SEARCH

SEARCH BY CITATION

Keywords:

  • alternative splicing;
  • nonsense-mediated decay;
  • post-transcriptional regulation;
  • mRNA surveillance;
  • premature termination codon;
  • aberrant mRNA

Summary

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

It has been reported that eukaryotic organisms have a nonsense-mediated mRNA decay (NMD) system to exclude aberrant mRNAs that produce truncated proteins. NMD is an RNA surveillance pathway that degrades mRNAs possessing premature translation termination codons (PTCs), thus avoiding production of possibly toxic truncated proteins. Three interacting proteins, UPF1, UPF2 and UPF3, are required for NMD in mammals and yeasts, and their amino acid sequences are well conserved among most eukaryotes, including plants. In this study, ‘The Arabidopsis Information Resource’ database was searched for mRNAs with premature termination codons. We selected five of these mRNAs and checked for the presence of PTCs in these mRNAs when translated in vivo. As a result we identified aberrant mRNAs produced by alternative splicing for each gene. These genes produced at least one alternative splicing variant including a PTC (PTC+) and another variant without a PTC (PTC−). We analyzed their PTC+/PTC− ratios in wild-type Arabidopsis and upf3 mutant plants and showed that the PTC+/PTC− ratios were higher in atupf3 mutant plants than wild-type plants and that the atupf3 mutant was less able to degrade mRNAs with premature termination codons than wild-type plants. This indicated that the AtUPF3 gene is required by the plant NMD system to obviate aberrantly spliced mRNA.


Introduction

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

Nonsense-mediated mRNA decay (NMD) is an RNA surveillance pathway that degrades mRNAs possessing premature translation termination codons (PTCs). This surveillance pathway is well conserved among eukaryotes, and it is now accepted that NMD has diverse roles in the control of gene expression (for reviews see Green et al., 2003; Maquat, 2004; Neu-Yilik et al., 2004; Singh and Lykke-Andersen, 2003; Wagner and Lykke-Andersen, 2002; Wilkinson and Shyu, 2002; Wilusz et al., 2001). Saccharomyces cerevisiae (Gonzalez et al., 2001), Drosophila melanogaster (Gatfield et al., 2003), Caenorhabditis elegans (Mango, 2001) and Homo sapiens (Maquat, 2004) are the main organisms in which analyses of NMD by UPF families have been performed. Three interacting proteins, UPF1, UPF2 and UPF3, are required for NMD and their sequences are found in most eukaryotes, including plants (Culbertson and Leeds, 2003). Drosophila melanogaster and S. cerevisiae recognize PTCs based on different rules from human cells (Gatfield et al., 2003; Gonzalez et al., 2001), while functional analyses of the UPF family and/or molecular mechanisms of NMD in plants have not been reported.

In plants, the mechanisms involved in NMD are unclear; however, it has been reported that PTCs reduce mRNA accumulation (Dehesh et al., 1993; Isshiki et al., 2001; Marchant and Bennett, 1998; Saito and Nakamura, 2004). Furthermore, it has also been reported that nonsense mutations on intronless genes reduce mRNA accumulation in plants, unlike in the human NMD system (van Hoof and Green, 1996; Jofuku et al., 1989; Petracek et al., 2000; Voelker et al., 1986). These results suggest that plants have an NMD-like surveillance system, as do other eukaryotes, but that the molecular mechanism that induces NMD may be different.

In this report, NMD target candidates were found by searching for the presence of PTCs in ‘The Arabidopsis Information Resource’ (TAIR) database. Then we examined the in vivo behavior of these transcripts, focusing on stability with and without PTCs. Five of the six NMD target candidate genes analyzed produced at least two variant mRNAs, one with a PTC (PTC+) and the other without a PTC (PTC−), possibly by alternative splicing. We therefore used a UPF3 knock-down mutant (atupf3-1) and analyzed the influence of a loss of UPF3 on PTC+ mRNA fate. PTC+ mRNAs were much less stable than PTC− mRNAs in wild-type plants. However, PTC+ mRNAs showed similar stability to PTC− mRNAs in the atupf3-1 mutant. It was suggested therefore that AtUPF3 is required by the plant NMD system to suppress aberrantly spliced mRNA with PTCs.

