Small-angle X-ray scattering (SAXS) is a popular structural and biophysical technique used to study the solution state of biological macromolecules and complexes. SAXS can provide information including, but not limited to, size and molecular weight, flexibility, degree of folding, overall shape, and even pseudo-atomic structural models. Over the past 25 years, driven by new high-brilliance synchrotron beamlines, advances in detector technology, and the development of a wide range of powerful software analysis tools, SAXS has become a mainstream structural biology/biophysical technique. Key to the growing popularity of SAXS have been new experimental approaches able to probe challenging systems, specifically in-line purification techniques coupled to SAXS, including size-exclusion chromatography, ion-exchange chromatography and, very recently, asymmetric flow field-flow fractionation (SEC-SAXS, IEC-SAXS, and AF4-SAXS). Having in-line purification allows high-quality measurements on less stable and more heterogeneous biological samples that were previously intractable for SAXS. The increasing automation of data collection and analysis and a growing awareness that SAXS is highly complementary to other structural and biophysical techniques such as X-ray crystallography (MX), nuclear magnetic resonance (NMR), cryo-electron microscopy (cryo-EM) and multi-angle and dynamic light scattering (MALS and DLS) also contribute to the popularity of the technique. Additionally, SAXS has proven invaluable for studying intrinsically disordered proteins (IDPs) and liquid-liquid phase separating (LLPS) systems, which are not amenable to common high resolution structural techniques such as MX and cryo-EM.
To learn more about designing and carrying out SAXS experiments at BioCAT see our user guides.
Contents
Equilibrium SAXS
Basic characteristics of the equilibrium SAXS setup at BioCAT:
| Energy | 12 keV |
| Flux | ~7*1012 ph/s |
| Energy Resolution (dE/E) | 2 x 10-4 (Si <111>) |
| Beam size (FWHM) | ~25 x 16 µm2 (V x H), focused at the detector |
| Sample to detector distance | ~3.7 m |
| q range | ~0.0027 - 0.43 Å-1 |
| Detector | EIGER2 XE 9M (Dectris) |
| Sample cell | Coflow cell with a 1.0 mm ID quartz capillary |
| Temperature range | 4 - 40 ˚C (more extreme temperatures may be possible upon request) |
While data acquisition during a typical SAXS experiment is relatively simple, data quality and interpretability are largely dependent on sample homogeneity and monodispersity, which is often compromised due to the limited shelf-lives of biological macromolecules. A majority of BioCAT users prefer to use the in-line SEC-SAXS instrument, which separates by size the sample from potential contaminants such as aggregates or breakdown products immediately before exposure to X-rays. BioCAT also provides other in-line purification techniques: IEC-SAXS for charged-based separation of samples and AF4-SAXS for size-based separation without the matrix of a column. AF4-SAXS is of particular interest for systems like lipid nanoparticles (LNPs) and liposomes, which tend to behave poorly on SEC columns but are of great interest currently for drug delivery purposes. BioCAT also has a newly built batch mode autosampler capable of low volume batch mode measurements from 96- and 384-well plates. For all of these experiments BioCAT offers in-line full-spectrum UV absorption measurements.
For SEC-SAXS and AF4-SAXS, BioCAT provides the option of additional coupled biophysical measurements, MALS, DLS, and RI, as part of the same experiment (SEC-MALS-SAXS or AF4-MALS-SAXS). MALS plus RI provides a more accurate determination of molecular weight than SAXS, which is particularly valuable for determining the oligomeric state of macromolecules and the stoichiometry of multicomponent complexes. Having both UV absorption and RI measurements allows the use of conjugate analysis to calculate the mass fraction of different components in a complex. This is particularly useful for determining the mass fraction of lipids in detergent solvated membrane protein systems, the degree of glycosylation for glycoproteins and binding stoichiometries for complexes between nucleic-acids and proteins. The Rh from DLS measurements in ratio with the Rg value from SAXS measurements provides information on the overall particle shape.
Besides these data collection modalities, BioCAT provides additional experimental capabilities that enhance SAXS data collection at the beamline. An in-vacuum coflow cell is used for all equilibrium techniques to minimize background scattering and radiation damage to samples, providing optimal signal-to-noise ratio measurements, and all systems are fully temperature controlled (including the HPLC, columns, and fluidic lines) for experiments anywhere between 4-40 ˚C. BioCAT also has a robust data processing pipeline that provides real-time data reduction, so experiments can be monitored while being run, and that provides initial automated analysis, including background subtraction, Guinier fits, P(r) calculation, and ab initio modeling, shortly after the experiment finishes, allowing users a quick look at preliminary results. Taken as a whole, BioCAT provides a mature, state-of-the-art facility for equilibrium SAXS capable of tackling the most challenging systems.
