Logo of plntcellLink to Publisher's site
Plant Cell. Dec 2012; 24(12): 4930–4947.
Published online Dec 7, 2012. doi:  10.1105/tpc.112.103697
PMCID: PMC3556967

LSM Proteins Provide Accurate Splicing and Decay of Selected Transcripts to Ensure Normal Arabidopsis Development[W]


In yeast and animals, SM-like (LSM) proteins typically exist as heptameric complexes and are involved in different aspects of RNA metabolism. Eight LSM proteins, LSM1 to 8, are highly conserved and form two distinct heteroheptameric complexes, LSM1-7 and LSM2-8,that function in mRNA decay and splicing, respectively. A search of the Arabidopsis thaliana genome identifies 11 genes encoding proteins related to the eight conserved LSMs, the genes encoding the putative LSM1, LSM3, and LSM6 proteins being duplicated. Here, we report the molecular and functional characterization of the Arabidopsis LSM gene family. Our results show that the 11 LSM genes are active and encode proteins that are also organized in two different heptameric complexes. The LSM1-7 complex is cytoplasmic and is involved in P-body formation and mRNA decay by promoting decapping. The LSM2-8 complex is nuclear and is required for precursor mRNA splicing through U6 small nuclear RNA stabilization. More importantly, our results also reveal that these complexes are essential for the correct turnover and splicing of selected development-related mRNAs and for the normal development of Arabidopsis. We propose that LSMs play a critical role in Arabidopsis development by ensuring the appropriate development-related gene expression through the regulation of mRNA splicing and decay.


During the last years, an increasing body of evidence indicates that posttranscriptional regulation plays an important role in modulating gene expression during development in eukaryotes (Halbeisen et al., 2008). Most eukaryotic genes are transcribed as precursors (pre-mRNAs) containing intron sequences. In order to yield correct translation products, introns need to be excised to generate mature mRNAs. This process, known as pre-mRNA splicing, is fundamental in both constitutive and regulated gene expression. Pre-mRNA splicing is precisely and efficiently performed by the spliceosome, a large ribonucleoprotein (RNP) complex machinery composed of five small nuclear RNP particles (U1, U2, U4/U6, and U5) and more than 200 polypeptides not tightly associated with snRNPs (Wahl et al., 2009). In many cases, however, the splicing process is flexible enough to allow the generation of alternative transcripts from a single gene by differential use of splicing sites. Site use may depend on the cell type, developmental stage, or physiological condition, thereby affecting protein diversity and transcript levels (Matlin et al., 2005). The general mechanism of splicing has been well studied in humans and yeast, being largely conserved between these organisms. In plants, the splicing process remains comparatively poorly understood, although the basic mechanisms of spliceosome assembly and intron excision appear to be as in the rest of eukaryotes (Lorković et al., 2000; Reddy, 2001). Consistent with this, the analysis of the Arabidopsis thaliana genome for the presence of known spliceosomal proteins indicated that the core of spliceosomal machinery is conserved between plants and animals (Wang and Brendel, 2004). Nonetheless, despite this conservation, incorrect splicing of mammalian pre-mRNAs in plant cells and vice versa denotes the existence of plant-specific splicing regulatory mechanisms requiring plant-specific splicing factors (Lorković et al., 2000; Reddy, 2001; Lorković, 2009). The characterization of different plant splicing proteins, including some Gly-rich RNA binding proteins, SR proteins, RNA helicases, and other RNA binding proteins, have revealed that they are essential for the accurate progress of diverse plant developmental processes (Raab and Hoth, 2007; Barta et al., 2008; Lorković, 2009; Deng et al., 2010; Zhang et al., 2011).

The control of mRNA turnover is another critical aspect in the regulation of eukaryotic gene expression. Two major pathways exist in yeast and mammals for mRNA decay, both of them being initiated by deadenylation through the CARBON CATABOLITE REPRESSION4/PGK PROMOTER DIRECTED OVERPRODUCTION2/NEGATIVE ON TATA-1 complex (Meyer et al., 2004; Parker and Song, 2004). Subsequently, transcripts can be processed by the 3′ to 5′ or the 5′ to 3′ decay pathways. In the first pathway, the deadenylated mRNA is degraded by a complex of proteins known as the exosome (Anderson and Parker, 1998). In the second pathway, the mRNA is decapped by the mRNA DECAPPING1 (DCP1)/DCP2 enzyme, making the mRNA susceptible to the EXORIBONUCLEASE1 (XRN1) (Beelman et al., 1996; Dunckley and Parker, 1999). Therefore, decapping is an important node in the regulation of mRNA lifespan and is modulated by a set of different proteins (Bonnerot et al., 2000; Coller et al., 2001). The decapping machinery accumulates in discrete cytoplasmic foci named processing bodies (P-bodies), which have been suggested to be functionally involved not only in mRNA decapping (Sheth and Parker, 2003; Cougot et al., 2004) but also in nonsense-mediated mRNA decay (Unterholzner and Izaurralde, 2004; Sheth and Parker, 2006), mRNA storage (Brengues et al., 2005), general translation repression (Coller and Parker, 2005), and microRNA-mediated repression (Bhattacharyya et al., 2006). Although the existence of both the 5′ to 3′ and the 3′ to 5′ decay pathways has been documented and their core components identified (Xu et al., 2006; Goeres et al., 2007; Belostotsky and Sieburth, 2009; Lange and Gagliardi, 2010), the governing principles of mRNA decay in plants, as in the case of the splicing process, are still poorly known. Moreover, genetic analyses have also uncovered plant-specific functional features in mRNA degradation pathways that are associated with plant-specific factors (Belostotsky and Sieburth, 2009; Xu and Chua, 2011). In Arabidopsis, for instance, no XRN1-like gene has been identified. Instead, the cytoplasmic 5′ to 3′ exoribonuclease activity is performed by XRN4 (Kastenmayer and Green, 2000; Souret et al., 2004). Plant P-bodies seem to function as yeast and human P-bodies. However, they also contain their own distinct protein components (Xu and Chua, 2011). Plants affected in mRNA turnover display severe developmental perturbations, indicating that proteins related to mRNA decapping and decay play important roles in regulating gene expression during plant development (Xu et al., 2006; Goeres et al., 2007; Belostotsky and Sieburth, 2009; Xu and Chua, 2009, 2011).

The SM-like proteins (LSMs) constitute a large family of proteins that function in multiple aspects of RNA metabolism. In yeast and animals, there are eight highly conserved LSM proteins (LSM1 to LSM8) that form two different heptameric ring complexes, LSM1-7 and LSM2-8, localized in the cytoplasm and nucleus, respectively. LSM1 and LSM8 define and confer the specificity to each complex, while the other proteins (LSM2 to LSM7) participate in both cytoplasmic and nuclear complexes. The LSM1-7 cytoplasmic complex binds to oligoadenylated mRNAs, promoting their decapping and subsequent degradation by the 5′ to 3′ pathway, and accumulates in P-bodies. The LSM2-8 nuclear complex binds to and stabilizes the U6 small nuclear RNA (snRNA), forms the core of the U6 small nuclear RNP, and functions in pre-mRNA splicing (reviewed in Beggs, 2005; Tharun, 2009). In silico approaches have allowed the identification of potential plant homologs of LSM proteins. Arabidopsis has homologs for the eight conserved LSMs, and three of them (LSM1, LSM3, and LSM6) are duplicated (Wang and Brendel, 2004). To date, however, plant LSMs have not been functionally characterized, and their role in RNA metabolism remains to be determined. Only Arabidopsis LSM5 and LSM4 genes have been experimentally studied, both of them being related to abscisic acid and osmotic stress signaling (Xiong et al., 2001; Deng et al., 2010; Zhang et al., 2011). Here, we report the molecular and functional characterization of the Arabidopsis LSM gene family. Our results indicate that Arabidopsis LSM proteins are also organized in two different heptameric complexes localized in the cytoplasm and nucleus. Whereas the cytoplasmic complex (LSM1-7) is involved in P-body formation, mRNA decapping, and, therefore, accurate mRNA decay, the nuclear complex (LSM2-8) is required for U6 snRNA stabilization and, consequently, proper pre-mRNA splicing. Genetic and molecular analyses reveal that LSM1-7 and LSM2-8 complexes are essential for the correct turnover and splicing of selected development-related mRNAs, respectively. Consistent with this, the absence of LSM1 and LSM8 proteins causes severe perturbations in Arabidopsis deveopment, which correlates with alterations in developmentally regulated gene expression. We conclude that LSMs play a critical role in Arabidopsis development by ensuring the appropriate development-related gene expression through the regulation of mRNA splicing and decay.


The Arabidopsis Genome Contains 11 Genes Encoding the Eight Highly Conserved LSM Proteins

Sequence comparisons and motif searches allowed the identification of 11 genes in the Arabidopsis genome encoding proteins related to the eight highly conserved proteins that in yeast and animals constitute the heptameric LSM complexes, LSM1-7 and LSM2-8, with three of them, the putative LSM1, LSM3, and LSM6, being duplicated (Wang and Brendel, 2004). The predicted Arabidopsis proteins contain the Sm1 and Sm2 motifs that are separated by a nonconserved linker region of variable length and conform the Sm bipartite domain typical of LSM proteins (Tharun, 2009) (see Supplemental Figure 1 online). LSM proteins have also been found in the genomes of different plant species (Proost et al., 2009; Goodstein et al., 2012). A phylogenetic analysis was performed with the LSM proteins from Arabidopsis and other representative plant species, including soybean (Glycine max), poplar (Populus trichocarpa), rice (Oryza sativa), and maize (Zea mays). Results revealed that all plant genomes analyzed contain genes encoding LSMs related to the eight conserved proteins from yeast and animals and that many of them are present in more than one copy as in the case of Arabidopsis LSM1, LSM3, and LSM6. The human LSM proteins were also included in the analysis as an internal control (see Supplemental Figure 2 and Supplemental Data Set 1 online).

