Trends in Genetics
Volume 23, Issue 1, January 2007, Pages 5-7
Journal home page for Trends in Genetics

Update
Research Focus
SR proteins: a foot on the exon before the transition from intron to exon definition

https://doi.org/10.1016/j.tig.2006.10.002Get rights and content

Two recent publications illuminate the evolution of alternative splicing, showing that a SR (serine–arginine-rich) protein that regulates alternative splicing in multicellular organisms is also found in a unicellular organism without alternative splicing, in which it can assist in the splicing of weak introns. Moreover, insertion of SR proteins into an organism lacking such proteins can restore the splicing of weak introns. These results imply that SR proteins had already facilitated the splicing of weak introns before the evolution of alternative splicing.

Introduction

Introns are found in all eukaryotic organisms. However, the number and size of introns and exons and their modes of recognition by the splicing machinery are different in unicellular compared with multicellular organisms, and also vary within these groups [1]. In the budding yeast Saccharomyces cerevisiae, only 3% of the genes contain introns; 99% of these contain only a single intron ∼270 nt long. By contrast, in the fission yeast Schizosaccharomyces pombe, 45% of the genes contain introns, ranging from 40 to 70 nt long; half of these genes contain more than one intron 2, 3. However, in multicellular organisms, most of the genes contain several introns (8.4 on average in humans), and comparative analyses among multicellular organisms have revealed both a greater number of exons per gene and larger intronic sequences in primates and other mammals compared with all other organisms 4, 5, 6. This indicates a great degree of variability in the intron–exon structure of genes among different eukaryotic organisms.

SR (serine–arginine-rich) proteins are splicing factors that regulate both alternative and constitutive splicing. They bind to short RNA sequences and mediate spliceosome assembly (Box 1). In metazoans, SR protein genes constitute nine families, of which six have two or more members in mammals. However, there are no SR proteins in S. cerevisiae and only two SR proteins in S. pombe. In other unicellular eukaryotes, there are one or two SR protein genes [7]. Thus, the diversity of SR proteins seems to have emerged with multicellularity.

There are two potential mechanisms for exon and intron selection by the splicing machinery, called intron and exon definition. These two models are still unproven, and all the indications for their existence are circumstantial. However, intron definition is presumably the ancient one, in which the splicing machinery recognizes an intronic unit and places the basal machinery across introns. Therefore, the size of the intron is under selection. Indeed in S. cerevisiae and S. pombe, almost all introns are less than 350 nt long, and all the information for accurate splicing is within the intron sequences 8, 9. This suggests that intron definition is the only system that directs the splicing machinery in these organisms [10]. In the second mechanism, exon definition, the basal splicing machinery is placed across exons. The length of exons must not exceed 300 nt. It was postulated that during evolution the enlargement of intronic sequences forced the splicing machinery to shift from the recognition of short intronic sequences to the selection of short exonic sequences – from intron to exon definition. This could explain the selective pressure to maintain short intronic sequences in yeast genes and short internal exons in the human genome (and other higher metazoans; see also Supplementary Data) [11].

To the best of our knowledge, there is no alternative splicing in S. cerevisiae and S. pombe, whereas alternative splicing is prevalent in multicellular organisms 2, 3. Therefore, one explanation might be that the ability to handle multi-intron genes and alternative splicing was lost in S. cerevisiae and S. pombe. Recent results that might support such a scenario imply a massive intron loss during the evolution of worms and flies, rather than intron gain in other organisms [12]. However, it is more likely that complex regulatory networks evolved from simple ones [13]. Pursuing this theme, it was estimated that the percentage of genes that undergo alternative splicing increases in higher metazoans compared with lower metazoans 4, 14. So, can we trace the steps leading to the appearance of alternative splicing (or to the loss of it in those yeasts)?

