Structural dynamics in the La-module of La-related proteins

Figures & data

Figure 1. Small Angle X-ray Scattering analysis of HsLa, HsLaRP7, HsLaRP6 and HsLaRP4A and HsLaRP4B La-modules. The domain boundaries of the La-motif (LaM) and RNA Recognition Motif 1 (RRM1), delineating the exact beginning and end of the structured domains, are indicated on top for each protein. In the case HsLaRP4B, for which structural analysis is underway (Conte et al., unpublished), this is an initial estimate from sequence alignment with LaRP4A proteins. (A-E) SEC elution profiles for the five La-modules in the apo state as labelled (black traces). The SEC elution profiles for HsLa, HsLaRP7 and HsLaRP6 in complex with U4 for HsLa and HsLaRP7 and with 48ntSL RNA for HsLaRP6 are shown as red traces in A, B and C. (F-J) Scattering curves obtained after buffer normalization and averaging (black traces for the apo La-modules, red traces for the complexes with RNA). (K-O) Normalized Kratky representations (in black for the apo La-modules and in red for the complexes with RNAs) calculated from data in the range q = 0.02–0.3. The typical values expected for globular proteins [I(q)/I(0)]⋅(q⋅R_g)² = 1.104, q⋅R_g = 1.73] are indicated by grey dashed cross lines

Table 1. SAXS-derived parameters and experimental details

Instrument		SEC-SAXS at SWING beamline SOLEIL
q range (Å ^−1)		0.0022–0.62
Temperature (°C)		25°C
Sample	Rg(Å) Reciprocal Space	I(0) (cm^−1/absorbance) Reciprocal Space	Rg(Å) Real Space	I(0) (cm^−1/absorbance) Real Space	Dmax (Å)	Porod Volume estimate (Å^3)	MW from Porod Volume/ 1.6 (kDa)	MW from sequence (kDa)
HsLa La-module	26.90 ± 0.40	00029 ± 1.1 10^−5	26.97 ± 0.09	0.002876 ± 8.285 10^−6	84	46400	29.0	22.3
HsLa La-module-U4 complex	23.82 ± 0.25	0.0022 ± 1.5 10^−5	22.76 ± 0.08	0.002013 ± 6.678 ± 10 ^−6	69	40300	25.0	22.3
HsLaRP7 La-module	32.08 ± 1.02	0.013 ± 2.6 10^−5	32.20 ± 0.10	0.01289 ± 2.965 10^−5	101	58200	36.4	24.0
HsLaRP7 La-module-U4 complex	26.79 ± 0.30	0.024 ±3.5 10^−5	27.40 ± 0.05	0.02369 ± 3.196 10^−5	92	48000	30.0	24.0
HsLaRP6 85–300 La-module	27.18 ± 0.18	0.0045 ± 3 10^−5	27.20 ± 0.07	0.004412 ± 1.539 10^−5	81	51900	32.4	23.9
HsLaRP6 85–300 La-module-48ntSL complex	31.48 ± 0.13	0.0032 ± 1.8 10^−6	31.60 ± 0.07	0.003138 ± 1.220 10^−5	90	94000	58.8	39.8
48ntSL RNA	22.29 ± 1.77	0.0091 ± 9.9 10^−6	22.47 ± 0.03	0.009076 ± 8.442 10^−6	75	20400	12.8	14.4
HsLaRP4A La-module	22.52 ± 0.60	0.0012 ± 2.5 10^−5	21.66 ± 0.16	0.001151 ± 8.423 10^−6	68	27800	17.4	20.5
HsLaRP4B La-module	21.33 ± 0.29	0.0018 ± 2.8 10^−5	21.37 ± 0.03	0.01821 ± 1.937 10^−5	67	28900	18.1	21.6
Software employed
Primary reduction		Foxtrot
Data processing		ATSAS 2.8 and Scatter
Ab initio analysis		DAMMIF/DAMMIN
Validation and averaging		DAMAVER
Computation of model intensities		CRYSOL
3D graphic representation		PyMOL