RNA gel blot analysis revealed that the 11 Arabidopsis LSM genes are expressed in all organs tested, including leaves, roots, flowers, and stems. Each pair of duplicated genes exhibited the same expression pattern (Figure 1A). To determine the expression of LSM genes at the tissue level, transgenic Arabidopsis plants containing fusions between all LSM promoters (LSMpro) and the β-glucuronidase (GUS) reporter gene were generated and assayed for GUS activity. Consistent with the results obtained from the RNA gel blot experiments, nearly constitutive GUS activity was observed in all cases. In leaves and cotyledons, GUS staining was preferentially detected in the vascular tissues. As representative examples, the expression of LSM8pro-GUS, LSM1Apro-GUS, and LSM1Bpro-GUS is shown (Figures 1B to to1E;1E; see Supplemental Figure 3 online). These results demonstrate that the 11 Arabidopsis LSM genes are active and ubiquitously expressed.

Figure 1.
Expression Patterns of Arabidopsis LSM Genes.

Subcellular Localization of Arabidopsis LSM Proteins

To investigate the subcellular localization of Arabidopsis LSM proteins, transgenic Arabidopsis expressing genomic LSM–green fluorescent protein (LSM-GFP) fusions driven by the corresponding LSMpro were obtained and analyzed. We first examined the subcellular localization of LSM1A, LSM1B, and LSM8, the Arabidopsis putative homologs of yeast and animal LSM proteins that differentiate the cytoplasmic and nuclear complexes, respectively. In root cells from seedlings expressing LSM1Apro-LSM1A-GFP or LSM1Bpro-LSM1B-GFP, green fluorescence suggested a cytoplasmic localization of LSM1A and LSM1B (Figure 2A). Conversely, in seedlings expressing the LSM8pro-LSM8-GFP fusion, green fluorescence was specifically localized in nuclei (Figure 2A). We also investigated the subcellular localization of Arabidopsis LSM3A, LSM3B, and LSM4, whose related yeast and animal proteins participate in both cytoplasmic and nuclear LSM complexes. In root cells from seedlings expressing LSM3Apro-LSM3A-GFP or LSM3Bpro-LSM3B-GFP, green fluorescence was detected in both nuclei and cytoplasm, indicating that LSM3A and LSM3B proteins simultaneously localize to these subcellular compartments (Figure 2A). Similar results were obtained when studying the subcellular localization of the LSM4-GFP fusion protein in seedlings expressing LSM4pro-LSM4-GFP (Figure 2A). These data strongly suggest that Arabidopsis LSM proteins have subcellular localizations similar to the LSM proteins from other eukaryotes.

Figure 2.
Subcellular Localization of Arabidopsis LSM Proteins.

Yeast and human LSM1-7 proteins have been described to accumulate in P-bodies (Ingelfinger et al., 2002; Sheth and Parker, 2003). We therefore examined whether Arabidopsis LSM proteins belonging to the cytoplasmic complex also localized in these cytoplasmic foci. P-bodies are rarely observed in plants growing under control conditions, whereas their number and size markedly increase under conditions that are associated with high levels of mRNA turnover, such as hypoxic or heat stress (Weber et al., 2008). When seedlings expressing LSM1Apro-LSM1A-GFP or LSM1Bpro-LSM1B-GFP were exposed to heat stress, LSM1A-GFP and LSM1B-GFP were largely localized to discrete cytoplasmic spots (Figure 2B). Under heat stress conditions, LSM3A-GFP, LSM3B-GFP, and LSM4-GFP fusion proteins also localized to cytoplasmic foci in root cells from seedlings expressing LSM3Apro-LSM3A-GFP, LSM3Bpro-LSM3B-GFP, or LSM4pro-LSM4-GFP, respectively (Figure 2B). Following cycloheximide treatment, which in yeast and humans results in the loss of P-bodies (Sheth and Parker, 2003), no cytoplasmic foci were observed in any case (Figure 2C), suggesting that the detected cytoplasmic spots of LSM-GFP fusion proteins corresponded to P-bodies. Consistent with its specific nuclear localization, LSM8-GFP did not accumulate in cytoplasmic spots in LSM8pro-LSM8-GFP seedlings exposed to heat stress (Figure 2B) or to heat stress plus cycloheximide (Figure 2C). To confirm that the foci defined by Arabidopsis cytoplasmic LSM proteins corresponded to P-bodies, we further analyzed their colocalization with DCP1, a protein that belongs to the Arabidopsis decapping complex and accumulates in P-bodies (Xu et al., 2006; Goeres et al., 2007). The examination of seedlings expressing LSM1Apro-LSM1A-GFP or LSM1Bpro-LSM1B-GFP cotransformed with a 35S–red fluorescent protein (RFP)–DCP1 fusion revealed that, in fact, LSM1A-GFP and LSM1B-GFP colocalized with RFP-DCP1 in root cells grown at room temperature (20°C) or exposed to 37°C (Figure 2D). Taken together, these observations demonstrate that Arabidopsis cytoplasmic LSM proteins accumulate in P-bodies.

DCP2 and VARICOSE (VCS), like DCP1, also belong to the Arabidopsis decapping complex and accumulate in P-bodies (Xu et al., 2006; Goeres et al., 2007). Accordingly, GFP-DCP2 and GFP-VCS fusion proteins localized to P-bodies within the cytoplasm of root cells from wild-type seedlings containing 35S-GFP-DCP2 or 35S-GFP-VCS constructs, respectively, exposed to heat stress (Figure 2E). When these constructs were introduced into an Arabidopsis mutant defective in LSM1 proteins (lsm1a lsm1b; see below) under the same stress conditions, the P-bodies were lost and the GFP-DCP2 and GFP-VCS signals were mostly dispersed in the cytosol (Figure 2E). From these results we conclude that, in addition to accumulating in P-bodies, LSM1 proteins are required for P-body formation in Arabidopsis.

Organization of Arabidopsis LSM Proteins

As mentioned above, yeast and animal LSM proteins typically exist as highly organized ring-shaped heptameric complexes (Figure 3A) (Tharun, 2009). Having established that Arabidopsis LSM proteins subcellularly localize as in other eukaryotes, we decided to study how they are organized. For this, we assayed in vivo LSM–LSM interactions by means of bimolecular fluorescence complementation (BiFC) (Hu et al., 2002; Walter et al., 2004) in Nicotiana benthamiana leaves. In yeast and animal LSM complexes, LSM1 and LSM8 are flanked by LSM2 and LSM4 (Figure 3A). Our experiments revealed that a significant proportion of cells cotransformed with LSM1A-nGFP and LSM2-cGFP or LSM4-cGFP, and LSM8-nGFP and LSM2-cGFP or LSM4-cGFP displayed green fluorescence (Figure 3B). Identical results were obtained cotransforming LSM1B-nGFP with LSM2-cGFP or LSM4-cGFP (see Supplemental Figure 4 online), indicating that Arabidopsis LSM1A, LSM1B, and LSM8 are capable of interacting in vivo with LSM2 and LSM4. Consistent with the typical cytoplasmic localization of LSM1 proteins in Arabidopsis (Figure 2A), LSM1(A or B)–LSM2 and LSM1(A or B)–LSM4 interactions mainly appeared in the cytoplasm of the N. benthamiana cells (Figure 3B; see Supplemental Figure 4 online). Conversely, interactions between LSM8 and LSM2 and LSM4 were essentially detected in the nucleus (Figure 3B), which is consistent with the characteristic nuclear localization of Arabidopsis LSM8 protein (Figure 2A). The specificity of all these interactions was demonstrated by the fact that, as expected from their different subcellular localization, we did not observe interaction between LSM1 proteins and LSM8 (Figure 3B; see Supplemental Figure 4 online). Interactions between LSM2 and LSM4 proteins were not found either (Figure 3B), in agreement with what has been proposed for yeast and animal LSM complexes (Figure 3A) (Beggs, 2005). However, we detected interactions between LSM2 and LSM3 (A or B), LSM3 (A or B), and LSM6 (A or B); LSM6 (A or B) and LSM5; LSM5 and LSM7; and LSM7 and LSM4 (Figure 3C; see Supplemental Figure 4 online). These interactions parallel those proposed for LSM complexes from other eukaryotes (Beggs, 2005) and were observed simultaneously in both cytoplasm and nucleus (Figure 3C; see Supplemental Figure 4 online), consistent with the subcellular localization of the corresponding LSMs (see above). In addition, also according to the interactions assumed in other LSM complexes (Beggs, 2005), we did not detect interactions between LSM2 and LSM7, LSM4 and LSM6 (A or B), LSM6 (A or B) and LSM7, and LSM5 and LSM3 (A or B) (Figure 3C; see Supplemental Figure 4 online). All of these data indicate that Arabidopsis LSMs are organized in two heptameric ring complexes localized in the cytoplasm (LSM1-7) and the nucleus (LSM2-8).

