets. For probe synthesis, poly(A)+ RNA was purified from total RNA using the mRNA purification kit from Amersham Biosciences (Heidelberg, Germany), according to the instructions of the manufacturer. To confirm proportionality between signal intensity and the amount of labelled transcript, the in vitro transcript of the nebulin control was added to each probe at 0.01% which was previously established as a spiked control (Bernard et al., 1996) for normalization purposes. Probe synthesis, mock hybridization, oligohybridization, and complex probe hybridization were carried out as described in Bellin et al. (2002), with the exception that the DNA was precipitated with 5 µl of 3 M Na-acetate pH 5.3, 5 µl of 10 mg ml–1 yeast tRNA (Invitrogen, Groningen, The Netherlands), and 60 µl isopropanol for 1 h at –20 °C to remove unincorporated nucleotides. After centrifugation for 30 min at room temperature, the pellet was resuspended in 100 µl H2O.
In total, 35 filters were used in one to four consecutive hybridization experiments.
Macroarray data analysis, statistics, and cluster formation
Hybridization signals were acquired by a phosphorimager (STORM; GE Healthcare, Heidelberg, Germany), quantified by the program Arrayvision (Imaging Research Inc., Haverhill, UK) and corrected by subtracting the local background. Based on the concept of spiked controls (Bernard et al., 1996), the median signal intensity of 96 spotted nebulin transcripts present in each of the six sectors of each filter was used as the normalization standard for the respective sector. The procedure of value normalization (Bellin et al., 2002) was automated by developing Excel macros.
Data analysis was based on four different hybridizations. As cDNAs were spotted in duplicate on each filter, eight independent data points per cDNA entered the evaluation in total. The small-sample procedure of the ArrayStat program package (Imaging Research Inc.) was selected to estimate the random error and to exclude outliers because of statistical reasons. Samples for which fewer than six reproducible values were left after outlier detection were excluded from the following analysis. An F-test in the same program was used to evaluate the significance of the differential expression during the time-course experiment.
To cluster profiles of differentially expressed transcripts the partition clustering algorithm k-means in the Genesis software (Sturn et al., 2002) was used for data of both years separately. In this procedure, k values from 2 to 15 were selected. Only expression values of transcripts with complete data sets for all time points evaluated were log2-transformed according to the Genesis software requirements, and included in the analysis. To avoid a bias by the very first, small sample at 452 °Cd and 514 °Cd in the first and second years, respectively, which revealed the highest variability between years, expression data sets only included samples collected later. The distance between expression patterns was measured using the Pearson correlation, and the median centre function was applied to highlight differential expression during the time course. For the data sets of either year, reproducible clusters were obtained with k=2 generating two clusters, cl1 and cl2. Best congruence was found between cluster members of either cl1 or cl2 in both years, and transcripts identified in the same cluster in the two years were selected as candidates. For each cluster member, the highest average expression value obtained at the time point of its strongest expression is referred to as the absolute value. The ratio of the highest and the lowest expression value within the time-course is the relative value which reflects the changes in abundance of the steady-state transcript levels during the time-course.
For the topological analysis, expression values for the inner and the outer beet zone were derived from four independent hybridizations to the same type of filters described above. Data of these eight technical replications were used to calculate average values for each tissue. Outliers were then detected by the ArrayStat program package (Imaging Research Inc.) using the pooled-curve fit procedure. Transcripts showing fewer than six reproducible values were eliminated for statistical reasons. Average values were tested for significant differences between the inner and outer beet using the z-test without threshold values. Finally, the candidates were ranked according to the ratios calculated from their expression values of the two beet regions.
Semi-quantitative RT-PCR analysis
First-strand cDNA was synthesized from 0.5 µg DNase-treated (Ambion Inc., Huntingdon, UK) poly(A)+ RNA using Superscript II reverse transcriptase (Invitrogen, Groningen, The Netherlands). Equal amounts of cDNA samples were estimated based on housekeeping genes, and the transcript pattern for the 60S ribosomal protein (accession number BQ488925) is given as an example in Fig. 2. Information on sequence-specific primers, PCR conditions, and expected and obtained fragment sizes is listed in Supplementary data (see Supplementary Table 1 at JXB online). For each transcript, the number of cycles providing the most dynamic range of the transcript profile was experimentally determined. RT-PCR products were separated in 1.5% agarose gels, scanned using the same phosphorimager as for the macroarray data acquisition, and densitometrically analysed using ImageQuant software (GE Healthcare, Heidelberg, Germany). For each transcript, all densitometric values were summed up over the annual time-course, and the amount of transcript in an individual sample was expressed as a percentage of the total value of expression.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Morphological and biochemical characterization of root development and sucrose accumulation
Sugar-beet development was analysed in field plots in two consecutive seasons. As temperature is a main factor determining seasonal effects on plant growth, cumulative thermal time (Thornley and Johnson, 1990) was calculated for each sample. The concept of summing up temperature over time is based on the observation that plants need certain amounts of warmth to reach a developmental stage. To evaluate sugar-beet development in the two years, morphological as well as biochemical characters were plotted against thermal time, covering the entire growth period from early stages of development starting at 500 °Cd up to beet maturity at a little more than 2000 °Cd (Fig. 1A–F). With the exception of beet mass, the largest increases in measured characters were found during the early stages. The number of cambium rings and sucrose content increased most between 452 °Cd and 1420 °Cd in the first year and 514 °Cd and 1293 °Cd in the second year. The number of cambium rings was already about half of the final number in the first sample, with a maximum of 13 or 14 at the end of the vegetative season. After only 657 °Cd in the first year and 514 °Cd in the second year, the sucrose concentration was 10.2% and 7.5% of the fresh weight, respectively. These values are equal to nearly 60% and 45% of the sucrose percentage at the end of the first and second season, 17.4% and 17.1%, respectively. At 1916 °Cd in the second year, the spatial distribution of sucrose in the beet was analysed. The core of the tap root, which contains fully expanded cambial rings, had a sucrose concentration of 17.2±0.6% while the outermost exterior zone contained 12.4±0.1% sucrose per gram fresh weight.
