Production, Crystallization, and Structure Determination of the IKK-binding Domain of NEMO

1. Design of construct for crystallography

1. Clone the sequence of NEMO-EEAA as in previous publication¹² in a vector for expression in E. coli using the T7 promoter, including a N-terminal hexa-histidine tag and a protease cleavage site.
   NOTE: In this protocol, we used a vector modified to include a N-terminal hexa-histidine tag and a Tobacco Etch Virus (TEV) cleavage site¹⁰. This vector facilitates cleavage of the His tag for protein crystallization and leaves only the short extension of GSW residues before the start of the desired protein sequence. The vector from which this was derived, and alternative vectors are listed in the Table of Materials. In this protocol, subsequent modifications to the original NEMO(44-111) sequence were introduced stepwise, as described earlier¹⁰, using side directed mutagenesis. We initially attempted to stabilize the NEMO coiled-coil dimer appending the ideal coiled-coil adaptors (in a length of at least three heptads) to the N-terminal or C-terminal end or to both. The double coiled coil was the most promising from earlier crystallization trials and it was subsequently modified introducing mutations to improve crystallization as described previously¹².

2. Large scale expression of His₆ tagged NEMO-EEAA

1. Transform construct into BL21(DE3) competent cells. Store at -80 °C as a cell glycerol stock.
2. Day 1 – Prepare a cell starter culture. In a 125 mL Erlenmeyer flask, add 20 mL of Terrific Broth solution and 20 µL of a 100 mg/mL stock of Ampicillin. Add a few microliters of cell glycerol stock (from -80 °C storage of BL21(DE3) competent cells transformed with vector).
3. Shake the 10 mL starter culture overnight at 37 °C, 220 rpm (approximately 15 h).
4. Day 2 – From the starter, dilute to an OD₆₀₀ = 0.1 in 250 mL of Terrific Broth. Add ampicillin to a final concentration of 100 µg/mL. Grow to an OD₆₀₀ = 0.8-1.0.
   1. Add isopropyl β-D-1-thiogalactopyranoside (IPTG) to 500 µM, and grow for 4 h at 37 °C.
   2. Measure OD₆₀₀ of induced culture after 4 h. Culture should reach an OD₆₀₀ = 6-10.

3. Purification of His₆ tagged NEMO-EEAA

1. Spin cell culture down at 3,800 x g for 20 min at 4 °C.
2. Save the cell pellet and discard the medium.
   NOTE: The cell pellet can be saved and stored at -20 °C at this point, for purification at a later time.
3. Resuspend the cells in 40 mL of lysis buffer containing 20 mM Tris, 150 mM NaCl, 10 mM imidazole, 2 mM MgCl₂, 0.5 mM phenylmethylsulfonyl fluoride, 2 mM Dithiothreitol (DTT), and 3 µL of Benzonase Nuclease.
   NOTE: The Nickel immobilized metal ion affinity chromatography (IMAC) column utilized is compatible with 2 mM DTT. Alternatively, 0.2 mM tris(2-carboxyethyl)phosphine (TCEP) can be utilized.
4. Split resuspended cells into two 20-25 mL aliquots.
5. Lyse the cells using French press (approximate pressure 25,000 psi), repeating 2-3 times for each aliquot (in the cold room).
   NOTE: Alternatively, cells could be lysed by sonication (not tested in this protocol).
6. Add urea to the cell lysate to a final concentration of 8 M, allow to incubate on a rocking platform for a minimum of 2 h or up to overnight. This and all following purification steps, with the exception of the dialysis, can be performed at room temperature.
7. Day 3 – Transfer the lysate to ultracentrifuge tubes and balance weight, ensuring lysate fills tubes to at least ¾ full. Spin the lysate at 125,000 x g for 45 min at 25 °C. Decant the supernatant into a 100 mL beaker for loading onto column.
   NOTE: Centrifuging at 4 °C will cause urea to crash out.
8. On a fast liquid chromatography system, remove ethanol from IMAC 5 mL column with 25 mL of ultrapure H₂O at 5 mL/min, followed by 25 mL of elution buffer containing 20 mM Tris, 150 mM NaCl, 500 mM imidazole, 2 mM DTT, pH 8.0, then 25 mL of binding buffer containing 20 mM Tris, 150 mM NaCl, 10 mM imidazole, 2 mM DTT, 8 M urea, pH 8.0.
9. Load urea incubated supernatant at 3 mL/min onto IMAC column, collecting the flow through. Wash the column for 10 column volumes with binding buffer at 3 mL/min.
10. Refold NEMO-EEAA construct on column by washing column with refolding buffer for 20 column volumes at 3 mL/min, containing 20 mM Tris, 150 mM NaCl, 10 mM imidazole, 2 mM DTT, pH 8.0.
11. Perform gradient elution of NEMO-EEAA, from 10 to 500 mM Imidazole over a 12-column volume gradient, collecting all eluate in fraction collection plate (1 mL fractions).
12. Continue elution at 500 mM imidazole for two column volumes, continuing to collect.
13. Run sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis (PAGE) of fractions to determine NEMO- EEAA presence in elution fractions.
    NOTE: We used 10% acrylamide, MES buffer.
14. Pool fractions containing pure target protein.
15. Measure protein concentration by Bradford assay¹⁷.
    NOTE: This is necessary to estimate the amount of protease for tag cleavage.

