Cloning and expression
The genes for full-length human Dnase1L3 and the Dnase1L3 ΔCTD were sub-cloned into BamHI/XhoI sites in a Tobacco-Etch Protease (TEV) cleavable, his-tagged, maltose-binding protein (MBP) expression vector for E. coli expression. The expression plasmid was transformed into Rosetta-Gami®2 (DE3) competent cells (Novagen) to maintain disulfide bond formation. Single colonies were picked and grown to confluence in 100 ml of LB media. 10 ml of culture was inoculated into 1 L of Terrific Broth (TB) media, and grown to OD600 of 2.0. The culture was induced with 400 μM IPTG and grown at 18 ∘C for 18 h. Cells were harvested by centrifugation at 18,000 rpm in an SS-40 rotor for 20 min. Cell pellets were frozen at −80 ∘C until used for purification.
SH3-CTD was expressed in BL-21-Gold (DE3) E. coli cells (Agilent) in 3 L of Terrific Broth, induced with 400 μM IPTG and grown at 18 ∘C for 18 h. Cells were harvested by centrifugation, frozen in liquid N2 and stored at −80 ∘C until ready for use.
Purification of full-length Dnase1L3, Dnase1L3ΔCTD, SH3-CTD and Dnase1L3-MBP fusion
15–20 g of cells were resuspended in 160 ml of lysis buffer (20 mM HEPES pH 7.4, 300 mM NaCl, 1 mM CaCl2). The cells were lysed using a microfluidizer and spun down at 19,500 rpm in a JA20 rotor for 50 min. The supernatant was collected and run through a newly regenerated Ni2+/NTA column. The column was washed with lysis buffer that contained 30 mM imidazole. The protein was eluted with lysis buffer that contains 250 mM imidazole, 5 mM maltose. 0.1 mM PMSF was added to the eluted sample. To remove the MBP fusion protein and to isolate the Dnase1L3, 2–3 mg of TEV protease were added to the sample and incubated at 4 ∘C overnight. An SDS gel was run to ensure the fusion protein was cut by TEV before moving on to the next purification step. The sample was diluted to 100 mM NaCl with Buffer A (20 mM Hepes pH 7.4, 1 mM CaCl2). An SP-Sepharose (cation exchange) column was prepared and the sample was run through the column while collecting the flow-through (which contains the MBP). The column was then washed with Buffer A until the UV returned to the baseline before the sample was loaded onto the column. The protein was eluted with a gradient of 0–100% Buffer B (20 mM Hepes pH 7.4, 1 M NaCl, 1 mM CaCl2) with a volume of at least three times the bed volume of the column. Purity was measured by SDS-PAGE. Finally, the sample was concentrated to run on an Superdex-75 gel filtration column equilibrated with Buffer C (20 mM HEPES pH 7.4, 400 mM NaCl, 1 mM CaCl2). The peak was then run on an SDS gel and imaged using a BioRad Stain-Free Gel imaging system.
To purify SH3-CTD, centrifuged bacteria were lysed by microfluidization and insoluble cell components were pelleted with 19,500 x g centrifugation. SH3-CTD was separated from soluble lysate with 6-His Ni2+/NTA affinity chromatography. The high-affinity fraction containing SH3-CTD was eluted at 250 mM Imidazole concentration. The elution was concentrated to 1 ml and injected into a 75 ml Superdex-75 size exclusion column. The monodisperse fraction was exchanged into 1 M NaCl and incubated overnight with 1 mg of TEV. The CTD was separated from the SH3 and TEV using Ni-NTA affinity chromatography, the CTD fraction came out in the FT. The CTD was buffer exchanged using 700 Da cutoff filters into 100 mM NaCl and concentrated with lyophilization.
The wild-type Dnase1L3 and Dnase1L3 R206C MBP fusion proteins were produced and purified in a similar manner as the full-length Dnase1L3 purification methods described above. However, after Ni2+/NTA affinity chromatography, no TEV protease was added to the sample. Instead the fusion protein was purified to homogeneity by Q-Sepharose anion exchange chromatography followed by size exclusion chromatography. At each step, the purity was assessed by SDS-PAGE.
HeLa cells (ATCC (Manassas, VA, USA), CCL-2) and HEK cells (ATCC CRL-1573) were cultured at 37 ∘C and 5% CO2 in DMEM (Corning, Corning, NY, USA) supplemented with 10% Equafetal bovine serum (Atlas, Fort Collins, CO), 1 × L-glutamine (D10) and 1 × penicillin/streptomycin. All cell lines were negative for mycoplasma.
