Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jan 15;113:46-55.
doi: 10.1016/j.ymeth.2016.10.008. Epub 2016 Oct 21.

Analysis of RNA Structure Using Small-Angle X-ray Scattering

Free PMC article

Analysis of RNA Structure Using Small-Angle X-ray Scattering

William A Cantara et al. Methods. .
Free PMC article


In addition to their role in correctly attaching specific amino acids to cognate tRNAs, aminoacyl-tRNA synthetases (aaRS) have been found to possess many alternative functions and often bind to and act on other nucleic acids. In contrast to the well-defined 3D structure of tRNA, the structures of many of the other RNAs recognized by aaRSs have not been solved. Despite advances in the use of X-ray crystallography (XRC), nuclear magnetic resonance (NMR) spectroscopy and cryo-electron microscopy (cryo-EM) for structural characterization of biomolecules, significant challenges to solving RNA structures still exist. Recently, small-angle X-ray scattering (SAXS) has been increasingly employed to characterize the 3D structures of RNAs and RNA-protein complexes. SAXS is capable of providing low-resolution tertiary structure information under physiological conditions and with less intensive sample preparation and data analysis requirements than XRC, NMR and cryo-EM. In this article, we describe best practices involved in the process of RNA and RNA-protein sample preparation, SAXS data collection, data analysis, and structural model building.

Keywords: Aminoacyl-tRNA synthetase; Native PAGE; RNA structure; SAXS; Size-exclusion chromatography; tRNA-like structures.


Figure 1
Figure 1. Common problems with SAXS data
Many issues related to polydispersity can be identified by initial visual inspection of the SAXS data such as (A) aggregation or radiation damage, which manifests as a gradual upward slope in the Guinier plot that increases with sample concentration or exposure time, (B) interparticle repulsion, identified by a gradual downward slope in the low-q region of the SAXS data, and (C) production of multimers or low-molecular weight aggregates with increasing concentration.
Figure 2
Figure 2. Identification of foldedness or globularity in Kratky plots
Compact globular macromolecules will exhibit a bell-shaped Kratky plot with non-globular molecules yielding a range of Kratky plot shapes where the high-q region does not regress back toward the X-axis.
Figure 3
Figure 3. Merging of SAXS data
To alleviate issues with detector saturation or large error in high-concentration or long-exposure data (black line) and poor signal to noise in low-concentration or short-exposure data (red line), data can be truncated (truncations denoted by barbells) and merged to yield a single curve with these complicating features absent (bottom panel).
Figure 4
Figure 4. Generating SAXS-derived models
(A) For generating ab initio models, the SAXS data is transformed to generate a PDDF that is a histogram of all inter-electron distances within a molecule (e distances) which is used to generate molecular envelopes. This method is not able to generate all-atom models (top right). (B) All-atom modeling starts with a secondary structure from which many de novo models are generated. These models are then screened for the ones that best fit the SAXS data. The model can then be refined using SAXS data as a restraint to obtain models that better fit the data.
Figure 5
Figure 5. Isosteric thymidine analog, BrdU, allows validation of all-atom model
Back calculated SAXS curve for an all-atom model of the primer binding site domain (PBS) in the HIV-1 5′UTR annealed to a complementary 18-nt DNA oligonucleotide designed to mimic the 3′ end of the tRNALys3 reverse transcription primer (blue line) closely matches experimental SAXS data (blue dots). Replacement of four thymidine residues in the 18-nt DNA with BrdU produces predictable changes in the back-calculated curve (red line) that also very closely matches the SAXS data of that complex (red dots).

Similar articles

See all similar articles

Cited by 3 articles

Publication types

MeSH terms


LinkOut - more resources