| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Plant Physiology 134:67-80 (2004) © 2004 American Society of Plant Biologists The Arabidopsis Root Transcriptome by Serial Analysis of Gene Expression. Gene Identification Using the Genome Sequence1Biochimie et Physiologie Moléculaire des Plantes, Unité Mixte de Recherche 5004, Agro-M/Centre National de la Recherche Scientifique/Institut National de la Recherche Agronomique/UM2, Place Viala, 34060 Montpellier cedex 1, France (C.F., S.M., C.C., P.N., J.B., F.G., P.D., H.S., A.G.); Institut de Génétique Humaine, Centre National de la Recherche Scientifique Unité Propre de Recherche 1142, University of Montpellier II, Place Eugène Bataillon, 34095 Montpellier cedex 5, France (D.P., T.C., J.M.); and Génome et Développement des Plantes, Unité Mixte de Recherche 5096, Centre National de la Recherche Scientifique/Institut de Recherche en Développement/University of Perpignan, 52 Av de Villeneuve, 66860 Perpignan cedex, France (V.D., R.C.)
Large-scale identification of genes expressed in roots of the model plant Arabidopsis was performed by serial analysis of gene expression (SAGE), on a total of 144,083 sequenced tags, representing at least 15,964 different mRNAs. For tag to gene assignment, we developed a computational approach based on 26,620 genes annotated from the complete sequence of the genome. The procedure selected warrants the identification of the genes corresponding to the majority of the tags found experimentally, with a high level of reliability, and provides a reference database for SAGE studies in Arabidopsis. This new resource allowed us to characterize the expression of more than 3,000 genes, for which there is no expressed sequence tag (EST) or cDNA in the databases. Moreover, 85% of the tags were specific for one gene. To illustrate this advantage of SAGE for functional genomics, we show that our data allow an unambiguous analysis of most of the individual genes belonging to 12 different ion transporter multigene families. These results indicate that, compared with EST-based tag to gene assignment, the use of the annotated genome sequence greatly improves gene identification in SAGE studies. However, more than 6,000 different tags remained with no gene match, suggesting that a significant proportion of transcripts present in the roots originate from yet unknown or wrongly annotated genes. The root transcriptome characterized in this study markedly differs from those obtained in other organs, and provides a unique resource for investigating the functional specificities of the root system. As an example of the use of SAGE for transcript profiling in Arabidopsis, we report here the identification of 270 genes differentially expressed between roots of plants grown either with NO3- or NH4NO3 as N source.
Serial analysis of gene expression (SAGE) is a sequence-based approach allowing the identification of a large number of transcripts present in tissues and the quantitative comparison of transcriptomes (Velculescu et al., 1995  ). The principle of SAGE is to generate a short specific tag (14 bp) from each mRNA present in a sample, resulting in the production of a SAGE tags library representative of this sample. The sequencing of these tags allows a high-throughput determination of their frequencies in the library, which are correlated with the relative amounts of the corresponding mRNAs. Thus, thousands of different transcripts can be analyzed, with a high specificity and most importantly, without any a priori on their identity. SAGE has proven to be a very powerful and robust method for investigating gene expression at the whole-genome scale (Velculescu et al., 1997; Boon et al., 2002; Liang, 2002) and to reflect the actual relative contents of mRNAs in a sample (Chrast et al., 2000, and refs. therein; Piquemal et al., 2002; Jung et al., 2003; Matsumura et al., 2003). As compared with cDNA arrays or oligochips, it has several advantages, such as the possibility to perform transcript profiling without the need of large technological investments, and the ability to obtain comprehensive transcriptomes from minute amounts of RNA (Virlon et al., 1999). The SAGE technology has been used extensively with animal systems, and more particularly in cancer research, where several hundred libraries and nearly 7 million SAGE tags have been obtained (Boon et al., 2002). Despite these developments, only very few studies have employed this methodology for transcript profiling in higher plants (Matsumura et al., 1999, 2003; Lorenz and Dean, 2002), and the first reports on SAGE in the model plant species Arabidopsis appeared only very recently (Jung et al., 2003; Lee and Lee, 2003).