|
With the exception of the fourth sample in the first year, root length increased from about 10 cm at the first measurement up to >20 cm at 1420 °Cd and 1293 °Cd in the first and second year and reached maximal length of 23 cm and 26 cm at 1668 °Cd and 1916 °Cd in the two seasons tested.
Leaf number and root diameter also showed the largest increase in the first part of the period monitored in the two years (Fig. 1D, E), but, considering their maximal values, the first measurements at around 500 °Cd were much smaller than the relative values of the previously mentioned traits, indicating a delay and a steep increase during the time monitored. The maximum leaf number of 43 was counted at 1668 °Cd in the first year, and in the second year this was at 1599 °Cd with 46 leaves. Root diameter reached the plateau phase at 1201 °Cd in the first year and, in the second year, a more linear increase in root diameter was observed until 1599 °Cd.
Mass increase appears to be further delayed in both years (Fig. 1F), and especially in the second year, the highest rates of increase were observed between the last time points reaching the maximum mass of about 1.1 kg at 1916 °Cd.
Tap root candidate transcripts differentially expressed during development and across the radial dimension of the beet
Macroarray results were obtained for 11 520 cDNA clones (Table 1A). In the time-course experiment, 3486 transcripts (30.26%) showed differential expression in the first year and 2052 (17.81%) in the second year. No differential expression was recorded for 7692 and 9076 transcripts, respectively. Data on 342 and 392 transcripts had to be discarded for statistical reasons in the two experiments.
|
Macroarray data analyses were performed for each year separately. Based on transcripts with complete sets of expression values, cluster analysis revealed two groups of differentially transcribed genes in both years. Transcripts of the first cluster, cl1, showed a higher expression during the first part of the development while transcripts of the second cluster, cl2, were characterized by higher expression values at the end of the season (Fig. 1G, H). The transition between high and low expression levels of cl1 genes was found at around 1420 °Cd in the first year and between 1293 °Cd and 1599 °Cd in the second year. Complementarily, the transition from low to high levels of cl2 transcripts occurred at around 1420 °Cd in the first year and between 1293 °Cd and 1599 °Cd in the second year. These time points, at which the expression profiles reverse, mark the transition from the early phase of rapid increase in most traits, apart from mass, to the plateau phase in which only minor changes in trait values were observed (Fig. 1A–F).
Crosswise comparisons between cl1 and cl2 members in both years revealed that 80.9% of cl1 transcripts in the first year were identical to 95% of the cl1 transcripts in the second. Concerning cl2, 91% of the transcripts identified in the first year were in common with 62% of the transcripts identified in the second. Considering both years, 543 putative genes revealed a reproducible preferential transcription during early stages of development, and 170 putative genes during late developmental stages (Table 1B). Profiles of selected transcripts were validated by semi-quantitative RT-PCR as shown for a representative set of six transcripts (Fig. 2).
The comparison of the two beet zones with differing sucrose content identified 373 transcripts with a higher expression in the inner beet region and 148 in the outer (Table 1B).
Classification of differentially expressed genes deduced from the time-course experiment
For 366 transcripts of cluster 1 and 102 transcripts of cluster 2, sequence analysis retrieved annotations with expected values below e–20 (Table 1B; see Supplementary Table 2 at JXB online). These transcripts were manually assigned to functional gene classes. The 40 transcripts with the highest absolute and relative expression (or both) are presented for both clusters (Table 2), with relative expression providing the factor by which transcript levels increase or decrease during the time-course. In cl1, 13 were among the 40 most expressed at both the absolute and relative scale, while in cl2, 16 of such transcripts were identified.
|
Differentially expressed genes with strong absolute or relative transcription during early development are predicted to contribute to high activity of cytoskeletal reorganization, transport processes, cell wall biogenesis, and defence responses
Among cluster 1 transcripts (Table 2A), 11 gene products are present which function in cytoskeletal reorganization, the secretory pathway, and vesicular trafficking, a group of transcripts absent from cluster 2. Encoded gene products such as tubulins and a KNOLLE-related putative syntaxin (Müller et al., 2003) play essential roles during cell division and cytokinesis. A predicted signal recognition particle receptor (Pool et al., 2002), as well as SEC14 family members (Bankaitis et al., 1989) and two C2 domain-containing proteins with similarity to SRC2 (Oufattole et al., 2005), are associated with synthesis and export of secretory proteins.
A second group of transcripts over-represented in cl1 are putatively involved in transport processes with a predominant group of major intrinsic proteins which are part of the transport systems for water and small non-electrolytes (Johansson et al., 2000).
A group of transcripts abundantly expressed during early, but also late development is predicted to function in cell wall biogenesis and modification. In the first phase of development, among others transcripts for glycoside hydrolase, xyloglucan:xyloglucosyl transferase and expansin proteins were identified which are potentially involved in the reorganization of pectins, components of the primary cell wall (Xu et al., 2004; Cosgrove, 2005; http://afmb.cnrs-mrs.fr/CAZY/).