Results

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

Phylogenic analysis of UPF3 ortholog candidates in plants

In the protein families database (Pfam) (Bateman et al., 2004), the SMG4_UPF3 motif family (Pfam accession number: PF03467) includes C. elegans Smg-4, S. cerevisiae Upf3p, D. melanogaster UPF3 and H. sapiens UPF3; proteins in this family have been shown to be involved in NMD (Aronoff et al., 2001; Gatfield et al., 2003; Lee and Culbertson, 1995; Lykke-Andersen et al., 2000). UPF3 ortholog candidates in the Oryza sativa and Arabidopsis thaliana genome were also found in the Pfam database. In addition, two UPF3 candidate sequences were used to search for similar proteins in other plants at the DDBJ website (http://www.ddbj.nig.ac.jp/search/blast-e.html) using TBLASTN (McGinnis and Madden, 2004). Such possible UPF3 ortholog protein sequences were subjected to phylogenetic analysis with UPF3 sequences characterized in different organisms (Figure 1).

image

Figure 1. Phylogenetic analysis of UPF3. The area within the rectangle indicates the plant group. The distance scale represents 0.1 fixed mutations per site. Bootstrap values, indicated at the nodes, were obtained from 1000 bootstrap replicates and are reported as percentages.

Download figure to PowerPoint

The clades constructed by the computer almost matched established phyla; that is, animal, plant and fungal groups. In addition, when we surveyed the full length of the amino acid sequences, proteins from each group had almost no similarity with those of other groups outside the SMG4_UPF3 motif region used for phylogenetic analysis. Hereafter we call the UPF3 candidate gene of Arabidopsis (At1g33980) AtUPF3 for convenience. Thus, we experimentally tested whether AtUPF3 is a functional ortholog involved in plant NMD.

Identification of a T-DNA tagged atupf3 candidate line

We selected a possible disruption mutant of the AtUPF3 gene from the Salk Institute T-DNA insertion mutant panel (Alonso et al., 2003; Yamada et al., 2003) (http://signal.salk.edu/cgi-bin/tdnaexpress) to experimentally test whether the AtUPF3 gene is involved in NMD. The SALK_025175 line had a T-DNA insertion at the 3′ end of AtUPF3 exon 5. Northern blot analysis showed that wild-type plants expressed AtUPF3 mRNA, but no accumulation of intact AtUPF3 mRNA was detected in homozygous T-DNA tagged plants (Figure 2). It was thus expected that this line had lost the normal function of the AtUPF3 gene, and therefore, it was used in subsequent experiments as an atupf3-1 mutant.

image

Figure 2. Structure of the T-DNA insertion in At1g33980. (a) Location of the T-DNA insertion in SALK_025175. pROK2 T-DNA left border inserted at position 1346 (the 3′ end of exon 5). (b) Northern blot analysis of polyA+ mRNA (500 ng) isolated from aboveground parts of individual plants. The arrow in (a) indicates the UPF3 mRNA. The UPF3 and GAPDH panels are the result of Northern blot analysis with a DIG-labeled antisense probe (line). The GAPDH panel represents the control; GAPDH mRNA is commonly used as an invariant control in studies of gene expression.

Download figure to PowerPoint

Prediction of mRNAs with PTCs from Arabidopsis genome information

If AtUPF3 were involved in NMD, the metabolic turnover of PTC+ mRNA would be influenced by AtUPF3. Therefore, to investigate the behavior of PTC+ mRNAs in plants, we identified natural PTC+ mRNA in Arabidopsis. If the termination codon mapped in an upstream exon, not the last exon of an mRNA, the mRNA could be a candidate for PTC+ mRNA as found with the human NMD system. We therefore tested the available database of Arabidopsis and collected, in silico, genes with termination codons located in the upstream exon.

We surveyed TAIR to find candidate PTC+ mRNAs from the 31 002 transcript splicing models registered as of October 2003. In total, 838 genes were identified as having termination codons in exons upstream of the correct termination codon. These were further narrowed down to 424 genes identified in the RIKEN Arabidopsis Full-Length cDNA library (Seki et al., 2002). Of these, we selected genes with a long open reading frame after the PTC and with PTCs located more than three exons upstream from the final exon. We arbitrarily selected six genes from the final pool and used them as PTC+ mRNAs (Table 1) for the subsequent experiments.