At present, a major limitation of SAGE is that in most species, tag to gene assignment (e.g. the identification of the gene the transcript of which has generated the SAGE tag) is based on EST clusters or on available cDNA sequences. This results in very incomplete identification of the transcripts revealed by SAGE tags, leaving many of them without any match in the databases (Lash et al., 2000
The first aim of our work was to overcome these strong limitations and to improve gene identification in SAGE using the annotated sequence of the Arabidopsis genome. To perform tag to gene assignment, the strategy was to generate in silico a reference database containing the virtual SAGE tags corresponding to the 26,620 Arabidopsis genes annotated in Munich Information Center for Protein Sequences (MIPS; http://mips.gsf.de). This database is thus expected to associate all SAGE tags that can be theoretically found in Arabidopsis transcriptomes with their related genes. The difficulty is that correct identification of virtual SAGE tags requires accurate determination of both 5'- and 3'-UTRs, which are not easily predictable from the genome sequence in plants (Zhu et al., 2003
Our second objective was to use the SAGE technology to obtain a much more comprehensive and exhaustive view of the genes expressed in the roots of higher plants. We focused our analysis on these organs because they are strongly under-represented in EST or cDNA libraries available for Arabidopsis (Seki et al., 2002
Summary of Sequenced SAGE Libraries Six SAGE libraries were generated from roots of Arabidopsis plants (ecotype Col-0) subjected to various mineral nutrient supplies. After elimination of sequences from linkers and vector, duplicate ditags, and ditags shorter than 20 bp, a total of 144,083 tags was obtained from the six combined libraries, representing 52,078 different SAGE tag species. Among these, 15,964 were detected more than once and up to 830 times (the whole set of data is available at http://genoplante-info.infobiogen.fr). Tags found only once were not retained for further analysis, due to the possibility of sequencing errors generating artifactual tags. As usual with SAGE, most of the unique tag species were present at low frequency (Table I). However, 80 different tags were at copy number of 100 or higher, and more than 4,700 tags were found at least five times.
To identify the genes corresponding to the 15,964 different tags found at least twice, we first used EST clusters from both The Institute for Genomic Research (TIGR; http://www.tigr.org, October 2002 release) and Unigene (http://www.ncbi.nlm.nih.gov/UniGene/UGOrg.cgi?TAXID=3702, November 2002 release) databases. From each EST cluster reported either in TIGR or Unigene, we extracted the virtual tag corresponding to the 10 bp downstream from the MboI recognition site (GATC) closest to the 3' end. This generated 31,002 and 24,456 virtual tags from TIGR and Unigene EST clusters, respectively (Table II). About 90% of these virtual tags were specific for a unique gene. However, when comparing these virtual tags to those found experimentally, a majority of the 15,964 different experimental tags did not match any EST cluster (56% and 60% for TIGR and Unigene, respectively; Table II). This finding is not unusual, and similar proportions of unmatched experimental tags were found in other species using EST clusters (see the introduction). This represents a strong limitation to the interpretation of SAGE data. Furthermore, many EST clusters do not correspond to full-length cDNAs. As a consequence, there is a risk of generating tag to gene mismatch because the virtual tag extracted from the EST cluster may not correspond to the actual SAGE tag of the gene (e.g. that corresponding to the MboI site closest to the 3' end of the transcript). For all of these reasons, the procedure of tag to gene assignment using EST clusters was found to be unsatisfactory.
To improve, both quantitatively and qualitatively, gene identification in SAGE libraries, we chose to develop an alternative procedure using the annotated sequence of the Arabidopsis genome (ftp://ftpmips.gsf.de/cress/arabidna/arabi_genomicplus500_v111102.gz). The strategy is summarized in Figure 1. It involves gene identification by matching experimental tags with the virtual ones expected to be generated from all predictable mRNAs. In brief, 10 different virtual tags databases were generated (see "Materials and Methods"), depending on the sizes of the 5'- and 3'-UTRs that were considered for generating virtual cDNAs and on the way virtual tags were extracted from these virtual cDNAs (only the last tag for "exclusive" lists or all tags between the last one in the open reading frame [ORF] and the last one in the 3'-UTR for "cumulative" lists). The total number of virtual tags obtained ranged from 24,432 to 24,999 in the exclusive lists and from 32,052 to 59,725 in the cumulative lists when increasing the UTR length from 100 to 500 bp (Table III). The proportion of virtual tags found in the 3'-UTR strongly depends on the size of this UTR (Table IV), illustrating the fact that the various options investigated led to markedly different databases.