The anti-Dnase1L3 rabbit polyclonal antibody was obtained from Abnova, and the anti-dsDNA monoclonal antibody was from the Developmental Studies Hybridoma Bank (DSHB, Iowa City, IA). Clone autoanti-dsDNA was deposited with the DSHB by Voss, E.W. (DSHB Hybridoma Product autoanti-dsDNA). HRP-conjugated secondary antibodies were from Jackson Immunoresearch (West Grove, Pennsylvania).
Plasmid degradation assay
Plasmid degradation assays were performed in a 10 μl reaction volume, 200 ng of plasmid DNA was incubated with varying concentrations of Dnase1L3 full length, Dnase1L3 ΔCTD, Dnase1, MBP-Dnase1L3, SH3-CTD or SH3 alone in Dnase assay buffer (20 mM Tris, pH 7.4, 5 mM MgCl2, 2 mM CaCl2) for 30 min at 37 ∘C51. The extent of DNA degradation was quantitated by measuring the integrated intensity of degraded and intact plasmid DNA from Gel Red-stained agarose gels using Photoshop Creative Suite (Adobe, San Jose, CA) and determining the percent degradation. The EC50 for plasmid degradation was calculated from the dose-response curve using logistic regression.
Barrier to transfection
HEK cells (ATCC CRL-1573) were plated 1 day prior to the assay at 5 × 105 cells per well in a 24 well plate. DNA-lipid complexes were prepared by incubating 100 ng of eGFP-N1 plasmid (Takara Biosciences) with Lipofectamine 2000 for 20 min. DNA-lipid complexes were then incubated with 100 ng each of full-length Dnase1L3, Dnase1L3 ΔCTD or Dnase1 at 37 ∘C, 5% CO2 for 30 min. HEK cells were then transfected with the control or the Dnase treated DNA-lipid complexes and incubated for 48 h. The cells were supplemented with fresh D10 media after 24 h. Cells were harvested, washed in FACS buffer (2% fetal calf serum, 0.05% NaN3 in 1x PBS), and analyzed on an Attune NxT flow cytometer. Transfection efficiency was 70% in control cells. Reduced transfection efficiency was calculated compared to control transfected cells.
Immune complex degradation
To measure immune complex degradation, a modified ELISA protocol was used. ELISA plates were precoated with 0.05 mg/mL poly-L-lysine at room temp for 20 min, washed with 1x nuclease-free water, and coated with 5 μg/ml calf thymus DNA (Sigma) overnight at 4 ∘C. After washing 3 x in PBS with 0.05% Tween (PBST) and blocking for 1 h at room temp with 1% BSA in PBST, 250 pg/mL anti-dsDNA antibody was added to all wells except the standard curve. The standard curve received 2-fold dilutions of anti-dsDNA antibody starting at 500 pg/mL. After 1 h, the plates were washed 3x in PBST, Dnase (diluted into 20 mM HEPES, pH 7.4, 300 mM NaCl, 1 mM CaCl2) was added, and plates incubated 37 ∘C for 2 h. Plates were washed 3x in PBST, incubated with HRP conjugated goat anti-mouse IgG antibody (1:20,000), and developed using 0.2 mg/ml TMB (Sigma), 0.015% H2O2 in 100 mM sodium acetate, pH 5.5. The reaction was stopped with 0.5 M H2SO4. A450 was measured and antibody concentration in each well calculated. The percentage of dsDNA antibody remaining in the well was calculated by comparison to control. Percentage immune complex degradation was 100 – %remaining antibody. The EC50 for each Dnase was calculated using logistic regression.
Generation of microparticles
Microparticles from HeLa cells (ATCC CCL-2) were generated by treatment with staurosporine to make the cells apoptotic52. Cells were cultured overnight and then treated with 1 mM staurosporine (Sigma-Aldrich) for 8 h in serum-free DMEM. Microparticles were harvested and debris removed by centrifugation for 5 min at 1500 rpm. The supernatant was collected and centrifuged at 20,000 x g for 30 min to pellet the microparticles.
Generation of liposomes
Synthetic liposomes were prepared by mixing 20:20:10:50 mol percent of phosphatidylcholine (Avanti), phosphatidylethanol-amine (Avanti), phosphatidylserine (Avanti) and cholesterol (Sigma). This lipid mixture was dried under N2 and resuspended to a final concentration of 4 mg/mL in liposome buffer (15 mM HEPES pH 7.2, 1 mM Mg(CH3COO)2 and 50 mM sorbitol). Liposomes were incubated at 37 ∘C for 1 h, freeze/thawed using dry ice/37 ∘C bath 5 times and stored at −80 ∘C until further use.
