Search

Certain introns significantly increase mRNA accumulation by a poorly understood mechanism. These introns have no effect when located upstream, or more than ~1 Kb downstream, of the start of transcription. We tested the ability of a formerly non-stimulating intron containing 11 copies of the sequence TTNGATYTG, which is over-represented in promoter-proximal introns in Arabidopsis thaliana, to affect expression from various positions. The activity profile of this intron at different locations was similar to that of a natural intron from the UBQ10 gene, suggesting that the motif increases mRNA accumulation by the same mechanism. A series of introns with different numbers of this motif revealed that the effect on expression is linearly dependent on motif copy number up to at least 20, with each copy adding another 1.5-fold increase in mRNA accumulation. Furthermore, 6 copies of the motif stimulated mRNA accumulation to a similar degree from within an intron or when introduced into the 5′-UTR and coding sequences of an intronless construct, demonstrating that splicing is not required for this sequence to boost expression. The ability of this motif to substantially elevate expression from several hundred nucleotides downstream of the transcription start site reveals a novel type of eukaryotic gene regulation.

To more precisely define the positions from which certain intronic regulatory sequences increase mRNA accumulation, the effect of a UBIQUITIN intron on gene expression was tested from six different positions surrounding the transcription start site (TSS) of a reporter gene fusion in Arabidopsis thaliana The intron increased expression from all transcribed positions but had no effect when upstream of the 5'-most TSS. While this implies that the intron must be transcribed to increase expression, the TSS changed when the intron was located in the 5'-untranslated region (UTR), suggesting that the intron affects transcription initiation. Remarkably, deleting 303 nucleotides of the promoter including all known TSSs and all but 18 nucleotides of the 5'-UTR had virtually no effect on the level of gene expression as long as an intron containing stimulatory sequences was included. Instead, transcription was initiated in normally untranscribed sequences the same distance upstream of the intron as when the promoter was intact. These results suggest that certain intronic DNA sequences play unexpectedly large roles in directing transcription initiation and constitute a previously unrecognized type of downstream regulatory element for genes transcribed by RNA polymerase II.

Introns are often added to transgenes to increase expression, although the mechanism through which introns stimulate gene expression in plants and other eukaryotes remains mysterious. While introns vary in their effect on expression, it is unknown whether different genes respond similarly to the same stimulatory intron. Furthermore, the degree to which gene regulation is preserved when expression is increased by an intron has not been thoroughly investigated. To test the effects of the same intron on the expression of a range of genes, GUS translational fusions were constructed using the promoters of eight Arabidopsis genes whose expression was reported to be constitutive (GAE1, CNGC2 and ROP10), tissue specific (ADL1A, YAB3 and AtAMT2) or regulated by light (ULI3 and MSBP1). For each gene, a fusion containing the first intron from the UBQ10 gene was compared to fusions containing the gene's endogenous first intron (if the gene has one) or no intron. In every case, the UBQ10 intron increased expression relative to the intronless control, although the magnitude of the change and the level of expression varied. The UBQ10 intron also changed the expression patterns of the CNGC2 and YAB3 fusions to include strong activity in roots, indicating that tissue specificity was disrupted by this intron. In contrast, the regulation of the ULI3 and MSBP1 genes by light was preserved when their expression was stimulated by the intron. These findings have important implications for biotechnology applications in which a high level of transgene expression in only certain tissues is desired.

Related motifs strongly increase gene expression when added to an intron located in coding sequences. Many introns greatly increase gene expression through a mechanism that remains elusive. An obstacle to understanding intron-mediated enhancement (IME) has been the difficulty of locating the specific intron sequences responsible for boosting expression because they are redundant, dispersed, and degenerate. Previously we used the IMEter algorithm in two independent ways to identify two motifs (CGATT and TTNGATYTG) that are candidates for involvement in IME in Arabidopsis. Here we show that both motifs are sufficient to increase expression. An intron that has little influence on expression was converted into one that increased mRNA accumulation 24-fold and reporter enzyme activity 40-fold relative to the intronless control by introducing 11 copies of the more active TTNGATYTG motif. This degree of stimulation is twice as large as that of the strongest of 15 natural introns previously tested in the same reporter gene. Even though the CGATT and TTNGATYTG motifs each increased expression, and CGATT matches the NGATY core of the longer motif, combining the motifs to make TTCGATTTG reduced the stimulating ability of the TTNGATYTG motif. Additional substitutions were used to test the contribution to IME of other residues in the TTNGATYTG motif. The verification that these motifs are active in IME will improve our ability to predict the stimulating ability of introns, to engineer any intron to increase expression to a desired level, and to explore the mechanism of IME by seeking factors that might interact with these sequences.