Among the 10 different lists of virtual tags generated by the strategy described above, one was selected to yield the reference database for high-throughput tag to gene assignment in our SAGE libraries. Three criteria were used for this selection: lowest level of unmatched tags (experimental tags with no match in the list of virtual tags), lowest level of nonspecific tags (experimental tags matching several genes), and highest level of reliability (correct tag to gene assignment, determined on a subset of genes for which cDNA sequences were available, see below). When all experimental tags (15,964 different) were matched against the virtual tag lists, marked differences were observed concerning the proportion of unmatched tags, depending on the virtual tag list (Fig. 2A). In all cases, the cumulative virtual tag lists yielded lower levels of unmatched tags, decreasing from 60% to 41% with increasing UTR length from 100 to 500 bp. With exclusive virtual tag lists, this proportion was always higher than 58%.
Considering the proportion of nonspecific tags (Fig. 2B), the tendency was reversed with better results obtained with the exclusive virtual tag lists. However, this proportion remained lower than 17% with the cumulative virtual tag lists, indicating that the large majority of experimental tags matching a virtual tag in these lists can be unambiguously associated with one single gene. Finally, the reliability of the tag to gene assignment was determined using a subset of genes for which cDNA sequences are available and experimentally confirmed. To perform this analysis, we selected 5,430 cDNA sequences (ecotype Col-0; http://signal.salk.edu), each corresponding to one gene identifier, according to the Arabidopsis Genome Initiative (AGI;http://www.arabidopsis.org/info/agi.jsp)uniformed nomenclature, and for which at least one MboI site was found. The actual SAGE tags deduced from these 5,430 cDNA sequences were compared with the virtual tags attributed to the corresponding genes in the 10 different virtual tag lists. An experimental tag was considered to be correctly assigned to a gene with our procedure when the actual tag of this gene (deduced from the cDNA sequence) is found in the virtual tag list associated with this gene. The results of this comparison led to drastically different conclusions between exclusive and cumulative virtual tag lists (Fig. 2C). For exclusive lists, the proportion of tags correctly assigned increased from 69% to 74%, with UTR length increasing from 100 to 200 bp, and markedly decreased to 37% when a 500-bp UTR was considered. In contrast, the reliability of gene identification with cumulative tag lists increased with increasing UTR length, to reach a maximum of 88% for 400- or 500-bp UTRs. The results of this analysis clearly indicate that a compromise has to be found between the lowest level of unmatched tags and the best reliability on the one hand (obtained when gene identification is performed using cumulative virtual tag lists; Fig. 2, A and C) and the lowest level of nonspecific tags on the other hand (obtained with exclusive virtual tag lists; Fig. 2B). We favored the objective of correct tag to gene assignment, and hence selected the cumulative virtual tag list generated with a 400-bp UTR length for building the reference database. The reasons for this precise choice is that considering 3'-UTRs shorter than 400 bp did not allow us to reach the maximum level of correct tag to gene assignment (Fig. 2C), whereas UTRs of 500 bp instead of 400 bp significantly increased the proportion of nonspecific tags (Fig. 2B). Our reference database for SAGE in Arabidopsis is available at http://genoplante-info.infobiogen.fr. It includes virtual tags for 26,490 genes (130 genes among the 26,620 annotated in MIPS had no MboI site in the cDNA sequence predicted with 400-bp UTRs).
The virtual tag reference database selected above was used to decipher the 15,964 different experimental tags identified from our SAGE libraries. From this total, 9,155 tags were assigned to one or several genes (43% of unmatched tags; Fig. 2A), with 7,776 of them matching one single gene (85% specificity; Fig. 2B). In this case, the gene identification was correct at 88% (Fig. 2C). The full list of these 7,776 genes can be viewed at http://genoplante-info.infobiogen.fr. Among them, 2,970 did not match any EST or cDNA in Unigene (November 2002 release). These genes were previously identified using gene prediction programs only, and our SAGE data thus provide strong experimental evidence for their expression in Arabidopsis. Eighty tags were found at a copy number of 100 or higher, with a majority of them (65) assigned to one or several genes (Table V). Eleven tags match several genes, but in most cases (8/11), this corresponds to two or three closely related genes, with the same function (Table V). The genes with highest levels of expression encode a late embryogenesis abundant protein (At4g02380), a thioglucosidase precursor (At3g09260), and an extensin (At1g21310). Various genes encoding ribosomal proteins and ubiquitin-related proteins were also found to be highly represented.
