Targeting NETs using dual-active DNase1 variants

Abstract

Background: Neutrophil Extracellular Traps (NETs) are key mediators of immunothrombotic mechanisms and defective clearance of NETs from the circulation underlies an array of thrombotic, inflammatory, infectious, and autoimmune diseases. Efficient NET degradation depends on the combined activity of two distinct DNases, DNase1 and DNase1-like 3 (DNase1L3) that preferentially digest double-stranded DNA (dsDNA) and chromatin, respectively.

Methods: Here, we engineered a dual-active DNase with combined DNase1 and DNase1L3 activities and characterized the enzyme for its NET degrading potential in vitro. Furthermore, we produced a mouse model with transgenic expression of the dual-active DNase and analyzed body fluids of these animals for DNase1 and DNase 1L3 activities. We systematically substituted 20 amino acid stretches in DNase1 that were not conserved among DNase1 and DNase1L3 with homologous DNase1L3 sequences.

Results: We found that the ability of DNase1L3 to degrade chromatin is embedded into three discrete areas of the enzyme's core body, not the C-terminal domain as suggested by the state-of-the-art. Further, combined transfer of the aforementioned areas of DNase1L3 to DNase1 generated a dual-active DNase1 enzyme with additional chromatin degrading activity. The dual-active DNase1 mutant was superior to native DNase1 and DNase1L3 in degrading dsDNA and chromatin, respectively. Transgenic expression of the dual-active DNase1 mutant in hepatocytes of mice lacking endogenous DNases revealed that the engineered enzyme was stable in the circulation, released into serum and filtered to the bile but not into the urine.

Conclusion: Therefore, the dual-active DNase1 mutant is a promising tool for neutralization of DNA and NETs with potential therapeutic applications for interference with thromboinflammatory disease states.

Keywords: DNase1; DNase1-like 3; NET degradation; NETosis; neutrophil extracellular traps (NETs); protein engineering; recombinant proteins; thromboinflammation.

Figure 1

Strategy for engineering dual-active DNase1/DNase1L3 mutants. (A) Scheme for the generation of dual-active DNase mutants based on addition of DNase1L3 activity to DNase1. DNase1 (I, orange) and DNase1L3 (II, blue) differ in their substrate specificities: DNase1 degrades dsDNA more efficiently than chromatin, and vice versa chromatin DNA is a preferred substrate for DNase1L3 as compared to dsDNA. The designed dual-active DNase1 variant (DNase1v, III, purple) contains sequences of DNase1 and DNase1L3 and efficiently digests both dsDNA, and chromatin. (B) Purified recombinant DNase1 (panel I) and DNase1L3 (panel II; 50 - 0.005 ng) were either incubated with dsDNA or chromatin isolated from salmon testes or from HEK293 cell nuclei, respectively. Diameters of the dark circles in zymography SRED assays present dsDNA degrading DNase1 activity (upper gel; dsDNA/D1activity). Agarose gel electrophoresis of digested chromatin shows chromatin degrading activity of DNase1L3. High-molecular weight chromatin complexes are cleaved into smaller fragments (lower gel; chromatin/D1L3 activity). (C) Co-incubation of DNase1L3 and DNase1 shows synergistic DNases activities. dsDNA and chromatin were incubated with increasing levels of DNase1 alone (panel III) or together with DNase1L3 (5 ng, panel IV) that alone (panel V) was not sufficient to degrade chromatin completely. Panel VI shows buffer treated dsDNA and chromatin. (D) Amino acid sequences of DNase1 and mutated regions of the 20 engineered DNase1 variants (DNase1^A – DNase1^T, which are detailed in Supplementary Table 1 ). Numbering relates to native DNase1 sequence. Residues conserved between DNase1 and DNase1L3 are depicted in grey and the signal sequence of DNase1 is highlighted in green. Non-conserved residues are highlighted in orange in the DNase1 backbone. DNase1L3 sequence stretches that were swapped into the DNase1 backbone are shown in blue. “cont.” indicates that DNase1^L and DNase1^R sequences continue from the lines above. D1 – DNase1, D1L3 – DNase1L3.