Table 1.  PTC+ mRNAs
At2g45670Calcineurin B subunit-related
At5g62760Similar to nuclear protein ZAP
At5g07910Leucine-rich repeat protein family
At1g51340MATE efflux family protein
At3g63340Protein phosphatase 2C-related
At4g30340Putative diacylglycerol kinase-related protein

Alternative splicing gave rise to PTC+ and PTC− mRNA variants

The existence of PTCs in vivo in the mRNAs selected was determined. The mRNAs were isolated from wild-type and atupf3-1 mutant plants and sequences were amplified by RT-PCR. Direct sequencing of the regions covering the predicted PTC and splice junctions of each mRNA was performed. In At4g30340, the PTC was identified in the predicted position and only one unique mRNA species was discernible (data not shown). In At2g45670, At5g62760, At5g07910, At1g51340 and At3g63340, we detected two or more species of mRNAs, possibly produced as splice variants. The splicing variants were observed by RT-PCR direct sequencing as mixed signals of major and minor peaks (Figure 3). At2g45670, At5g62760, At5g07910 and At1g51340 showed one alternative splicing variant including a PTC (PTC+) in its reading frame and another variant without a PTC (PTC−) (Figure 4). Furthermore, At3g63340 had two alternative splicing sites; one site did not lead to a frameshift nor expression of PTC+ mRNA (Figure 4, At3g63340 position a), but the other produced both PTC+ and PTC− mRNAs (Figure 4, At3g63340 position b).

image

Figure 3. (a) Raw sequence data of PTC+ mRNAs in wild-type and atupf3-1 mutant plants. The indicated regions include alternative splice sites. At3g63340 has two alternative exon–exon junctions, indicated as points A and B, respectively. (b) Ratios of PTC+/PTC− mRNA by measuring the peak heights of the sequences. White bars represent the NMDTs PTC+/PTC− mRNA ratios in the wild-type plants and the black bars represent those of the atupf3 mutant. The asterisk indicates ‘not detected’.

Download figure to PowerPoint

image

Figure 4. Schematic representation of the PTC+ mRNA with alternative splice sites (asterisks); indicated ‘PTCs’ were produced by abnormal exons, and ‘stops’ were produced by normal exons.

Download figure to PowerPoint

Increased PTC+/PTC− mRNA ratios in atupf3-1

We supposed that the higher and lower peaks proportionally reflected the mRNA sequences of major and minor populations described in the analysis above, respectively, considering that the ratio of the two sequences should be constant during RNA purification, RT-PCR and sequencing. To validate our supposition, we tested whether the low and high peak heights proportionally reflected minor and major mRNA ratios, respectively. PCR direct sequencing was performed using arbitrary ratios of mixed plasmids as the PCR templates; they were cloned into each splicing variant at At3g63340 point B (see Experimental procedures and Supplementary Material).

The results supported our supposition (Supplementary Material), showing that the sequence peak ratios closely reflected the actual PTC+/PTC− mRNA ratios. Consequently, we concluded that this method would give an accurate picture of PTC+/PTC− ratios in wild-type and atupf3-1 mutant plants.

The PTC+/PTC− ratios measured for the five genes were higher in the atupf3-1 mutant than in the wild-type plants. Thus, the presence or absence of PTCs determines mRNA fate. This conclusion is also supported by the fact that one of the two alternative splicing products at At3g63340 point A, which would not produce the PTC+ variant, showed no differences between wild-type and atupf3-1 mutant plants (Figure 3).

We detected variation in the PTC+/PTC− ratios of At3g63340 position B depending on the growth stage and tissues. Many PTC+ mRNAs were observed in younger stage tissues from wild-type (Supplementary Material), but generally, in all stages and tissues, PTC+/PTC− ratios in the atupf3 mutant were higher than in the wild-type plant. These results suggest that AtUPF3 participates in the specific suppression of PTC+ mRNA accumulation, which might be produced by alternative splicing.

