JXB Advance Access originally published online on July 12, 2005
Journal of Experimental Botany 2005 56(419):2495-2505; doi:10.1093/jxb/eri242
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press [2005] on behalf of the Society for Experimental Biology.
RESEARCH PAPER |
Molecular analysis of Arabidopsis endosperm and embryo promoter trap lines: reporter-gene expression can result from T-DNA insertions in antisense orientation, in introns and in intergenic regions, in addition to sense insertion at the 5' end of genes
1Plant Molecular Biology Laboratory, Department of Plant and Environmental Sciences, University of Life Sciences, PO Box 5003, N-1432 Ås, Norway
2Department of Molecular Biosciences, University of Oslo, Oslo, PO Box 1041 Blindern, N-0316 Oslo, Norway
3School of Life Sciences, University of Skövde, SE-54128 Skövde, Sweden
To whom correspondence should be addressed. Fax: +47 2285 4443. E-mail: reidunn.aalen{at}imbv.uio.no
Received 12 January 2005; Accepted 4 June 2005
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Random insertions of promoterless reporter genes in genomes are a common tool for identifying marker lines with tissue-specific expression patterns. Such lines are assumed to reflect the activity of endogenous promoters and should facilitate the cloning of genes expressed in the corresponding tissues. To identify genes active in seed organs, plant DNA flanking T-DNA insertions (T-DNAs) have been cloned in 16 Arabidopsis thaliana GUS-reporter lines. T-DNAs were found in proximal promoter regions, 5' UTR or intron with GUS in the same (sense) orientation as the tagged gene, but contrary to expectations also in inverted orientation in the 5' end of genes or in intergenic regions. RT-PCR, northern analysis, and data on expression patterns of tagged genes, compared with the expression pattern of the reporter lines, suggest that the expression pattern of a reporter gene will reflect the pattern of a tagged gene when inserted in sense orientation in the 5' UTR or intron. When inserted in the promoter region, the reporter-gene expression patterns may be restricted compared with the endogenous gene. Among the trapped genes, the previously described nitrate transporter gene AtNRT1.1, the cyclophilin gene ROC3, and the histone deacetylase gene AtHD2C were found. Reporter-gene expression when positioned in antisense orientation, for example, in the SLEEPY1 gene, is indicative of antisense expression of the tagged gene. For T-DNAs found in intergenic regions, it is suggested that the reporter gene is transcribed from cryptic promoters or promoters of as yet unannotated genes.
Key words: Embryo, endosperm, GUS, promoter trapping
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Several approaches can be taken to identify genes involved in developmental processes in specific tissues. A classic forward genetics approach relies on the identification of observable mutant phenotypes. The success of this approach may be limited in two ways: a mutation that leads to an early lethal phenotype may mask gene function in later developmental stages, or a mutation may lead to no phenotype at all, due to the presence of additional genes, redundant in function. A more random approach is the identification of tissue-specific transcripts, for instance by differential screening of cDNA libraries. This method relies on the feasibility of isolating sufficient amounts of the tissues of interest. A third method, promoter-trapping, i.e. the random insertion of a promoterless reporter gene in a genome followed by screening for tissue-specific reporter-gene expression, has the advantage that it, in theory, can be used for any tissue at any developmental stage. In addition, this approach does not rely on the generation of mutant phenotypes.
The endosperm of the Arabidopsis seed is surrounded by the maternal seed coat and is eventually consumed by the developing embryo (Berger, 1999
, 2003; Brown et al., 1999; Olsen, 2001, 2004). Therefore endosperm tissue, as well as early developmental stages of the embryo, are very difficult to isolate, for example, for a differential screening approach. Extensive screening for mutant phenotypes is also demanding and will not pick up redundant genes. Promoter trapping has thus been used successfully in Arabidopsis to isolate reporter lines with tissue-specific expression in seeds (Topping et al., 1994; Topping and Lindsey, 1997; Casson et al., 2002; Stangeland et al., 2003). Theoretically, transcription of a promoterless reporter gene (e.g. ß-glucuronidase, GUS) inserted at a given position in the plant genome would indicate that a promoter has been trapped, and that the reporter gene is driven by regulatory elements normally governing an endogenous gene. A few tagged genes have been identified from these collections so far, i.e. POLARIS (PLS; Casson et al., 2002) and EXORDIUM (EXO; Farrar et al., 2003). Several enhancer trap lines with seed-specific GUS expression have also been reported (Vielle-Calzada et al., 2000), although without any detailed molecular description of the trapped enhancer elements or the integration patterns of the insertions relative to endogenous genes.
A collection of Arabidopsis promoter trap lines have been screened and a number of lines expressing GUS have been identified in the seed (Stangeland et al., 2003). The distinctive patterns of GUS expression in these lines provide molecular markers for embryo and endosperm compartments, as well as for other seed organs at all developmental stages. The lines have been divided into four main groups according to the GUS-expression patterns (Fig. 1, Table 1). GNOCCHI1/760 belongs to the first group and shows expression in the chalazal endosperm and embryo at the embryo globular stage and endosperm-specific GUS activity at early heart stage (Fig. 1A). The second group consists of the embryo-specific GUS markers LINGUINE1-6 (Fig. 1B, C). The third group was called SENAPE, and expressed GUS predominantly in the endosperm and embryo with some activity in other seed organs (Fig. 1D), and in some lines in somatic tissue as well. The markers of group 4 included lines with strong activity in the whole plant, for example, STINCO and PESCE (Fig. 1E, F), while PESTO/462 was preferentially expressed in pedicel stomata.
|
|
|
To identify seed-expressed Arabidopsis genes and their regulatory elements conferring this specificity, the position and pattern of integration of the T-DNA have been analysed in 16 of the lines. In some of them the T-DNA was found in an expected position, i.e. close to the transcription start site of a gene with the GUS reporter in the same direction as the trapped gene. However, in about half of the lines the T-DNA was inserted in reverse orientation with respect to a gene or was in intergenic regions. In the GNOCCHI1/760 line the T-DNA was inserted in an intron, resulting in a fusion transcript of the tagged gene and GUS, but not allowing a translational fusion between the encoded protein and GUS. Reverse transcriptase–polymerase chain reaction (RT-PCR), northern hybridization, and, for the tagged gene in the GNOCCHI1/760 line, a promoter–GUS fusion, was used to investigate the expression patterns of tagged genes and compare these with the GUS-expression patterns of the respective marker lines.