Interestingly, only 15 tags of the 80 with copy numbers of 100 or more had no match in the virtual tag database, which is much lower than expected from the proportion of unmatched tags estimated with the whole set of experimental tags (43%; Fig. 2A). The proportion of unmatched tags was clearly dependent on the tag copy number, decreasing from 57% for tags present only twice to 18% for those found more than 20 times (Fig. 3). In contrast, the proportion of nonspecific tags was not affected by the tag copy number (data not shown).
To illustrate the power of SAGE for discriminating between various members of multigene families, the experimental tags corresponding to ion transporter or channel genes were compiled from the six SAGE libraries (Table VI). The families investigated included genes encoding nitrate, ammonium, sulfate, phosphate, potassium and iron transporters, and potassium channels. More than 50 of these genes were found to be expressed in roots. In most instances, the experimental tag was specific for one gene in the family (Table VI), even in cases where a high sequence homology was found between two or more members. The two most highly expressed genes were NRT2.1, encoding a high-affinity NO3- transporter, and At1g32450, a putative NO3- transporter, member of the large NRT1/PTR family.
To investigate the usefulness of SAGE for identifying global modifications of gene expression triggered by environmental changes, we compared two libraries obtained from plants grown with either NO3- or NH4NO3 as N source (Fig. 4). The numbers of sequenced tags were 20,886 and 31,354 for the NO3- and NH4NO3 libraries, respectively. This yielded a total of 6,715 different tags found at least twice in the two combined libraries, among which 4,001 matched one single gene. Statistical analysis of the data indicated that 270 of these 4,001 genes were differentially expressed at P < 0.01 between NO3- and NH4NO3 libraries. A large variety of functions are associated with these genes, and a significant proportion of them (26%) encode putative, hypothetical, or unknown proteins. As a limited example, Table VII presents a subset of selected genes for which a response was found between the two conditions (the whole set of data is available at http://genoplanteinfo.infobiogen.fr). These include genes involved in nitrogen metabolism, carbon metabolism, and water transport. Gene expression of the NR1 isoform of nitrate reductase was strongly repressed on NH4NO3 as compared with NO3-, as is also the case for the high-affinity nitrate transporter NRT2.1. In contrast, transcript accumulation of two Glu dehydrogenase isoforms increased when NH4+ was present in addition to NO3- in the nutrient solution. Several genes encoding malate dehydrogenases, malic enzyme, and NAD-dependent isocitrate dehydrogenase were also induced in presence of NH4+. Finally, at least four genes encoding aquaporins were found to be strongly overexpressed with NH4NO3 as a N source.
The Advantages of Tag to Gene Assignment Based on the Annotated Genome Sequence
Our initial attempts to perform gene identification for SAGE transcript profiling using EST/cDNA databases resulted in a large proportion of unmatched tags (Table II), as always reported in other species (Lee et al., 2002
Tag to gene assignment based on the exclusive virtual tag lists was found not to be satisfactory because less than 40% of the experimental tags could be assigned to one or several genes (Fig. 2A) and because this assignment was in many cases not in agreement with that performed using cDNA sequences, especially when UTRs longer than 200 bp were considered (Fig. 2C). Interestingly, the best results with exclusive lists were obtained with 200-bp UTRs (Fig. 2, A and C). This is consistent with the idea that the mean length of 3'-UTRs for Arabidopsis genes is 210 ± 95 bp (Mathé, 2000 Much more exhaustive and correct gene identification was obtained with cumulative virtual tag lists, when 3'-UTRs of 200 bp or longer were considered (Fig. 2, A and C). This can be explained by the fact that the actual tags of most genes are already present in those selected with 200-bp UTRs. On the other hand, including the virtual tags found between 200 and 500 bp after the stop codon in the list improved tag to gene assignment both qualitatively and quantitatively because it allowed us to take into account correctly genes with unusually long 3'-UTRs (Fig. 2, A and C). However, cumulative virtual tag lists gave less satisfactory results than exclusive ones concerning the specificity of tag to gene assignment (Fig. 2B). This is understandable because in cumulative lists, several virtual tags are generally associated with each gene (from 1.2 to 2.3 on average with 3'-UTR from 100 to 500 bp). Thus, this increases the possibility that an experimental tag matches more than one gene.