Activities in stress defence and pathogenesis response are indicated by the transcription of a major latex protein (Nessler et al., 1985), an osmotin- and a thaumatin-like protein, as well as Bet v I allergen. Preferential transcription of gene products such as an aspartyl protease, a nucellin member (Chen and Foolad, 1997), and two transcripts with deduced U-box domains characteristic of the ubiquitination machinery (Azevedo et al., 2001) suggest high protein degradation.
Among gene products involved in signal transduction, five transcripts for phosphate-responsive proteins were preferentially expressed in the early phase, and a similar gene product, EXO, has been associated with cell division in meristem development (Farrar et al., 2003).
Metabolism-related transcripts were represented by three which encode mutase family proteins catalysing the formation of C–P bonds in phosphonates (Anzai et al., 1987). Analogues are known from carnation, but their functions are as yet unclear (Wang et al., 1993). These transcripts also contain a signature of isocitrate lyase, an enzyme of the glyoxylate cycle. The transcript for isocitrate lyase correlated with seedling vigour in sugar beet (de los Reyes et al., 2003). In the study cited, germination-related transcripts participate in lipid metabolism. Also cluster 1 contains eight transcripts expressed at a relatively low level for enzymes predicted to be involved in fatty acid oxidation (see Supplementary Table 2 at JXB online). Additionally, the presence of three different transcripts for adenosylhomocysteinase, and 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase indicates the relevance of methionine biosynthesis in this developmental period. The amino acid regulates one-carbon metabolism and methylation capacity (Mudd and Datko, 1990; Ranocha et al., 2001; Kocsis et al., 2003). The importance of one-carbon metabolism is further supported by the high expression of the transcript encoding GTP cyclohydrolase, a key enzyme in the synthesis of the C1-unit transferring cofactor folate (Hanson and Gregory, 2002).
Preferential expression of genes during late development indicates activity in protein biosynthesis, plant growth factor metabolism, cell wall biogenesis, and defence responses
Transcripts for ribosomal proteins and translation initiation factors associated with ribosome and protein biosynthesis were over-represented in cl2 (Table 2B). Also genes involved in plant growth-factor regulation such as gibberellin 20-oxidase (Fleet and Sun, 2005), zeaxanthin epoxidase catalysing a crucial step in ABA synthesis (Seo and Koshiba, 2002), and two ethylene-responsive transcriptional coactivators belong to the over-represented group.
Transcripts for gene products related to signal transduction, defence, stress, protein degradation, and cell-wall biogenesis were of similar abundance to cl1, even if the composition of the group was different. During late development, transcription of a presumed vacuolar processing enzyme, a cysteine proteinase known to be induced in senescing organs associated with programmed cell death (Rojo et al., 2004; Hara-Nishimura et al., 2005; Thompson and Vierstra, 2005), showed abundant expression. The same pattern was evident for a transcript encoding glutamate-cysteine ligase/gamma-glutamylcysteine synthetase which catalyses a rate-limiting step in the biosynthesis of the antioxidant glutathione (Noctor et al., 2002). A transcript for a putative galactinol synthase involved in the heat stress-dependent synthesis of the trisaccharide raffinose (Panikulangara et al., 2004) showed high absolute and relative expression in cl2.
Four of the most overrepresented transcripts during late development encode chlorophyll A-B binding proteins.
Transcription profile of the inner tap-root zone reflects high protein turnover whereas transcriptional activity in the outer zone is more in line with exposure to pathogens and light
Among preferentially core-expressed transcripts, 240 were annotated with expected values below e–20; among transcripts with preferential expression in the outer beet, 97 fulfilled this criterion (Table 1B). Transcripts with differential expression in the inner and outer beet region were ranked according to their expression value ratios, and the 40 most extreme were grouped according to their predicted functions in Table 3.
|
Both beet regions shared gene products in most functional categories, but there were a few exceptions. The category of proteolysis including a cullin and an F-box protein, as well as an ubiquitin-conjugating enzyme and an ATP-dependent Clp protease, is unique to the inner zone. The associated protein turnover may be refilling substrate pools required for reorganization of metabolic activity also involving transcription and translation. Transcriptional activity is indicated by the presence of transcription-factor domains such as zinc finger- or MADS-box elements in signal transduction-related gene products preferentially expressed in the inner zone. Transcripts encoding ribosomal proteins were also over-represented at moderate levels (see Supplementary Table 2 at JXB online), and a putative transcript for 1-phosphatidylinositol-4-phosphate 5-kinase regulating development of procambial cells (Elge et al., 2001) was found preferentially in the inner region.
Among the most prominent gene products in the outer beet region there are some related to chloroplast or photosynthetic activity and plant growth factor regulation. Transcripts encoding gibberellin 2-oxidase and a gibberellin-responsive protein, both members of cluster 1, point to a particular role for this growth factor. Further signal transduction-related transcripts in the outer region comprise a group of phosphatases with functions at the post-translational level.
Transcripts encoding transporters are present in both zones. However, those encoding multi-drug transporters such as MATE efflux (Brown et al., 1999; Eckardt, 2001) and ABC1 proteins (Martinoia et al., 2002) possibly involved in detoxification, were only found in the external root zone, whereas a sugar transporter was identified in the inner sucrose-rich region.