Increased stability of PTC+ mRNAs in atupf3-1 mutants

Next, we experimentally determined the accumulation ratios of PTC+ and PTC− mRNAs that affected the AtUPF3 gene levels. Leaf sections from young leaves of wild-type and atupf3-1 plants were soaked in medium containing actinomycin D (AcD), an inhibitor of de novo mRNA synthesis, to observe the stability of once-transcribed PTC+ mRNA. Total mRNA was isolated 4 h after drug application and each PTC+/PTC− ratio was measured by RT-PCR direct sequencing (Figure 5; At3g63340 point B and Table 2). In the wild-type plant, PTC+ mRNAs of At2g45670, At5g62760, At1g51340 and At3g63340 point B PTC+ were not detected after AcD treatment. PTC+ mRNA of At5g07910 could not be detected irrespective of AcD treatment, and the ratio of splicing variants at At3g63340 point A without PTC did not change. These data suggest that, after being synthesized in wild-type plants, PTC+ mRNA is rapidly degraded compared with PTC−. In contrast, in the atupf3-1 line, the PTC+/PTC− ratio was almost constant over the 4 h of AcD treatment. These results from leaf sections suggested that PTC-dependent mRNA instability after transcription resulted in less accumulation and required AtUPF3 function.

image

Figure 5. Analysis of At3g63340 mRNA, point B. (a) Raw sequence data of PTC+ mRNAs in wild-type and atupf3-1 mutant plants 4 h after AcD and CHX treatments. (b) Ratios of PTC+/PTC− mRNA by measuring the peak heights of the sequences.

Download figure to PowerPoint

Table 2.  Change in the PTC+/PTC− ratio with AcD and CHX treatment
GeneSplicing variantWild-type (col-0)atupf3-1
ControlActCHXControlActCHX
  1. −: Detection limit; +: PTC+ mRNA was detectable; ++: more PTC+ mRNA than (+) was detected. In At3g63340 position a, the PTC+ variant was not produced, indicating the existence of a minor variant.

At2g45670PTC+/−+++++++++
At5g62760PTC+/−+++++++++
At5g07910PTC+/−++++
At1g51340PTC+/−+++++
At3g63340 position aPTC− only++++++
At3g63340 position bPTC+/−+++++++++

We applied cycloheximide (CHX), an inhibitor of translation and sometimes used as an NMD inhibitor (Carter et al., 1995), to leaf sections of wild-type and atupf3-1 mutant plants. PTC+/PTC− mRNAs did not change in the atupf3-1 mutant, however, CHX treatment remarkably elevated the PTC+/PTC− mRNA ratios in wild-type Arabidopsis to the same levels as in the atupf3-1 mutant. As with AcD treatment, CHX treatment did not influence the ratio of splicing variants at At3g63340 point A without PTC. These results support the observation that instability of PTC+ mRNA is coupled with translation. Thus, AtUPF3 is the functional ortholog in plants.

We suspected that the atupf3-1 mutant accumulated more PTC+ mRNAs than wild-type Arabidopsis due to a defect in the destabilization of PTC+ mRNAs. We harvested polyA+ RNA from three independent plants of the wild-type and atupf3-1 mutants, and Northern blot analysis was performed to compare the six PTC+ mRNAs in the wild-type and atupf3-1 plants (Figure 6). These signals showed all splicing variants, as PTC+ and PTC− mRNAs of these genes could not be separated by agarose gel electrophoresis. Rather expectedly, it was shown that the atupf3-1 mutant accumulated a much enhanced level of At3g63340 mRNA, as was the case with At2g45670, At5g62760 and At5g07910 mRNAs. Remarkably, At1g51340 mRNA decreased in the atupf3-1 mutant.

image

Figure 6. Northern blot analysis of polyA+ mRNA (500 ng) isolated from aboveground parts of individual plants with each PTC+ mRNA probe. The left lane represents the wild-type plant and the right is the atupf3-1 mutant. Ratios of signal intensity between wild-type and atupf3-1 plants were measured by Scion image (Scion Corporation, Frederick, MD, USA). Standard divisions were calculated using three independent experiments.

