Figure 1.
Overview of RNA interaction sites mapped on ORF578–120 and ALYREF1–155 using different approaches.
Backbone amide chemical shift change (δCS) and saturation transfer (ST) data are shown for ORF578–120 (A) and ALYREF1–155 (B) with different ligands added as labeled. Crosses indicate residues with signal broadened beyond detection. For δCS, large (>0.1) and moderate (>0.04) chemical shift changes of each protein relative to signals in its free state, are shown as solid and broken bars, respectively. For ST data, signal intensity ratios significantly different from the background mean values are represented by solid (>6 SD) and broken (>3 SD) bars, respectively. Labels indicate the state of the protein for each dataset, data shown for interactions with non-specific RNA (green), ORF57 specific RNA (red/orange) and ALYREF-ORF57-RNA complex (blue).
Figure 2.
Structure of the ALYREF – ORF57 complex.
(A) Ribbon representation showing ORF57 colored blue, and ALYREF RRM colored green, red and yellow for looped, α-helical and β-sheet regions, respectively. Positions of N- and C-termini of polypeptide chains are labeled. (B) Overlay of 20 lowest energy structures with backbone shown in the same orientation. The best-fit superposition is made using heavy backbone atoms of structurally defined regions aa74–152 of ALYREF and aa108–119 of ORF57. Color-coding is the same as on panel A. (C) Alternative view of ALYREF54–155-ORF57103–120 complex showing the hydrophobic sidechains involved in the interaction. (D) Schematic of the ALYREF and ORF57 binding site. ORF57 residues are colored blue and ALYREF in black; hydrophobic and electrostatic interactions are indicated by green and red dashes, respectively.
Figure 3.
Comparison of ALYREF54–155-ORF57103–120 with ALYREF54–155-ICP27103–138 and U2AF complex structures.
(A) Overlay of the RRM domains of ALYREF in complex with ICP27103–138 (green and magenta, PDB code 2kt5 [30]) and ORF57-bound ALYREF determined here (cyan and orange), demonstrating the shift in α-helix 1 position. (B) ALYREF54–155 in complex with ORF57103–120, determined here. (C) Backbone amide weighted chemical shift changes (δCS) in the ALYREF54–155-ORF57103–120 complex are emphasized by color. δCS>0.3 are red, 0.15–0.299 orange, 0.05–0.149 yellow, prolines are blue and regions unaffected are green. (D) U2AF35 in complex with U2AF65 (PDB code 1jmt, [50]). (E) ALYREF54–155 in complex with ICP27103–138, previously determined (pdb 2kt5). (F) Backbone amide weighted chemical shift changes (δCS) in the ALYREF54–155-ICP27103–120 complex with the same coloring as panel C.
Table 1.
NMR calculation statistics for an ensemble of the 20 lowest energy structures of ALYREF fragment (ALYREF54–155) bound to ORF57103–120 (PDB code 2YKA).
Figure 4.
Obtaining site-specific information on RNA binding to protein-protein complex.
(A) Scheme illustrating the principle of the ST-IDIS-TROSY method proposed here. RNA is added to a differentially labeled pair of interacting proteins, and selective saturation of RNA protons by RF pulses is transferred through space to the adjacent amides within both proteins in the complex. (B) The example of the resultant ST-IDIS-TROSY spectra: signals from amide groups situated within 5 Å of the RNA are selectively weakened (colored red), as detected independently and simultaneously in the spectra of both proteins.
Figure 5.
Typical effects of complex formation and RNA→protein ST on selected signals of ALYREF1–155 and ORF578–120.