Section snippets

Srp2p in S. pombe supports splicing of weak introns

Recent results from the Wise laboratory suggest that the SR proteins were already involved in enhancing splicing of suboptimal introns in organisms that support only intron definition [15]. These authors found that one of the two SR proteins in S. pombe, Srp2p, improves the recognition of a suboptimal 3′ splice site (3′ss) and thus facilitates the splicing of the cognate intron. Webb et al. demonstrated that Srp2p binds to an exonic sequence that is rich in purines and is located downstream of

SR proteins facilitate splicing of weak introns in an SR-free organism

How might the evolution of SR proteins facilitate alternative splicing? The transition from intron to exon definition, probably occurring between S. pombe and multicellular organisms in the course of evolution, was the major selective pressure leading to the proliferation of SR genes in multicellular organisms. Proliferation of such genes can give an advantage in assisting the basal splicing machinery in finding short exons in large intronic sequences. The transition from intron to exon

How do SR proteins relate to the evolution of alternative splicing?

According to these findings, we can add another layer to the hypothesis regarding the origin of alternative splicing [3]. The proliferation of SR proteins during the evolution of multicellular organisms from unicellular organisms released the burden from the basal machinery of having to bind efficiently to the four splice signals. Consequently, natural selection permitted mutations that result in suboptimization of certain splice sites (such as U1 base pairing to the 5′ss). This causes the

A correlation between intron size and exon skipping

The appearance of alternative splicing is presumably related to the expansion of intronic sequences beyond the maximal recognition length of introns by the intron definition system. This could explain the selective pressure leading to shorter internal exons in higher compared with lower eukaryotes 10, 11.

The Hertel laboratory found a correlation between the size of the flanking introns and the ability of the internal exon to undergo alternative splicing [19]. They demonstrated that in the fruit

References (24)

  • G. Yeo

    Variation in sequence and organization of splicing regulatory elements in vertebrate genes

    Proc. Natl. Acad. Sci. U. S. A.

    (2004)
  • N.L. Barbosa-Morais

    Systematic genome-wide annotation of spliceosomal proteins reveals differential gene family expansion

    Genome Res.

    (2006)
  • Cited by (37)

    • SRSF1-Regulated Alternative Splicing in Breast Cancer

      2015, Molecular Cell
      Citation Excerpt :

      Thus, in agreement with our SRSF1 regulatory maps, both examples indicate that binding of SRSF1 near the 5′SS but not the 3′SS is associated with promoting exon inclusion. SR protein binding to constitutive exons plays a role in SS selection and exon definition (Ram and Ast, 2007). Accordingly, we observed an inverse correlation between the strength of both SSs and the number of SRFS1 motifs in the 113 CA-exons used to build the regulatory maps (Table S5).

    • Differential GC Content between Exons and Introns Establishes Distinct Strategies of Splice-Site Recognition

      2012, Cell Reports
      Citation Excerpt :

      Indeed, in lower eukaryotes, where introns are generally short and can be directly recognized by the splicing machinery, intron definition is probably the dominant mode of splicing. On the other hand, in higher eukaryotes, such as vertebrates, where most introns are long, the splicing machinery had to adapt to identify short exons among the long introns via the exon definition mechanism (Ast, 2004; Berget, 1995; Hertel, 2008; Keren et al., 2010; Niu, 2008; Ram and Ast, 2007). Our results suggest that GC content differences are associated with two different types of alternative splicing: intron retention and exon skipping.

    • Co-evolution of the branch site and SR proteins in eukaryotes

      2008, Trends in Genetics
      Citation Excerpt :

      Variations in the splicing properties are related to the lifestyle of organisms [22] and this probably triggered a simplification of the splicing machinery in some single-cell eukaryotes. The current hypothesis for the evolution of splicing is that the proliferation of splicing factors during the evolution of eukaryotes, especially in metazoans, probably released the spliceosome of the efficient recognition of the splicing signals [14,23,24], enabling the degeneracy of splicing signals. Our results indicate that the expansion of RS repeats in SR proteins had a fundamental role in the relaxation of the splicing signals and in the origin of regulated splicing.

    View all citing articles on Scopus
    View full text