Figure 3.
Organization of Arabidopsis LSM Proteins.

In yeast and animals, cytoplasmic and nuclear LSM complexes are determined by the presence of LSM1 and LSM8 proteins, respectively (Tharun, 2009). The occurrence of a similar structural requirement in Arabidopsis complexes was examined by analyzing the subcellular distribution of LSM4, a protein marker of both Arabidopsis cytoplasmic and nuclear LSM complexes, in plants deficient in LSM1 and LSM8 proteins. As described above, in root cells from Arabidopsis seedlings containing the LSM4pro-LSM4-GFP construct, the LSM4-GFP fusion protein was simultaneously detected in both cytoplasm and nucleus (Figure 3D). Interestingly, however, in mutant seedlings for LSM1 and LSM8 (lsm1a lsm1b and lsm8, respectively) bearing the same construct, the fusion protein preferentially localized in nuclei or cytoplasm, respectively (Figure 3D). These observations strongly support the notion that LSM1 and LSM8 proteins are essential for the formation of the cytoplasmic and nuclear LSM complexes, respectively, in Arabidopsis.

LSM1- and LSM8-Deficient Arabidopsis Mutants Display Severe Developmental Alterations

The results described above indicated that Arabidopsis LSMs are also organized in cytoplasmic and nuclear complexes determined by the presence of LSM1 and LSM8 and suggested similar functions as the complexes from yeast and animals. To test this assumption, we first searched for T-DNA insertion mutants in LSM1 and LSM8 genes. Plants containing single T-DNA insertions located in the fourth exon of LSM1A or in the second intron of LSM1B were identified (Figure 4A). LSM1A or LSM1B mRNAs were undetectable in plants homozygous for the insertions (Figure 4B), revealing that these new LSM1A and LSM1B alleles (lsm1a and lsm1b) were null or highly hypomorphic. Intriguingly, lsm1a and lsm1b plants did not present any obvious morphological or developmental abnormality, being indistinguishable from their corresponding wild-type plants, Nossen-0 (No-0) and Columbia-0 (Col-0) ecotypes, respectively (see Supplemental Figure 5 online).

Figure 4.
Phenotypic Analysis of lsm1a lsm1b Double Mutant.

Since LSM1A and LSM1B are 80% identical (see Supplemental Figure 1 online), they might be functionally redundant, which would explain the wild-type phenotypes exhibited by lsm1a and lsm1b single mutant plants. Therefore, we decided to obtain the lsm1a lsm1b double mutant that was subsequently backcrossed four times with Col-0 to have both mutations within this genetic background. As expected, lsm1a lsm1b plants did not accumulate LSM1A and LSM1B mRNAs (Figure 4B). Remarkably, in contrast with single mutants, the lsm1a lsm1b double mutant showed severe developmental alterations. Seed germination in lsm1a lsm1b was delayed compared with the wild type and disturbed, producing epinastic, chlorotic, and small cotyledons (Figures 4C and and4D;4D; see Supplemental Figure 6A online). Cotyledonary veins were disorganized with disruptions, preventing the formation of closed loops as in wild-type veins (Figure 4E; see Supplemental Figure 6C online). lsm1a lsm1b rosette and cauline leaves were smaller than wild-type leaves, more serrated, and presented an abnormal venation phenotype and smaller petioles (Figures 4F and and4G;4G; see Supplemental Figures 6D to 6F online). The root system was also altered in lsm1a lsm1b plants, with the root length and the number of secondary roots being reduced (Figure 4H; see Supplemental Figures 6G and 6H online). On the other hand, the elongation of primary and secondary inflorescences ceased prematurely in the double mutant, altering plant architecture and given rise dwarf plants (Figure 4I). lsm1a lsm1b plants flowered earlier than wild-type plants under both long- and short-day photoperiods, though this phenotype was much more pronounced under noninductive photoperiodic conditions (see Supplemental Figures 6I and 6J online). Finally, mutant plants produced few siliques that were shorter and contained less seeds than those of wild-type plants (Figure 4J; see Supplemental Figures 6K and 6L online). Moreover, these seeds were small and frequently presented morphological alterations (Figure 4K). lsm1a lsm1b plants transformed with either LSM1Apro-LSM1A-GFP (c-lsm1a) or LSM1Bpro-LSM1B-GFP (c-lsm1b) were rescued for all of the above phenotypes (Figures 4C to to4K;4K; see Supplemental Figures 6 and 7 online), confirming that LSM1A and LSM1B are, in fact, functionally redundant and that the mutant phenotypes displayed by the double mutant were due to the absence of LSM1A and LSM1B expression.

In addition, two transgenic lines were identified that contained single T-DNA insertions located in the fifth exon of LSM8 (Figure 5A). In homozygous plants for the insertions, LSM8 mRNA was undetectable, indicating that these new LSM8 alleles (lsm8-1 and lsm8-2) were null or highly hypomorphic (Figure 5B). lsm8-1 and lsm8-2 mutants also exhibited developmental defects (Figure 5; see Supplemental Figure 6 online). Both of them showed the same phenotypes, but they were more pronounced in lsm8-1. Seeds from lsm8 mutants germinated as wild-type seeds, although a significant percentage of mutant seedlings exhibited alterations in the shape and number of their cotyledons, and veins formed more closed loops in lsm8 than in wild-type cotyledons (Figures 5C and and5D;5D; see Supplemental Figures 6A to 6C online). lsm8-1 and lsm8-2 rosette leaves had short petioles and were smaller and flatter than wild-type leaves, but their vasculature and margins were normal (Figure 5E; see Supplemental Figures 6D to 6F online). Regarding the radicular system, the root length and the number of secondary roots were reduced in lsm8 mutants compared with the wild type (Figure 5F; see Supplemental Figures 6G and 6H online). The length of primary and secondary inflorescences was not affected in the mutants (Figure 5G). Nonetheless, they flowered significantly earlier than wild-type plants under short-day photoperiods (see Supplemental Figures 6I and 6J online). Although the number of siliques produced in lsm8 mutants was as in the wild type, they were shorter and contained fewer seeds that frequently aborted (Figures 5H and and5I;5I; see Supplemental Figures 6K and 6L online). lsm8-1 mutant plants transformed with the construct LSM8pro-LSM8-GFP (c-lsm8) exhibited wild-type phenotypes (Figures 5C to to5I;5I; see Supplemental Figure 6 online), confirming that their mutant phenotypes were due to the lack of LSM8 expression. Altogether, these data provide direct evidence that LSM1 and LSM8 proteins are required to ensure correct developmental transitions in Arabidopsis, from germination to flowering, and also in seed formation.

Figure 5.
Phenotypic Analysis of lsm8 Mutants.

Accumulation of Capped Transcripts and mRNA Stability Are Affected in lsm1a lsm1b Mutants

The possibility that the Arabidopsis LSM1-7 cytoplasmic complex functions in mRNA degradation, as described in yeast and animals (Bouveret et al., 2000; Tharun et al., 2000), was tested by analyzing the decay rates of several mRNAs that have been reported to be unstable transcripts, such as EXPANSIN-LIKE1 (EXPL1), ARABIDOPSIS ORTHOLOG OF HS1 PRO1-2 (ATHSPRO2), JASMONATE-ZIM-DOMAIN PROTEIN6 (JAZ6), ARABIDOPSIS NITRATE REDUCTASE2 (NIA2), JAZ1, and RELATED TO ABI3/VP1 1 (RAV1) (Gutiérrez et al., 2002), in lsm1a lsm1b and wild-type plants. As a control, we also analyzed the turnover of EUKARYOTIC TRASLATION INITIATION FACTOR 4A1 (EIF4A1) mRNA, which is considered a stable transcript (Gutiérrez et al., 2002). Decay rates were assayed by comparing relative levels of mRNAs following cordycepin-induced transcriptional arrest (Gutiérrez et al., 2002). Our results confirmed the instability of the former mRNAs and the stability of the latter in the wild type (Figure 6A). In lsm1a lms1b, however, the steady state levels of all unstable transcripts analyzed were higher than in wild-type plants and their rates of decay clearly reduced, their estimated half-lives (the time required for an mRNA to be reduced to half its initial value) being at least two times longer (Figures 6A and and6B).6B). As expected, the steady state levels and the decay rate of EIF4A1 RNA were similar in mutant and wild-type plants (Figures 6A and and6B).6B). The analysis of the stability of EXPL1, ATHSPRO2, JAZ6, and EIF4A1 transcripts in c-lsm1a and c-lsm1b plants confirmed that LSM1A and LSM1B are functionally redundant and demonstrated that the increased mRNA stability noticed in lsm1a lms1b was caused by the simultaneous absence of LSM1A and LSM1B expression (Figures 6C and and6D;6D; see Supplemental Figures 8A and 8B online). We also examined the stability of EXPL1, JAZ6, and EIF4A1 mRNAs in lsm1a, lsm1b, and lsm8-1 single mutants. As presumed, all mRNAs showed similar turnover in cordycepin-treated wild-type and mutant plants (see Supplemental Figures 8C to 8H online), confirming again the functional redundancy of LSM1A and LSM1B and establishing that the Arabidopsis LSM2-8 nuclear complex does not play a role in cytoplasmic mRNA degradation.

Figure 6.
mRNA Stability and Accumulation of Capped Transcripts in the lsm1a lsm1b Double Mutant.