SEC-SAXS, SEC-MALS-SAXS, and IEC-SAXS
BioCAT provides a custom designed HPLC system with two Agilent 1260 Infinity II bioinert pumps, an Agilent 1260 Infinity II bioinert multisampler, and customized valves and control software that allow for measurement by a column on one pump while simultaneously equilibrating a column on the other. Users can switch between the two flow paths with the click of a button, providing highly automated sample measurement. The entire system is housed in a temperature controlled and variable incubator, and water jacketed lines feed the temperature controlled sample cell for optimal stability and variability. Full spectrum UV absorbance measurements are collected during every elution using a fiber coupled flow cell, a balanced Halogen-Deuterium light source (Ocean Insight) and a Black-Comet spectrometer (StellarNet). This custom HPLC is used for SEC-SAXS, SEC-MALS-SAXS, and IEC-SAXS.
The beamline also has the following SEC columns available for users, though users may also bring their own columns to address potential cross contamination and reproducibility issues:
- Superdex 30 Increase 10/300 (MW ~0.1-7 kDa)
- Superdex 75 Increase, both 10/300 and 5/150 (MW ~3-70 kDa)
- Superdex 200 Increase, both 10/300 and 5/150 (MW ~10-600 kDa)
- Superose 6 Increase, both 10/300 and 5/150 (MW ~5-5,000 kDa)
- Wyatt 010S5 100Å (MW range 0.1-100 kDa)
- Wyatt 015S5 150Å (MW range 0.5-150 kDa)
- Wyatt 030S5 300Å (MW range 5-1,250 kDa)
SEC-SAXS
The standard mode of SAXS data collection uses in-line Size Exclusion Chromatography (SEC) coupled to SAXS (SEC-SAXS). The sample runs through a size exclusion column to separate potential aggregates or different oligomeric states immediately before flowing through the capillary for X-ray exposure.
SEC-MALS-SAXS
BioCAT provides a data collection mode where SEC is coupled to MALS (multi-angle light scattering), DLS (dynamic light scattering), and RI (refractive index) detectors in addition to the SAXS flow-cell, a technique called SEC-MALS-SAXS. A Wyatt DAWN HELEOS II MALS+DLS (17 channels LS, plus 1 DLS) detector, and a Wyatt Optilab T-rEX dRI detector are used for these measurements. The MALS and RI detectors are temperature controlled from 4-40 C. This approach provides all of the sample quality benefits of SEC-SAXS and eliminates possible ambiguity about differences between non-identical separate SEC-SAXS and SEC-MALS-DLS-RI measurements. There is a small loss of resolution compared to SEC-SAXS due to the additional detectors, so this method should only be used if the MALS and DLS data are needed for the experiment.
IEC-SAXS
Ion exchange chromatography (IEC) allows separation of particles by charge rather than size, making it useful in cases where SEC cannot resolve between different components in solution. BioCAT offers IEC in-line with SAXS (IEC-SAXS) for samples that are not separable by SEC-SAXS. Because IEC requires a changing buffer during elution (typically a slope or step gradient in salt or pH), data analysis is more involved than for SEC-SAXS, but analysis algorithms are now widely available, making this a routine technique at the beamline. These experiments are more involved, and often require some work to optimize an appropriate gradient, so it is important to discuss your potential IEC-SAXS experiments with beamline personnel before requesting beamtime.
The custom HPLC used for SEC-SAXS is also capable of IEC-SAXS as each pump is a quaternary pump able to create the necessary gradient for elution. BioCAT has the following IEC columns available for users:
- Capto HiRes Q 5/50
- Capto HiRes S 5/50
Batch Mode
Batch mode samples are directly loaded into the sample cell, rather than first passing through a sizing column. This reduces the volume and concentration required, but aggregates and other large species are not separated from the sample, increasing requirements on sample prep. BioCAT has a custom built batch mode autosampler capable of loading samples from 96 and 384 well plates. The plates are temperature controlled, and the measurements are done in the coflow sample cell to minimize radiation damage. This autosampler can run one sample every ~3 minutes.
At BioCAT, these measurements batch mode measurements take ~10 µl of sample. However, in some cases smaller volumes can also yield usable SAXS data.
AF4-MALS-SAXS
Asymmetric flow field-flow fractionation (AF4) is a method for size-based separation of macromolecules with no stationary phase and almost no shear. This makes it a power alternative separation technique for systems that are challenging to separate on an SEC, either due to column interactions, or stability issues due to the shear and high hydrostatic pressures of columns. BiocAT profiles AF4 in-line with MALS, DLS, RI, and SAXS (AF4-MALS-SAXS). AF4-SAXS is of particular interest for systems like lipid nanoparticles (LNPs) and liposomes, which tend to behave poorly on SEC columns but are of great interest currently for drug delivery purposes.
