Peering through the pore

Abstract

In eukaryotes, most genetic material resides in a separate membrane-bound compartment known as the nucleus. Transport of cargo, such as RNA and protein, across this barrier is facilitated by the nuclear pore complex (NPC). In the July issue of Plant Physiology, we showed that a component of the NPC, Arabidopsis thaliana TPR (AtTPR), is required for normal development. Two striking phenotypes of attpr mutants are that they are early flowering and show an accumulation of polyadenylated RNA in the nucleus. In addition, the expression of several microRNAs (miRNAs) is reduced in attpr mutants. In this addendum, we have examined the effect of AtTPR on the expression of miRNA targets. Our results show that miRNA targets are more likely to be upregulated than other transcripts in attpr mutants. For example, when comparing the nuclear RNA pool between wild-type and attpr plants, we found that 75% of the miRNA targets showing a significant change in transcript level are upregulated in attpr mutants. Although the targets of some miRNAs were upregulated, other miRNA targets were relatively unaffected by attpr mutations. Thus it appears that AtTPR may be required for the proper expression or localization of a subset of miRNAs.

Introduction

Because the nucleus is physically separated from the rest of the cell by the nuclear membrane, the transport of molecules between the nucleus and the cytoplasm is facilitated by the nuclear pore complex (NPC). The NPC is a large structure that traverses the nuclear membrane and is composed of approximately 30 proteins arranged in an eight-fold radially symmetric complex.¹ The nuclear face of the NPC is associated with a filamentous structure called the nuclear basket, the fibers of which extend 30–60 nm into the nucleus.² One of the major components of the nuclear basket in vertebrates is TRANSLOCATED PROMOTER REGION (TPR). TPR is a 267-kDa coiled-coil protein that is thought to constitute a scaffold for the assembly of other nuclear-basket components.² Two recent papers have examined the Arabidopsis homolog of TPR, Arabidopsis thaliana TPR (AtTPR³)/(NUCLEAR PORE ANCHOR (NUA⁴). Consistent with the localization of TPR in animals, AtTPR localizes to the inner side of the nuclear membrane.⁴ attpr mutants have a number of phenotypes including early flowering (Fig. 1A), reduced plant size and fertility, and alterations in auxin signaling. These pleiotropic phenotypes may be a result of reduced mRNA export from the nucleus; attpr mutants contain approximately eight times more polyadenylated RNA in the nucleus than wild type.³

Figure 1.

Effect of AtTPR on development and miRNA-target levels. (A) Wild type (left) and attpr mutant plants (right). (B) Effect of attpr mutations on the expression of all genes and miRNA targets. Shown is the percentage of genes, among genes showing a statistically significant (p < 0.05) change in expression, with increased (black bars) or decreased expression (white bars). The numbers above the bars indicates the number of transcripts. p values indicate the statistical significance of the difference in gene regulation between all transcripts and miRNA targets. (C) Number of target genes showing no change in expression (grey bars), upregulation (black bars), or downregulation (white bars) for each of the indicated miRNAs.

Effects of AtTPR on Flowering Time

In our laboratory, attpr mutants were isolated in a genetic screen for suppressors of the late-flowering phenotype of luminidependens (ld) mutants.⁵ LD is a member of the “autonomous” floral-promotion pathway (AP), which acts to repress expression of the floral inhibitor FLOWERING LOCUS C (FLC).⁶ Loss-of-function mutations in any of the genes in the AP results in elevated FLC levels and delayed flowering. The late-flowering phenotype of AP mutants can be eliminated by loss-of-function mutations in FLC or by a prolonged exposure to cold temperatures, a process known as vernalization. Vernalization results in an epigenetic shut off of FLC expression that is mediated by changes in chromatin structure at the FLC locus.⁷