We next assessed whether the reduction of mRNA decay observed in lsm1a lms1b could be due to a deficiency in its mRNA decapping capacity. For this, rapid amplification of cDNA ends (RACE)-PCR that allows detection of capped forms of specific mRNAs was used. PCR experiments with low and high number of cycles were performed. In both cases, we found that EXPL1, ATHSPRO2, and JAZ6 mRNAs accumulated in their capped form in the lsm1a lsm1b mutant compared with the wild type (Figure 6E). These effects were corrected by expression of LSM1A and LSM1B transgenes in c-lsm1a and c-lsm1b plants, respectively (Figure 6E). The capped forms of mRNAs corresponding to the above genes were found not to be changed in lsm1a, lsm1b, and lsm8-1 mutants (see Supplemental Figure 8I online). These results indicated that the Arabidopsis LSM1-7 complex operates in cytoplasmic mRNA degradation by promoting decapping.

Loss of LSM8 Influences U6 snRNA Stability and Results in pre-mRNA Splicing Defects

In yeast and animals, the LSM2-8 nuclear complex acts in pre-mRNA splicing by stabilizing the spliceosomal U6 snRNA (Beggs, 2005). To determine whether the Arabidopsis LSM nuclear complex has a similar function, we first analyzed the effects of LSM8 on pre-mRNA splicing at the genome-wide level using tiling arrays (Affymetrix Arabidopsis Tiling 1.0R) and total RNAs from wild-type and lsm8-1 mutant plants. Two-week-old plants were selected for these experiments as they represent an intermediate stage of development. We searched for introns with significantly higher hybridization signals in mutant than in wild-type plants. Thus, we identified 469 introns, belonging to 453 genes, with increased hybridization signals in lsm8-1 (see Supplemental Data Set 2A online). The increased hybridization signals detected in lsm8-1 should reflect intron retention since hybridization signals in other introns and exons of the genes did not differ between wild-type and lsm8-1 plants. These results were validated by RT-PCR for a subset of genes appertaining to different ontology categories, including protein metabolism (AT1G17960), intracellular transport (AT3G59390), developmental processes [ARABIDOPSIS THALIANA PROTEIN ARGININE METHYLTRASFERASE 4A (PRMT4A)] or signal transduction [CASEIN KINASE-LIKE5 (CKL5) and PROTEIN KINASE AME3 (AME3)], in both lsm8-1 and lsm8-2 mutants (Figure 7A; see Supplemental Figure 9 online). However, the intron retention events in these genes were not detected in c-lsm8 and lsm1a lsm1b plants (Figure 7A), confirming that the splicing defects unveiled in lsm8 mutants were specifically due to the loss of LSM8 function and that the Arabidopsis LSM1-7 cytoplasmic complex is not involved in pre-mRNA splicing. As a control, tiling array data were also validated by analyzing the retention of an intron of ACTIN-RELATED PROTEIN4 (ARP4), a gene that did not display any intron retention event in the array, in lsm8-1 and lsm8-2 mutants. As expected, the intron was not retained in these plants (Figure 7A).

Figure 7.
Intron Retention and U6 snRNA Stability in lsm8 Mutants.

Next, we investigated the possible role of the Arabidopsis LSM nuclear complex in U6 snRNA stability by assessing the levels of this snRNA in cordycepin-treated lsm8 mutant and wild-type plants. Results revealed that the steady state levels of U6 snRNA were lower in mutants than in wild-type plants and that after cordycepin treatment these levels were maintained in wild-type plants but decreased rapidly in lsm8-1 and lsm8-2 mutants (Figure 7B). Therefore, the stability of U6 snRNA is dependent on the presence of LSM8 and, consequently, on the LSM2-8 nuclear complex. The effect of LSM8 on U6 snRNA stability seems to be highly specific since the levels of U3 small nucleolar RNA (snoRNA), which is transcribed by RNA polymerase III like the U6 snRNA, and U4 snRNA, which is synthesized by RNA polymerase II, did not decrease in cordycepin-treated lsm8 mutants (Figure 7B). As expected, c-lsm8 and lsm1a lsm1b plants showed similar levels of U6 snRNA, U3 snoRNA, and U4 snRNA as the wild type before and after cordycepin treatment (Figure 7C; see Supplemental Figure 10 online). Therefore, it was concluded that the Arabidopsis LSM2-8 nuclear complex is essential for accurate splicing of selected mRNAs through the stabilization of the spliceosomal U6 snRNA.

Arabidopsis Mutants Deficient in LSM1 or LSM8 Proteins Exhibit Altered Development-Related Gene Expression

In an attempt to understand the function of LSM complexes in Arabidopsis development, we studied the global impact of lsm1 and lsm8 mutations on gene expression. The comparison of mRNA profiles from lsm1a lsm1b and the wild type was performed using Agilent Arabidopsis Oligo Microarrays v4 and total RNAs extracted from 2-week-old plants. Transcript levels of 358 genes were found to be higher, by at least twofold, in lsm1a lsm1b than in the wild type (see Supplemental Data Set 2B online). On the other hand, transcripts corresponding to 316 genes were reduced by more than twofold in lsm1a lsm1b compared with the wild type (see Supplemental Data Set 2B online). Gene ontology analysis of deregulated genes in the double mutant unveiled that 72 of them were implicated in developmental processes, including seed germination, root and leaf development, inflorescence development, flowering, and embryogenesis (see Supplemental Data Set 2B online), which is consistent with its severe mutant phenotype (Figures 4C to to4K).4K). The microarray data were validated, confirming the altered expression of several overexpressed and underexpressed genes related to different developmental processes in lsm1a lsm1b plants by quantitative RT-PCR (Figures 8A and and8B).8B). On the other hand, c-lsm1a and c-lsm1b plants exhibited wild-type expression patterns for all validated genes (Figures 8A and and8B),8B), demonstrating that the LSM1-7 cytoplasmic complex is required for the accurate expression of development-related genes in Arabidopsis.

Figure 8.
Accumulation of Development-Related Transcripts in the lsm1a lsm1b Double Mutant.

Since the Arabidopsis LSM cytoplasmic complex functions in mRNA degradation by promoting decapping (see above), the high levels of some development-related mRNAs detected in the absence of LSM1 proteins might be due to a selective stabilization of the corresponding transcripts as a result of the retention of their 5′ cap. This possibility was first examined by measuring the degradation rates of five development-related mRNAs (YELLOW-LEAF-SPECIFIC GENE9 [YLS9], UDP-GLUCOSYL TRANSFERASE 87A2 [UGT87A2], ARABIDOPSIS THALIANA EXPANSIN-LIKE14 [ATEXP14], MATERNAL EFFECT EMBRYO ARREST14 [MEE14], and ARABIDOPSIS THALIANA HOMEOBOX12 [ATHB12]), whose levels were elevated in the lsm1a lsm1b double mutant, in cordycepin-treated wild-type and lsm1a lsm1b plants. Interestingly, the decay of all transcripts, except ATHB12, was significantly slower in the mutant than in wild-type plants (Figure 8C). In addition, all transcripts, but not ATHB12, retained their 5′ cap in lsm1a lsm1b (Figure 8D), providing evidence that, in fact, the Arabidopsis LSM1-7 cytoplasmic complex is essential for correct developmental gene expression by regulating the decapping and, therefore, the stabilization of specific development-related transcripts. Accordingly, the degradation rates and cap levels of YLS9, UGT87A2, ATEXP14, MEE14, and ATHB12 transcripts in c-lsm1a and c-lsm1b plants were as in the wild type (Figures 8C and and8D8D).

The effect of lsm8 mutations on gene expression at a genome-wide level was determined analyzing the above-mentioned tiling arrays, which, in addition to allowing splicing analysis, constitute a robust platform for detection of transcriptional activity (Laubinger et al., 2008). Compared with the wild type, 65 and 193 annotated genes were found to be at least twofold up- and downregulated, respectively, in the lsm8-1 mutant (see Supplemental Data Set 2C online). Gene ontology categorization of these deregulated genes revealed that a considerable number (17 upregulated and 50 downregulated) were related to developmental processes throughout the Arabidopsis life cycle (see Supplemental Data Set 2C online), which could explain the mutant phenotypes exhibited by lsm8 mutants (Figures 5C to to5I).5I). Microarray results were validated by assaying the expression of a group of deregulated genes implicated in different developmental processes in wild-type, lsm8-1, lsm8-2, and c-lsm8 plants by quantitative RT-PCR (Figures 9A and and9B).9B). These data demonstrated that the Arabidopsis LSM2-8 nuclear complex is also crucial for appropriate development-related gene expression.

Figure 9.
Intron Retention in Developmental Genes in lsm8 Mutants.

Considering that the LSM2-8 nuclear complex regulates development-related gene expression and functions in pre-mRNA splicing (see above), it was presumed that a number of genes involved in development might display splicing defects. Remarkably, 65 out of the 453 genes that showed intron retention events were found to be related to different developmental processes (see Supplemental Data Set 2A online). The inefficient splicing of some of these genes, including ABNORMAL SUSPENSOR1/DICER-LIKE1 (ASU1/DCL1), OLIGOCELLULA2 (OLI2), EMBRYO DEFECTIVE 2785 (EMB2785), EMBRYO DEFECTIVE 2016 (EMB2016), and GLUCURONOXYLAN GLUCURONOSYLTRANSFERASE (GUT2), in the absence of LSM8 was confirmed by RT-PCR analysis with appropriate primers (Figure 9C). As expected, the splicing of other developmental-related genes, such as AINTEGUMENTA (ANT), was not affected (Figure 9C). These findings indicate that the Arabidopsis LSM2-8 nuclear complex ensures the accurate splicing of specific development-related mRNAs, allowing correct developmental gene expression.