Lipid and microparticles binding assay
Microparticles prepared from HeLa cells (ATCC CCL-2) or 4 mg/ml liposomes were incubated with 10 mg/mL bovine serum albumin carrier protein and 6 μM full length Dnase1L3, Dnase1L3 ΔCTD or Dnase1 in 1x PBS with Ca2+ and Mg2+ on ice for 1 h. Microparticles or liposomes were then centrifuged at 20,000 x g 4 ∘C for 15 min and the supernatant saved for SDS-PAGE analysis. The pellet was washed with 1x PBS with Ca2+ and Mg2+ and centrifuged at 20,000 x g 4 ∘C for 15 min. The resulting pellet was resuspended in SDS sample buffer. Samples were resolved on a 12.5% gel and either Coomassie-stained or transferred to a 0.45-μm nitrocellulose membrane. Immunoblotting with anti-Dnase1L3 (1:1000) was followed by anti-rabbit IgG conjugated to HRP (1:10,000). Antibody staining was visualized on a FluorChemE (Protein Simple, San Jose, CA, USA) using enhanced chemiluminescence reagent [0.01% H2O2 (Walmart, Bentonville, AR, USA), 0.2 mM p-Coumaric acid (Sigma-Aldrich), 1.25 mM Luminol (Sigma-Aldrich), and 0.1 M Tris, pH 8.4]. Immunoblots were analyzed using Photoshop Creative Suite 3.
The dsDNA binding assays with the CTD were performed using a FAM (carboxyfluorescein)-labeled 40-mer DNA (5’-FAM-GTGTTCGGACTCTGCCTCAAGACGGTAGTCAACGTGCTTG-3’ and 5’-CAAGCACGTTGACTACCGTCTTGAGGCAGAGTCCGAACAC-3’, IDT)53. The fluorescence polarization binding assays for Dnase1, Dnase1L3 WT and Dnase1L3 R132A were performed using the same 40mer sequence labelled with FITC (fluorescein 5(6)-isothiocyanate). The dsDNA was made by annealing the fluorophore-labeled ssDNA to an unlabeled complementary strand at 95 ∘C. Fluorescence polarization (FP) experiments for the CTD were performed in binding buffer (50 mM Hepes pH 7.5, 100 mM NaCl, 0.1 mM EDTA, 1%Glycerol, 1 mM TCEP, 0.1% PEG 6000 w/v), with 20 nM dsDNA, and an increasing amount of isolated CTD for a final volume of 25 μL. FP experiments for Dnase1L3 R132A were performed similarly, but the binding buffer contained increased EDTA, 1 mM to inhibit DNA degradation. Samples were incubated for 30 min at 37 ∘C in a 384-well black plate (Fisher Scientific) before reading for FAM or FITC-FP in a Synergy Neo2 plate reader using the FP 485/530 filter. Three technical replicates were performed. The apparent dissociation constants were calculated by fitting the data to a modified version of the Hill equation54.
Each of the protein samples were diluted into 20 mM HEPES, 400 mM NaCl, 1 mM CaCl2 pH 7.4 to a concentration of 2.2 μM. The protein was incubated for 30 min at 25 ∘C with 25 μM ANS (8-anilino-1-naphthalenesulfonic acid). ANS was excited at 365 nm and fluorescence was recorded from 400 to 600 nm for buffer, the MBP control sample, Dnase1L3 WT fusion protein, and Dnase1L3 R206C fusion protein in a 96 well plate using the BioTek Synergy4 Multimode Plate Reader. The excitation/emission slit width was set to 5 nm. The mean fluorescence for each wavelength was taken from three independent measurements, and the mean ± standard deviation is reported for each datapoint.
Crystallization, data collection, structure solution and refinement
Purified Dnase1L3 ΔCTD was concentrated to 8 mg/ml and was screened with multiple sparse matrix crystallization kits. Initial crystal hits were expanded to improve crystal quality and yield more useful crystals for X-ray analysis. The final crystallization condition was 19% PEG 8000, 250 mM MgCl2, 100 mM Tris, pH 8.5 at 10 ∘C. The final Dnase1L3 ΔCTD crystals grew to a final size of ~0.5 mm in 60 days and were obelisk-shaped. The Dnase1L3 ΔCTD crystallized in the P1 space group with four molecules in the asymmetric unit (Table 1). The final X-ray dataset was collected at the SLAC beamline 14-1. The Dnase1L3 ΔCTD crystals diffracted to 1.9 Å .
