Structure and occurrence of SR proteins

SR proteins have a typical domain organization containing either one or two RNA recognition motifs (RRMs) at their N terminus and an argine-serine-rich region at their C terminus (the so-called RS domain). There are also SR-related proteins which either do not possess RRMs or contain an unusual domain distribution. Most of all previous papers focussed on the differences between SR proteins across metazoan[2].
Both RS domain and RRMs are modular, for example RRMs can be exchanged between SR proteins[5] and may still bind to RNA without the presence of the RS domain[4]. On the other hand the RS domain can be bound to a heterologous RNA-binding domain and be still functional[6].
A number of proteins being involved in cell metabolism and containing an RS domain are collectivly referred as RS proteins. There are also a number of SR-related proteins, which are displayed with the SR proteins and their structure in human hereafter[17]:

Figure 2: Structure of SR proteins

Open image in full size: images/structure1.png

All SR proteins consist of at least 2 domains: a N-terminal RNP[18] (RRM; Superfamily SCOP: 54929[26]), an optional zinc knuckle motif (only in SRp20 9G8), and the C-terminal RS domain[19] (Family SCOP: 52065[25]), which is the functional domain that interacts with the spliceosome.
Additional information for the SR proteins:

1. SRp20 9G8: [12]
   The protein is extensively phosphorylated on serine residues in the RS domain and is mediating protein-protein interaction within the intron during spliceosome assembly, independently binding to exonic enhancer sequences and recruiting components to adjacent introns for splice-site recognition and alternative splicing.

2. p54 SRp86: (no additional citation information found)
   The p54/SRp86 family has homologs in Caenorhabditis elegans, Drosophila melanogaster, Danio rerio and Arabidopsis thaliana having two RRM. One of the RRMs is similiar to the human RRM in either p54 (D. rerio, A. thaliana) or SRp86 (C. elegans).

3. RY1 (SNUT3 HUMAN): [27] [14]
   This family is also known to be found in C. elegans, D. melanogaster, and A. thaliana with a homolog in D. rerio (ENSDARG00000035625). According to Genecards, RY1 has only one domain: DUF1777[21], which is not annotated in SCOP currently. Pfam list the protein as protein of unknown function[22], but links to SR-25 stating it's related to RNA splicing[23].

4. SRp1: SC35 and SRp46: [28]
   Also note that SRp1 is not essential in fungi[7] and has not been implicated in splicing[11]. In Rhizopus oryzae (R. oryzae) the SRp1 homologs order is inverted - the RS domain is located at the N terminal of the protein instead of the C terminal.

5. SRm300: [15]
   The SRm300 family has no RRM, but instead an RS domain tagged as IPR013170 mRNA_splic_Cwf21[15] [20]. Current research shows that no protist or fungi has a homolog of SRm300[9]. The original reference[17] is listing this family as splicing co-activator.

6. SRp30c-ASF: [16]
   Recent studies show that ASF is not present in Gallus gallus but a homolog can be found in Cryptosporidium parvum[9].

7. SRp2: SRp40, SRp55, SRp75: [29]
   The SRp2 family is homolog to the human SRp40, SRp55 and SRp75. In contrast to SRp1, SRp2 is essential in fungi and has been implicated in the regulation of splicing[11]. There is at least one SRp2 homolog (in Ustilago maydis) which possess a second RRM domain which is located at the scaffold, so there is no information on the C-terminal region available. Also noteworthy is that R. oryzae possess two non-exact copies of SRP2, preassumably the result of recent duplications.

Hereafter is a structural view of SR-related proteins:

Figure 3: Structure of SR-related proteins

Open image in full size: images/structure2.png