The analyses suggest that promoter trap screening may identify unannotated genes, cryptic promoters, or regulatory elements driving expression of antisense transcripts, as well as annotated genes displaying tissue-specific expression.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material and growth conditions
Plants were grown under conditions reported by Stangeland et al. (2003)
. Except for 13, 97, 134, and 462, each line contained a single T-DNA insert. In the former, genetic analyses and outcrossing identified the T-DNA inserts that co-segregated with the GUS activity.
Isolation and analysis of flanking sequences
Plant sequences adjacent to the T-DNA left and right borders were isolated as described previously (Meza et al., 2002), or by long-range inverse PCR as follows. Genomic DNA was digested with HindIII and circularized by ligation. Ligated DNA was incubated at 65 °C for 10 min and 94 °C for 30 s prior to PCR using Elongase®Enzyme Mix (Invitrogen). 2-STEP PCR with the primers RBREV1 5'-GCG ATC CAG ACT GA ATG CCC ACA GGC CG-3' or RBREV2 5'-TGA TTT CAC GGG TTG GGG TTT CTA CAG GAC GG-3' together with RBDIR 5'-GTT ACT AGA TCG GGA AGA TCC TCT AGA GTC GAC CTG CAG-3' was used for the isolation of plant sequences adjacent to the right T-DNA border of pMHA2, and LBREV 5'-CAA TCG TTG CGG TTC TGT CAG TTC C-3' and LBDIR 5'-GCG ACA ACA TGT CGA GGC TCA GC-3' for the isolation of plant DNA flanking the left pMHA2 T-DNA border by 3-STEP PCR. For the isolation of plant sequences adjacent to the right T-DNA border of p
GUSBin19, RBDIRBin19 5'-GGA ACA CGG CGG CAT CAG AGC AGC CGA TTG-3', RBREVBin19 5'-CCT TCC CTC CCT CAA TCG GTT GAA TGT CGC CC-3' and RBSEQBin19 5'-GCC TCT CCA CCC AAG CGG-3' were used.
Northern analysis
mRNA isolated from 100 mg of different Arabidopsis tissues using magnetic oligo d(T) beads (GenoPrepTM mRNA beads, GenoVision, Qiagen) was separated on denaturing RNA gels, blotted, and hybridized as described by Galau et al. (1986) using probes labelled with [32P]dCTP using the Random Primed DNA Labeling Kit (Roche Diagnostics GmbH, Mannheim) with biotinylated single-stranded templates bound to magnetic streptavidin-coated beads (GenoprepTM Streptavidin Beads, GenoVision). Alternatively, PCR fragments were labelled using Rediprime II Labelling Kit (Amersham Biosciences). The following gene-specific primers were used: 18S rDNA – NS3-F: 5'-GCA AGT CGT GTG CCA GCA GCC-3' and NS4-R: 5'-CTT CCG TCA ATT CCT TTA AG-3'; At5g05100 – 775-B, 5'-AAT CCG ATA CCG ATT CCG C-3' and 775-3' UTR, 5'-TGC TGA ATA CGA CCA AAA GG-3'; LDC – 5'-GGA CTT AGC CCT ATT TGC TCC ATC TCC CAG C-3' and 5'-GCT GGT AAT AAA GTC AGT TAC AAA GAT GCT GCT ATC G-3'; AtHD2C – 5'-TAG AAC CTT CTC CTT CTC CTT CGA CGC ATC GT-3' and 5'-TGG CAA CTG AAA GTT CAC CTG TTT CGC A-3'.
RT-PCR
Total RNA was isolated by Qiagen plant RNA isolation kit, and incubated for 30 min at 37 °C with RQ1 RNase-free DNase in M-MLV buffer prior to the RT reaction using M-MLV reverse transcriptase (Promega). The following primers were used: GUS – GUSDIR: 5'-TTA ACG ATC AGT TCG CCG ATG CAG AT-3' and GUSREV: 5'-TTC GCC CTT CAC TGC CAC TGA CC-3'; actin – ACT11F: 5'-AAC TTT CAACAC TCC TGC CAT G-3' and ACT11R: 5'-CTG CAA GGT CCA AAC GCA GA-3'; At2g16600 (ROC3) – 16RTDIR: 5'-CGG TGG CAA ATC CGC CG-3' and 16RTREV: 5'-CTG AGA TCC ATT CGT GTT CGC ACC-3'; At3g05120 – 134RTDIR: 5'-GCG AGC GAT GAA GTT AAT CTT ATT GAG AGC-3' and 134RTREV: 5'-GGG TAT GGA TTC TCT GGT GCA CGC-3'; line 760 fusion transcript – 760DIR: 5'-GCT GGT AAT AAA GTC AGT TAC AAA GAT GCT GCT ATC GAA CTC GGA ACC-3' for the first exon of At2g28305 and the GUS primer RBREV1 (see isolation of flanking sequences).