Download figure to PowerPoint

Discussion

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

Possible PTC+ mRNAs were selected by in silico prediction from the TAIR databases. Next, we verified whether in silico-selected genes produced PTC+ mRNAs in vivo. At3g63340, At2g45670, At5g62760, At5g07910 and At1g51340 produced at least two alternative splicing variants, one of which had a PTC while the other did not (Figure 4). The ratios of PTC+/PTC− in these genes increased in the atupf3-1 mutant compared with the wild-type Arabidopsis (Figure 3), and few PTC+ mRNAs were detected in wild-type plants. Thus, the involvement of AtUPF3 in elimination of PTC+ mRNA was confirmed by studying the atupf3-1 mutant.

Both RT-PCR and Northern blot analyses have shown total mRNA accumulation in At2g45670, At5g62760 and At5g07910 increased in the atupf3-1 mutant. However, Northern blot analysis showed a substantial increase in At3g63340 mRNA and decrease in At1g51340 mRNA in the atupf3-1 mutant (Figure 6) compared with the expected result from RT-PCR direct-sequencing. It is probable that transcriptional levels of At3g63340 and At1g51340 were altered in the atupf3-1 mutant by an unknown mechanism. In yeast, it has been reported that many genes are indirectly influenced by NMD (Lelivelt and Culbertson, 1999).

The PTC+/PTC− mRNA ratio was greater, especially in younger tissue (Supplementary Material), however, it is unknown whether truncated proteins were produced from these PTC+ mRNA. As the transcription levels are high in young tissue, pre-degraded PTC+ mRNA might have accumulated.

The PTC+ mRNA was relatively stable for some time after synthesis in the atupf3-1 mutant, while its stability in wild-type Arabidopsis was apparently low. PTC+ mRNA-specific degradation was inhibited even in wild-type plants by CHX. Thus, it was confirmed that PTC+ mRNA-specific degradation is coupled by concomitant protein translation. On the other hand, accumulation of the splicing variant that did not produce PTC was not influenced by the existence of AtUPF3, AcD treatment or CHX treatment.

We were also able to test whether a gene showing significant homology with UPF3 genes in other organisms was the functional UPF3 ortholog gene in plants. It can now be concluded that the above-mentioned degradation is specific for PTCs and that AtUPF3 is involved in the plant NMD system, as in animal and yeast NMD. However, the sequences outside the homologous amino acid stretch were very divergent among different organisms (Culbertson and Leeds, 2003). Possibly due to this, it was reported that human cells, D. melanogaster and S. cerevisiae recognize PTC+ mRNA using different rules (Gatfield et al., 2003; Gonzalez et al., 2001); however, it is not clear what rules control Arabidopsis recognition of PTCs. In addition, NMD of intronless genes was reported in plants (van Hoof and Green, 1996; Jofuku et al., 1989; Petracek et al., 2000; Voelker et al., 1986). The requirements for NMD targets in plants should therefore be tested in the future, and more detailed analyses of plant NMD machinery is anticipated; molecular and functional analyses of AtUPF3 would solve the questions regarding plant NMD machinery.

Alternative splicing produces multiple transcripts from one gene, resulting in expression of multiple proteins with diverse and sometimes antagonistic functions. In humans, alternative splicing occurs widely in 42% of total genes (Modrek et al., 2001), and it is now accepted that alternative splicing is important in human gene expression, with abnormalities resulting in several human diseases (review, Faustino and Cooper, 2003).

Genome-wide analysis indicated that 11.6% of transcriptional units would produce alternative spliced transcripts in A. thaliana, and alternative splicing profiles were shown to be affected by environmental stress conditions (Iida et al., 2004). Recently, progress has been made in the elucidation of alternative splicing regulation (Reddy, 2004; Savaldi-Goldstein et al., 2003, review), and the functional significance of alternative splicing in plants has been partly reported. Proper function of the circadian autoregulatory loop requires alternative splicing of the circadian-clock-regulated RNA-binding protein, AtGRP7 (Staiger et al., 2003), and the disease-resistant genes N and RPS4 need alternative splicing variants to function (Dinesh-Kumar and Baker, 2000; Zhang and Gassmann, 2003).

