Online Size-exclusion and Ion-exchange Chromatography on a SAXS Beamline

The overall goal of implementing online size-exclusion and ion-exchange chromatography on small angle x-ray scattering beam lines is to obtain better quality low resolution model data by ensuring minimal delays between sample preparation and data acquisition. These methods can help to answer key questions in the field of structural biology. Like the radius operation, the largest intra-particular distance, the particle shape, the degree of folding denaturation or disorder.

SAXS requires molar dispersed samples. The main advantage of this technique is that the time between the purification and the measurement is short, helping to avoid sample denaturation or advication. Before going to the beamline facility, prepare a concentrated sample using previously published methods and perform SEC test runs.

At the beamline facility, place one milliliter of gel filtration buffer in a 1.5 milliliter reaction tube. Then, connect the bottle of gel filtration buffer to an HPLC system in line with the flow-through capillary of the SAXS set-up. Set the flow rate, maximum column pressure, and acquisition time.

Note the back pressure without the column. Purge the lines, and then connect and equilibrate the size exclusion chromotography column. Load the one milliliter buffer sample into the sample changer.

Exit and secure or intra-lock the experimental hutch, then log into the beamline control software in sample changer mode. Send the buffer sample in the sample changer directly through the SAXS flow-through capillary. Perform a short SAXS test run and check that the sound x-ray scattering intensity shows no signs of radiation damage.

Switch the beamline control system to HPLC mode and perform a SAXS test run of buffer through the full HPLC SAXS system. If the data does not match the first measurement, re-equilibrate and perform another test run. Next, in the laboratory, centrifuge the protein sample at 13, 000 times G for 10 minutes to remove aggregates.

Transfer the supernatant to a HPLC compatible glass vial. Load the vial into the auto-sampler, noting the sample position. Interlock the hutch and create a folder for the experiment data in the beamline control system.

Then, in the HPLC software, click on quick batch. Set the file path and activate automatic ascii conversion of UV Vis data. Enter the injection volume and the sample position in the auto-sampler.

Click start'to add the run to the queue. Enter the file name, but do not click save'yet. In the beamline control software, set the number of data frames so that the SAXS data collection time is slightly longer than the HPLC acquisition time.

Open the beamline shutter, and start SAXS data collection. Switch to the HPLC software, and click save'to inject the sample. Note the frame number at the beginning of UV Vis data acquisition, and monitor the summed intensity over time.

During sample preparation, perform an IEC test run with a linear sodium chloride gradient from 25 millimolar to one molar. For each peak of interest, determine the corresponding amount of high salt buffer B in the mobile phase. To begin the IEC SAXS experiment, set aside one milliliter samples of buffers A and B, and then connect the buffers to the HPLC system.

Set the HPLC parameters, and then purge the system and flush the lines with low salt buffer A.Connect and equilibrate the ion exchange column with buffer A.Load the one milliliter samples of buffers A and B into the sample changer. Interlock the hutch and perform a short SAXS test run for each buffer sample in sample changer mode to check the buffer susceptibility to radiation damage. Then, switch to HPLC mode and perform another SAXS test run for buffer A.Compare the scattering patterns to those obtained from the sample changer test run.

Next, create a folder for the experiment data. Then, design step wise HPLC gradient that starts with zero percent high salt buffer B, and increases five percent at intervals of two column volumes to reach 100 percent buffer B.Set up a single gradient run in which no sample is injected. Open the beamline shutter, start SAXS data collection, and start the HPLC run.

Note the number of frames elapsed when UV Vis data collection begins. After data collection, check that the summed intensity remained constant for at least 100 frames, and matches the values obtained from the samples of buffers A and B.Then, make another step wise gradient HPLC method, starting from zero percent buffer B.Create a step for each percent B value determined in the off-line IEC gradient test run. Add gradient steps one percentage point below, and 1.5 points above each of those values.

Set each step interval to 2.5 column volumes, and the final step in the gradient to 100 percent B.Dilute the supernatant as needed to achieve the composition of buffer A.Stop the HPLC pumps, disconnect the column tubing from the detectors, and place the end of the line into a container to collect the column flow. Then, transfer the buffer A line into the diluted sample, ensuring that no air bubbles are trapped in the tube. Start the pumps to draw the sample into the column.

Once the sample container is nearly empty, stop the pumps and put the buffer A bottle back on the line. Start the pumps and run buffer A through the column for two column volumes, then stop the pumps and reconnect the column to the detectors. Using the new step wise method, set up a single HPLC run with no auto-sample injection.

Open the beamline shutter and acquire the SAXS and UV Vis data. Monitor the summed intensity over time. Then, use post processing software to convert the HPLC data to an ascii format.

Using an online data manager, identify the SAXS frames corresponding to the sample elution peak. Verify that the radius of giyration is stable throughout the peak. Import the peak frames into a SAXS data processing program, and create an average non-corrected scattering curve.

For SCC SAXS, confirm that the high Q region of the average buffer, not just the non-corrected scattering curve. Load and scale the auto-subtracted frames from the peak, and check that there are no systematic differences. Average all curves to obtain the final SAXS curve.

For IEC SAXS, average 50 frames from each gradient step. Compare these averages to the non-corrected curve, and select the gradient step with the best high Q match. Subtract the chosen average from the non-corrected curve to create the final SAXS curve.

Finally, determine the pair distance distribution and perform ab initio modeling to determine the most representative model of the protein. The structural data parameters obtained from SEC and IEC SAXS of the D5 protein were very similar. The molecular mass estimates were both in agreement with the predicted mass.

Ab initio modeling was performed with no imposed symmetry, and with C6 symmetry. Both models are compatible with the experiment data, indicating that the protein structure possesses C6 symmetry. Molecular modeling yields a hexogonal pyramid-like structure with a partly obstructed central channel.

The averaged SEC and IEC models were very similar. Upon examining individual models, the channel obstructions were found likely to be an artifact of the averaging process. SAXS is becoming reasonable more standard, but for many purification and other ion exchange chromotography steps needed for getting rid of aggregates or contaminants.

In this protocol we presented both techniques, online size exclusion chromotography, and online ion exchange chromotography on a bio SAXS beamline. A caveat of both techniques is that the precise protein concentration is not known, so we cannot determine the molecular mass scattering. However, for globular proteins, we can still estimate the mass volume.

A linear gradient might be applied instead of the step wise gradient we used here in the ion exchange chromotography, although this requires an optimization of the buffer condition for an optimal separation of the peaks. These techniques allow us to collect good quality SAXS data, even on complicated systems.