Although LSM-related proteins have been found in the genomes of different plant species (Proost et al., 2009; Goodstein et al., 2012), they have not yet been biochemically characterized and their function in RNA metabolism remained to be established. In this study, we used genetic, molecular, cell biology, and biochemical studies to demonstrate that Arabidopsis LSMs are organized in two heptameric complexes. More importantly, our results reveal that these complexes are essential for normal Arabidopsis development, and this role seems to be performed by controlling the proper turnover and splicing of selected developmental-related mRNAs that, in turn, ensures the appropriate gene expression during plant development.

Subcellular localization and BiFC experiments strongly support the idea that Arabidopsis LSM proteins assemble into two heteroheptameric complexes that differ by a single subunit, LSM1A/B or LSM8, and localize in cytoplasm (LSM1-7) and nucleus (LSM2-8). First, in Arabidopsis, as in other eukaryotes (Beggs, 2005), LSM1 proteins (LSM1A and LSM1B) specifically accumulate in the cytoplasm, while LSM8 has a nuclear localization and the rest of LSM proteins are simultaneously localized in cytoplasm and nucleus. Second, Arabidopsis LSM proteins do not interact promiscuously with each other. Instead, each LSM specifically interacts with two other LSM proteins following the same pattern of interaction as in the yeast and human heptameric complexes (Beggs, 2005). Consistent with their different subcellular localization and with the assumption that they define the two Arabidopsis LSM complexes, LSM1 and LSM8 proteins do not interact with each other. Moreover, while all interactions involving LSM1 proteins take place in the cytoplasm, those involving LSM8 occur in the nucleus and those involving LSM2-7 proteins occur simultaneously in both subcellular compartments. Third, LSM1 and LSM8 proteins are required for the formation of the Arabidopsis cytoplasmic and nuclear LSM complexes, respectively.

Our genetic and molecular analyses allowed us to establish the function of Arabidopsis LSM complexes. In Arabidopsis plants deficient in LSM1 proteins, several transcripts accumulate in their capped forms and show a reduced degradation rate with the corresponding increase in their half-lives, indicating that the Arabidopsis LSM1-7 complex function in the 5′ to 3′ pathway of mRNA decay as an activator of decapping. As expected from their high amino acid identity (80%), LSM1A and LSM1B are functionally redundant. lsm1a and lsm1b single null mutants are not perturbed in mRNA decapping and decay, and LSM1A and LSM1B, individually, are able to complement the alterations in mRNA decapping and decay displayed by the lsm1a lsm1b double mutant. On the other hand, Arabidopsis plants lacking LSM8 are affected in the stability of the spliceosomal U6 snRNA which, accordingly, results in pre-mRNA splicing defects. Compared with the wild type, at least 469 intron retention events distributed among 453 genes were detected in the lsm8-1 mutant under our experimental conditions, demonstrating that the LSM2-8 complex regulates genome-wide pre-mRNA splicing. Although intron retention constitutes the most frequent splicing defect in plant genes (Syed et al., 2012), it is probable that other mRNA splicing defects, including exon skipping, alternative 5′ splicing, and alternative 3′ splicing, also occur in the absence of LSM8. Unfortunately, however, the detection of these defects is unreliable when using tiling arrays to analyze pre-mRNA splicing at global level (Ner-Gaon and Fluhr, 2006). The existence of splicing defects has also been reported in some genes of plants harboring a postembryonic lethal mutation in LSM4 (Zhang et al., 2011). Nonetheless, only one of these genes (AT1G28060) has been found in our tiling analysis of the lsm8-1 mutant, in all likelihood because of the plants used being at different developmental stages and the different methods of analysis being used. Furthermore, only a few genes were analyzed for splicing defects in the lsm4 mutant (Zhang et al., 2011). The fact that not all mRNAs exhibit reduced degradation rates in lsm1a lsm1b plants nor splicing defects in lsm8 mutants indicates that the cytoplasmic and nuclear LSM complexes from Arabidopsis, as described for other components of the Arabidopsis machineries involved in mRNA degradation and processing (Xu et al., 2006; Goeres et al., 2007; Xu and Chua, 2009; Kim et al., 2010; Rymarquis et al., 2011), act on selected targets. How selected mRNAs are targeted to these complexes remains largely unknown. According to their relevant function in mRNA decapping and degradation, the lack of LSM1 proteins has a deep impact on Arabidopsis gene expression, the levels of more than 600 transcripts being significantly altered, 358 increased and 316 reduced, in lsm1a lsm1b plants. Similarly, the expression of at least 250 genes is significantly affected in null mutants for LSM8. Consistent with the implication of Arabidopsis LSM8 in pre-mRNA splicing, in this case the number of downregulated (193) genes is much higher than that of upregulated (65) ones. These data indicate that, like other factors involved in the Arabidopsis decapping 5′ to 3′ decay pathway, including DCP2, DCP5, and XRN4, or in pre-mRNA splicing, such as STABILIZED1 (STA1) (Lee et al., 2006; Xu et al., 2006; Goeres et al., 2007; Xu and Chua, 2009; Rymarquis et al., 2011), the Arabidopsis LSM1-7 and LSM2-8 complexes also play a major role in maintaining appropriate levels of gene expression. Interestingly, however, the result of the absence of these factors on Arabidopsis gene expression seems to be highly specific.

In eukaryotic cells, P-bodies appear as cytoplasmic foci containing RNP complexes associated with translational repression, mRNA storage, and cytoplasmic mRNA decay pathways (Xu and Chua, 2011). Under conditions promoting high levels of mRNA turnover, such as osmotic, hypoxic, or heat stress conditions, P-bodies increase in number and size, becoming more apparent (Teixeira et al., 2005; Weber et al., 2008). Nevertheless, it is not yet clear how P-bodies are formed and what their function is in eukaryotic cells. Human LSM4 localizes in P-bodies and loses this localization when mutations are introduced in residues involved in interacting with other LSM proteins (Ingelfinger et al., 2002). In yeast, LSM2 and LSM7 fail to localize to P-bodies in LSM1-deficient cells (Tharun et al., 2005), and it has been shown that LSM4 plays a role in the localization of the LSM1-7 complex in P-bodies and in P-body assembly (Decker et al., 2007; Reijns et al., 2008). Our findings demonstrate that the Arabidopsis LSM1-7 complex not only accumulates in P-bodies, which is consistent with its function in cytoplasmic mRNA decapping and decay but is also essential for their formation. As expected from its specific nuclear localization, LSM8 does not localize in P-bodies. The implication of Arabidopsis LSM2-LSM7 proteins in P-body formation is difficult to assess due to the absence of viable lsm2-lsm7 null mutants (see below). To date, only few proteins have been related with P-bodies in plants, including DCP5, an Arabidopsis protein indirectly implicated in regulating mRNA decapping that has a function in P-body formation (Xu and Chua, 2009), and DCP1, DCP2, and VCS, three proteins that constitute a decapping complex and colocalize with P-bodies in Arabidopsis (Xu et al., 2006; Goeres et al., 2007). In addition, Arabidopsis proteins XRN4, ARABIDOPSIS THALIANA TANDEM ZINC FINGER PROTEIN1, and POLYPYRIMIDINE TRACT BINDING PROTEINS have also been found in plant P-bodies (Weber et al., 2008; Pomeranz et al., 2010; Stauffer et al., 2010). The identification of additional P-body components will certainly help to understand how they are formed and what their function is in plant cells.

It has been described that the proteins involved in mRNA decapping, DCP1, DCP2, VCS, and DCP5, as well as the splicing factors STA1, U11/U112-31K, and ERECTA MRNA UNDER-EXPRESSED, play an essential role in Arabidopsis development, their absence being lethal (Lee et al., 2006; Xu et al., 2006; Goeres et al., 2007; Xu and Chua, 2009; Furumizu et al., 2010; Kim et al., 2010). Arabidopsis plants deficient in LSM1 and LSM8 proteins also display quite severe development alterations, but they are viable. lsm1a lsm1b and lsm8 mutants are affected in both vegetative and reproductive developmental traits, indicating that cytoplasmic and nuclear LSM complexes are required for the normal development of Arabidopsis throughout the different phases of its life cycle. Nevertheless, consistent with the different function of the two LSM complexes, the phenotypes exhibited by lsm1a lsm1b and lsm8 mutants are different. Recently, T-DNA insertional mutations for LSM4 and LSM7 have been described to show postembryonic and embryonic lethality, respectively (Zhang et al., 2011; http://www.seedgenes.org/). We have observed the same lethal phenotype not only for the lsm4 and lsm7 null mutations but also for the lsm3a lsm3b and lsm6a lsm6b doubles, with the corresponding single mutants exhibiting wild-type phenotypes, as well as for the lsm1a lsm1b lsm8 triple mutations (C. Perea-Resa, T. Hernández-Verdeja, and J. Salinas, unpublished data). Moreover, we have not found any insertion abolishing the expression of LSM2 and LSM5 genes in the available T-DNA collections, which suggests that, probably, lsm2 and lsm5 null mutations are also lethal. However, weak mutant alleles for these genes do not appear to be lethal. In fact, a point mutation in LSM5 (sad1) that causes the conversion of a Glu residue to a Lys makes mutant plants much smaller than the wild type but does not result in lethality (Xiong et al., 2001). Altogether, these data indicate that the presence of at least one LSM complex is essential in Arabidopsis. In yeast, it has been proposed that LSM2-7 proteins might associate, in the apparent absence of LSM1 or LSM8, with other proteins, including related SM proteins, to form complexes that would remain at least partially active (Beggs, 2005). We cannot exclude that this could be the case in lsm1a lsm1b and lsm8 mutants. Further studies are required to understand how Arabidopsis can develop and reproduce with just one LSM complex.