Alternative splicing has the potential not only to produce variegated proteins, but also harmful truncated proteins resulting from PTC insertion. Unfavorable alternative mRNA variants with PTCs should be degraded by the NMD system shown in this study, at least in part. Further experiments are necessary to analyze the physiological roles of AtUPF3 in plants.

Experimental procedures

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

Sequence analysis

A sequence similarity search was performed using TBLASTN with the AtUPF3 protein sequence (TrEMBL accession number Q9FX10). Phylogenetic analyses were carried out using the protein sequences of Mus musculus (XP_110787, XP_356061), H. sapiens (AAG60691, AAG60690), D. melanogaster (NP_611844), Anopheles gambiae (TrEMBL accession number Q7Q545), C. elegans (T22286), S. cerevisiae (P48412), Schizosaccharomyces pombe (NP_593705), Zea mays (CD438629), O. sativa (CAE03133), Brassica napus (CD814697), Glycine max (BG789939), Lactuca sativa (BU003829), Lycopersicon esculentum (BG134758), Populus balsamifera (CN520557), and A. thaliana (NP_174660); the numbers shown are the GenBank accession numbers if not otherwise described. The SMG4_UPF3 region of these sequences, with gaps excluded, were aligned using Clustal X (Thompson et al., 1994), and a phylogenetic tree was generated by the neighbor-joining method (Saitou and Nei, 1987). The unrooted tree was visualized using Tree View (Page, 1996).

Plant materials

Arabidopsis thaliana plants were grown at 22°C under long-day conditions (16 h light and 8 h dark). Seeds were planted on Jiffy-7 (Jiffy, Batavia, IL, USA) and acclimated for 2 days at 4°C.

Identification of a T-DNA tagged atupf3-1 line

We screened two possible AtUPF3 insertion mutants, SALK_161923 and SALK_025175. Plants from each line were screened by PCR with pairs of specific primers for the AtUPF3 or T-DNA sequence. Template DNA was prepared from single leaves as described previously (Liu et al., 1995). The DNA was suspended in 50 μl of TE buffer [10 mm Tris (pH 7.5), 1 mm EDTA] and 1 μl of DNA was used for the PCR. The following wild-type-specific primer sets and insertion mutant-specific primer sets were then used: for the T-DNA insertion in SALK_025175, primers AtUPF3-4F and AtUPF3-6R, and AtUPF3-4F and LBb1, and in SALK_161923, primers AtUPF3-4F and AtUPF3-6R, and LBb1 and AtUPF3-6R (Table 3). A T-DNA homozygote was only identified in SALK_025175, and a T-DNA heterozygote was only identified in SALK_161923. The heterozygote SALK_161923 did not give rise to any homozygous plants in the PCR genotype screens.

Table 3.  Primer sets
Target geneRegionPrimer nameOrientationSequence
At5g07910668–687NMDT1-FForwardAGTTTCAGCTGATGGAAGGA
1068–1043NMDT1-RReverseAAGTAGAATCGTCATCTCAATATAAG
At3g633401698–1719NMDT-2FForwardCCCTGTAGCAACTGTACGCTTC
2109–2089NMDT2-RReverseTTCTTCAGCACATAGCTTGAG
At1g51340997–1016NMDT3-F2ForwardTGTGTAACTCTCTCCGCGTC
1520–1500NMDT3-RReverseTGTCCCAATCCGCCAGAATCC
At5g62760917–936NMDT5-FForwardCCTCATCGATCTACTCGTCC
1471–1451NMDT5-RReverseTCCATGTCCATATCCACCTCC
At2g45670751–774NMDT6-FForwardAAGAATGCTGTGCATGAAATAAAG
1243–1223NMDT6-RReverseCTGCTTACATGGAATAGCGAC
At4g30340602–623NMDT7-FForwardAAAAGATACGGATCATGGTAGC
1307–1286NMDT7-RReverseCAGAGAGATTATAAGTGAAGCC
AtUPF3 Genome DNA1195–1215AtUPF3-4FForwardACTTCTATTGTTGATCTCTGG
1838–1858AtUPF3-6RReverseagaaagctgggtATGCTGTTCCGGTTGTGGTGG
T-DNA6285–6244LBb1ReverseAACCAGCGTGGACCGCTTGCTG
AtUPF3 mRNA827–848AtUPF3-5FForwardaaaaagcaggctACCGAGACAATCCTGATAACCC
1315–1294AtUPF3-7RReverseagaaagctgggtGACTAGAATGTCTGGCAGAACC
GATEWAY attB1ForwardGGGGACAAGTTTGTACAAAAAAGCAGGCT
attB2ReverseGGGGACCACTTTGTACAAGAAAGCTGGGT