When considering all data, we made a compromise and chose the cumulative virtual tag list with 400-bp UTRs for merging with the list of experimental tags. In this case, the percentage of unassigned tags was limited to 43%. This allowed matching of 9,155 experimental SAGE tags to one or several genes, instead of 7,038 or 6,356 using TIGR or Unigene EST clusters, respectively. In agreement with other studies (Jones et al., 2001
Clearly, the use of the annotated sequence of the Arabidopsis genome markedly improves gene identification in SAGE, as found also in C. elegans or fruitfly (Pleasance et al., 2003
The finding of a fairly large proportion (more than 40%) of experimental tags that could not be assigned to any gene deserves detailed discussion. This was commonly observed in SAGE studies, even when using full genome sequence for gene identification (Pleasance et al., 2003
Artifactual tags can arise at many steps of the SAGE procedure. A first cause for generating artifactual tags is production of internal tags upstream from the correct SAGE tag. This may be due either to incomplete digestion by the anchoring enzyme (MboI in our case) or to mispriming of the oligo(dT) to internal stretches of poly(A) during cDNA synthesis (Welle et al., 1999
The second option, i.e. that unassigned tags are true tags originating from transcripts with an incorrect/absent virtual tag in the reference database, is much more likely. This may also have several causes. First, our virtual tag database is most certainly not fully correct, in particular because it relies on the assumption that 5'- or 3'-UTRs are not longer than 400 bp. In fact, this database matches that generated from the available sequences of 5,430 cDNAs for only 88% of the genes, suggesting that 12% of the virtual tags included in our list may be incorrect. However, it is not clear whether all selected cDNA sequences correspond to truly full-length cDNAs, leaving the possibility that wrong virtual tags can also be extracted from partial cDNA sequences. Second, errors in the Arabidopsis genome sequence can generate wrong virtual tags if the error is located in the GATC MboI site or in the downstream 10 bp. Polymorphism can also explain part of the lack of tag to gene assignment as well, if SNPs are located in the areas corresponding to SAGE tags. This should be limited in our case because we used Col-0 ecotype for our study, which is the one selected for sequencing the genome. Furthermore, polymorphism seems to be quite limited, even between different Arabidopsis ecotypes (Haas et al., 2002
Our set of 144,083 sequenced tags represents to date the largest set of SAGE data available in plants. It allowed us to identify 7,776 genes expressed in roots, with nearly 3,000 of them with no EST or cDNA match in Unigene. This constitutes an important advance in the characterization of the Arabidopsis root transcriptome. This transcriptome appears to be quite specific. At the exception of a few genes encoding late embryogenesis abundant protein, glutathione S-transferase, and methallothionein, none of the genes highly expressed in the roots (Table V) were found to be significantly represented in SAGE libraries of Arabidopsis leaves (Jung et al., 2003
As a first example to illustrate the usefulness of SAGE for transcript profiling in roots, with respect to the key function of these organs in nutrient acquisition from the soil, we show that SAGE allows specific investigation of the expression levels of the various members of ion transporter multigene families (Table VI). This point remains one of the shortcoming of investigations based on cDNA arrays. The two transporter genes with the highest expression levels, At1g08090 and At1g32450, belong to NRT families of putative nitrate transporters. At1g08090 is AtNrt2.1, which has been found to encode a major component of the high-affinity nitrate uptake system (Cerezo et al., 2001
The second example relates to the characterization of differential gene expression in roots in response to environmental changes such as a modification of the N source supplied to the plants. Among the various genes found to be differentially expressed between NO3- and NH4NO3 libraries, several were already known to be affected by the nature of the N source and allow at least partial validation of our data. For instance, NO3- uptake and assimilation is known to be markedly repressed in the presence of NH4+, and our observation that both nitrate reductase and the high-affinity nitrate transporter NRT2.1 were down-regulated at the transcript level in NH4NO3-grown plants (Table VII) is in complete agreement with previous studies with these genes (Vincentz et al., 1993
Plant Culture and RNA Isolation
The Arabidopsis plants (ecotype Col-0) were grown hydroponically as described previously (Lejay et al., 1999
To obtain the six SAGE libraries, we followed the SADE protocol (SAGE Adaptation for Downsized Extracts; a SAGE variant) described by Virlon et al. (1999 Final concatemers were cloned in pBluescript II KS(-) from Stratagene (La Jolla, CA), digested by EcoRV, dephosphorylated, and purified on agarose gel. Ligation was performed overnight at 16°C and ElectroMAX DH10B Escherichia coli cells (Invitrogen, Rockville, MD) were then used for transformation by electroporation. Plasmids were prepared using the R.E.A.L. Prep 96 Plasmid kit (Qiagen, Courtaboeuf, France). The six SAGE libraries were sequenced separately. Cycle sequencing reactions were carried out using the ABI PRISM Dye Terminator kit (Applied Biosystems, Foster City, CA), and the products were run on an Applied Biosystems Prism 3100 DNA sequencer.
