Structural basis for recognition of AU-rich element RNA by the HuD protein

by Sang Jee

Introduction:

The RNA recognition motif (RRM), also known as RNA-binding domain (RBD) or ribonucleoprotein domain (RNP) is one of the most abundant protein domains in eukaryotes and the most common nucleic acid binding protein motifs. The structure of RRM mainly has two identical binding domains that bind on opposite sides of the RNA loop. Each RRM has a beta-alpha-beta-beta-alpha -beta fold RRM consists of a compact globule made of a 4-stranded antiparallel (beta) sheet that forms the RNA-binding surface backed by two helices (alpha). Thus, specific interactions are found in the space between the two RRM domains that serve as a region of recognition element.

Hu proteins bind to adenosine-uridine (AU)-rich elements (AREs) in the 3` untranslated regions of mRNAs where they stabilize the RRM domain structure.

These RRM domain structures also consist of four beta antiparallel stranded sheet and two alpha helices.
Here is picture of the alignment of the human HuD 1,2 amino acid sequence with the other Hu Proteins.

It clearly shows the beta-alpha-beta-beta-alpha -beta fold with the position of the ribonucleoprotein motifs RNP1 and RNP2.

In comparison to structures of other RRM domain-nucleic acid complexes, the RNA recognition motif domains of the HuD exhibits two identical base recognition pockets that interact with bases at the C-terminal ends of RRM domains.

The button will show protein in color and DNA in blue

at the N-terminus,

at the C-terminus.

spin OFF

spin ON

zoom OFF

Overall Structure:

AREs are divided into three classes. Classes 1 AREs contain one to three copies of scattered AUUUA repeats while class 2 AREs contain multiple copies of clustered AUUUA repeats. Class 3 AREs are less well-defined, but have a long continuous U-rich regions.

The role of Hu proteins is to stabilize ARE-containing mRNAs with increases to the translation efficiency of ARE-containing mRNAs. Hu Proteins contain three highly conserved RNA recognition motif (RRM) domains. The first two RRM domains are sufficient for binding to ARE sequence while the function of the third RRM domain is uncertain. However, it has been reported to bind to long poly(A) tails.

The crystal structures of the first two RRM recognition motif domains of the HuD protein in complex with an 11-nucleotide fragment are class 1 ARE(c-fos AREs) and class 2 ARE (tumor necrosis factor alpha AREs).

Class 1 ARE(the c-fos ARE)

1.STRUCTURE

Out of the total 27-nucleotide segment of the c-fos ARE , an 11-nucleotide fragment binds to HuD with high affinity. This nucleotide fragment, known as cfos-11(AUUUUUAUUUU) forms a 1:1 complex with the HuD 1,2 protein.
that formed a 1:1 complex with the HuD1,2 protein.

A cleft is formed with the c-fos-11RNA bound between the basic surfaces of the opposing beta-sheets of two RRM domains. The only interaction between the two domains is a hydrogen bond between the side chain of Lys 111 and the main chain carbonyl oxgen of Ile 152.

The turn of backbone in the AREs region is mediated by direct hydrogen bonds between the 2-OH groups of U4 and U5 and the phosphate group of A7 and between the 2-OH of U6 and the phospahte group of U4.

A water molecule bridges the phosphate groups of U5 and U6, and another water molecule interacts with the phosphate groups of A7 and U8. Thus, this conformation of RNA allows specific recognition by the HuD 1,2 protein.

2. RNA RECOGNITION

Of the eight bases observed in the crystal structure, only U6, which is stacked with A7, is not specifically recognized by HuD1,2. RRM domain 1 interacts with U5 through U10 while domain 2 interacts mainly with U3 and U4. Thus, RRM domain 1 is the primary ARE binding domain.

CLICK HERE TO SEE THE SUMMARY OF CONTACTS BETWEEN HUD1,2 PROTEIN AND CFOS-RNA

Recognition of U3 and U4 by domain 2 is identical to that of U9 and U10 by domain 1. Tyr 128 forms a stacking interaction with U3 and Ile 42 provides a hydrophobic environment for U9. Similarly, Asn 126 and Phe 170 in domain 2 form side chain and stacking interactions with U4; the structurally equivalent Asn 40 and Phe 84 in domain 1 interact with U10.

Class 2 ARE(the tumor necrosis factor ARE)

1.STRUCTURE

HuD1,2 binds to the sequence AUUUAUUU where the single substitution of an A for U3 are found in class 2 AREs. This structure is very identical to the complex in class 1 AREs, although an additional nucleotide (U2) and ribose ring (U11) provide the different nucleotide sequence.

2. RNA RECOGNITION

A consensus RNA recognition sequence for HuD1,2 protein will be predictable by analyzing the difference between cfos-11 and TNF-alpha-11 AREs .

For example, a uracil would be preferred at the U3/A3 position since adenine at this positions in the HuD1,2-TNF-alpha-11 structure is involved in only a stacking interaction.

At the U4 and U10 positions, Asn 126 and Asn 40 can tolerate a uracil or cytosine to form hydrogen bonds through their OD1 atoms with the N4 positions of cytosines .

The A7 positions and the U6 position may tolerate any base of proteins because it does not make any contact with the protein.

At the U5 and U9 positions uracil is strongly prefered since the main chain interactions do not tolerate a substitution for cytsine. Thus, the U5 and U9 positions is the most specific since it correspons to the central uracil residues in the AUUUA repeat.

CLICK HERE TO SEE THE SUMMARY OF CONTACTS BETWEEN HUD1,2 PROTEIN AND TNF-alpha-11 RNA

Conclusion:

A comparison of structures of RRM domains exhibits that the proteins folds are very similiar. However the conformations and orientations of the bound nucleic acids are different

All the residues at structurally equivalent positions form main chain and stacking interactions with bases in the U3/U9 binding pocket and side chain and stacking interactions with bases in the U4/U10 binding pockets.

Furthermore, the bases that bind in roughly the same positions as U3/U9 or U4/U10 in the HuD 1,2 structures were recognized in a similar manner, mainly by residues in the ribonucleoprotein motifs RNP1 and RNP2.

Overall, the ineteraction between main chain and side chain with additional hydrogen bond interactions allows specific recognition of many bases at the C-terminal ends of the RRM domains.

CLICK HERE TO SEE THE SUMMARY OF CONTACTS BETWEEN HUD1,2 PROTEIN AND RNA RECOGNITION MOTIFS

Reference:

Wang, X and Tanaka, T. M. (2001) "Structural basis for recognition of AU-rich element RNA by he HuD protein" Nature Struc. Bio. 8, 141-145.

Allain, H.A., Bouvet, P., Dieckmann, T., Feigon, J (2000) "Molecular basis of sequence-specific recognition of pre-ribosomal RNA by nucleolin" THe EMBO Journal. 19, 6870-6881.

PDB Code: 1FXL, 1G2E