The 1H dimension slices through 1H-15N-correlation spectra are displayed for three representative signals of each protein, on the left for ALYREF and on the right for ORF57. The residue assignments in the free form are labeled at the top, and same signals are shown below each other for different complexes, as indicated. First type of signal (ALYREFM1 and ORF57E24) is not significantly affected by any complex formation, or ST. Second type (ALYREFA34 and ORF57Y81) is not affected much by protein and marginally affected by RNA binding, but is altered or displays significant ST effect in the ternary complex (percentage drop in signal intensity is indicated in blue, and ST spectral traces shown in red). These are residues likely contributing to cooperative ternary complex formation, forming contacts with RNA. Third type of signals originates from the structured regions of proteins (ALYREFA104 and ORF57R111). ALYREFA104 signal is not significantly affected in protein-protein complex, but shows significant increase in ST effect in the ternary complex, suggesting that ORF57 recruits RNA to the proximity of this residue. ORF57R111 signal is broadened beyond detection in protein-protein complex, and remains broadened in the ternary complex. For this signal strong ST effect is observed when in complex with RNA. This residue is involved in initial viral RNA recognition, but then RNA is displaced from this site by ALYREF binding.
Figure 6.
Fluorescent studies and simulations of the ternary complex formation between ORF57, O, fragment of viral RNA, R, and ALYREF, A.
(A) Simultaneous non-linear fit of normalized fluorescence-derived parameters ΔλNbcm (blue shift of emission signal) and ΔIN (fluorescence quenching) to the non-redundant three-equation model using DynaFit software, to obtain Kd for the ternary complex. The experimental values of thus determined KdOA+R, as well as Kd's for other complexes measured earlier, are summarized on two illustrative thermodynamic cycles for the ternary complex assembly presented on panels (B) and (C). Simulations (using COPASI software [67]) for these two possible cycles illustrate an increase in the concentration of ternary ORF57-RNA-ALYREF complex when ORF57 is added to 10 µM equimolar mixture of ALYREF and RNA, assuming the simplest four-state equilibrium model (B), or six-state model which additionally takes into account weak nonspecific ALYREF-RNA binding (C). The arrow marks a point where all the components of ternary complex are present in equimolar amounts. The presence of equimolar ORF57 significantly increases the concentration of RNA in complex with ALYREF (i.e., [OAR] vs [AR][O] = 0). The baseline concentrations [AR][O] = 0 are indicated on the panel (C) on the left, assuming two different conservative estimates for values of Kd for nonspecific ALYREF-RNA binding.
Figure 7.
Probing ORF57-RNA binding by mutations and UV cross-linking, and ALYREF-ORF57-RNA remodeling assay.
(A) Purified hexa-histidine tagged GB1 (negative control) or ORF578–120 WT and point mutants (as labeled) were incubated with end-labeled 14merS RNA oligonucleotide before the mixture was subjected to UV cross-link (+) or not (−). Similarly in the remodeling assay, WT ORF578–120 was incubated with end-labeled 7merS (B) or 14merS (C), before the mixture was added to purified GST-ALYREF immobilized onto glutathione coated beads. Purified complexes were eluted in native conditions and UV cross-linked (+) or not (−). All samples were finally analyzed on 15% SDS-PAGE stained with Coomassie blue and by PhosphoImaging.
Figure 8.
Model of the passage of RNA between ORF57 and ALYREF.
Local protein interactions with RNA are detected by moderate (orange) and large (red) saturation transfer effects (represented by ‘lightning bolts’) observed by ST-HSQC or ST-IDIS-TROSY and mapped onto respective regions. Broadened residues are colored light-yellow. Linked black circles represent a position for transiently bound RNA. ALYREF (green) binds RNA 14merS weakly via its RRM and N-terminal regions (A), whereas ORF57 (blue) binds 14merS tightly mainly via the R-b helix and also the aa81–92 region (B). Interaction of ORF57-RNA complex with ALYREF partially displaces the RNA from the R-b helix, while RNA maintains contact with ORF57 aa81–92 and also forms new contacts with ALYREF's aa22–48 and helix-2 of the RRM domain (C). The RNA contacts with ALYREF within the ALYREF-ORF57-RNA ternary complex are more abundant than for just ALYREF-RNA, and thus ORF57 enhances the interaction of viral RNA with ALYREF.