Promoter::GUS construct for the lysine decarboxylase gene At2g28305
The transgene was made by amplifying 1509 bp of the promoter region and 5' UTR of the lysine decarboxylase gene (pLDC) At2g28305 using primers 5'-ATG AAC CAA ATG TGG AAT TG-3' and 5'-ACA CAA AGT TTT GTT TTA AG-3' with attB1 and attB2 Gateway sequences in the 5' end, respectively. The PCR fragment was recombined into pDONR207 vector using Gateway Cloning Technology (Invitrogen Life Technologies) and further recombined into a pPZP211G-GAWI (Butenko et al., 2003) destination vector. The pLDC::GUS-pPZP211G construct, verified by sequencing, was introduced into Agrobacterium tumefaciens strain C58C1 pGV2260 and C24 Arabidopsis plants were transformed by floral dipping (Clough and Bent, 1998).
GUS assays
Histochemical GUS assays were performed according to Stangeland and Salehian (2002) or for the pLDC::GUS transformants according to Grini et al. (2002). Dissected ovules were observed with a Zeiss Axioplan Imaging2 microscope system equipped with Nomarski optics and a cooled LCD imaging facilities.
Bioinformatics
BLAST searches were run using NCBI (http://www.ncbi.nlm.nih.gov) BLAST and SiGnAL (http://signal.salk.edu) BLAST. Searches for open reading frames (ORFs) were performed using the EMBOSS Transeq tool (http://www.ebi.ac.uk/emboss/transeq/).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Identification of T-DNA insertion sites in promoter trap lines
The analysed lines were identified in a collection transformed with the pMHA2 (Mandal et al., 1995
) or pGUSBin19 (Topping et al., 1991) promoter trapping vectors (Fig. 2). Plant DNA fragments flanking the right and left T-DNA borders were cloned (see Materials and methods), and based on the analyses of the position and orientation of the T-DNA insertions (Table 1), the lines could be sorted into four categories (Fig. 3): (i) lines with T-DNA inserted in proximal promoter regions or 5' UTRs with the GUS gene in sense orientation relative to the tagged gene; (ii) lines with the GUS gene inserted in inverted orientation in the 5' end of genes; (iii) lines with T-DNA inserted in intergenic regions (IGRs); and (iv) lines with T-DNA inserted in introns with the GUS gene in sense orientation.
|
Reporter gene insertions in sense orientation in proximal promoter regions or 5' UTRs reflect, at least in part, the expression pattern of the trapped gene
In four lines (PESTO1/462, PESCE1/134, SENAPE2/16, and LINGUINE6/619), the T-DNAs were found in proximal promoter regions as indicated from cloned cDNAs, presumably upstream of the basal promoter of the tagged genes (Table 1, Fig. 3A). In the PESTO1/462 line, three tandemly repeated p
GUSBin19 T-DNAs behaved as a single Mendelian trait and co-segregated with the GUS activity. Plant DNA upstream of the At2g34310 gene was found flanking one of the three right borders. In the other lines a single T-DNA was found.
GUS expression in these lines was confirmed by northern hybridization (Fig. 4A), or in cases with weak expression levels, by RT-PCR (data not shown). The PESCE1/134 line, harbouring the pGUSBin19 T-DNA, exemplifies transcription start close to the T-DNA insertion point, i.e. based on the size of the GUS transcript (about 2.1 kb corresponding to the 1.8 kb CDS plus UTRs and polyA tail) (Fig. 4A). In this line the T-DNA was located upstream of the transcription start site of At3g05120, a gene of unknown function. RT-PCR showed that the tagged gene was active in wild-type (wt) plants (Fig. 4B). RT-PCR also amplified At3g05120 cDNA from mRNA of plants homozygous for the PESCE1/134 insertions (Fig. 4B), demonstrating that the T-DNA insertion in the proximal promoter region had not knocked out the expression of the At3g05120 endogenous gene. In SENAPE2/16, in which the pMHA2 T-DNA was integrated near the transcription start of the ROC3 gene encoding a cytosolic cyclophilin (Chou and Gasser, 1997), a similar situation was found, i.e. ROC3 was expressed in plants homozygous for the T-DNA insertion (Fig. 4C).
|
The pMHA2 T-DNA in LINGUINE6/619 line was inserted in the upstream region of the AtHD2C histone deacetylase gene (Wu et al., 2000
; Pandey et al., 2002). AtHD2C was shown by northern hybridization to be predominantly expressed in flowers, and weakly in siliques, seedlings, and leaves (Fig. 4D), although this reporter line did not show GUS expression in the latter two organs (data not shown). In order to investigate further the requirement of the AtHD2C in seed development, a line from the SALK Insertion Sequence Database (Alonso et al., http://signal.salk.edu/) reported to carry a T-DNA insertion in the seventh exon of AtHD2C was genotyped for the insertion (SALK_039784), and inspected phenotypically. However, no apparent phenotype was found in seeds or other organs in the homozygous plant identified. By contrast to the latter four lines, the T-DNA of the SENAPE4/775 line was integrated in a 5' UTR (Table 1; Fig. 3A). Northern hybridization using mRNA from different tissues showed the presence of the GUS transcript in the same tissues and in the same relative abundance as the At5g05100 transcripts in wt plants (shown for siliques and flowers in Fig. 4E). The expression level of At5g05100 was significantly lower in the siliques and flowers of the SENAPE4/775 line than in wt plants (Fig. 4E).
The T-DNA of LINGUINE2/13 was inserted upstream of the translation start site of At2g27660 (Table 1, Fig. 3A). Since only an expressed sequence tag (EST) is available for this gene, it is not known how long the 5' UTR is and whether the T-DNA is integrated in the 5' UTR or the proximal promoter region of the gene.