BioCAT uses a Wyatt Eclipse NEON with dilution control module (DCM) for optimal separation. This is coupled to a Wyatt DAWN HELEOS II MALS+DLS (17 channels LS, plus 1 DLS) detector, and a Wyatt Optilab T-rEX dRI detector for the MALS-DLS-RI measurements. It is run off of an Agilent 1260 Infinity II HPLC with a multisampler and mulit-wavelength UV detector. BioCAT provides the following channels, spacers, and membranes for users:
- Wyatt Short Channel
- Wyatt Long Channel
- Wyatt Dispersion Inlet Channel
- Polyethersulfone (PES) membranes with 10 and 30 kDa MW cutoffs
- Regenerated cellulose (RC) membranes with 10 and 30 kDa MW cutoffs
- 275, 400 and 525 µm spacers
These experiments are more involved, and often require some work to optimize an appropriate separation program, so it is important to discuss your potential AF4-MALS-SAXS experiments with beamline personnel before requesting beamtime.
Time-Resolved SAXS
Basic characteristics of the time-resolved SAXS setup at BioCAT:
| Energy | 12 keV |
| Flux | ~2*1012 ph/s |
| Energy Resolution (dE/E) | 2 x 10-4 (Si <111>) |
| Beam size (FWHM) | ~3 x 2 µm2 (V x H), focused at the sample |
| Sample to detector distance | ~2 m |
| q range | ~0.01 - 0.65 Å-1 |
| Detector | EIGER2 XE 9M (Dectris) |
| Temperature | RT |
| Time range (chaotic mixer) | 45 μs to 60 ms |
| Time range (laminar mixer) | 5 ms to 1.5 s |
| Time range (stopped flow) | >1 ms |
While studying biological macromolecules in equilibrium remains the predominant use of SAXS, the focus of considerable effort at BioCAT over the past 15 years has been the development of instruments for time-resolved SAXS (TR-SAXS) that allow real time measurement of macromolecules while they undergo conformational transformations. Of particular interest are transient states too short lived to be studied using most techniques. In order to answer these questions, BioCAT provides general users access to routine experiments with two different continuous flow microfluidic mixers, which together provide access to time ranges after mixing from sub-50 μs to 1.5 s. BioCAT also has a stopped flow instrument and on-plate mixing in the batch mode autosampler for access to longer time ranges. BioCAT provides all of the necessary hardware and the control and analysis software, so all users have to do is provide their sample and buffer, just like an equilibrium SAXS experiment.
Mixing-based TR-SAXS experiments occupy a unique spot in the landscape of X-ray techniques studying time resolved changes in biological macromolecules. While laser pump-probe SAXS/WAXS experiments at XFELs and synchrotrons can access much faster time ranges (fs and ps respectively), these types of experiments are limited in the range of reactions that can be studied to only samples with light induced conformation changes or, more recently, temperature jump and caged compound triggered reactions (though these tend to be on slower timescales). While not amenable to ultra-fast timescales, mixing can study any process initiated by a change in solution conditions, such as these conditions explored by BioCAT users: salt jump, pH jump, addition of a ligand/substrate/co-factor, addition of another macromolecule and renaturation by dilution of denaturant. The sample consumption for the mixer experiments is also typically much lower than pump probe experiments, often <1 mg for slower reactions, whereas XFEL and synchrotron pump probe experiments can use 50-100 mg or more. While higher-resolution techniques such as time resolved MX, freeze-quench time-resolved cryo-EM, and time-resolved NMR are also used to study these types of reactions and can provide vital information, they all have limitations such as restrictions on large conformation changes (MX), limited time ranges to generally >1 ms (mixing-based MX, cryo-EM), and restrictions on size (NMR, cryoEM) and flexibility (cryo-EM, MX). As with most structural biology questions, there is no perfect technique for every system, but mixing-initiated TR-SAXS serves an important role by providing fast time points, a wide range of conditions, and a time range well suited to many interesting biological dynamics, including many large conformational motions, oligomerization, complexation, and refolding in macromolecules of essentially any size.
The biological questions that originally motivated the development of TR-SAXS technology at BioCAT were protein and RNA folding, key to understanding many human diseases like Alzheimer’s disease, Parkinson’s disease and ALS which are caused due to mis-folding of proteins. This included user projects looking at the transient intermediate structural states and the refolding kinetics of various proteins and RNA. This early focus has expanded, and now a wide variety of systems are studied, including published work on conformational changes induced by ligand binding, kinetics of micelle formation, and the kinetics of liquid-liquid phase separation.