RNA purification and RT-PCR

Total RNA was purified from wild-type and atupf3-1 plants as previously described (Shirzadegan et al., 1991). Samples of total RNA (2 μg) were used to synthesize cDNA using Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA) and oligo-dT as a primer (Invitrogen). PCR was performed in a 20-μl mixture with 0.5 μl of reverse transcriptase reaction products, Ex Taq polymerase (Takara, Kyoto, Japan) and the respective PTC+ mRNA primer sets (Table 3). Initial denaturation at 94°C for 3 min was followed by 30 cycles of PCR amplification at 94°C for 30 sec, 55°C for 30 sec, 72°C for 1 min, then 5 min incubation at 72°C.

Probes and their template plasmids

The GATEWAY conversion cassette B (including attB1, ccdB, Cmr and attB2 sites; Invitrogen) was inserted into the EcoRV site of pBluescript II SK+ (Stratagene, San Diego, CA, USA) to construct pBlueGATE. The AtUPF3 cDNA fragment (827–1315 nt in GenBank accession no. AC079286-T15K4.3 mRNA) was amplified from cDNA of a wild-type plant by PCR (AtUPF3-5F and AtUPF3-7R or attB1 and attB2 were used for the first and second rounds of PCR, respectively) and subcloned in a pDONR201 vector (Invitrogen) to obtain a pDONR201/AtUPF3 probe with the aid of BP clonase (Invitrogen). Next, the AtUPF3 fragment was transferred to pBlueGATE by LR clonase (Invitrogen) to generate pBlueGATE/AtUPF3si. These PCR, BP and LR reactions were performed according to the manufacturer's instructions.

Each PTC+ mRNA was amplified from atupf3-1 mutant cDNA by PCR using the primers shown in Table 3, and subcloned in the pCRII vector (Invitrogen) with a TOPO TA Cloning Kit Dual Promoter (Invitrogen). Subcloning was performed according to the manufacturer's instructions. Each DIG-labeled PTC+ mRNA sequence and AtUPF3 antisense probes was obtained from pCRII harboring a respective PTC+ mRNA sequence or pBlueGATE/AtUPF3si template DNA using a DIG-RNA labeling kit (Roche, Grenzacherstrasse, Switzerland).

Evaluation of PTC+/PTC− mRNA ratios by RT-PCR direct sequencing

We tested whether or not the PCR direct sequencing was semi-quantitative. PTC− and PTC+ At3g63340 mRNA sequences cloned in pCRII were mixed at ratios of 1.0:0.1, 1.0:0.5 and 1.0:1.0, respectively. These mixtures were diluted and 50-pg equivalents were used for PCR in a 20-μl mixture of Ex Taq polymerase and NMDT2-F and NMDT2-R primer sets (Table 3). Initial denaturation at 94°C for 3 min was followed by 30 cycles of PCR amplification at 94°C for 30 sec, 55°C for 30 sec, 72°C for 1 min, then 5 min incubation at 72°C. Fifteen microliters of PCR product was then precipitated with 15 μl of isopropanol and 1.5 μl of 10 m ammonium acetate, and the resultant pellet was rinsed with 70% ethanol and diluted in 10 μl TE buffer. Approximately 50 ng of PCR products was used for each sequence analysis using BigDye terminator v3.1 and an ABI310 sequencer (Applied Biosystems, Foster City, CA, USA). The NMDT2-R primer used for PCR was also used for sequencing. The ratios of the average sequencing signal peak heights were compared with the ratio of PTC+ mRNA/PTC− mRNA. The data are shown in Supplementary Material. RT-PCR direct sequencing was examined independently three times (Supplementary Material). The results indicated that the rates of average peak heights reflected the PTC+/PTC− mRNA ratios, thus RT-PCT direct sequencing could be used for semi-quantitative analysis.