Introns that elevate mRNA accumulation have been found in a wide range of eukaryotes. However, not all introns affect gene expression, and direct testing is currently the only way to identify stimulatory introns. Our genome-wide analysis in Arabidopsis thaliana revealed that promoter-proximal introns as a group are compositionally distinct from distal introns and that the degree to which an individual intron matches the promoter-proximal intron profile is a strong predictor of its ability to increase expression. We found that the sequences responsible for elevating expression are dispersed throughout an enhancing intron, as is a candidate motif that is overrepresented in first introns and whose occurrence in tested introns is proportional to its effect on expression. The signals responsible for intron-mediated enhancement are apparently conserved between Arabidopsis and rice (Oryza sativa) despite the large evolutionary distance separating these plants.

To investigate how plants acquire and assimilate sulfur from their environment, we isolated and characterized two mutants of Arabidopsis thaliana deficient in sulfate transport. The mutants are resistant to selenate, a toxic analogue of sulfate. They are allelic to each other and to the previously isolated sel1 (selenate-resistant) mutants, and have been designated sel1-8 and sel1-9. Root elongation in these mutants is less sensitive to selenate than in wild-type plants. Sulfate uptake into the roots is impaired in the mutants under both sulfur-sufficient and sulfur-deficient conditions, but transport of sulfate to the shoot is not affected. The sel1 mutants contain lesions in the sulfate transporter gene Sultr1;2 located on the lower arm of chromosome 1. The sel1-1, sel1-3 and sel1-8 mutants contain point mutations in the coding sequences of Sultr1;2, while the sel1-9 mutant has a T-DNA insertion in the Sultr1;2 promoter. The Sultr1;2 cDNA derived from wild-type plants is able to complement Saccharomyces cerevisiae mutants defective in sulfate transport, but the Sultr1;2 cDNA from sel1-8 is not. The Sultr1;2 gene is expressed mainly in roots, and accumulation of transcripts increases during sulfate deprivation. Examination of transgenic plants containing the Sultr1;2 promoter fused to the GUS-reporter gene indicates that Sultr1;2 is expressed mainly in the root cortex, the root tip and lateral roots. Weaker expression of the reporter gene was observed in hydathodes, guard cells and auxiliary buds of leaves, and in anthers and the basal parts of flowers. The results indicate that Sultr1;2 is primarily involved in importing sulfate from the environment into the root.

Nine blue fluorescent mutants of the flowering plant Arabidopsis thaliana were isolated by genetic selections and fluorescence screens. Each was shown to contain a recessive allele of trp1, a previously described locus that encodes the tryptophan biosynthetic enzyme phosphoribosylanthranilate transferase (PAT, called trpD in bacteria). The trp1 mutants consist of two groups, tryptophan auxotrophs and prototrophs, that differ significantly in growth rate, morphology, and fertility. The trp1 alleles cause plants to accumulate varying amounts of blue fluorescent anthranilate compounds, and only the two least severely affected of the prototrophs have any detectable PAT enzyme activity. All four of the trp1 mutations that were sequenced are G to A or C to T transitions that cause an amino acid change, but in only three of these is the affected residue phylogenetically conserved. There is an unusually high degree of sequence divergence in the single-copy gene encoding PAT from the wild-type Columbia and Landsberg erecta ecotypes of Arabidopsis.

Many introns significantly increase gene expression through a process termed intron-mediated enhancement (IME). Introns exist in the transcribed DNA and the nascent RNA, and could affect expression from either location. To determine which is more relevant to IME, hybrid introns were constructed that contain sequences from stimulating Arabidopsis thaliana introns either in their normal orientation or as the reverse complement. Both ends of each intron are from the non-stimulatory COR15a intron in their normal orientation to allow splicing. The inversions create major alterations to the sequence of the transcribed RNA with relatively minor changes to the DNA structure. Introns containing portions of either the UBQ10 or ATPK1 intron increased expression to a similar degree regardless of orientation. Also, computational predictions of IME improve when both intron strands are considered. These findings are more consistent with models of IME that act at the level of DNA rather than RNA.