4. His₆ tag cleavage and purification

1. Add TEV in a 1:10 weight ratio of TEV:NEMO-EEAA protein to cleave the His₆ tag. The TEV protease was purified in house.
   NOTE: Optimize the amount of TEV, time and temperature required for complete cleavage separately.
   1. Express TEV protease, S219V mutant¹⁸ in BL21(DE3)-RIL cells and purify as described earlier¹⁹. Briefly grow the cells as described in step 2, lyse by French press and purify using an IMAC column. Store final protein in 25 mM sodium phosphate buffer, pH 7.8, 150 mM NaCl, 1 mM EDTA, 2 mM DTT, 20% v/v glycerol.
2. Dialyze the sample overnight (approximately 15 h) in 4 L of 20 mM Tris, 150 mM NaCl, 2 mM DTT, pH 8.0, to allow for cleavage and to remove excess imidazole from the sample for the subsequent purification.
3. Day 4 – Remove the sample from dialysis. Run an SDS-PAGE gel of the sample from TEV cleavage to ensure cleavage is completed.
4. On a fast liquid chromatography system, remove ethanol from an IMAC 5 mL column with 25 mL of ultrapure H₂O at 5 mL/min, followed by 25 mL of elution buffer containing 20 mM Tris, 150 mM NaCl, 500 mM imidazole, 2 mM DTT, pH 8.0, then binding buffer containing 20 mM Tris, 150 mM NaCl, 10 mM imidazole, 2 mM DTT, pH 8.0.
5. Load the column at 1 mL/min with TEV-cleaved NEMO-EEAA. Cleaved NEMO-EEAA will elute in the flow-through: collect in a 96 well fraction collection plate (1 mL fractions). Wash the column for five column volumes of 20 mM Tris, 150 mM NaCl, 10 mM imidazole at 1 mL/min, continuing to collect in fraction collection plate.
6. Elute TEV and uncleaved His₆-NEMO-EEAA with three column volumes of 20 mM Tris, 150 mM NaCl, 500 mM imidazole, 2 mM DTT, pH 8.0, collecting elution in a 50 mL flask.
7. Run SDS-PAGE gel of flow-through fractions to determine presence of cleaved NEMO-EEAA.
8. Pool flow-through fractions containing cleaved NEMO-EEAA construct, and concentrate using a stirred-cell concentrator to 5 mL. Membrane molecular weight cut-off (MWCO) = 3 kDa.
9. Using a 3 kDa MWCO membrane, dialyze concentrated sample in 2 L of 20 mM Tris, 100 mM NaCl, 2 mM DTT, pH 8.0 for 2 h. Change the dialysis buffer to 2 L of fresh dialysis buffer for overnight dialysis (approximately 15 h), at 4 °C.
10. Day 5 – Load 5 mL of the dialyzed sample on a size exclusion chromatography (SEC) 16 mm x 60 cm column (34 μm average particle size) at 1 mL/min in 2 mM Tris, 100 mM NaCl, 2 mM DTT, pH 8.0. Repeat with additional columns depending on sample volume.
11. Pool the fractions corresponding to dimeric NEMO-EEAA.
    NOTE: NEMO-EEAA elutes between 60-65 mL, corresponding to a larger molecular weight protein, due to the elongated nature of the dimeric coiled coil.
12. Concentrate using a stirred-cell concentrator and a MWCO = 3 kDa membrane to a final concentration of 113 µM (1.65 mg/mL).
13. Aliquot the protein and store at 4 °C (stable for over 1 month).

5. Sparse matrix screening

NOTE: The protocol performs crystallization trials using commercially available screens and setting up sitting drop experiments using a crystallization robot. Crystal images are collected automatically by an imager.