Reporter genes inserted in inverted orientation in the 5' end of genes suggest antisense transcription of trapped genes
In three lines (STINCO3/439, SENAPE1/18, and STINCO2/97) the T-DNA was found integrated in the 5' end of a gene, however with the GUS gene in antisense orientation (Table 1; Fig. 3B). The GUS-expression patterns in these three lines (e.g. STINCO3/439 in Fig. 1E) leave no doubt that the promoterless GUS gene is transcribed. In the STINCO3/439 line where the T-DNA was located near the transcription start of SLEEPY1 (SLY1) that regulates gibberellin signalling (McGinnis et al., 2003), northern analysis with a GUS probe hybridized to a transcript 
700 bp longer than the regular GUS transcript found in several other lines (Fig. 4A). This indicates that the transcription start site was distal to the T-DNA integration site and downstream of the 453 bp long ORF of the intronless SLEEPY gene.
Reporter genes inserted in intergenic regions suggest the presence of unannotated genes or cryptic promoters
The T-DNAs of four lines (LINGUINE5/194, SENAPE3/58, LINGUINE3/636, and LINGUINE1/747) were integrated in IGRs at least 800 bp from the nearest annotated gene (Table 1; Fig. 3C). The T-DNA of the LINGUINE5/194 line was integrated downstream from a tRNA gene, with the GUS gene in the same direction. A GUS transcript of the expected length was found in this line using northern hybridization (Fig. 4A), suggesting transcription to start close to the T-DNA insertion site. One accession of the BAC clone F22G5 (accession number AC022464) covering this genomic region, suggests the existence of a gene with two exons (F22G5.4), encoding a protein of 65 amino acids (aa) (AAF79579), located 238 bp downstream of, and in the same orientation as, the tRNA gene. The T-DNA of line LINGUINE5/194 would be situated 10 bp downstream of the stop codon of the putative gene. However, attempts to confirm the expression of this putative gene by RT-PCR, using different primers on several wt tissues, failed. It is not supported by the presence of cDNAs or ESTs, and in later annotations of this region it has been omitted.
In the other three lines (SENAPE3/58, LINGUINE3/636, and LINGUINE1/747) the gene upstream of the T-DNA had opposite orientation relative the GUS gene, thus GUS expression could not result from read-through from these genes. Northern hybridization was used to investigate these lines. GUS transcripts of the expected length were found, both for SENAPE3/58 (Fig. 4A) and LINGUINE3/636 (weak signal, not shown), again suggesting transcription start close to the T-DNA insertion sites. The regions flanking the T-DNA insertions were analysed more thoroughly in order to detect unannotated genes. Gene finder programs and database searches failed to detect ORFs, genes, or indications of gene expression around the T-DNA insertion points of these two lines.
However, for line LINGUINE1/747, an EST generated from mRNA of green siliques (F14301 [GenBank] ) in which the longest ORF is 38 aa (data not shown), does fit the genomic sequence 450 bp downstream of the T-DNA insertion site.
Reporter genes inserted in introns in sense orientation may be expressed due to alternative splicing
In three lines (STINCO1/607, SENAPE5/637, GNOCCHI1/760) the T-DNA was inserted in an intron with the GUS gene in the same orientation as the tagged gene (Table 1; Fig. 3D). Thus, the insertions are likely to disrupt the use of the normal splice acceptor site of the respective pre-mRNAs. The T-DNA of the STINCO1/607 line was integrated in the first intron found in the 5' UTR of a gene of unknown function, suggesting that an mRNA containing the GUS ORF is generated through the lack of splicing of this intron.
In the two other lines, the T-DNAs were integrated in introns positioned in coding regions. The pGUSBin19 T-DNA of the SENAPE5/637 line was inserted 770 bp from the start of the fourth and the largest intron of the gene encoding the nitrate transporter AtNRT1.1 (Guo et al., 2001). The LB region of the T-DNA had suffered a deletion of 651 bp, thus the splice acceptor site and the stop codons of the GUS gene leader sequence had been removed during T-DNA integration (cf. Fig. 2). Use of an alternative splice acceptor site present in the 3' end of the remaining part of the intron, or in the 5' end of the GUS ORF, generating an in-frame fusion of GUS with the fourth exon, would result in a translational fusion between the transporter protein and GUS, thus explaining the GUS expression detected in embryo, endosperm, funicle, hydathodes, and strongest in roots (Stangeland et al., 2003). This pattern is consistent with that reported for AtNRT1.1 constructs with fusions to GUS/GFP in the first or fourth exon (Guo et al., 2001), except that the expression pattern in seeds has not been reported previously.
The GNOCCHI1/760 line had the pGUSBin19 T-DNA inserted after the 145th bp of the third and longest intron of a gene encoding a short protein of 213 aa, belonging to the family of lysine decarboxylases (LDC). Northern hybridization with a GUS-specific probe against mRNA from siliques of this line revealed a very weak transcript, 700 bp longer than expected (Fig. 4A). The increased size indicated a transcriptional fusion with the upstream mRNA. This was confirmed by sequencing of a 1 kb cDNA fragment amplified by RT-PCR using a GUS primer and a gene-specific primer. The cloned cDNA contained the first three exons of the gene, encoding the first 78 aa of the LCD protein, and sequences of the pGUSBin19 downstream from nucleotide 101 immediately after a splice acceptor site present in the T-DNA (Fig. 2). The GUS ORF, however, starts at position 546 of the T-DNA, and the ATG start codon is found in position 684 (Topping et al., 1991). A fusion protein between LCD and GUS thus cannot be made from the fusion transcript. It is therefore suggested that the LCD exons function as untranslated leader sequences, and that translation of GUS starts from the ATG of the GUS gene at a frequency sufficient to allow histochemical detection of GUS expression.