In both the inner and the outer part of the beet, transcripts involved in cytoskeleton formation, vesicular trafficking, and protein secretion, as well as those related to defence and stress, were identified. Gene products involved in cell-wall biogenesis showed a similar distribution, but the annotations point to different functional contexts: whereas in the inner region annotations for phosphomannomutase or cinnamoyl-CoA reductase indicated biosynthetic activities in cell-wall differentiation, the outer root zone contained transcripts for a glycoside hydrolase family 28 protein with polygalacturonase activity, and a glycosyl hydrolase family 18 protein with chitinase activity, both associated with processes such as degradation, reorganization, and possibly defence.
It is noticeable that, among the genes preferentially expressed in the outer part of the mature beet, more cl1 members than cl2 members were present (Table 3). In the complete set of genes transcriptionally monitored (see Supplementary Table 2 at JXB online), the ratio of cl1:cl2 transcripts in the outer region was 22:5, whereas the ratio for the inner region was nearly even with 14 cl1 and 13 cl2 transcripts.
Genes involved in carbohydrate metabolism reveal preferential transcription during early beet development and in the core of the tap root
Based on their predicted function, 32 differentially expressed genes were assigned to glycolysis, the oxidative pentose phosphate cycle, the reductive tricarboxylic acid cycle, or to the biosynthesis pathway of sugars (Table 4).
|
Transcripts encoding hexokinase 1, two types of phosphofructokinase, fructose-bisphosphate aldolase, and glyceraldehyde 3-phosphate dehydrogenase were found co-regulated in cluster 1 with moderate absolute and relative expression values. Transcripts for dihydrolipoamide S-acetyltransferase, a component of the pyruvate dehydrogenase complex, and ATP-citrate synthase, aconitate hydratase, and malate dehydrogenase, followed a similar transcription pattern. The latter three enzymes act in the reductive tricarboxylic acid cycle, and aconitate hydratase is also a key enzyme of the glyoxylate pathway. Preferential expression during early development was also found for the transcripts encoding five different members of a mutase family characterized by the isocitrate lyase signature. Isocitrate lyase is a further key enzyme of the glyoxylate pathway but, contrary to the aconitate hydratase transcript, most of the mutase transcripts were about 2.5-fold more strongly expressed in the outer region of maturing beets than in the core.
Cl1-transcription patterns were found for an alcohol dehydrogenase, indicating alcoholic fermentation in young beet tissue, 6-phosphogluconate dehydrogenase, an enzyme typical of the oxidative pentose phosphate cycle, and sucrose synthase 2 (Bellin et al., 2002; BQ490013 [GenBank] ), a candidate not contained in the unique cDNA set, but added to the macroarray analysis whose expression pattern was confirmed by semi-quantitative RT-PCR (Fig. 2C). With additional transcripts for two sugar transporters and a sugar isomerase domain-containing gene product, cluster 1 included the majority of putatively carbohydrate metabolism-related transcripts, whereby the transcripts for the sugar isomerase domain-containing protein and one transporter were also preferentially expressed in the core of the beet. Among transcripts with preferential expression in the inner part of the beet, a second putative glyceraldehyde 3-phosphate dehydrogenase and an enolase were identified. Transcripts for a further malate dehydrogenase and succinyl-CoA ligase followed a similar profile, as did those for the preferentially beet-expressed sucrose synthase SBSS1 (Hesse and Willmitzer, 1996; Bellin et al., 2002), a fructose-1, 6-bisphosphatase and a fructose-6-phosphate 2-kinase/fructose-2, 6-bisphosphatase, key enzymes involved in sucrose catabolism and glucose (re)synthesis important in regulating sink strength (Nielsen et al., 2004). A transcript for a possible pyruvate decarboxylase involved in fermentation and a further sugar transporter showed a similar regulation.
Concerning invertases regulating sink strength (Sturm and Tang, 1999), three neutral isoforms and a further beta-fructofuranosidase produced data which could be analysed, but no differential expression was found. The same was true for two transcripts encoding sucrose phosphate synthase catalysing the final step in sucrose biosynthesis.
In addition to genes with a role in carbohydrate metabolism, the majority of differentially expressed genes with predicted functions in the metabolism of amino and fatty acids, as well as in lipid degradation, were identified in cl1 (see Supplementary Table 2 at JXB online). Among the 40 most expressed transcripts there were two for a 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase, an essential enzyme in methionine biosynthesis, and a putative 3-hydroxy-3-methylglutaryl-CoA reductase 1 catalysing a key step in the synthesis of the isoprenoid precursor mevalonic acid (Table 2A).
The level of transcription of carbohydrate metabolism-related genes was in agreement with the expression of 11 genes involved in fatty and amino acid metabolism in the inner beet region (see Supplementary Table 2 at JXB online). The high preferential transcription of a putative acyl-CoA dehydrogenase-related gene product in the core of the beet indicated the importance of beta-oxidation of fatty acids in this tissue (Table 3).
Only two transcripts encoding gene products potentially involved in the oxidative pentose phosphate cycle, a transketolase and a glucosamine-6-phosphate isomerase/6-phosphogluconolactonase, showed preferential expression during the late developmental stages. The latter was strongly overexpressed in the outer beet region, in which also a plastidic dihydrolipoamide dehydrogenase 2, a sucrase, and the above-mentioned mutase family proteins were preferentially transcribed. During late beet development and in the peripheral, sucrose-poor region, no fatty or amino acid metabolism-related transcripts were identified, apart from two with predicted functions in lipid oxidation.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
In the present work, sugar beet samples of different developmental stages and topological zones were analysed with respect to growth characters, sucrose content, and typical transcript profiles. It was the aim to correlate changes in phenotypic characters, and, in particular, in sucrose content as a molecular phenotype, with alterations in transcript levels ultimately to identify the underlying candidate genes for the breeding process.