Introns in a wide range of organisms including plants, animals and fungi are able to increase the expression of the gene that they are contained in. This process of intron-mediated enhancement (IME) is most thoroughly studied in Arabidopsis thaliana, where it has been shown that enhancing introns are typically located near the promoter and are compositionally distinct from downstream introns. In this study, we perform a comprehensive comparative analysis of several sequenced plant genomes. We find that enhancing sequences are conserved in the multi-cellular plants but are either absent or unrecognizable in algae. IME signals are preferentially located towards the 5′-end of first introns but also appear to be enriched in 5′-UTRs and coding regions near the transcription start site. Enhancing introns are found most prominently in genes that are highly expressed in a wide range of tissues. Through site-directed mutagenesis in A. thaliana, we show that IME signals can be inserted or removed from introns to increase or decrease gene expression. Although we do not yet know the specific mechanism of IME, the predicted signals appear to be both functional and highly conserved.

Introns are often required for full expression of genes in organisms as diverse as plants, insects, nematodes, yeast, and mammals. To explore the potential mechanisms of intron-mediated enhancement in Arabidopsis thaliana, the effect of varying the position of an intron was determined using a series of reporter gene fusions between TRYPTOPHAN BIOSYNTHESIS1 (TRP1) and GUS. Two introns that differ in the degree to which they stimulate expression were individually tested at six locations within coding sequences and two positions in the 3'-UTR. The ability of the first introns from both the TRP1 and POLYUBIQUITIN10 (UBQ10) genes to elevate mRNA accumulation in transgenic plants was found to decline with distance from the promoter, despite their being efficiently spliced from all coding sequence locations. Neither intron significantly enhanced mRNA accumulation when positioned 1.1 kb or more from the start of transcription. In addition, measurements of GUS enzyme activity revealed that both introns at all locations elevated GUS activity more than they enhanced mRNA accumulation. The stimulation mediated by two of four other introns tested at the position nearest the promoter was also greater at the level of GUS activity than mRNA accumulation. These findings support a model in which introns increase transcription and promote translation by two distinct mechanisms.

To explore possible mechanisms of intron-mediated enhancement of gene expression, the features of PAT1 intron 1 required to elevate mRNA accumulation were systematically tested in transgenic Arabidopsis. This intron is remarkably resilient, retaining some ability to increase mRNA accumulation when splicing was prevented by mutation of 5' and 3' splice sites, branchpoint sequences, or when intron U-richness was reduced. Enhancement was abolished by simultaneously eliminating branchpoints and the 5' splice site, structures involved in the first two steps of spliceosome assembly. Although this suggests that the splicing machinery is required, intron splicing is clearly not enough to enhance mRNA accumulation. Five other introns were all efficiently spliced but varied widely in their ability to increase mRNA levels. Furthermore, PAT1 intron 1 was spliced but lost the ability to elevate mRNA accumulation when moved to the 3' UTR. These findings demonstrate that splicing per se is neither necessary nor sufficient for an intron to enhance mRNA accumulation, and suggest a mechanism that requires intron recognition by the splicing machinery but also involves nonconserved intron sequences.

Summary The expression of the Arabidopsis thaliana PAT1 gene, which encodes the tryptophan biosynthetic enzyme phosphoribosylanthranilate transferase, was investigated using translational fusions of the PAT1 promoter to the GUS reporter gene. Independent stably transformed A. thaliana lines containing a single copy of a fusion that includes the entire plastid transit peptide and the first two introns of PAT1 had on average 30 times more GUS enzyme activity than plants transformed with a construct in which GUS was fused a short distance downstream of the PAT1 start codon. Plants containing the construct without introns or leader peptide accumulated undetectable amounts of PAT1-GUS fusion protein and mRNA, even though the transcriptional rate of both fusion constructs was comparable. Fusions containing the entire transit peptide and either of the first two introns yield as much GUS activity as constructs containing both introns, but constructs containing the transit peptide but no introns give rise to much lower levels. Therefore, introns greatly enhance the expression of PAT1-GUS fusions, and they act post-transcriptionally to increase the steady-state level of mRNA.

A long term goal for this laboratory is to obtain a detailed understanding of the biochemical and genetic mechanisms that plants use to regulate the synthesis of the amino acid tryptophan (Last et al., 1992; Last, 1993). In addition to its role in the production of tryptophan for protein synthesis, this pathway gives rise to many indolic secondary metabolites, including the auxin, indole-3-acetic acid (Wright et al., 1991) and the Arabidopsis phytoallexin called camalexin (Tsuji et al., 1992). Thus, we anticipate that plants regulate the enzymes of tryptophan biosynthesis in response to requirements for both protein and secondary product biosynthesis.