The GUS activity in the GNOCCHI1/760 line was predominantly localized to the chalazal endosperm and in the globular embryo proper (Fig. 1A). To investigate whether this expression pattern was a reflection of the expression pattern of the tagged LDC gene, Arabidopsis plants were transformed with a construct consisting of 1509 bp of the LDC promoter fused to GUS (LDC::GUS; see Materials and methods). Transgenic plants were investigated for LDC::GUS expression in seedlings, mature roots, rosette leaves, stem, flowers, and developing seeds. LDC::GUS expression was found in the vascular tissue of flower petals and stamen filaments (Fig. 5A) and in the vascular tissue and veins of rosette leaves (Fig. 5B). In addition, a strong LDC::GUS staining was observed in maternal chalazal tissue shortly after fertilization, and persisted throughout seed development (Fig. 5C–F). In seeds at embryo heart stage an LDC::GUS signal could also be observed in the chalazal endosperm but it could not be resolved whether this pattern reflected diffusion of the substrate or true expression of the transgene (Fig. 5E). The earliest developmental expression found after fertilization was in syncytial micropylar endosperm at the early globular embryo stage (Fig. 5D). No expression was found in the embryo at this stage. Concurrent with endosperm cellularization, when the embryo is at the heart stage, LDC::GUS expression was found in the embryo proper, preferentially stronger in the epidermis and cells lining the epidermis (Fig. 5E). In addition, LDC::GUS was expressed equally and weakly in micropylar, central, and chalazal endosperm. Expression was more obvious in aleurone cells, the endosperm cells lining the maternal seed coat (Fig. 5E). At near mature stages, when most of the endosperm is consumed, LDC::GUS expression was found in the cotyledons of the bent-cotyledon-stage embryo (Fig. 5F). This expression pattern clearly verifies the expression of LDC in the embryo and endosperm. However, the observed pattern is different from the observations made in the GNOCCHI1 reporter line (Fig. 5G, H). In particular, the pronounced GNOCCHI1 chalazal pattern could not be verified (compare Fig. 5E and H).
|
The LDC expression pattern in flowers, rosette leaves, and seeds was confirmed by northern hybridization (Fig. 5I) and RT-PCR (data not shown). The weaker expression in siliques compared with flowers and leaves, probably reflects the restricted expression of LDC::GUS to specific cells of the seed in the siliques.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Tagged genes
Promoter trapping vectors are utilized with the intention of identifying genes expressed in specific tissues. The molecular characterization of our GUS-reporter lines shows that such identification is not necessarily straightforward. In the lines where the reporter gene is found in sense orientation in the transcribed region of a gene, the GUS expression pattern is likely to reflect the expression pattern of the tagged gene. This was demonstrated here for SENAPE4/775 (Fig. 4E). The gene tagged in this line, At5g05100, encodes a protein with a putative R3H domain suggested to bind single-stranded nucleic acids (Grishin, 1998
), and may be of importance in seed organs, as well as other parts of the plant. However, no mutant phenotype was observed, probably due to redundancy, since Arabidopsis encodes three other highly similar proteins (genes At2g40960, At3g10770, At3g56680). Likewise, the protein encoded by At5g20060, tagged in STINCO1/607, has high identity to proteins encoded by two other genes in the Arabidopsis genome (At3g5650 and At1g52700).
Another tagging example is SENAPE5/637, where a previously reported GUS expression pattern is consistent with the expression pattern reported for AtNRT1.1, i.e. pronounced expression in roots, and weaker expression in leaves (Guo et al., 2001; Stangeland et al., 2003). AtNRT1.1 mutants are inhibited in primary root growth and display a late-flowering phenotype even in the absence of nitrate (Guo et al., 2001). The reporter-gene expression in the endosperm and embryo of the SENAPE5/637 line, suggests a role for the nitrate transporter also during seed development. However, no obvious mutant phenotype was visible during seed development.
The expression pattern of the LDC gene (Fig. 5) was not fully consistent with that of the GNOCCHI1/760 marker line since GUS was not observed in seedlings, flowers, or leaves. This is most likely due to a limited use of the ATG start codon of the GUS gene, which in the LCD-GUS fusion transcript is not in frame with the upstream ATG of the LDC gene. In addition, the LDC::GUS transgene showed no obvious expression in the chalazal endosperm (compare Fig. 5E and H). One possible explanation to this is that the LDC promoter fragment used was too short to include all the gene regulatory sequences responsible for LDC expression in planta. Another possibility is that the fusion intron in the GNOCCHI1/760 line has generated a novel regulatory context, also allowing translation to start preferably from the GUS ATG in the chalazal endosperm.
The presence of at least five Arabidopsis genes encoding proteins similar to the tagged LDC (>60% identity), may explain why no consistent mutant phenotype was seen in the GNOCCHI1/760 line.
Trapped promoter elements
Available cDNA data and large-scale sequencing of short mRNA-derived tags (Meyers et al., 2004), imply that the T-DNAs in the lines 16, 619, 134, and 462 are inserted proximal promoter regions, upstream of the transcription start point of genes. T-DNA insertions in such positions do not necessarily down-regulate the tagged genes, e.g. ROC3 in the SENAPE2/16 line and At3g05120 in the PESCE1/134 line (Fig. 4B, C). The present results indicate that these genes, at least in leaves, can be driven by very short promoter fragments (159 bp and 194 bp, respectively). This is in line with studies showing that the first few hundred base pairs of a promoter can confer strong and tissue-specific gene expression, e.g. promoters of the embryo-specific ß-phaseoline (phas) gene of bean, and the aleurone- and embryo-expressed 1-Cys peroxiredoxin gene AtPER1 (van der Geest and Hall, 1996; Chandrasekharan et al., 2003; Haslekås et al., 2003). In these two promoters a CACGTG motif has been shown to be important for high-level expression.