For better comparability of beet development between years which are subject to climatic differences, sampling dates were expressed as thermal time which accounts for the accumulated warmth as a main growth promoter. The rates of increase in shoot and root growth during the vegetation period (Fig. 1A–F) are in line with previous reports (Elliott and Weston, 1993). With the exception of mass, all measured characters rapidly increased in the first phase, indicating the immediate start of ring initiation and sucrose accumulation in beet development. Mass accumulated mainly at a later stage. Largely in line with these findings, no more than two expression clusters of transcripts with almost complementary expression patterns were identified in the time-course experiment (Fig. 1G, H). The transition from high to low expression or vice versa was around 1420 °Cd in the first year and 1293–1599 °Cd in the second year, when maximal values for most morphological characters, as well as for sucrose percentage, had almost been achieved.
As the beet sucrose concentration remains nearly constant from 1300 °Cd to 1400 °Cd onwards, the net increase in sugar appears to be due to growth and mass increase in the late phase and sucrose accumulation of the newly generated cells in the outer zone. Thus beet sink strength is maintained, and plant metabolism may be adjusted to further sucrose accumulation and to the imminent winter period at the end of the first season reflected by high protein turnover and biosynthesis activity, especially in the inner zone in late development. Plant growth factors, as indicated by the respective class in cl2, may assume regulatory functions of the developmental programme (Table 2B), and candidates encoding gene products involved in pathogen defence and senescence which were over-represented among the cl2 members indicate their increasing importance in maturing, sucrose-rich beets.
The notion that there is a spatial gradient in cell expansion and sucrose accumulation throughout the beet from inner to outer zones during development is complemented by the finding that transcripts belonging to cl1 are over-represented in the outer beet zone. This implies that this region may be developmentally delayed, a condition which is also indicated by the low sucrose content and as yet unexpanded cambial rings. In both young beets and the outer region of more mature beets, cells derived from the cambia are considered small and undifferentiated, but preparing for sucrose accumulation and expansion, a situation arguing for the expression of related genes. Following this rationale, the temporal gradient of transcript abundance in beet development may be partly superimposed on the spatial gradient. As a result, maturing beets contain transcripts typical of the innermost sucrose-rich cells and typical of differentiating sucrose-poor cells in the outer parts simultaneously. Based on this reasoning, selected processes and the gene products involved can be presented according to their transcript levels as follows.
An example of a factor typical of expanding cells, both during early development and in the maturing outer zone, appears to be gibberellin. This is indicated by preferential expression of two transcripts related to gibberellin synthesis and activity in both cl1 and the outer region (Tables 2, 3). Gibberellin-like activity also correlated well with the period of rapid cell expansion at the temporal scale (Elliott and Weston, 1993).
The nearly 9-fold increase in beet diameter between 500 °Cd and 1500 °Cd indicates a high growth rate and biosynthetic activity. This is reflected at the transcript level with high expression values for cl1 transcripts encoding gene products with functions in cytoskeletal reorganization, which is seen in the context of cell division and differentiation, and transport processes, which serve directed distribution of metabolites and cellular homeostasis in the case of membrane intrinsic proteins (Table 2A).
The largest number of differentially expressed beet transcripts encoding enzymes involved in primary carbohydrate metabolism is predicted to function in either the reductive tricarboxylic acid or the glyoxylate cycle, and shows moderate expression levels (Table 4). Both cycles are anaplerotic pathways refilling pools of substrates for biosynthetic reactions typical of actively growing tissue. Correspondingly, preferential expression is found in the young beets and/or the inner region, with the exception of four transcripts, showing similarity to mutase family proteins with an isocitrate lyase domain, which are preferentially expressed in the outer region of the root. This spatial type of expression makes it unlikely that the predicted enzymes participate in the same glyoxylate cycle. Either the identified gene products act in locally separated, but functionally similar pathways, or the mutase family proteins assume different functions (Anzai et al., 1987; Wang et al., 1993).
According to the model for sucrose accumulation in sugar-beet roots (Fieuw and Willenbrink, 1989), sucrose phosphate synthase and sucrose synthase play major roles in the regulation of sink strength; invertases are considered further candidates in this respect (Sturm and Tang, 1999). None of the two isoforms of sucrose phosphate synthase (Hesse et al., 1995) or the invertases showed preferential expression in the time-course analysis, and the possibility exists that regulation takes place at a different level in these cases. However, the sucrose synthase isoform which is preferentially transcribed in roots and inflorescences (accession number BQ490013; Bellin et al., 2002), showed the highest expression level during early beet development (Fig. 2C), correlating with the period of highest sucrose accumulation. In addition, a transcript for a sucrase-related protein, which is predicted to increase the sucrose flow by cleaving sucrose at the sink, is preferentially expressed in the outer region of the mature beet (Table 3). The preferentially root-expressed sucrose synthase isoform SBSS1 (accession numbers X81974 [GenBank] /BQ490130/BQ594479; Hesse and Willmitzer, 1996; Bellin et al., 2002) was over-represented in the inner sucrose-rich region at maturity (Fig. 2, Table 3). Its pattern of expression was shared by transcripts encoding fructose-1, 6-bisphosphatase and fructose-6-phosphate 2-kinase/fructose-2, 6-bisphosphatase, two further enzymes mediating sink strength (Nielsen et al., 2004).