The results presented in this work demonstrate that posttranscriptional regulation has an important role in controlling gene expression related to plant development. In fact, we show that several selected genes involved in both vegetative and reproductive development are targets of the Arabidopsis LSM complexes. Thus, the LSM1-7 cytoplasmic complex ensures the precise half-life of the transcripts corresponding to its targets, for instance, UGT87A2 (floral transition), MEE14 (embryo development), or YLS9 (leaf development), and, consequently, their adequate temporal expression patterns. The LSM2-8 nuclear complex, in turn, guarantees the correct splicing of its targets, such as ASU1/DCL1 (flower development), OLI2 (leaf development), or EMB2785 (embryo development) and, therefore, the accurate translation of the corresponding transcripts. Furthermore, we also show that, consistent with their role in turnover and splicing of development-related mRNAs, the Arabidopsis LSM complexes regulate the expression levels of many genes that are implicated in different developmental processes, including seed germination, root development, leaf development, floral transition, flower development, and embryogenesis. In particular, the expression levels of 72 and 67 specific genes involved in development were found to be altered in lsm1a lsm1b and lsm8-1 mutants, respectively. However, it is obvious that these numbers should be considerably higher taking into account that only plants from one developmental stage (2 weeks old) were analyzed by microarray experiments. We propose that the cumulative defects in gene expression are responsible for the abnormal developmental phenotypes observed in these plants.

In conclusion, the findings presented here reveal the organization and function of Arabidopsis LSM proteins and demonstrate that these proteins are crucial for plant growth and development. Understanding the molecular mechanisms that regulate the function of LSMs and confer their target specificity constitutes an interesting challenge for the future.


Plant Material, Constructs, and Growth Conditions

Arabidopsis thaliana Col-0 ecotype and mutants lsm8-1 (Salk-025064) and lsm8-2 (Salk-048010) were obtained from the Nottingham Arabidopsis Stock Center. Mutant lsm1b is a Gabi-kat line from Max Plank Institute (GK 391E05). Arabidopsis No-0 ecotype and mutant lsm1a (12-2253-1) were obtained from Riken Institute. lsm1a is a Ds transposon insertion line in the No-0 background. lsm1a and lsm1b single mutants were crossed to generate a lsm1a lsm1b double mutant that was subsequently backcrossed four times with Col-0 to have both mutations within this genetic background. Transgenic Col-0 plants containing the 35S-GFP-DCP2 and 35S-GFP-VCS constructs (Goeres et al., 2007) were kindly provided by Leslie Sieburth (University of Utah, Salt Lake City, UT). These plants were crossed with lsm1a lsm1b to obtain double mutants with the 35S-GFP-DCP2 and 35S-GFP-VCS constructs in homozygosis. All mutant and transgenic lines were genotyped using the primers listed in Supplemental Data Set 2D online.

To obtain the LSMspro-GUS fusions, at least 1-kb promoter fragment from each of the 11 Arabidopsis LSM genes was cloned into the pBI101 binary vector (Clontech). For the LSMspro-LSM-GFP fusions, genomic regions containing the LSM1A, LSM1B, LSM3A, LSM3B, LSM4, and LSM8 genes, including at least 1 kb of the corresponding promoters, were cloned into the pGWB4 gateway binary vector (Nakagawa et al., 2007). All fusions were verified by sequencing and introduced in Col-0 via Agrobacterium tumefaciens C58C1 using the floral dip method (Clough and Bent, 1998). Fusions LSM1Apro-LSM1A-GFP, LSM1Bpro-LSM1B-GFP, and LSM8pro-LSM8-GFP were also introduced in lsm1a lsm1b and lsm8-1 mutants. Similarly, the LSM4pro-LSM4-GFP fusion was used to transform lsm1a lsm1b and lsm8-1 mutants. Finally, the fusion 35S-RFP-DCP1 (Weber et al., 2008), kindly provided by Markus Fauth (Johann Wolfgang Goethe-University Frankfurt, Germany), was introduced in transgenic lines containing LSM1Apro-LSM1A-GFP or LSM1Bpro-LSM1B-GFP. All transgenic lines were genetically determined to have the constructs integrated at a single locus in homozygosis. For BiFC assays, full-length cDNAs corresponding to the 11 LSM genes were amplified with appropriate primers (see Supplemental Data Set 2D online) to incorporate convenient restriction sites at their 5′ and 3′ ends. Fragments were cloned into the pSPYNE-35S and pSPYCE-35S binary vectors (Walter et al., 2004), kindly provided by Jörg Kudla (Westfälische Wilhelms-Universität Münster, Germany), sequenced, and introduced in Agrobacterium C58C1 for subsequent agroinfiltration. Agroinfiltration was performed in leaves from 3-week-old plants of Nicotiana benthamiana grown at 25°C, essentially as described (English et al., 1997), without using a silencing suppressor. The expression of fusion proteins was subsequently assayed 3 d after agroinfiltration.

Plants were grown at 20°C under long-day photoperiods (16 h of cool-white fluorescent light, photon flux of 90 µmol m−2 s−1) in pots containing a mixture of organic substrate and vermiculite (3:1 [v/v]) or in Petri dishes containing Murashige and Skoog medium supplemented with 1% Suc and solidified with 0.8% (w/v) agar. Plants used to estimate flowering time in short-day conditions were grown under an 8-h light regime.

Gene Expression Analysis

Total RNA was extracted using the Purezol reagent (Bio-Rad) according to the manufacturer’s protocol. RNA samples were treated with DNase I (Roche) and quantified with a Nanodrop spectrophotometer (Thermo Scientific). RNA blot hybridizations were performed according to standard procedures. Specific probes were obtained by PCR with the primers described in Supplemental Data Set 2D online and labeled with [α-32P]dCTP using the Megaprime DNA labeling systems kit (GE Healthcare). Equal RNA loading in the experiments was monitored by rRNA staining. RNA samples for each experiment were analyzed in at least three independent blots, and each experiment was repeated at least twice. For real-time RT-PCRs, cDNAs were prepared with the iScript cDNA synthesis kit (Bio-Rad) and then amplified using the Bio-Rad iQ2 thermal cycler, the SsoFast EvaGreen Supermix (Bio-Rad), and gene-specific primers (see Supplemental Data Set 2D online). The relative expression values were determined using the AT4G24610 gene as a reference (Czechowski et al., 2005). All reactions were realized in triplicate employing three independent RNA samples.

Determination of GUS Activity

GUS activity in Arabidopsis transgenic plants containing the fusion LSMspro-GUS was detected and measured as previously described (Medina et al., 2001).

Microscopy Analysis

Subcellular localization of fusion proteins in transgenic Arabidopsis was performed in roots from 6-d-old seedlings grown in vertical position on Murashige and Skoog medium supplemented with 1% Suc and solidified with 0.8% (w/v) agar. Heat treatment was performed by transferring seedlings to 37°C for 2 h. Treatment with cycloheximide was performed by incubating seedlings in liquid Murashige and Skoog medium supplemented with 200 µg/mL of cycloheximide for 2 h at 37°C. Transient expression of fusion proteins in leaves of 3-week-old plants of N. benthamiana was assayed 3 d after agroinfiltration as described above. Microscopy images were collected using a TCS SP2 confocal laser spectral microscope (Leica Microsystems). The excitation lines for imaging GFP and RFP fusions were 488 and 561 nm, respectively.

Cordycepin Treatments, mRNA Half-Life Estimations, and Capped mRNA Analysis

Six-day-old seedlings and 2-week-old plants were used for cordycepin treatment, essentially as described (Gutiérrez et al., 2002). Samples were collected at the indicated time points, and total RNA was extracted using the Purezol reagent (Bio-Rad). Gene expression was analyzed by RNA blot hybridizations or real-time RT-PCR as described above. To examine U6 snRNA, U3 snoRNA, and U4 snRNA decay, additional cordycepin was added to the samples at 9 and 24 h to ensure transcriptional repression. For graphical representation of mRNA stability and mRNA half-life estimation, the hybridization bands were quantified with the ImageJ software (NIH), and values were normalized to wild-type time 0.

To determine if accumulating mRNAs were capped, RNA ligase–mediated RACE was performed using the First Choice RLM-RACE kit (Ambion) following the manufacturer’s specifications. RNAs were extracted from 6-d-old seedlings or 2-week-old plants with the RNeasy kit (Qiagen), and PCRs were performed using a low (20 to 25) or high (30 to 32) number of cycles. Specific primers for the 5′ RACE adapter and for the genes tested are described in Supplemental Data Set 2D online.