In the following video from the 2022 Everything BioSAXS 8 workshop, Dr. Jesse Hopkins discusses the current state of time resolved SAXS at BioCAT (get slides)
Continuous Flow
BioCAT has been developing advanced microfluidic mixers, including a chaotic/turbulent mixer and a laminar flow mixer, to collect SAXS data on reactions as fast as ~45 µs. Rapid mixing devices for SAXS have fallen into two broad categories — chaotic/turbulent and laminar. These devices facilitate rapid and efficient mixing events between multiple fluid streams containing the biological macromolecule of interest and small solutes that engender structural changes in the macromolecule.
Laminar mixing utilizes hydrodynamic focusing to reduce the central flow channel to a narrow (typically ~1-10 µm) sheath. A version of this mixer based on a design from the Pollack group at Cornell) is currently available at BioCAT and can provide access to time ranges from ~5 ms to 1.5 s. These experiments use modest amounts of sample, ~1 mg per time series (~100 time points).
In chaotic/turbulent mixing, chaotic/turbulent flow breaks the solution into eddies small enough for reactants to diffuse rapidly. Mixing can be much more rapid than in laminar flow mixers, but requires much higher flow rates. In its current iteration, the BioCAT mixer (developed in collaboration with the Matthews group at U. Mass.) can access time points from ~45 µs to 60 ms and a complete experiment can be performed with ~10 mg of sample.
Time-resolved experiments are more involved than equilibrium experiments, so interested users should discuss possible experiments with the SAXS scientific contact.
Both of BioCAT’s mixers are microfabricated in quartz by Translume, with X-ray observation regions with 1 mm (laminar) or 250 μm (chaotic) deep channels and ~50 μm thick quartz windows. For both mixers, the mixer is mounted on vertical and horizontal scanning motors, and pumps are connected to the inlet ports on the mixer. The mixer is placed at the focal point of the microfocusing CRL optics, so that the small beam passes through the microfluidic channel without scattering from the edges of the channel. A typical measured time series consists of ~100 timepoints.
After setup and calibration, the basic data collection procedure is as follows: First, flow of mixing and sample buffers is started. Then, simultaneously, the X-ray exposure and a continuous scan of the mixer is started. The X-ray beam is scanned along the observation region, and images are measured while the mixer is moving (a continuous/fly scan rather than step scan), which is important for minimizing radiation damage and maximizing throughput. Each exposure along the observation region corresponds to a different timepoint after mixing. Repeated scans along the mixer add additional data at the same timepoints, which improves the signal to noise of the measurement at the timepoint. After sufficient buffer scans, the sample is injected into the mixer via an injection valve. Measurements are carried out while all the sample flows through the mixer, yielding multiple scans with mixed sample measured at every timepoint. Once all the sample has passed through the mixer, additional scans of just buffer are measured, yielding pre- and post-sample buffer and sample measurements at every timepoint as part of the same experiment. If necessary, the measurement is repeated multiple times to provide good data at each time point. Typically, at least 3 such measurements are made.
Background scattering varies with position on the mixer, so reproducible scanning and detector triggering are paramount for data quality. The horizontal motor, ‘down’ the channel to different time points is a Newport XMS160-S capable of 1 nm incremental motions, 0.03 μm bidirectional repeatability, and an optimally level scan. The vertical motor, ‘across’ the channel for alignment and during the scan to account for any small angular offset of the channel from horizontal, is a Newport GTS30V with similar precision but with lower speeds. Triggering exposure and metadata collection, such as transmitted intensity for normalization, on the encoder positions provides highly reproducible data, to the accuracy of the encoders (1 nm) and jitter in the various trigger systems (<120 ns).
Fluid handling and monitoring for the laminar mixer uses syringe pumps (Pump 11 Pico Plus Elite, Harvard Apparatus), stand-alone chromatography injection valves (IDEX MXP9900-00), and flow meters (BFS-1 Elveflow). The chaotic mixer uses much of the same setup, but is driven by three HPLC pumps (ReaXus LD012PRX Teledyne).
Stopped Flow
The BioCAT stopped flow setup uses a Biologic SFM-400 stopped flow mixer with an MEC 22998 micro-volume mixer, allowing ~1 ms dead time, and an x-ray observation cell. Because of the limitations in time resolution and possibility of radiation damage, unless you specifically know your experiment requires stopped flow mixing, BioCAT recommends using the continuous flow systems.
Instrumentation for SAXS
In addition to the instrumentation described above, BioCAT has a fully equipped wet lab for sample preparation. In addition to the beamline instrumentation described elsewhere, two sets of scatterless in-vacuum JJ x-ray slits are used as the collimating and anti-scatter beam slits, and a two sets of in-vacuum Xenocs scatterless x-ray slits are used as the guard slits. An in-line sample camera is located just after the guard slits, using a mirror with a 6 mm through hole for the x-ray beam. Normalization of data is done using an active beamstop which uses indirect detection on a photodiode.