1. Using commercially available sparse matrix screens (see Table of Materials), pipette 60 µL of sparse matrix solution into each of the 96 wells of a 2 drop-chamber crystallization plate for sitting drop vapor diffusion (reservoir solution).
2. Using a robotic drop setter, dispense 100 nL of protein solution at 1.65 mg/mL in a 1:1 ratio with reservoir solution in drop 1 for a final volume of 200 nL; then 66 nL of protein solution with 134 nL of reservoir solution for a final volume of 200 nL in drop 2 (1:2 ratio).
3. Seal the plate using 3-inch-wide sealing tape immediately after dispensing.
   NOTE: Drops will dry out if left exposed to atmosphere for longer than 2-3 min.
4. Store the trays in the crystallization imager storage, at 20 °C, checking the images collected automatically for crystal presence, starting after two days.
   NOTE: Crystallization screening proceeded in parallel with construct optimization. Initial crystals formed in the following conditions (commercial screens listed in the Table of Materials): a) 0.1 M Tris, pH 8.0, 30 % v/v polyethylene glycol (PEG) MME 550, 5% poly-γ-glutamic acid, 200-400 kDa low molecular weight polymer (PGA -LM); b) 0.1 M Tris, pH 7.8, 20% w/v PEG MME 2k, 5% PGA-LM; c) 0.1 M Tris, pH 7.8, 20% w/v PEG 3350, 5% PGA-LM. Sparse matrix screen crystals will have poor lattice uniformity; therefore, they will look poor using cross-polarized imaging. Use UV imaging to ensure that the crystals contain protein. The following seed stock generation step is necessary for obtaining the final diffraction quality crystals.

6. Seed stock generation

NOTE: We reproducibly obtain crystals for seed generation in 0.1 M Tris pH 8.0, 5% PGA-LM, 3.6% w/v PEG 20k. However, crystals will show high mosaicity and are unsuitable for data collection at this stage.

1. Using a kit for seed generation, prepare seed stock by pipetting out entire drop with crystal present, and place into 50 µL of crystallization condition solution in the provided vial.
2. Vortex the seed stock for 3 min, pulsing 20 s on and 10 s off.
3. Serially dilute seed stock in 1:10 increments down to 1:10,000. Store all dilutions at 4 °C, for further use.

7. Fine screens

1. Design fine screens varying the conditions for Tris, PGA-LM, and PEG. Vary PEG length for individual trays.
   NOTE: The fine screen that produced the crystal utilized for structure determination of NEMO-EEAA employed the following conditions: 0.1 M Tris pH 8; PGA-LM varied from 8 to 0% in columns 1-12. Rows A-H screened different PEGs, with concentrations varying in columns 1-12 as follows. A: PEG 200 (0-40% v/v); B: PEG 400 (0-40% v/v); C: PEG MME 500 (0-40% v/v); D: PEG 1000 (0-30% w/v); E: PEG 3350 (0-30% w/v); F: PEG 6k (0-30% w/v); G: PEG 10k (0-20% w/v); H: PEG 20k (0-20% w/v). The crystal appeared in well H4 (5.45% w/v PEG 20k, 5.8% w/v PGA-LM). In this protocol, a protein crystallography screen builder liquid handler was used to build the screens.
2. Seed a protein stock in a 1:25 volume ratio of 1:1,000 seed dilution.
   NOTE: Concentration of NEMO-EEAA will drop slightly, but crystals will still form.
3. Repeat step 2 using 1:10,000 seed dilution, same 1:25 volume ratio.
4. Using the drop setter, dispense 100 nL of protein solution at 1.65 mg/mL, with 1:1,000 dilution of seed stock into a 1:1 ratio with reservoir solution in drop 1 for a final volume of 200 nL. Repeat for drop 2, but with 1:10,000 dilution of seed stock present.
5. Seal the plate using 3-inch-wide sealing tape immediately after dispensing.
   NOTE: Drops will dry out if left exposed to atmosphere for longer than 2-3 min.
6. Store the trays in the imager storage, at 20 °C, checking images collected after two days for crystal presence.
7. Check the crystals with the cross-polarizer imager for lattice uniformity, to select for single conditions to set up following trays.

8. Generation of crystals for data collection

1. Design single condition screens around Tris, PGA-LM, and PEG condition which produced uniform crystals as analyzed by cross-polarized images, and the largest crystals possible.
   NOTE: Crystals from these conditions are irregular in shape, mostly rectangular thin sheets. It is key to select a condition where the edges of the crystal are well defined and have the greatest thickness possible.
2. Make 20 mL of crystallization condition by hand.
3. Using a multi-channel pipettor, dispense 60 µL per well in a 2 drop-chamber, 96 well crystallization plate for sitting drop vapor diffusion.
4. Prepare seed stock as described in step 6.2.
5. Dispense the protein as described in step 6.3.
6. Seal the plate using 3-inch-wide sealing tape immediately after dispensing.
   NOTE: Drops will dry out if left exposed to atmosphere for longer than 2-3 min.
7. Store the trays in the imager at 20 °C and check the images for crystal presence every day.
8. Check cross-polarized images of the crystals for lattice uniformity, to select crystals for data collection.