The raw sequences obtained from concatemer clones were analyzed using PHRED (Ewing et al., 1997
The statistical analysis of SAGE data for identification of genes differentially expressed was performed as described by Piquemal et al. (2002
Upon request, all novel materials described in this publication will be made available in a timely manner for noncommercial research purposes.
In a very recent paper, Yamada et al. (2003
We gratefully acknowledge the technical assistance of Michèle Laudié and Christel Llauro for sequencing of the SAGE libraries. Received July 18, 2003; returned for revision September 7, 2003; accepted October 22, 2003.
www.plantphysiol.org/cgi/doi/10.1104/pp.103.030536.
1 The work was supported by Génoplante (project nos. Af 1999 064 and Bi 1999 065) and by the Montpellier LR Génopole.
2 These authors contributed equally to this work.
3 Present address: Unité de Génétique et d'Amélioration des Fruits et Légumes, UR 1052 Institut National de la Recherche Agronomique, Domaine St Maurice, BP 94, 84 143 Montfavet cedex, France. * Corresponding author; e-mail gojon{at}ensam.inra.fr; fax 33–4–67–52–57–37.
Aubourg S, Rouzé P (2001) Genome annotation. Plant Physiol Biochem 39: 181-193[CrossRef]
Audic S, Claverie JM (1997) The significance of digital gene expression profiles. Genome Res 7: 986-995 Beevers L, Hageman RH (1980) Nitrate and nitrite reduction. In BJ Miflin, ed, The Biochemistry of Plants, Vol 5. Academic Press, New York, pp 115-168 Boheler KR, Stern MD (2003) The new role of SAGE in gene discovery. Trends Biotechnol 21: 55-57[CrossRef][ISI][Medline]
Boon K, Osorio EC, Greenhut SF, Schaefer CF, Shoemaker J, Polyak K, Morin PJ, Buetow KH, Strausberg RL, De Souza SJ et al. (2002) An anatomy of normal and malignant gene expression. Proc Natl Acad Sci USA 99: 11287-11292
Cerezo M, Tillard P, Filleur S, Munos S, Daniel-Vedele F, Gojon A (2001) Major alterations of the regulation of root NO3- uptake are associated with the mutation of Nrt2.1 and Nrt2.2 genes in Arabidopsis. Plant Physiol 127: 262-271
Chen J, Sun M, Lee S, Zhou G, Rowley JD, Wang SM (2002) Identifying novel transcripts and novel genes in the human genome by using novel SAGE tags. Proc Natl Acad Sci USA 99: 12257-12262
Chrast R, Scott HS, Papasavvas MP, Rossier C, Antonarakis ES, Barras C, Davisson MT, Schmidt C, Estivill X, Dierssen M et al. (2000) The mouse brain transcriptome by SAGE: differences in gene expression between P30 brains of the partial trisomy 16 mouse model of Down syndrome (Ts65Dn) and normals. Genome Res 10: 2006-2021 Curie C, Alonso JM, Le Jean M, Ecker JR, Briat JF (2000) Involvement of NRAMP1 from Arabidopsis thaliana in iron transport. Biochem J 347: 749-755 Ewing B, Hillier LD, Wendl M, Green P (1997) Base-calling of automated sequencer traces using Phred. I. Accuracy assessment. Genome Res 8: 175-185 Filleur S, Dorbe MF, Cerezo M, Orsel M, Granier F, Gojon A, Daniel-Vedele F (2001) An Arabidopsis T-DNA mutant affected in Nrt2 genes is impaired in nitrate uptake. FEBS Lett 489: 220-224[CrossRef][ISI][Medline] Gansel X, Muños S, Tillard P, Gojon A (2001) Differential regulation of the NO3- and NH4+ transporter genes AtNrt2.1 