With only low levels of transcripts related to carbohydrate metabolism, but high transcript levels for fibre biogenesis- and stress defence-related gene products, a similar situation was found in sugar-cane stems (Casu et al., 2003, 2004). Just one sugar transporter, PST2, showed high abundance in the maturing sucrose-rich stem and was associated with maintaining sugar fluxes into mature-stem sink tissues and regulating the osmotic equilibrium, an important function in tissues with high sucrose levels. Interestingly, three transcripts for sugar transporters have been found preferentially expressed in cluster 1 and/or the inner beet region, in agreement with the highest sucrose increase rate and/or the highest absolute sucrose content (Table 4). The particular relevance of the putative gene products identified in the present analysis, however, cannot be supported without their detailed functional analysis carried out at the cellular level.
Cluster-1 transcripts preferentially expressed in the outer zone of maturing beets like those encoding the cell wall proteins xyloglucan:xyloglucosyl transferase and three glycoside hydrolase family 28 proteins (Tables 2A, 3) may alternatively be boundary-specific with constant expression during development, but due to the decreasing proportion of the boundary zone in the growing beet, which means a low ratio of surface area to total volume, lower transcript levels may have been detected in whole mature beets used for the kinetic study. In line with this hypothesis, most of the cl1 genes in the outer zone are not characterized by high absolute expression, but by high relative transcript levels (Table 2A), indicating a strong decrease in measured transcripts caused by dilution from early to late stages.
The few transcripts common to the outer beet region and cl2 are assumed to be regulated based on the spatial exposition of the maturing beet to exogenous pathogens and to light. Apart from transcripts for two chitinases implicated in pathogen defence, transcripts were found for a subunit of ribulose-bisphosphate carboxylase and a chlorophyll A-B binding protein, both being induced by light in leaves (Lu et al., 2002).
The present work shows that monitoring molecular phenotypes such as the metabolite sucrose in combination with gene expression profiling can be an effective tool to associate genes with functions and to identify key genes in metabolic processes which may be relevant in the plant breeding process. In natural or experimental populations, transcript levels of individual genes can be treated as quantitative factors to be mapped as genetic loci, a concept first introduced as ‘genetical genomics’ (Jansen and Nap, 2001). If the expression level is indicative of an agronomically important phenotype trait, selection in the breeding process may be complemented by transcript profiling. Such effects were recently demonstrated for transcriptionally regulated genes in barley (Potokina et al., 2004, 2006). Quantitative trait analyses for sugar quality and yield are feasible (Weber et al., 1999, 2000; Schneider et al., 2002), and linkage analyses of the genes identified in the present work should determine their genetic relevance for the traits considered.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Supplementary data can be found at JXB online. Table 1: PCR conditions and fragment sizes of candidate transcripts used for validation on macro array results in semi-quantitative RT-PCR experiments shown in Fig. 2. Table 2: Complete data set on cDNA sequence annotations and macroarray expression values for 11 520 unique sugar beet cDNAs.
|
Acknowledgements |
|---|
The work was financially supported by the BMBF-funded SWEET GABI project (grant number 0312283E) which also funded DB and KS. We would like to thank Silke Möhring for providing the primers to amplify the candidate with the accession number BQ588168 and Susanne Horstmann-Schwarz for technical support.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Anzai H, Murakami T, Imai S, Satoh A, Nagaoka K, Thompson CJ. (1987) Transcriptional regulation of bialaphos biosynthesis in Streptomyces hygroscopicus. Journal of Bacteriology 169 3482–3488.
Azevedo C, Santos-Rosa MJ, Shirasu K. (2001) The U-box protein family in plants. Trends in Plant Science 6 354–358.[CrossRef][Web of Science][Medline]
Bankaitis VA, Malehorn DE, Emr SD, Greene R. (1989) The Saccharomyces cerevisiae SEC14 gene encodes a cytosolic factor that is required for transport of secretory proteins from the yeast Golgi complex. Journal of Cell Biology 108 1271–1281.
Bellin D, Werber M, Theis T, Schulz B, Weisshaar B, Schneider K. (2002) EST sequencing, annotation and macroarray transcriptome analysis identify preferentially root-expressed genes in sugar beet. Plant Biology 4 700–710.[CrossRef]
Bernard K, Auphan N, Granjeaud S, Victorero G, Schmitt-Verhulst AM, Jordan BR, Nguyen C. (1996) Multiplex messenger assay: simultaneous, quantitative measurement of expression of many genes in the context of T cell activation. Nucleic Acids Research 24 1435–1442.