Measurement of RNA degradation

Expanded leaves of approximately 7-week-old plants were cut into rectangular sections (about 5 mm wide) and used for inhibitor treatment. Leaf sections were then suspended in a 10-fold dilution of MS medium [0.46 g l−1 Murashige and Skoog Plant Salt Mixture (WAKO, Osaka, Japan), 3 g l−1 sucrose, pH 5.8]. The concentrations of each inhibitor tested were: cycloheximide, 20 μm and ActinomycinD, 100 μg ml−1, as previously described (Holtorf et al., 1999; Lambein et al., 2003). The leaf sections were vacuum-infiltrated with the inhibitors and incubated at room temperature. Total RNA was harvested and purified 4 h after inhibitor treatment as previously described (Shirzadegan et al., 1991). Two-microgram samples of total RNA were used to synthesize the cDNA. PCR was performed in a 20-μl mixture with 0.5 μl of reverse transcriptase reaction products. Fifteen microliters of PCR product was used for direct sequencing.

Northern blot analysis

PolyA+ mRNA was purified using Magextractor mRNA (Toyobo, Osaka, Japan) from 100 μg total mRNA. PolyA+ RNAs (500 ng) were treated with 50% formamide and 2.2 m formaldehyde in MOPS buffer (20 mm 4-morpholinepropanesulfonic acid, 5 mm sodium acetate, 1 mm EDTA, pH 7.0) and separated in 1% agarose gels containing 20 mm MOPS buffer. The RNA was then transferred onto Hybond N+ membranes by capillary blotting. After drying at room temperature, the membranes were irradiated with a UV lamp at 70 000 μJ cm−2 using a UV crosslinker (Stratagene).

Each PTC+ mRNA or AtUPF3 probe was hybridized with the membranes in modified Church buffer (50% formamide, 5× SSC, 500 mm Na2HPO4, pH 7.0, 7% SDS, 0.1%N-lauroylsarcosine) at 68°C overnight. The membranes were then washed twice with 2× SSC containing 0.1% SDS each for 15 min at 68°C, and twice with 0.5× SSC containing 0.1% SDS each for 15 min at 68°C. The hybridized DIG-RNA probes were detected using anti-digoxigenin-Ab Fab fragments (Roche, Mannheim, Germany) and CDP-star (Amersham Biosciences, Piscataway, NJ, USA).

Acknowledgements

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

We thank the Salk Institute Genome Analysis Laboratory for providing the sequence-index Arabidopsis T-DNA insertion mutants, and the Arabidopsis Biological Resource Center Stock Center (Ohio State University, Columbus) for providing the seeds.

K.H. was supported by a Research fellowship from The Japanese Society for the Promotion of Science for Young Scientists. This work was supported in part by a Spatiotemporal Network of RNA Information Flow Grant-in-Aid for scientific research, priority area (A), awarded by the Ministry of Education, Culture, Sports, Science, and Technology, Japan (no. 14035215) to Y.W.

References

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

Supporting Information

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

Supplemental data 1 evaluation of PTC+/PTC- mRNA ratios by RT-PCR direct sequencing. (A) Measurement point of PTC+ and PTC- At2g63340 mRNA. Red arrow heads indicate measurement points for PTC- mRNA. Black arrow heads indicate measurement points for PTC+ mRNA. These measurement points were unaffected by other nucleotide sequences of PTC+ and PTC- mRNAs. The PTC+/PTC- ratio was determined using the average of peak height ratios at each point. (B) Reproducibility and quantitative precision of PTC+/PTC- mRNA evaluations by PCR direct sequencing. ?Input? indicates the mixing ratios of control PCR template and ?Output? indicates the measurement results. Results for three experiments are denoted in graph (C). Supplemental data 2-B The difference in PTC+/PTC- ratios in whole ground parts by plant age.

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

FilenameFormatSizeDescription
TPJ_2473_sm_FigureS1.ppt74KSupporting info item
TPJ_2473_sm_FigureS2.ppt74KSupporting info item
TPJ_2473_sm_Suppmat.ppt74KSupporting info item
TPJ_2473_sm_Suppmat_2.ppt74KSupporting info item

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