and AtAmt1.1 in Arabidopsis: relation with long-distance and local controls by N status of the plant. Plant J 26: 143-155[CrossRef][ISI][Medline] Haas BJ, Volfovsky N, Town CD, Troukhan M, Alexandrov N, Feldmann KA, Flavell RB, White O, Salzberg SL (2002) Full-length messenger RNA sequences greatly improve genome annotation. Genome Biol 3: research 0029.1-0029.12 Hoarau J, Barthes L, Bousser A, Deléens E, Prioul JL (1996) Effect of nitrate on water transfer across roots of nitrogen pre-starved maize seedlings. Planta 200: 405-415
Jones SJ, Riddle DL, Pouzyrev AT, Velculescu VE, Hillier L, Eddy SR, Stricklin SL, Baillie DL, Waterston R, Marra MA (2001) Changes in gene expression associated with developmental arrest and longevity in Caenorhabditis elegans. Genome Res 11: 1346-1352 Jung SH, Lee JY, Lee DH (2003) Use of SAGE technology to reveal changes in gene expression in Arabidopsis leaves undergoing cold stress. Plant Mol Biol 52: 553-567[CrossRef][ISI][Medline]
Kal AJ, van Zonneveld AJ, Benes V, van den Berg M, Koerkamp MG, Albermann K, Strack N, Ruijter JM, Richter A, Dujon B et al. (1999) Dynamics of gene expression revealed by comparison of serial analysis of gene expression transcript profiles from yeast grown on two different carbon sources. Mol Biol Cell 10: 1859-1872 Lagarde D, Basset M, Lepetit M, Conejero G, Gaymard F, Astruc S, Grignon C (1996) Tissue specific expression of Arabidopsis AKT1 gene is consistent with a role in K+ nutrition. Plant J 9: 195-203[CrossRef][ISI][Medline]
Lash AE, Tolstoshev CM, Wagner L, Schuler GD, Strausberg RL, Riggins GJ, Altschul SF (2000) SAGEmap: a public gene expression resource. Genome Res 10: 1051-1060
Lee JY, Lee DH (2003) Use of serial analysis of gene expression technology to reveal changes in gene expression in Arabidopsis pollen undergoing cold stress. Plant Physiol 132: 517-529 Lee S, Clark T, Chen J, Zhou G, Scott LR, Rowley JD, Wang SM (2002) Correct identification of genes from serial analysis of gene expression tag sequences. Genomics 79: 598-602[CrossRef][ISI][Medline] Lejay L, Tillard P, Lepetit M, Olive F, Filleur S, Daniel-Vedele F, Gojon A (1999) Molecular and functional regulation of two NO3- uptake systems by N- and C-status of Arabidopsis plants. Plant J 18: 509-519[CrossRef][ISI][Medline]
Liang P (2002) SAGE Genie: a suite with panoramic view of gene expression. Proc Natl Acad Sci USA 99: 11547-11548 Lobreaux S, Massenet O, Briat JF (1992) Iron induces ferritin synthesis in maize plantlets. Plant Mol Biol 19: 563-575[CrossRef][ISI][Medline] Lorenz WW, Dean JF (2002) SAGE profiling and demonstration of differential gene expression along the axial developmental gradient of lignifying xylem in loblolly pine (Pinus taeda). Tree Physiol 22: 301-310[ISI][Medline]