Given the strong suppression of the late-flowering phenotype of ld, we investigated the effect of attpr on FLC expression in wild type, ld, or FRIGIDA (FRI)-containing backgrounds. FRI is a dominant promoter of FLC found in naturally occurring late-flowering vernalization-responsive accessions of Arabidopsis. Thus, both ld and FRI- containing plants are late flowering due to elevated levels of FLC transcript. attpr mutations resulted in earlier flowering in all backgrounds tested (Fig. 1A). In ld and FRI-containing backgrounds, FLC transcript levels were significantly reduced, however, FLC levels remained higher than wild type. This result is significant because ld attpr and FRI attpr lines flower earlier than ld flc or FRI flc, thus the early-flowering phenotype of attpr cannot be entirely explained by a reduction in FLC levels and suggests that attpr may target other flowering time genes in addition to FLC. Three candidates for additional targets of attpr are the floral promoters FT, TWIN SISTER OF FT (TSF) and SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1). The expression of these genes is upregulated by attpr mutations. However, because these genes are also negatively regulated by FLC, it is not possible to determine if the upregulation of FT, TSF and SOC1 by attpr mutations is direct or a result of reduced FLC expression (of course, these two models are not mutually exclusive). It may be possible to further define the relationship between AtTPR, FLC, FT, TSF, and SOC1 by examining the effect of attpr mutations in the presence of a constitutively expressed copy of FLC (e.g., 35S::FLC). If an attpr mutation increases the expression of FT, TSF, or SOC1 in a 35S::FLC background, it would provide strong evidence that AtTPR has FLC-independent effects on the expression of these floral promoters.

miRNA Targets are Upregulated in attpr Mutants

An interesting phenotype of attpr mutants is the accumulation of polyadenylated RNA in the nucleus.^3,4 Similar phenotypes have been reported in mutants for other nuclear pore components.^8,9 Presumably, the increase in nuclear polyadenylated RNA levels observed in these mutants is a result of reduced export through the nuclear pore. To better characterize the effect of AtTPR on RNA transport between the nucleus and the cytoplasm, we used micro-arrays to compare the nuclear and total mRNA pools between wild-type and attpr plants. Interestingly, we found that, although the total amount of polyadenylated RNA in the nucleus of attpr mutants was approximately eight times higher than wild type, the composition of the RNA pools between the two genotypes was relatively similar (which may help to explain the fact that attpr mutants are still viable). Thus it appears that attpr mutations have a fairly general effect on the export of mRNAs from the nucleus.

One group of genes whose expression we have examined in greater detail is those that are predicted to be targets of miRNAs. Because we have previously shown that the expression of several miRNAs is reduced in attpr mutants,³ we predicted that miRNA targets may be upregulated in attpr mutants. For this analysis we compared the expression of all transcripts present on the arrays to a set of 226 genes predicted to be targets of miRNAs (Arabidopsis Small RNA Project, http://asrp.cgrb.oregonstate.edu/). In each comparison, we determined the percentage of transcripts showing a statistically significant (p > 0.05) change in expression between the two conditions. To eliminate errors due to low signal strength, transcripts were considered only if, in at least one condition, the transcript was called “present” in at least four of the five replicates.

In a comparison of total mRNA between wild type and the attpr mutant, of the genes showing a significant change in expression, the fraction of miRNA targets showing an increase in gene expression in attpr (58.7%) was found to be significantly higher than the fraction of all genes showing an increase in gene expression (52.8%) (Fig. 1B). The difference in expression was even more dramatic in the nuclear mRNA fraction; in the attpr mutant 75.3% of miRNA targets showed increased expression, compared to 55.1% for all transcripts. This increase in the levels of miRNA targets in the nucleus in attpr suggests that these transcripts are either preferentially retained in the nucleus or their stability/transcription is increased, possibly by reduced nuclear import of components required for miRNA-mediated elimination.

In our previous work, we showed that a subset of miRNAs had reduced levels in the attpr mutant.³ Therefore, we were curious to determine if the differences in miRNA target levels observed in the nuclear RNA fractions are consistent with the downregulation of only a subset of miRNAs. Many miRNAs are predicted to target multiple transcripts; therefore lower expression of a single miRNA should result in the increased expression of multiple targets. For this analysis, we chose miRNAs with at least five detectable targets and determined the number of targets showing no change, increased, or decreased expression in the attpr mutant. The expression of the groups of miRNA targets was quite variable (Fig. 1C). The majority of the targets of miR165, miR172, miR156 and miR396 were increased in expression, whereas the targets of other miRNAs, such as miR164, miR400, miR859, miR319 and miR846 most often showed no change in expression. Thus, overall, these results are consistent with AtTPR being required for a proper expression or localization of only a subset of miRNAs. It should be noted, however, that it is not known if all miRNAs are expressed during the developmental stage at which RNAs were extracted for the microarray analysis (whole plants with one open flower). Thus it is possible that some miRNA targets show no change in expression in the attpr mutant simply because a particular miRNA is not expressed at this stage of development.

Addendum to: Jacob Y, Chareerat M, Veley KM, Kim SY, Michaels SD. The nuclear pore protein AtTPR is required for RNA homeostasis, flowering time, and auxin signaling. Plant Physiol. 2007;144:1383–1390.