Microarray Analysis

Total RNA from 2-week-old Col-0, lsm1a lsm1b, and lsm8-1 plants was extracted using the RNeasy kit (Qiagen), and three biological replicates were independently hybridized per transcriptomic comparison. For microarray analysis of the lsm1a lsm1b mutant, RNA amplification and labeling were performed basically as described (Goda et al., 2008). Hybridization was performed on Agilent Arabidopsis Oligo Microarrays v4 (catalog number G2519F-V4021169) in accordance with the manufacturer’s specifications. The statistical significance of the results was evaluated with FIESTA software (http://bioinfogp.cnb.csic.es). Genes with an false discovery rate–corrected P value lower than 0.05 and a fold change of more or less than 2 were selected for consideration. Data from these microarray experiments have been deposited in the Gene Expression Omnibus database under accession number GSE39630.

For microarray analysis of the lsm8-1 mutant, double-stranded cDNAs were synthesized, processed, and labeled with the GeneChip whole-transcript double-stranded target assay kit (Affymetrix) following the manufacturer’s instructions. Labeled cDNAs were used to hybridize Affymetrix Arabidopsis Tiling 1.0R arrays (catalog number 900594). Data were analyzed with Tiling Analysis Software from Affymetrix using TAIR7 as reference annotation (BPMAP file). To detect altered gene expression, genes with at least one exon identified as significantly over- or underexpressed (P value lower than 0.05 and a fold change of more or less than 2) were considered. A gene was accepted as differentially expressed when the 10% trimmed mean of the signals of all probes in its exons and UTRs was at least twofold higher or lower in the mutant than in the wild type. For those genes with splicing variants, only the constitutive exons were considered. Similarly, introns with significantly higher signals in the mutant than in the wild type were initially considered to be intron retention events. For high confidence, only the introns covered with a minimum of three probes and average signals over twofold were selected (see Supplemental Data Set 2A online). Data from these microarray experiments have been deposited in the Gene Expression Omnibus database under accession number GSE39617.

Intron Retention Analysis

Total RNA from 2-week-old plants was extracted with Purezol (Bio-Rad) and used for cDNAs generation with the iScript cDNA synthesis kit (Bio-Rad). Intron retention was revealed by RT-PCR using a pair of specific primers for each gene tested (see Supplemental Data Set 2D online). One primer was situated inside the retained intron and the second one in an adjacent exon. All PCR reactions were performed using RNA with (+RT) or without (−RT) reverse transcriptase to detect genomic DNA contaminations. Genomic DNA was included in all reactions as a positive control, and TUBULIN expression level was used as a loading control.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL data libraries under the accession numbers listed in Supplemental Data Set 2E online. The microarray data were submitted to the Gene Expression Omnibus site (www.ncbi.nlm.nih.gov/geo) under accession numbers GSE39630 and GSE39617.

Supplemental Data

The following materials are available in the online version of this article.


We thank Markus Fauth for the 35S-RFP-DCP1 fusion, Leslie Sieburth for the transgenic plants containing the fusions 35S-GFP-DCP2 and 35S-GFP-VCS, Jörg Kudla for the pSPYNE-35S and pSPYCE-35S vectors, Jose Manuel Franco, Javier Forment and Pablo Gonzalez-Garcia for tiling array data analysis, and Jose Antonio Jarillo and Roberto Solano for discussions and comments. This work was supported by grants CSD2007-00057, EUI2009-04074, and BIO2010-17545 from the Spanish Secretary of Research, Development, and Innovation.


C.P.-R. and T.H.-V. designed the research, performed research, and analyzed data. R.L.-C. and M.M.C. performed research. J.S. designed the research, analyzed data, and wrote the article.


precursor mRNA
green fluorescent protein
red fluorescent protein
bimolecular fluorescence complementation
rapid amplification of cDNA ends
small nuclear RNA
small nucleolar RNA