Although many regulatory elements are found in proximal promoter regions, GUS expression in our lines with T-DNA insertions in promoters (Fig. 1C, D) may partially reflect the expression pattern of the tagged gene. This has previously been shown for embryo, leaf, and root expression of the EXO gene, tagged in the promoter trap line AtEM201 (Farrar et al., 2003). However, the EXO promoter was also active in some tissues where no GUS expression had been detected in AtEM201.
A similar situation was found for SENAPE2/16 where GUS expression was detected predominantly in the peripheral endosperm and in the late heart stage of embryo development (Fig. 1C), suggesting a role for the tagged ROC3 in seed development. ROC3 is, however, also expressed in seedlings, stems, and leaves in wt plants (Chou and Gasser, 1997). The weak but specific GUS expression patterns seen in this line (Fig. 1C), imply T-DNA integration downstream of elements conferring organ-specific expression, but upstream of stronger enhancer elements. Computer-based analysis (http://www.dna.affrc.go.jp/htdocs/PLACE) identified the CACGTG motif just downstream of the T-DNA insertion point of SENAPE2/16.
In LINGUINE6/619, this motif was also found downstream of the T-DNA insertion point in the promoter of the AtHD2C histone deacetylase gene. The GUS activity detected in the embryo of this line (Fig. 1D) is consistent with in situ hybridization data for AtHDA2 (Zhou et al., 2004). However, the northern blot analysis shows that the AtHD2C is also weakly expressed in seedlings and leaves (Fig. 4D), even though GUS expression could not be detected in these organs in LINGINE6/619.
Another member of the HD2 gene family, AtHD2A, has been shown to be required for seed development (Wu et al., 2000; Zhou et al., 2004). Although this proposes such a role also for AtHD2C, no defects were found in seed development in the SALK_039784 line that carries a T-DNA insert in the seventh exon, suggesting some of the HD2 family members to be redundant in function.
Antisense transcripts
The T-DNAs of STINCO2/97, SENAPE1/18, and STINCO1/439 were positioned in inverted orientation at the 5' end of the genes. Northern hybridization showed a longer GUS transcript than expected in the STINCO1/439 line, indicating transcription start downstream of the tagged SLY1 gene, which itself is expressed in all tissues examined (McGinnis et al., 2003). A recent whole genome transcriptional analysis in Arabidopsis revealed significant antisense expression of about one-third of the genes (Yamada et al., 2003), among them SLY1 in flower and root samples, and At5g20060 (tagged in SENAPE1/18) in all four samples tested (see http://www.ncbi.nlm.nih.gov/geo/ accession GSE637). Antisense expression of these genes has also been detected using massively parallel signature sequencing (MPSS) (Meyers et al., 2004), cf. http://mpss.udel.edu/at/. Thus, GUS expression in the lines studied (e.g. Fig. 1E) is consistent with reporter-gene transcription driven by antisense promoters.
Regulatory sequences have been isolated both at the 3' end and in introns of transcribed genes (Sieburth and Meyerowitz, 1997; Kloti et al., 1999), where they might function as silencers or enhancers. Alternatively, they might act as promoters for the generation of naturally transcribed antisense RNAs that possibly have a gene regulatory function (Terryn and Rouze, 2000; Yamada et al., 2003).
Cryptic promoters and/or unannotated genes
The GUS activity in line LINGUINE1/747 is very strong and localized to the cells of the basal embryo developing into the root tip (Fig. 1B), resembling the previously identified PLS GUS reporter (Topping and Lindsey, 1997). The PLS gene encodes a small polypeptide (36 aa) involved in regulation of the root growth and leaf vascular patterning (Casson et al., 2002). No gene annotations could be found in the region around the T-DNA integration site in the line studied. However, the nearby EST (F14301
[GenBank]
) might suggest that a short protein is generated from this region.
The annotation of the Arabidopsis genome is incomplete since genes encoding regulatory RNAs like micro-RNAs (Kidner and Martienssen, 2003), or genes with very short ORFs are difficult to identify by gene finder programs (Casson et al., 2002; Butenko et al., 2003). Re-examination of the regions where T-DNA integration occurred in lines SENAPE/58 and LINGUINE3/636 lines failed to identify significant ORFs or ESTs, and for LINGUINE/194 the existence of a putative small gene annotated in early accession could not be substantiated. In an attempt to identify genes of non-coding RNAs, the relevant intergenic regions have also been screened for the presence of small RNAs using the small RNA database (http://cgrb.orst.edu/smallRNA/db/), but no matches were found. Thus GUS expression in these lines may be due to so-called cryptic promoters, defined as elements normally inactive that have the ability to direct transcription when positioned adjacent to genes. Both constitutive and tissue-specific cryptic promoters have been isolated from tobacco (Fobert et al., 1994; Foster et al., 1999) and seem to be functionally equivalent to elements responsible for the expression of plant genes (Wu et al., 2003a).
The biological significance of promoter trap lines as screening tools
The molecular analysis reported here suggests that the expression pattern from promoterless reporter genes inserted in sense orientation in transcribed regions of genes, will reflect the expression pattern of the tagged gene. When using a tagging vector with a splice acceptor site, for example, pGUSBin19, expression can also be achieved when T-DNAs are inserted in introns positioned in the coding region.