The structure was solved with molecular replacement, as implemented in Phenix55, using the 1DNK structure as the target. X-ray data to 2.2 Å was used for refinement. Electron density was fit using COOT56 prior to additional rounds of structural refinement in Phenix. Each of the four molecules in the asymmetric unit was refined independently using Phenix55.
A commercially synthesized peptide corresponding to the last 23 amino acids of Dnase1L3 (i.e., C-terminal domain (SSRAFTNSKKSVTLRKKTKSKRS) (United BioSystem Inc, Herndon, VA, US) was resuspended from powder in milli-Q water to a stock concentration of 50 μM. The stock was diluted to 20 nM in 5 mM NaCl with increasing concentrations of DNA from 0 nM to 293 μM of a self-complementary DNA substrate (GCGATCGCGCGATCGC). The circular dichroism data were collected on a Jasco J-815 CD spectrophotometer from 190 nm to 350 nm. Buffer subtractions were the equivalent concentration of oligomer DNA in 5 mM NaCl. The spectra were recorded in CD units of mdeg and then converted to molar ellipticity. The CTD was disordered throughout the measurements; however, the background subtraction became increasingly different from the Sample CD at high concentrations of DNA (293 μM) likely corresponding to differences in the secondary structure of the DNA, even though the peptide remained disordered.
For NMR experiments, transformed Rosetta BL21 cells were grown in 2x minimal M9 media using15NH4Cl (1 g/L) and unlabeled or13C D-glucose (3 g/L) as sole nitrogen and carbon sources, respectively. Protein expression was induced with 0.5 mM isopropyl-β-d-thiogalactopyranoside for 18 h at 25 ∘C. Bacterial cells were suspended in SH3 lysis buffer (20 mM HEPES pH 7.4, 300 mM NaCl). Cell suspensions were lysed in a microfluidizer. Lysates were clarified by centrifugation at 19,500 rpm in a JA-20 rotor for 45 min at 4 ∘C and loaded on Ni-NTA resin pre-equilibrated with lysis buffer. SH3-CTD constructs were eluted in 250 mM imidazole lysis buffer. The purification was followed by size exclusion chromatography in SH3 lysis buffer. SH3-CTD constructs were then buffer exchanged into 20 mM Hepes pH 7.4, 300 mM NaCl, and 10% D2O and concentrated to a desirable concentration (0.7 mM) for NMR spectroscopy using a 5 kDa molecular weight cutoff centrifugal concentrator (Millipore). All NMR experiments were performed on an Agilent 600 MHz (14.1 T) DD2 NMR spectrometer equipped with a room temperature HCN z-axis gradient probe. NMR data were processed with NMRPipe/NMRDraw57 and analyzed with CCPN Analysis58. Backbone 13Cα,13Cβ,13C’,15N, and 1HN resonance assignments of SH3-CTD were obtained from standard gradient-selected triple-resonance HNCACB, HN(CO)CACB, HNCO, HN(CA)CO59, HCCH-TOCSY60, and NOESY HSQC61 experiments at 22 ∘C. Assignment data were collected with a random nonuniform sampling (NUS) scheme and reconstruction of NUS spectra was performed using Sparse Multidimensional Iterative Lineshape-Enhanced (SMILE) program62. The CTD and the linker between CTD and SH3 domains assignment were isolated by excluding the SH3 domain assigned residues using an assignment reference from the BMRB (code = 18054)40.
SAXS data collection and processing
SAXS data were collected on a Xenocs BioXolver configured for SAXS/WAXS/GISAXS with a Genix 3D Cu HFVL source and a DECTRIS EIGER 1M detector. Purified full-length Dnase1L3 and ΔCTD were concentrated to 4 mg/ml. The enzyme was monodisperse, as determined by size exclusion chromatography, and was in 400 mM NaCl, 20 mM HEPES, 1 mM CaCl2. The filter concentrator flowthrough buffer was used for SAXS buffer subtraction. The proteins were irradiated five times for 300 s each and averaged before buffer subtraction with BioXtas-RAW. The Indirect Fourier Transform was performed using GNOM63 with the limits based on the GNOM analysis of IFT quality. Bead models were constructed using Dammif/n64. MultiFoXS42 was used in conjunction with the pairwise distribution function to refine the crystal structure of Dnase1L3 ΔCTD grafted with the CTD structure from NMR using comparative structure modeling65.