  • Anderson J.S.J., Parker R.P. (1998). The 3′ to 5′ degradation of yeast mRNAs is a general mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3′ to 5′ exonucleases of the exosome complex. EMBO J. 17: 1497–1506 [PMC free article] [PubMed]
  • Barta A., Kalyna M., Lorković Z.J. (2008). Plant SR proteins and their functions. Curr. Top. Microbiol. Immunol. 326: 83–102 [PubMed]
  • Beelman C.A., Stevens A., Caponigro G., LaGrandeur T.E., Hatfield L., Fortner D.M., Parker R. (1996). An essential component of the decapping enzyme required for normal rates of mRNA turnover. Nature 382: 642–646 [PubMed]
  • Beggs J.D. (2005). Lsm proteins and RNA processing. Biochem. Soc. Trans. 33: 433–438 [PubMed]
  • Belostotsky D.A., Sieburth L.E. (2009). Kill the messenger: mRNA decay and plant development. Curr. Opin. Plant Biol. 12: 96–102 [PubMed]
  • Bhattacharyya S.N., Habermacher R., Martine U., Closs E.I., Filipowicz W. (2006). Relief of microRNA-mediated translational repression in human cells subjected to stress. Cell 125: 1111–1124 [PubMed]
  • Bonnerot C., Boeck R., Lapeyre B. (2000). The two proteins Pat1p (Mrt1p) and Spb8p interact in vivo, are required for mRNA decay, and are functionally linked to Pab1p. Mol. Cell. Biol. 20: 5939–5946 [PMC free article] [PubMed]
  • Bouveret E., Rigaut G., Shevchenko A., Wilm M., Séraphin B. (2000). A Sm-like protein complex that participates in mRNA degradation. EMBO J. 19: 1661–1671 [PMC free article] [PubMed]
  • Brengues M., Teixeira D., Parker R. (2005). Movement of eukaryotic mRNAs between polysomes and cytoplasmic processing bodies. Science 310: 486–489 [PMC free article] [PubMed]
  • Clough S.J., Bent A.F. (1998). Floral dip: A simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16: 735–743 [PubMed]
  • Coller J., Parker R. (2005). General translational repression by activators of mRNA decapping. Cell 122: 875–886 [PMC free article] [PubMed]
  • Coller J.M., Tucker M., Sheth U., Valencia-Sanchez M.A., Parker R. (2001). The DEAD box helicase, Dhh1p, functions in mRNA decapping and interacts with both the decapping and deadenylase complexes. RNA 7: 1717–1727 [PMC free article] [PubMed]
  • Cougot N., Babajko S., Séraphin B. (2004). Cytoplasmic foci are sites of mRNA decay in human cells. J. Cell Biol. 165: 31–40 [PMC free article] [PubMed]
  • Czechowski T., Stitt M., Altmann T., Udvardi M.K., Scheible W.R. (2005). Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol. 139: 5–17 [PMC free article] [PubMed]
  • Decker C.J., Teixeira D., Parker R. (2007). Edc3p and a glutamine/asparagine-rich domain of Lsm4p function in processing body assembly in Saccharomyces cerevisiae. J. Cell Biol. 179: 437–449 [PMC free article] [PubMed]
  • Deng X., Gu L., Liu C., Lu T., Lu F., Lu Z., Cui P., Pei Y., Wang B., Hu S., Cao X. (2010). Arginine methylation mediated by the Arabidopsis homolog of PRMT5 is essential for proper pre-mRNA splicing. Proc. Natl. Acad. Sci. USA 107: 19114–19119 [PMC free article] [PubMed]
  • Dunckley T., Parker R. (1999). The DCP2 protein is required for mRNA decapping in Saccharomyces cerevisiae and contains a functional MutT motif. EMBO J. 18: 5411–5422 [PMC free article] [PubMed]
  • English J., Davenport G., Elmayan T., Vaucheret H., Baulcombe D. (1997). Requirement of sense transcription for homology-dependent virus resistance and trans-inactivation. Plant J. 12: 597–603
  • Furumizu C., Tsukaya H., Komeda Y. (2010). Characterization of EMU, the Arabidopsis homolog of the yeast THO complex member HPR1. RNA 16: 1809–1817 [PMC free article] [PubMed]
  • Goda H., et al. (2008). The AtGenExpress hormone and chemical treatment data set: Experimental design, data evaluation, model data analysis and data access. Plant J. 55: 526–542 [PubMed]
  • Goeres D.C., Van Norman J.M., Zhang W., Fauver N.A., Spencer M.L., Sieburth L.E. (2007). Components of the Arabidopsis mRNA decapping complex are required for early seedling development. Plant Cell 19: 1549–1564 [PMC free article] [PubMed]
  • Goodstein D.M., Shu S., Howson R., Neupane R., Hayes R.D., Fazo J., Mitros T., Dirks W., Hellsten U., Putnam N., Rokhsar D.S. (2012). Phytozome: A comparative platform for green plant genomics. Nucleic Acids Res. 40(Database issue): D1178–D1186 [PMC free article] [PubMed]
  • Gutiérrez R.A., Ewing R.M., Cherry J.M., Green P.J. (2002). Identification of unstable transcripts in Arabidopsis by cDNA microarray analysis: Rapid decay is associated with a group of touch- and specific clock-controlled genes. Proc. Natl. Acad. Sci. USA 99: 11513–11518 [PMC free article] [PubMed]
  • Halbeisen R.E., Galgano A., Scherrer T., Gerber A.P. (2008). Post-transcriptional gene regulation: From genome-wide studies to principles. Cell. Mol. Life Sci. 65: 798–813 [PMC free article] [PubMed]
  • Hu C.D., Chinenov Y., Kerppola T.K. (2002). Visualization of interactions among bZIP and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol. Cell 9: 789–798 [PubMed]
  • Ingelfinger D., Arndt-Jovin D.J., Lührmann R., Achsel T. (2002). The human LSm1-7 proteins colocalize with the mRNA-degrading enzymes Dcp1/2 and Xrnl in distinct cytoplasmic foci. RNA 8: 1489–1501 [PMC free article] [PubMed]
  • Kastenmayer J.P., Green P.J. (2000). Novel features of the XRN-family in Arabidopsis: evidence that AtXRN4, one of several orthologs of nuclear Xrn2p/Rat1p, functions in the cytoplasm. Proc. Natl. Acad. Sci. USA 97: 13985–13990 [PMC free article] [PubMed]
  • Kim W.Y., Jung H.J., Kwak K.J., Kim M.K., Oh S.H., Han Y.S., Kang H. (2010). The Arabidopsis U12-type spliceosomal protein U11/U12-31K is involved in U12 intron splicing via RNA chaperone activity and affects plant development. Plant Cell 22: 3951–3962 [PMC free article] [PubMed]
  • Lange H., Gagliardi D. (2010). The exosome and 3′-5′ RNA degradation in plants. Adv. Exp. Med. Biol. 702: 50–62 [PubMed]
  • Laubinger S., Zeller G., Henz S.R., Sachsenberg T., Widmer C.K., Naouar N., Vuylsteke M., Schölkopf B., Rätsch G., Weigel D. (2008). At-TAX: A whole genome tiling array resource for developmental expression analysis and transcript identification in Arabidopsis thaliana. Genome Biol. 9: R112. [PMC free article] [PubMed]
  • Lee B.H., Kapoor A., Zhu J., Zhu J.K. (2006). STABILIZED1, a stress-upregulated nuclear protein, is required for pre-mRNA splicing, mRNA turnover, and stress tolerance in Arabidopsis. Plant Cell 18: 1736–1749 [PMC free article] [PubMed]
  • Lorković Z.J. (2009). Role of plant RNA-binding proteins in development, stress response and genome organization. Trends Plant Sci. 14: 229–236 [PubMed]
  • Lorković Z.J., Wieczorek Kirk D.A., Lambermon M.H., Filipowicz W. (2000). Pre-mRNA splicing in higher plants. Trends Plant Sci. 5: 160–167 [PubMed]
  • Matlin A.J., Clark F., Smith C.W. (2005). Understanding alternative splicing: Towards a cellular code. Nat. Rev. Mol. Cell Biol. 6: 386–398 [PubMed]
  • Medina J., Catalá R., Salinas J. (2001). Developmental and stress regulation of RCI2A and RCI2B, two cold-inducible genes of Arabidopsis encoding highly conserved hydrophobic proteins. Plant Physiol. 125: 1655–1666 [PMC free article] [PubMed]
  • Meyer S., Temme C., Wahle E. (2004). Messenger RNA turnover in eukaryotes: Pathways and enzymes. Crit. Rev. Biochem. Mol. Biol. 39: 197–216 [PubMed]
  • Nakagawa T., Kurose T., Hino T., Tanaka K., Kawamukai M., Niwa Y., Toyooka K., Matsuoka K., Jinbo T., Kimura T. (2007). Development of series of gateway binary vectors, pGWBs, for realizing efficient construction of fusion genes for plant transformation. J. Biosci. Bioeng. 104: 34–41 [PubMed]
  • Ner-Gaon H., Fluhr R. (2006). Whole-genome microarray in Arabidopsis facilitates global analysis of retained introns. DNA Res. 13: 111–121 [PubMed]
  • Parker R., Song H. (2004). The enzymes and control of eukaryotic mRNA turnover. Nat. Struct. Mol. Biol. 11: 121–127 [PubMed]
  • Pomeranz M., Lin P.C., Finer J., Jang J.C. (2010). AtTZF gene family localizes to cytoplasmic foci. Plant Signal. Behav. 5: 190–192 [PMC free article] [PubMed]
  • Proost S., Van Bel M., Sterck L., Billiau K., Van Parys T., Van de Peer Y., Vandepoele K. (2009). PLAZA: A comparative genomics resource to study gene and genome evolution in plants. Plant Cell 21: 3718–3731 [PMC free article] [PubMed]
  • Raab S., Hoth S. (2007). A mutation in the AtPRP4 splicing factor gene suppresses seed development in Arabidopsis. Plant Biol. (Stuttg.) 9: 447–452 [PubMed]
  • Reddy A.S. (2001). Nuclear pre-mRNA splicing in plants. Crit. Rev. Plant Sci. 20: 523–571
  • Reijns M.A., Alexander R.D., Spiller M.P., Beggs J.D. (2008). A role for Q/N-rich aggregation-prone regions in P-body localization. J. Cell Sci. 121: 2463–2472 [PMC free article] [PubMed]
  • Rymarquis L.A., Souret F.F., Green P.J. (2011). Evidence that XRN4, an Arabidopsis homolog of exoribonuclease XRN1, preferentially impacts transcripts with certain sequences or in particular functional categories. RNA 17: 501–511 [PMC free article] [PubMed]
  • Sheth U., Parker R. (2003). Decapping and decay of messenger RNA occur in cytoplasmic processing bodies. Science 300: 805–808 [PMC free article] [PubMed]
  • Sheth U., Parker R. (2006). Targeting of aberrant mRNAs to cytoplasmic processing bodies. Cell 125: 1095–1109 [PMC free article] [PubMed]
  • Souret F.F., Kastenmayer J.P., Green P.J. (2004). AtXRN4 degrades mRNA in Arabidopsis and its substrates include selected miRNA targets. Mol. Cell 15: 173–183 [PubMed]
  • Stauffer E., Westermann A., Wagner G., Wachter A. (2010). Polypyrimidine tract-binding protein homologues from Arabidopsis underlie regulatory circuits based on alternative splicing and downstream control. Plant J. 64: 243–255 [PubMed]
  • Syed N.H., Kalyna M., Marquez Y., Barta A., Brown J.W. (2012). Alternative splicing in plants - Coming of age. Trends Plant Sci. 17: 616–623 [PMC free article] [PubMed]
  • Teixeira D., Sheth U., Valencia-Sanchez M.A., Brengues M., Parker R. (2005). Processing bodies require RNA for assembly and contain nontranslating mRNAs. RNA 11: 371–382 [PMC free article] [PubMed]
  • Tharun S. (2009). Roles of eukaryotic Lsm proteins in the regulation of mRNA function. Int. Rev. Cell Mol. Biol. 272: 149–189 [PubMed]
  • Tharun S., He W., Mayes A.E., Lennertz P., Beggs J.D., Parker R. (2000). Yeast Sm-like proteins function in mRNA decapping and decay. Nature 404: 515–518 [PubMed]
  • Tharun S., Muhlrad D., Chowdhury A., Parker R. (2005). Mutations in the Saccharomyces cerevisiae LSM1 gene that affect mRNA decapping and 3′ end protection. Genetics 170: 33–46 [PMC free article] [PubMed]
  • Unterholzner L., Izaurralde E. (2004). SMG7 acts as a molecular link between mRNA surveillance and mRNA decay. Mol. Cell 16: 587–596 [PubMed]
  • Wahl M.C., Will C.L., Lührmann R. (2009). The spliceosome: Design principles of a dynamic RNP machine. Cell 136: 701–718 [PubMed]
  • Walter M., Chaban C., Schütze K., Batistic O., Weckermann K., Näke C., Blazevic D., Grefen C., Schumacher K., Oecking C., Harter K., Kudla J. (2004). Visualization of protein interactions in living plant cells using bimolecular fluorescence complementation. Plant J. 40: 428–438 [PubMed]
  • Wang B.B., Brendel V. (2004). The ASRG database: Identification and survey of Arabidopsis thaliana genes involved in pre-mRNA splicing. Genome Biol. 5: R102. [PMC free article] [PubMed]
  • Weber C., Nover L., Fauth M. (2008). Plant stress granules and mRNA processing bodies are distinct from heat stress granules. Plant J. 56: 517–530 [PubMed]
  • Xiong L., Gong Z., Rock C.D., Subramanian S., Guo Y., Xu W., Galbraith D., Zhu J.K. (2001). Modulation of abscisic acid signal transduction and biosynthesis by an Sm-like protein in Arabidopsis. Dev. Cell 1: 771–781 [PubMed]
  • Xu J., Chua N.H. (2009). Arabidopsis decapping 5 is required for mRNA decapping, P-body formation, and translational repression during postembryonic development. Plant Cell 21: 3270–3279 [PMC free article] [PubMed]
  • Xu J., Chua N.H. (2011). Processing bodies and plant development. Curr. Opin. Plant Biol. 14: 88–93 [PMC free article] [PubMed]
  • Xu J., Yang J.Y., Niu Q.W., Chua N.H. (2006). Arabidopsis DCP2, DCP1, and VARICOSE form a decapping complex required for postembryonic development. Plant Cell 18: 3386–3398 [PMC free article] [PubMed]
  • Zhang Z., et al. (2011). Arabidopsis floral initiator SKB1 confers high salt tolerance by regulating transcription and pre-mRNA splicing through altering histone H4R3 and small nuclear ribonucleoprotein LSM4 methylation. Plant Cell 23: 396–411 [PMC free article] [PubMed]

Articles from The Plant Cell are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • BioProject
    BioProject links
  • Gene
    Gene links
  • GEO DataSets
    GEO DataSets
    GEO DataSet links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...