Our understanding of promoter elements in the regulation of the tissue- and organ-specific expression is limited. One well-studied example is, however, the promoter of the phas gene (van der Geest and Hall, 1996; Chandrasekharan et al., 2003), which has a seed-specific enhancer, a region that down-regulates seed- and stem/root-specific expression and a basal promoter (van der Geest and Hall, 1996). Similarly, 427 bp of the promoter of the AtPER1 is sufficient for seed-specific expression in Arabidopsis and contains elements of importance both for endosperm and embryo expression (Haslekås et al., 2003). Therefore, in lines with T-DNAs inserted in the middle of promoters, GUS expression in embryo and endosperm is considered to be biologically significant, although the tagged gene may be expressed in a broader range of tissues. As shown for the EXO gene, such lines give valuable information about the tagged gene (Farrar et al., 2003), and—if crossed to mutant lines—can also be useful in characterizations of mutants.
The disproportion between the number of isolated marker lines and the number of identified seed-expressed genes may partially be due to trapping of cryptic regulatory elements. Although cryptic promoters may not be functional in wt plants, such promoters may be useful for research and applied purposes (Wu et al., 2003b). The present study also suggests that promoter trap vectors can be used to identify promoters conferring endogenous antisense gene transcription.
Instead of ignoring promoter-trap data that are unexpected or difficult to explain, such results should be collected, analysed thoroughly, and reported. Systematic analysis of such data may turn out to provide important information about the distribution and frequency of genomic regions capable of promoting transcription.
|
Acknowledgements |
|---|
This work was supported by the Research Council of Norway (grants to BS, AB, and ZS project no. 129525/420 and to PEG project no. 104429/130). The authors thank Zanina Grieg, Solveig H Engebretsen, and Roy Falleth for technical assistance. We would also like to thank the Salk Institute Genomic Analysis Laboratory for providing the sequence-indexed Arabidopsis T-DNA insertion mutants. This work has been facilitated by the Norwegian Arabidopsis Research Center, a national technology platform supported by the functional genomics programme (FUGE) in the Research Council of Norway.
|
Footnotes |
|---|
* Present address: The National Hospital, Section for Gene Therapy, N-0027 Oslo, Norway.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Berger F. 1999. Endosperm development. Current Opinion in Plant Biology 2, 28–32.[CrossRef][ISI][Medline]
Berger F. 2003. Endosperm: the crossroad of seed development. Current Opinion in Plant Biology 6, 42–50.[CrossRef][ISI][Medline]
Brown RC, Lemmon BE, Nguyen H, Olsen OA. 1999. Development of endosperm in Arabidopsis thaliana. Sexual Plant Reproduction 12, 32–42.
Butenko MA, Patterson SE, Grini PE, Stenvik GE, Amundsen SS, Mandal A, Aalen RB. 2003. Inflorescence deficient in abscission controls floral organ abscission in Arabidopsis and identifies a novel family of putative ligands in plants. The Plant Cell 15, 2296–2307.
Casson SA, Chilley PM, Topping JF, Evans IM, Souter MA, Lindsey K. 2002. The POLARIS gene of Arabidopsis encodes a predicted peptide required for correct root growth and leaf vascular patterning. The Plant Cell 14, 1705–1721.
Chandrasekharan MB, Bishop KJ, Hall TC. 2003. Module-specific regulation of the beta-phaseolin promoter during embryogenesis. The Plant Journal 33, 853–866.[CrossRef][ISI][Medline]
Chou IT, Gasser CS. 1997. Characterization of the cyclophilin gene family of Arabidopsis thaliana and phylogenetic analysis of known cyclophilin proteins. Plant Molecular Biology 35, 873–892.[CrossRef][ISI][Medline]
Clough SJ, Bent AF. 1998. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. The Plant Journal 16, 735–743.[CrossRef][ISI][Medline]
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][ISI][Medline]
Fobert P, Labbe H, Cosmopoulos J, Gottlob-McHugh S, Ouellet T, Hattori J, Sunohara G, Iyer VN, Miki BL. 1994. T-DNA tagging of a seed coat-specific cryptic promoter in tobacco. The Plant Journal 6, 567–577.[CrossRef][ISI][Medline]
Foster E, Hattori J, Labbe H, Ouellet T, Fobert PR, James LE, Iyer VN, Miki BL. 1999. A tobacco cryptic constitutive promoter, tCUP, revealed by T-DNA tagging. Plant Molecular Biology 41, 45–55.[CrossRef][ISI][Medline]
Galau GA, Hughes DW, Dure IL. 1986. Abscisic acid induction of cloned cotton late embryogenesis-abundant (Lea) mRNAs. Plant Molecular Biology 7, 155–170.
Grini PE, Jürgens G, Hülskamp M. 2002. Embryo and endosperm development is disrupted in the female gametophytic capulet mutants of Arabidopsis. Genetics 162, 1911–1925.
Grishin NV. 1998. The R3H motif: a domain that binds single-stranded nucleic acids. Trends in Biochemical Science 23, 329–330.
Guo FQ, Wang R, Chen M, Crawford NM. 2001. The Arabidopsis dual-affinity nitrate transporter gene AtNRT1.1 (CHL1) is activated and functions in nascent organ development during vegetative and reproductive growth. The Plant Cell 13, 1761–1777.