Statistics and reproducibility
Prism 5.0 (GraphPad, La Jolla, CA, USA) or Excel were used for statistical analysis. Data are represented as mean ± SEM or standard deviation as indicated. The EC50 was calculated by logistic regression. Statistical significance was determined by one-way ANOVA or repeated measures ANOVA; p < 0.05 was statistically significant. Graphs were generated in R ggplot2, Excel and Photoshop.
All molecular dynamics input files were generated with CHARMM-GUI for NAMD with the CHARMM36m force field66. The high-resolution crystal structure of Dnase1L3 was used throughout for MD. The total charge of the system was neutralized by randomly substituting water molecules with Mg2+ and Cl− to obtain neutrality with 0.15 M salt concentration. The TIP3 model for water was used throughout. A switching function was applied to the van der Waal’s potential energy from 10 to 12 Å to calculate nonbonded interactions. The Particle Mesh Ewald (PME) algorithm was used to calculate electrostatic interactions. Equilibration runs used the NVT ensemble at 300 K. Energy minimization was performed for 10,000 steps to avoid any bad contacts generated while solvating the system. C-α restraints were generated from VMD. The simulations were analyzed using VMD. Production runs were performed on the HPCC Nocona cluster at Texas Tech University.
The structure for the CTD, as determined by solution-state NMR, was isolated from the SH3 domain along with DNA based on the 1BNA 12-mer of dsDNA. The solution box was built with the CTD and a 12-mer (5’-CGCGAATTCGCG-3’) dsDNA molecule that was placed ~30 Å and 50 Å from the CTD. The distances were required to minimize the bias of the simulation. The simulation was run for a total of 1 μs with neutralizing KCl. VMD and the Bio3D library for R67 were used to analyze the trajectories.
DNA was added to the system based on a superposition of the 1DNK crystal structure of Dnase1 with DNA. Chain C of the Dnase1L3 high-resolution crystal structure was the starting structure for the MD runs with DNA. A 75Å x 75Å x 75Å water box with 150 mM MgCl2 (charge neutralized) was sized to the input Dnase1L3 molecule and was built using periodic boundary conditions. The simulated system contained ~35,000 solvent atoms, ~69 counter ions, and 4336 protein atoms. The crystallographic Mg2+ were retained. Three simulations were run for 1 μs each, starting from separate equilibration runs.
The R206C mutation was generated with PyRosetta using Chain A of the crystal structure and Rosetta relaxed 3 times with the lowest energy score chosen prior to MD simulation. The solution box was built with 150 mM MgCl2 and charge neutralization. The crystallographic Mg2+ were retained. A total of three production runs of 500 ns for Arg-206 and Cys-206 were generated for analysis. The last 100 ns (400–500 ns) were sufficient to reach a stable snapshot of each repeat. Ten frames, each 10 ns apart from the last 100 ns of the MD simulations were analyzed for free energy of folding using Rosetta, with the lowest energy score, REF-2015, after three full Rosetta relax cycles compared between Arg-206 and Cys-206. A one-way ANOVA, (alpha = 0.05) was performed to highlight the difference between the single comparison of significance, Arg-206 or Cys-206 on free energy of folding. The average computed free energy of folding from 10 timestamps across 3 simulations, for both WT and R206C Dnase1L3, was measured as −664.99 (n = 30) for WT (Arg206) and −650.96 (n = 30) for R206C. The standard deviation for WT free energy of folding was 13.633, and 15.642 for R206C. The difference in folding was determined to be 14.03 ± 3.79 (s.d.), p = 0.000476.
Iso-electric point calculations
Iso-electric calculations were computed using H++68,69,70. For the actin-binding loop following the actin-binding helix in Dnase124, and the homologous loop in Dnase1L3, the regions of interest were isolated in Pymol and processed using the H++ web server at a pH of 7, with a salinity of 0.15 M, internal dielectric of 0.1, and external dielectric of 80.
pKa of active site residues
The computational determination of acid dissociation constant was done with a local installation of DelPhi-Pka38. All four Dnase1L3 molecules in the asymmetric unit and each available Dnase1 structure10,24,33,34,35,36,37, were processed for pKa calculation in DelPhi-Pka after being cleaned of any hetero-atoms.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.