Figure 2. Distance distribution functions and ab-initio models. (A-E) Distance distribution functions for the La-modules of La, HsLaRP7, HsLaRP6, HsLaRP4A and HsLaRP4B in the apo form (black traces) and for the La-modules of HsLa, HsLaRP7 and HsLaRP6 in complex with their target RNAs in red traces. The distance distribution function for free 48ntSL RNA is shown in orange in C. (F-N) Low-resolution ab-initio models were generated from the distance distribution functions for the La-modules (F-J, grey), for HsLa and HsLaRP7 and HsLaRP6 in complex with RNA (K-M, red) and for the 48ntSL RNA (N, orange). A similar orientation for each model was chosen (approximately as in Fig. 4A, left) based on superposition on the HsLa structure following SUPCOMB fitting of the ab-initio models to the respective atomic structures, when available, or to the HsLa structure for HsLaRP6 and HsLaRP4B

Table 2. CRYSOL and EOM analysis of the SAXS data for the La-modules using available high-resolution structural information

	CRYSOL		EOM
SAXS data	Structural model	χ^2	Rigid body model definition	χ^2
HsLa La-module-U4	X-ray Structure PDB 2VOP	1.8	-	-
HsLa La-module	X-ray Structure PDB 2VOP (with the RNA removed)	7.1	PDB 2VOP; dummy residues: 99–107	1.03
HsLaRP7 La-module-U4	X-ray Structure PDB 4WKR	≫10	-	-
HsLaRP7 La-module	X-ray Structure PDB 4WKR (with the RNA removed)	≫10	PDB 4WKR; dummy residues: 1–27, 117–121 and 189–208	1.30
HsLaRP6 La-module-48ntSL	No structure available	-	-	-
HsLaRP6 La-module	No structure available	-	PDB 2MTF for LaM (70–178) and 2MTG for RRM1 (181–195); dummy residues: 178–179 and 296-300	2.55
HsLaRP6 LaM	NMR structure PDB 2MTF (removing residues 70–84)	1.2	-	-
HsLaRP6 RRM1	NMR structure PDB 2MTG	1.2	-	-
HsLaRP4A La-module	NMR structure PDB 6I9B	1.5	PDB 6I9B; dummy residues: 117–121 and 275-287	1.01
HsLaRP4A LaM	NMR structure PDB 6I9B	1.6	-	-
HsLaRP4A RRM1	NMR structure PDB 6I9B	2.2	-	-
HsLaRP4B	No structure available	-	-	-

Figure 3. Ensemble Optimization Method (EOM) for the HsLa, HsLaRP7, HsLaRP6, and HsLaRP4A La-modules. (A-D) Distribution of the R_g (red) of the 10.000 conformations generated by allowing flexibility to the linker residues (see methods) and R_g for the ensemble of models that best fit the scattering data (black). The most populated state is marked with an asterisk (*) and the population percentages are shown. (E-H) Distribution of the D_max of all possible conformations (red) and those of the ensemble that best fit the scattering data (black). The most populated state is marked with an asterisk (*) and the population percentages are shown. (I-L) Fitting of the representative structures of the ensemble to the experimentally recorded scattering curves. (M-P) Representative structures of the ensemble of conformation in the context of the ab-initio models generated by DAMMIF, where the LaM domain (orange) is shown in a fixed orientation, to highlight the different positioning of the RRM1 domain (in pink, blue and green). The colour code used is: pink, blue, green from the most to least populated conformation. The most populated conformation of the RRM1 (pink) is marked with an asterisk (*). The linker regions, input as flexible residues in the EOM process, are represented as dashed lines

Figure 4. Comparison of crystal structures of RNA-bound forms with EOM-derived models of apo La-modules. (A) Superposition of the X-ray structure of HsLa-U4 (PDB 2VOP, protein in grey and RNA in green) with the most abundant apo HsLa EOM ensemble structure (LaM in orange and RRM1 in light pink, same colour-code used in Fig. 3). (B) Superposition of the X-ray structure of HsLaRP7-U4 (PDB 4WKR, protein in grey and RNA in green) with the preferred apo HsLaRP7 EOM ensemble structure (LaM in orange and RRM1 in light pink). The percentages indicate the fraction of the molecular population that exists in a particular state. The structures in (A) and (B) are aligned on the LaM. Although their RNA-bound forms are highly similar, the apo EOM models of HsLa and HsLaRP7 differ. For HsLa a more similar domain arrangement in the RNA-bound crystal structure and in the EOM model of the apo La-module could be observed (see also Supplementary Fig. S8)