Haslekås C, Grini PE, Nordgard SH, Thorstensen T, Viken MK, Nygaard V, Aalen RB. 2003. ABI3 mediates expression of the peroxiredoxin antioxidant AtPER1 gene and induction by oxidative stress. Plant Molecular Biology 53, 313–326.[CrossRef][ISI][Medline]
Kidner CA, Martienssen RA. 2003. Macro effects of microRNAs in plants. Trends in Genetics 19, 13–16.[CrossRef][ISI][Medline]
Kloti A, Henrich C, Bieri S, et al. 1999. Upstream and downstream sequence elements determine the specificity of the rice tungro bacilliform virus promoter and influence RNA production after transcription initiation. Plant Molecular Biology 40, 249–266.[CrossRef][ISI][Medline]
Mandal A, Sandgren M, Holmström K-O, Gallois P, Palva ET. 1995. Identification of Arabidopsis thaliana sequences responsive to low temperature and abscisic acid by T-DNA tagging and in-vivo gene fusion. Plant Molecular Biology Reporter 13, 243–254.
McGinnis KM, Thomas SG, Soule JD, Strader LC, Zale JM, Sun T-p, Steber CM. 2003. The Arabidopsis SLEEPY1 gene encodes a putative F-box subunit of an SCF E3 ubiquitin ligase. The Plant Cell 15, 1120–1130.
Meyers BC, Tej SS, Vu TH, Haudenschild CD, Agrawal V, Edberg SB, Ghazal H, Decola S. 2004. The use of MPSS for whole-genome transcriptional analysis in Arabidopsis. Genome Research 14, 1641–1653.
Meza TJ, Stangeland B, Mercy IS, et al. 2002. Analyses of single-copy Arabidopsis T-DNA-transformed lines show that the presence of vector backbone sequences, short inverted repeats and DNA methylation is not sufficient or necessary for the induction of transgene silencing. Nucleic Acids Research 30, 4556–4566.
Olsen OA. 2001. Endosperm development: cellularization and cell fate specification. Annual Review of Plant Physiology and Plant Molecular Biology 52, 233–267.[CrossRef][ISI][Medline]
Olsen OA. 2004. Nuclear endosperm development in cereals and Arabidopsis thaliana. The Plant Cell 16, Suppl. S214–S227.
Pandey R, Muller A, Napoli CA, Selinger DA, Pikaard CS, Richards EJ, Bender J, Mount DW, Jorgensen RA. 2002. Analysis of histone acetyltransferase and histone deacetylase families of Arabidopsis thaliana suggests functional diversification of chromatin modification among multicellular eukaryotes. Nucleic Acids Research 30, 5036–5055.
Sieburth LE, Meyerowitz EM. 1997. Molecular dissection of the AGAMOUS control region shows that cis elements for spatial regulation are located intragenically. The Plant Cell 9, 355–365.[Abstract]
Stangeland B, Salehian Z. 2002. An improved clearing method for GUS assay in Arabidopsis endosperm and seed. Plant Molecular Biology Reporter 20, 107–114.
Stangeland B, Salehian Z, Aalen R, Mandal A, Olsen OA. 2003. Isolation of GUS marker lines for genes expressed in Arabidopsis endosperm, embryo, and maternal tissues. Journal of Experimental Botany 54, 279–290.
Terryn N, Rouze P. 2000. The sense of naturally transcribed antisense RNAs in plants. Trends in Plant Science 5, 394–396.[CrossRef][ISI][Medline]
Topping JF, Agyeman F, Henricot B, Lindsey K. 1994. Identification of molecular markers of embryogenesis in Arabidopsis thaliana by promoter trapping. The Plant Journal 5, 895–903.[CrossRef][ISI][Medline]
Topping JF, Lindsey K. 1997. Promoter trap markers differentiate structural and positional components of polar development in Arabidopsis. The Plant Cell 9, 1713–1725.[Abstract]
Topping JF, Wei W, Lindsey K. 1991. Functional tagging of regulatory elements in the plant genome. Development 112, 1009–1019.[Abstract]
van der Geest AH, Hall TC. 1996. A 68 bp element of the beta-phaseolin promoter functions as a seed-specific enhancer. Plant Molecular Biology 32, 579–588.[CrossRef][ISI][Medline]
Vielle-Calzada JP, Baskar R, Grossniklaus U. 2000. Delayed activation of the paternal genome during seed development. Nature 404, 91–94.[CrossRef][Medline]
Wu K, Hu M, Martin T, Wang C, Li XQ, Tian L, Brown D, Miki B. 2003a. The cryptic enhancer elements of the tCUP promoter. Plant Molecular Biology 51, 351–362.[CrossRef][ISI][Medline]
Wu K, Tian L, Malik K, Brown D, Miki B. 2000. Functional analysis of HD2 histone deacetylase homologues in Arabidopsis thaliana. The Plant Journal 22, 19–27.[CrossRef][ISI][Medline]
Wu K, Tian L, Zhou C, Brown D, Miki B. 2003b. Repression of gene expression by Arabidopsis HD2 histone deacetylases. The Plant Journal 34, 241–247.[CrossRef][ISI][Medline]
Yamada K, Lim J, Dale JM, et al. 2003. Empirical analysis of transcriptional activity in the Arabidopsis genome. Science 302, 842–846.
Zhou C, Labbe H, Sridha S, Wang L, Tian L, Latoszek-Green M, Yang Z, Brown D, Miki B, Wu K. 2004. Expression and function of HD2-type histone deacetylases in Arabidopsis development. The Plant Journal 38, 715–724.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Ungru, M. K. Nowack, M. Reymond, R. Shirzadi, M. Kumar, S. Biewers, P. E. Grini, and A. Schnittger Natural Variation in the Degree of Autonomous Endosperm Formation Reveals Independence and Constraints of Embryo Growth During Seed Development in Arabidopsis thaliana Genetics, June 1, 2008; 179(2): 829 - 841. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||