Figure 2

Screening of dual-active DNase1 variants for DNase1 and DNase1L3 activity. (A) dsDNA (upper gel; D1 activity) or chromatin (lower gel; D1L3 activity) was incubated with supernatants of HEK293 cell-expressed DNase1 variants, DNase1 or DNase1L3. DNase1 and DNase1L3 activities of the various DNases were analyzed as in Figure 1B above. (B) Densitometric scans of the digested dsDNA quantified DNase1 activity. Enzymatic activity is given relative to DNase1 (D1) levels set to 100%. (C) Residual chromatin signals as a measure for DNase1L3 activity is blotted relative to DNase1L3 (D1L3) activity (100%). (D) Sum of DNase1 and DNase1L3 activities derived from (B) and (C). p-value by paired one-way ANOVA with Dunnett’s multiple comparisons test. Data are mean ± SEM, n = 3. D1 – DNase1, D1L3 – DNase1L3.

Figure 3

Characterization of dual-active variant DNase1^G,K,O. (A) Amino acid sequence of the dual-active DNase1 variant DNase1^G,K,O that combines the identified DNase1L3-derived mutations of DNase1^G (five residues; positions 86, 88-89, 91-92), DNase1^K (two residues; positions 136-137) and DNase1^O (eight residues; positions 199-202, 204-205, 209, 211). Signal peptide sequence is shown in green. Non-mutated DNase1 sequence is in orange and swapped DNase1L3 derived residues in blue. (B) Dose dependency of dsDNA (upper gel; D1 activity) and chromatin (lower gel; D1L3 activity) degrading activity of DNase1^G,K,O. Serial 1:10 dilution of supernatants of transfected HEK293 cells were tested. (C) dsDNA (upper gel; D1 activity) and chromatin (lower gel; D1L3 activity) degrading activity of recombinant DNase1^G, DNase1^K, DNase1^O, and DNase1^G,K,O variants, and native DNase1, and DNase1L3. (D) Quantification of DNase1, (E) DNase1L3, and (F) combined DNases activities of various DNases from densitometric scans as in (C). Data is given relative to native enzyme activity (100%) in (D) and (E). p-value by paired one-way ANOVA with Dunnett’s multiple comparisons test. Columns represent mean ± SEM, n = 3. D1 – DNase1, D1L3 – DNase1L3.

Figure 4

DNase activities in serum, bile, and urine of DNase1^G,K,O transgenic mice. We produced mouse lines with transgenic expression of DNase1^G,K,O, DNase1 and DNase1L3 in hepatocytes on a Dnase1^-/-Dnase1l3^-/- deficient background. DNase1 (upper gel; D1 activity) and DNase1L3 (lower gel; D1L3 activity) activities in serum (A), bile (B), and urine (C) of DNase1^G,K,O transgenic mice were measured by dsDNA and chromatin degradation, respectively. Representative gels of n = 3. DNase1 (D), DNase1L3 (E) and summarized DNase1 and DNase1L3 (F) activities in serum, bile, and urine of DNase1^G,K,O (purple) DNase1 (orange) and DNase1L3 (blue) transgenic mice. Activities are normalized to native DNase levels (100%). Data represent mean ± SEM, p-value by two-way ANOVA with Dunnett’s multiple comparisons test. Each data point represents an individual animal, DNase1^G,K,O (n = 4), DNase1 (n = 5), DNase1L3 (n = 3). D1 – DNase1, D1L3 – DNase1L3.

Figure 5

Dual-active DNase1^G,K,O efficiently degrades NETs. Human neutrophils were PMA-stimulated to produce NETs. Formed NETs were treated with supernatants of DNase1^G,K,O, DNase1, DNase1L3, or of mock vector transfected HEK293 cells. (A) Kinetics of DNase1^G,K,O, DNase1 and DNase1L3-mediated NET degradation. Data represent mean ± SEM. (B) Residual NET fibers upon DNase1^G,K,O, DNase1, DNase1L3 incubation (n = 32, each) were quantified by Sytox Green fluorescence at 90 min. Recombinant human DNase1 (1 U/ml, Pulmozyme; n = 16) and baseline fluorescence of unstimulated neutrophils (n = 20) served as controls. Data represent mean ± SEM, p-value, Kruskal-Wallis with Dunn’s multiple comparisons test. (C) Representative fluorescence images of NETs after 90 min DNase treatment. White arrowheads indicate undigested NET remnants. A representative image of n = 3 is shown. Scale bar 100 µm.