Brown MH, Paulsen IT, Skurray RA. (1999) The multidrug efflux protein NorM is a prototype of a new family of transporters. Molecular Microbiology 31 394–395.[CrossRef][Web of Science][Medline]
Casu RE, Dimmock CM, Chapman SC, Grof CP, McIntyre CL, Bonnett GD, Manners JM. (2004) Identification of differentially expressed transcripts from maturing stem of sugarcane by in silico analysis of stem expressed sequence tags and gene expression profiling. Plant Molecular Biology 54 503–517.[CrossRef][Web of Science][Medline]
Casu RE, Grof CP, Rae AL, McIntyre CL, Dimmock CM, Manners JM. (2003) Identification of a novel sugar transporter homologue strongly expressed in maturing stem vascular tissues of sugarcane by expressed sequence tag and microarray analysis. Plant Molecular Biology 52 371–386.[CrossRef][Web of Science][Medline]
Chen F and Foolad MR. (1997) Molecular organization of a gene in barley which encodes a protein similar to aspartic protease and its specific expression in nucellar cells during degeneration. Plant Molecular Biology 35 821–831.[CrossRef][Web of Science][Medline]
Cosgrove DJ. (2005) Growth of the plant cell wall. Nature Reviews Molecular Cell Biology 6 850–861.[CrossRef][Web of Science][Medline]
de los Reyes BG, Myers SJ, McGrath JM. (2003) Differential induction of glyoxylate cycle enzymes by stress as a marker for seedling vigor in sugar beet (Beta vulgaris). Molecular Genetics and Genomics 269 692–698.[CrossRef][Web of Science][Medline]
Desprez T, Amselem J, Caboche M, Hofte H. (1998) Differential gene expression in Arabidopsis monitored using cDNA arrays. The Plant Journal 14 643–652.[CrossRef][Web of Science][Medline]
Eckardt NA. (2001) Move it on out with MATEs. The Plant Cell 13 1477–1480.
Elge S, Brearley C, Xia HJ, Kehr J, Xue HW, Mueller-Roeber B. (2001) An Arabidopsis inositol phospholipid kinase strongly expressed in procambial cells: synthesis of PtdIns(4, 5)P2 and PtdIns(3, 4, 5)P3 in insect cells by 5-phosphorylation of precursors. The Plant Journal 26 561–571.[CrossRef][Web of Science][Medline]
Elliott MC and Weston GD. (1993) Biology and physiology of the sugar-beet plant. In Cooke DA and Scott RK (Eds.). The sugar beet crop: science into practiceLondon Chapman and Hall pp. 37–66.
Ewing B and Green P. (1998) Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Research 8 186–194.
Ewing B, Hillier L, Wendl MC, Green P. (1998) Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Research 8 175–185.
Farrar K, Evans IM, Topping JF, Souter MA, Nielsen JE, Lindsey K. (2003) EXORDIUM: a gene expressed in proliferating cells and with a role in meristem function, identified by promoter trapping in Arabidopsis. The Plant Journal 33 61–73.[CrossRef][Web of Science][Medline]
Fieuw S and Willenbrink J. (1989) Sucrose synthase and sucrose phosphate synthase in sugar beet plants (Beta vulgaris L. ssp altissima). Journal of Plant Physiology 131 153–162.
Fieuw S and Willenbrink J. (1990) Sugar transport and sugar-metabolizing enzymes in sugar storage roots (Beta vulgaris ssp. altissima). Journal of Plant Physiology 137 216–223.
Fleet CM and Sun TP. (2005) A DELLAcate balance: the role of gibberellin in plant morphogenesis. Current Opinion in Plant Biology 8 77–85.[CrossRef][Web of Science][Medline]
Hanson AD and Gregory JF. (2002) Synthesis and turnover of folates in plants. Current Opinion in Plant Biology 5 244–249.[CrossRef][Web of Science][Medline]
Hara-Nishimura I, Hatsugai N, Nakaune S, Kuroyanagi M, Nishimura M. (2005) Vacuolar processing enzyme: an executor of plant cell death. Current Opinion in Plant Biology 8 404–408.[CrossRef][Web of Science][Medline]
Herwig R, Schulz B, Weisshaar B, et al. (2002) Construction of a ‘unigene’ cDNA clone set by oligonucleotide fingerprinting allows access to 25 000 potential sugar beet genes. The Plant Journal 32 845–857.[CrossRef][Web of Science][Medline]
Hesse H, Sonnewald U, Willmitzer L. (1995) Cloning and expression analysis of sucrose-phosphate synthase from sugar beet (Beta vulgaris L.). Molecular and General Genetics 247 515–520.
Hesse H and Willmitzer L. (1996) Expression analysis of a sucrose synthase gene from sugar beet (Beta vulgaris L.). Plant Molecular Biology 30 863–872.[CrossRef][Web of Science][Medline]
Jansen RC and Nap JP. (2001) Genetical genomics: the added value from segregation. Trends in Genetics 17 388–391.[CrossRef][Web of Science][Medline]
Johansson I, Karlsson M, Johanson U, Larsson C, Kjellbom P. (2000) The role of aquaporins in cellular and whole plant water balance. Biochimica et Biophysica Acta 1465 324–342.[Medline]
Kocsis MG, Ranocha P, Gage DA, et al. (2003) Insertional inactivation of the methionine S-methyltransferase gene eliminates the S-methylmethionine cycle and increases the methylation ratio. Plant Physiology 131 1808–1815.
Lu C, Koroleva OA, Farrar JF, Gallagher J, Pollock CJ, Tomos AD. (2002) Rubisco small subunit, chlorophyll a/b-binding protein and sucrose:fructan-6-fructosyl transferase gene expression and sugar status in single barley leaf cells in situ: cell type specificity and induction by light. Plant Physiology 130 1335–1348.
Martinoia E, Klein M, Geisler M, Bovet L, Forestier C, Kolukisaoglu U, Muller-Rober B, Schulz B. (2002) Multifunctionality of plant ABC transporters: more than just detoxifiers. Planta 214 345–355.[CrossRef][Web of Science][Medline]
Milford GFJ. (1973) The growth and development of the storage root of sugar beet. Annals of Applied Biology 75 427–438.