9. Determination of cryo-protectant

1. To test the cryo-protectants, create stock solutions corresponding to the crystallization conditions but containing 30%, 20%, 10%, and 5% higher concentration of each component. The addition of the cryo-protectant volume will result in a final concentration of components that is the same as the crystallization conditions.
   NOTE: For testing a cryo-protectant at a 30% concentration by volume for a 0.1 M Tris crystallization condition, start with a stock solution of 0.143 M Tris, before adding the cryo-protectant.
2. From a cryo-reagents kit, create 10 µL of sample for cryo-protectant test by mixing 30% by volume of cryo condition in 70% of crystallization condition stock solution, for a final concentration of 30% cryo-protectant in original crystallization conditions. Mix thoroughly.
   1. Using a 10 µL pipette, take 5 µL of test cryo-protected solution and plunge the pipet tip into liquid nitrogen. If ice is observed, discard.
   2. Test all cryo-protectants at 30%. For the successful solutions, repeat the process at 20%, then 10% and 5% of cryo-protectant.
   3. Utilizing the successful cryo-protectant solution with the smallest percentage of cryo-protectant, add 0.5 µL of solution into a test drop with crystals present. Observe under microscope, timing how long crystal lasts in the condition, if not indefinite.
      NOTE: These crystals are for test only and will be discarded. For NEMO-EEAA, 12% 1,2-propanediol is the optimal cryo-protectant solution. 12% accounts for the dilution the cryo-protectant solution will experience when added to the crystal drop of approximately 100 nL, for an approximate final concentration of 1,2-propanediol of 10%.

10. Crystal looping

1. Loop crystals 1-2 days before shipment to synchrotron.
   NOTE: Crystals will be roughly 60-100 µm in diameter: 0.05 – 0.10 mm loops are ideal for looping.
2. Cut the tape from the top of the well.
3. Add 0.5 µL of crystallization solution containing 12% 1,2-propanediol cryo-protectant directly to the well.
   NOTE: Final concentration of 1,2-propanediol is now at approximately 10%, due to dilution by 100 nL of drop solution (approximate drop volume reduction from initial 200 nL, due to vapor diffusion).
4. Loop the crystal from the well.
   NOTE: Crystals often grow on the bottom of the well but will dislodge with a gentle nudge from the loop. Once dislodged, loop.
5. Store the crystal containing cryo-loops in pucks immersed in liquid N₂.
6. Store these pucks in liquid N₂ in Dewar flask until they are ready for shipment for X-ray diffraction at the synchrotron.

11. Data collection

1. Collect X-ray diffraction data. In this protocol, use AMX beamline (ID: 17-ID-1), National Synchrotron Light Source II.
   NOTE: Data was collected on site but can be collected remotely. Collect data in the region of the crystal which showed lattice uniformity in cross-polarized images. Position the crystals in the loop so that data collection does not involve "poor" regions of the crystal while the crystal rotates in the goniometer. Data which provided the best resolution came from collection on the edge of an extension of a crystal about 5 x 5 μm in size. Use rastering to identify the best areas on the crystal to collect²⁰.

12. X-ray data processing

1. Process dataset to highest resolution collected (1.8 Å) with XDS IDXREF program to determine space group, unit cell, and solvent content.
2. Integrate data using XDS INTEGRATE program.
3. Process scaled intensities from XDS XSCALE program using STARANISO Server, using I/σI cutoff mean of 1.2 for diffraction-limit surface for the data.
   NOTE: Data will not be cut in true ellipsoid. Retain all data above the I/σI of 1.2 cutoff. The statistics on the data calculated for spherical completeness will be poor, due to the non-spherical truncation. The elliptical completeness was 88% with highest resolutions of: 1.88 Å, 2.10 Å and 2.55 Å along the a*, b* and c axis, respectively.

13. Structure solution

1. Utilize the X-ray structure of GCN4 (PDB: 4DMD)¹⁵ as a search model for molecular replacement using MRage²¹ in PHENIX²². The 4DMD structure was defined in MRage as an "ensemble", and the MRage solution successfully built the structure portion corresponding to the N-terminal coiled-coil adaptor of NEMO-EEAA, homologous to the search model, for both chains in the dimer.
   NOTE: Turn off anisotropy correction in PHASER, or data will be further scaled back.
2. Utilize successive rounds of Autobuild²³ in PHENIX and manual building (using 2F_o-F_c and F_o-F_c maps, in Coot²⁴) to build the remainder of the structure.
3. Manually build the residues still missing into the model based on 2F_o-F_c and F_o-F_c maps, using Coot²⁴.
   NOTE: The last stage involved building the 4 N-terminal residues and 4 C-terminal residues for each monomer. Sidechain placement was also manually adjusted as needed.
4. Calculate a composite omit map using PHENIX and a 10% omission of the structure.

14. Structure refinement

1. Refine the structures with PHENIX Refine. Run the initial refinements against bulk-solvent and stereochemistry weights, relaxing RMSbond and RMSangle constraints to 0.01 and 1.0, respectively. Continue refinement on Individual B-factors, TLS parameters, and Occupancies.