Margulies EH, Kardia SL, Innis JW (2001) A comparative molecular analysis of developing mouse forelimbs and hindlimbs using serial analysis of gene expression (SAGE). Genome Res 11: 1686-1698 Mathé C (2000) Analyse in silico des gènes d'Arabidopsis thaliana: description, classification, et modélisation pour aider à la prédiction des gènes. PhD thesis. Paris VII University, Paris Matsumura H, Nirasawa S, Kiba A, Urasaki N, Saitoh H, Ito M, Kawai-Yamada M, Uchimiya H, Terauchi R (2003) Overexpression of Bax inhibitor suppresses the fungal elicitor-induced cell death in rice (Oryza sativa L.) cells. Plant J 33: 425-434[CrossRef][ISI][Medline] Matsumura H, Nirasawa S, Terauchi R (1999) Transcript profiling in rice (Oryza sativa L.) seedlings using serial analysis of gene expression (SAGE). Plant J 20: 719-726[CrossRef][ISI][Medline] Piquemal D, Commes T, Manchon L, Lejeune M, Ferraz C, Pugnere D, Demaille J, Elalouf J, Marti J (2002) Transcriptome analysis of monocytic leukemia cell differentiation. Genomics 80: 361-371[CrossRef][ISI][Medline]
Pleasance ED, Marra MA, Jones SJM (2003) Assessment of SAGE in transcript identification. Genome Res 13: 1203-1215 Saha S, Sparks AB, Rago C, Akmaev V, Wang CJ, Vogelstein B, Kinzler KW, Velculescu VE (2002) Using the transcriptome to annotate the genome. Nat Biotechnol 20: 508-512[CrossRef][ISI][Medline]
Seki M, Narusaka M, Kamiya A, Ishida J, Satou M, Sakurai T, Nakajima M, Enju A, Akiyama K, Oono Y et al. (2002) Functional annotation of a full-length Arabidopsis cDNA collection. Science 296: 141-145 Shibagaki N, Rose A, McDermott JP, Fujiwara T, Hayashi H, Yoneyama T, Davies J (2002) Selenate-resistant mutants of Arabidopsis thaliana identify Sultr1;2, a sulfate transporter required for efficient transport of sulfate into roots. Plant J 29: 475-486[CrossRef][ISI][Medline] Stitt M (1999) Nitrate regulation of metabolism and growth. Curr Opin Plant Biol 2: 178-186[CrossRef][ISI][Medline]
van den Berg A, van der Leij J, Poppema S (1999) Serial analysis of gene expression: rapid RT-PCR analysis of unknown SAGE tags. Nucleic Acids Res 27: e17
Velculescu VE, Zhang L, Vogelstein B, Kinzler KW (1995) Serial analysis of gene expression. Science 270: 484-487 Velculescu VE, Zhang L, Zhou W, Vogelstein J, Basrai MA, Bassett DE Jr, Hieter P, Vogelstein B, Kinzler KW (1997) Characterization of the yeast transcriptome. Cell 88: 243-251[CrossRef][ISI][Medline]
Vert G, Grotz N, Dedaldechamp F, Gaymard F, Guerinot ML, Briat JF, Curie C (2002) IRT1, an Arabidopsis transporter essential for iron uptake from the soil and for plant growth. Plant Cell 14: 1223-1233 Very AA, Sentenac H (2003) Molecular mechanisms and regulation of K+ transport in higher plants. Annu Rev Plant Biol 54: 575-603[CrossRef][Medline] Vincentz M, Moureaux T, Leydecker MT, Vaucheret H, Caboche M (1993) Regulation of nitrate and nitrite reductase expression in Nicotiana plumbaginifolia leaves by nitrogen and carbon metabolites. Plant J 3: 315-324[CrossRef][ISI][Medline]
Virlon B, Cheval L, Buhler JM, Billon E, Doucet AJ, Elalouf JM (1999) Serial microanalysis of renal transcriptomes. Proc Natl Acad Sci USA 96: 15286-15291 Wang SM (2003) Response: the new role of SAGE in gene discovery. Trends Biotechnol 21: 57-58[CrossRef]
Welle S, Bhatt K, Thornton CA (1999) Inventory of high-abundance mR NAs in skeletal muscle of normal men. Genome Res 9: 506-513
Zhang L, Zhou W, Velculescu VE, Kern SC, Hruban RH, Hamilton SR, Volgelstein B, Kinzler KW (1997) Gene expression profiles in normal and cancer cells. Science 276: 1268-1272 |
TOP
ABSTRACT
RESULTS