Mudd SH and Datko AH. (1990) The S-methylmethionine cycle in Lemna paucicostata. Plant Physiology 93 623–630.
Müller I, Wagner W, Volker A, Schellmann S, Nacry P, Kuttner F, Schwarz-Sommer Z, Mayer U, Jurgens G. (2003) Syntaxin specificity of cytokinesis in Arabidopsis. Nature Cell Biology 5 531–534.[CrossRef][Medline]
Nessler CL, Allen RD, Galewsky S. (1985) Identification and characterization of latex-specific proteins in opium poppy. Plant Physiology 79 499–504.
Nielsen TH, Rung JH, Villadsen D. (2004) Fructose-2, 6-bisphosphate: a traffic signal in plant metabolism. Trends in Plant Science 9 556–563.[CrossRef][Web of Science][Medline]
Noctor G, Gomez L, Vanacker H, Foyer CH. (2002) Interactions between biosynthesis, compartmentation and transport in the control of glutathione homeostasis and signalling. Journal of Experimental Botany 53 1283–1304.
Oufattole M, Park JH, Poxleitner M, Jiang L, Rogers JC. (2005) Selective membrane protein internalization accompanies movement from the endoplasmic reticulum to the protein storage vacuole pathway in Arabidopsis. The Plant Cell 17 3066–3080.
Panikulangara TJ, Eggers-Schumacher G, Wunderlich M, Stransky H, Schoffl F. (2004) Galactinol synthase1: a novel heat shock factor target gene responsible for heat-induced synthesis of raffinose family oligosaccharides in Arabidopsis. Plant Physiology 136 3148–3158.
Pool MR, Stumm J, Fulga TA, Sinning I, Dobberstein B. (2002) Distinct modes of signal recognition particle interaction with the ribosome. Science 297 1345–1348.
Potokina E, Caspers M, Prasad M, Kota R, Zhang J, Sreenivasulu Wang M, Graner A. (2004) Functional association between malting quality trait components and cDNA array based expression patterns in barley (Hordeum vulgare L.). Molecular Breeding 14 153–170.[CrossRef]
Potokina E, Prasad M, Malysheva L, Roder MS, Graner A. (2006) Expression genetics and haplotype analysis reveal cis regulation of serine carboxypeptidase I (Cxp1), a candidate gene for malting quality in barley (Hordeum vulgare L. Functional and Integrative Genomics 6 25–35.
Ranocha P, McNeil SD, Ziemak MJ, Li C, Tarczynski MC, Hanson AD. (2001) The S-methylmethionine cycle in angiosperms: ubiquity, antiquity and activity. The Plant Journal 25 575–584.[CrossRef][Web of Science][Medline]
Rojo E, Martin R, Carter C, et al. (2004) VPEgamma exhibits a caspase-like activity that contributes to defense against pathogens. Current Biology 14 1897–1906.[CrossRef][Web of Science][Medline]
Schneider K, Schäfer-Pregl R, Borchardt DC, Salamini F. (2002) Mapping QTLs for sucrose content, yield and quality in a sugar beet population fingerprinted by EST-related markers. Theoretical and Applied Genetics 104 1107–1113.[CrossRef][Web of Science][Medline]
Seo M and Koshiba T. (2002) Complex regulation of ABA biosynthesis in plants. Trends in Plant Science 7 41–48.[CrossRef][Web of Science][Medline]
Sturm A and Tang GQ. (1999) The sucrose-cleaving enzymes of plants are crucial for development, growth and carbon partitioning. Trends in Plant Science 4 401–407.[CrossRef][Web of Science][Medline]
Sturn A, Quackenbush J, Trajanoski Z. (2002) Genesis: cluster analysis of microarray data. Bioinformatics 18 207–208.
Telles G and da Silva F. (2001) Trimming and clustering sugarcane ESTs. Genetics and Molecular Biology 24 17–23.
Thompson AR and Vierstra RD. (2005) Autophagic recycling: lessons from yeast help define the process in plants. Current Opinion in Plant Biology 8 165–173.[CrossRef][Web of Science][Medline]
Thornley JHM and Johnson IR. (1990) Empirical threshold models of developmental rate: the temperature-sum rule. Plant and crop modelling: a mathematical approach to plant and crop physiologyOxford Clarendon Press pp. 166–172.
Wang H, Brandt AS, Woodson WR. (1993) A flower senescence-related mRNA from carnation encodes a novel protein related to enzymes involved in phosphonate biosynthesis. Plant Molecular Biology 22 719–724.[CrossRef][Web of Science][Medline]
Weber WE, Borchardt DC, Koch G. (1999) Combined linkage maps and QTLs in sugar beet (Beta vulgaris L.) from different populations. Plant Breeding 118 193–204.
Weber WE, Borchardt DC, Koch G. (2000) Marker analysis for quantitative traits in sugar beet. Plant Breeding 119 97–106.[CrossRef]
Willenbrink J, Doll S, Getz HP, Meyer S. (1984) Zuckeraufnahme in isolierten Vakuolen und Protoplasten aus dem Speichergewebe von Beta-Rüben. Berichte der Deutschen Botanischen Gesellschaft 97 27–39.
Xu Z, Escamilla-Trevino L, Zeng L, et al. (2004) Functional genomic analysis of Arabidopsis thaliana glycoside hydrolase family 1. Plant Molecular Biology 55 343–367.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||