ANIMAL MODEL FOR DRUG DEVELOPMENT

Abstract

The present invention relates to a non-human mammalian animal which has been modified to have in the blood, plasma and/or serum (a) an increased number of leukocytes and/or neutrophils, and (b) a reduced activity of the DNase 1 and/or DNase 1-like 3 enzymes. The non-human mammalian animal is particularly suitable for studying inflammation and/or a disease associated with inflammation. In a further aspect, the invention relates to the use of the non-human mammalian animal as a model for identifying therapeutic or diagnostic targets of inflammation and/or a disease associated with inflammation. In a still further aspect, the invention relates the use of the non-human mammalian animal as a model for drug candidate testing. In addition, a method for testing an anti-inflammatory drug candidate against extracellular DNA is provided. Finally, a method for testing an anti-inflammatory drug candidate for modifying the formation or degradation of neutrophil extracellular traps is provided. In still another aspect, the present invention relates to a non-human mammalian animal, which has been modified to have an increased number of neutrophils in blood.

Claims

1-36. (canceled)

37. A method for making a non-human animal model of inflammatory disease, comprising: expressing in a non-human animal a heterologous G-CSF polynucleotide effective to induce neutrophilia in said animal, and providing at least one additional pro-inflammatory stimulus in said animal selected from: (1) a deficiency of one or more of DNase1 enzyme activity and DNase1L3 enzyme activity; (2) a genetic background or modification associated with an inflammatory disease; and (3) administration of a low dose of lipopolysaccharide (LPS).

38. The method of claim 37, wherein the non-human animal is a rodent.

39. The method of claim 38, wherein the non-human animal model is a rat, mouse, hamster, rabbit, or guinea pig.

40. The method of claim 37, wherein the non-human animal is a mouse.

41. The method of claim 37, wherein the heterologous G-CSF is expressed in the liver of said animal.

42. The method of claim 41, wherein the heterologous G-CSF is expressed from an injected plasmid.

43. The method of claim 37, comprising injecting said animal with a low dose of lipopolysaccharide to induce an inflammatory phenotype.

44. The method of claim 43, wherein the non-human animal exhibits disseminated intravascular coagulation (DIC).

45. The method of claim 37, wherein the non-human animal has a deficiency of one or more of DNase1 or DNase1L3 enzyme activity.

46. The method of claim 45, wherein the non-human animal exhibits intravascular accumulation of neutrophil extracellular traps (NETs).

47. The method of claim 46, wherein the non-human animal exhibits intravascular DNA clots.

48. The method of claim 46, wherein the non-human animal has a deletion or inactivation of DNase1 and/or DNase1L3 genes.

49. The method of claim 48, wherein the non-human animal has a deletion or inactivation of DNase1 and DNase1L3 genes.

50. The method of claim 37, wherein the non-human animal model has a genetic modification associated with an inflammatory disease.

51. The method of claim 50, wherein the inflammatory disease is systemic lupus erythematosus (SLE).

52. The method of claim 50, wherein the non-human animal model develops arthritis in response to G-CSF expression.

53. A method for drug target identification or validation for an inflammatory disease, comprising: providing the non-human animal model of claim 37, modifying the activity or expression of one or more target genes in cells of said animal, and determining whether an inflammatory phenotype is reduced.

54. The method of claim 53, wherein the inflammatory phenotype is evaluated in tissues isolated from said animal.

55. The method of claim 53, wherein the inflammatory phenotype is selected from one or more of accumulation of NETs, intravascular DNA clots, disseminated intravascular coagulation (DIC), and arthritis.

56. A method for selecting a pharmaceutical composition for treating an inflammatory disease, the method comprising: providing the non-human animal model of claim 37; administering a candidate drug for the inflammatory disease to said animal or tissue isolated therefrom; determining whether the candidate drug reduces an inflammatory phenotype of said animal, and selecting a candidate drug that reduces the inflammatory phenotype for treatment of inflammatory disease.

57. The method of claim 56, wherein the inflammatory phenotype is selected from one or more of accumulation of NETs, intravascular DNA clots, disseminated intravascular coagulation (DIC), and arthritis.

58. The method of claim 56, wherein the candidate drug is a small molecule drug candidate.

59. The method of claim 56, wherein the candidate drug is a DNase enzyme.

60. The method of claim 56, wherein the selected candidate is formulated for administration to a human patient.

Description

DESCRIPTION OF THE FIGURES

[0117] FIG. 1 shows that DNase1 and DNase1l3 in murine serum degrade NETs. (a) DNA-staining of activated neutrophils shows NETs (arrows) in samples incubated with buffer, whereas incubation with 10% WT serum for 3 hours degrades NETs. (b) DPZ and (c) SRED detects the activity of DNase1 and DNase1l3 in WT serum, DNase1l3 in DNase1.sup.−/− serum, DNase1 in DNase113/serum. No DNase activity is detected in DKO serum. (d) Quantification of DNase activity shown in panel c over time (n=5; §: p<0.001 vs. all other groups). (e) NETs are degraded by serum from WT, DNase1.sup.−/−, DNase1l3.sup.−/− mice, but stable in serum from DKO mice (Scale bar: 50 μm). (f) Quantification of cell-free DNA fragments in the supernatant of NETs incubated with sera from indicated strains. Sera from DKO mice do not generate cell-free DNA. (n=4; §: p<0.001 vs. all other groups). (g) Supplementation of DKO serum with recombinant DNase1 or DNase1l3 restores NET-degradation (Scale bar: 50 μm) and (h) the generation of cell-free DNA to WT levels (§: p<0.001 vs. all other groups).

[0118] FIG. 2 shows that extracellular DNases are required for survival of experimental neutrophilia. (a) Four-week old WT mice were injected with a CSF3-expression plasmid (pCSF3) or with a control plasmid (pCtrl). (a) G-CSF levels are stably increased 3 days after the injection (§: p<0.001 vs. base line, BL). (b) CD11b/Ly6G-double positive cells in blood steadily increase after G-CSF expression, indicating neutrophilia (#: p<0.01, §: p<0.001 vs. BL). (c) Mice treated with pCSF3 and pCtrl show a similar growth in body weight (n/s, not significant, pCSF3 vs. pCtrl). (d) Mice injected with pCSF3 develop splenomegaly after 4 weeks (scale bar: 1 cm; §: p<0.001 vs. BL). (e) Mice treated with pCSF3 and pCtrl show similar plasma levels of LDH, ALT, AST, BUN, and creatinine. (f) Expression of pCSF3 is lethal in DKO mice, but not in WT, DNase1.sup.−/−, DNase1l3.sup.−/− mice. DKO survive the injection of pCtrl. (§: p<0.001 vs. all other groups). (g) Expression of pCSF3 causes in DKO mice a sudden and rapidly progressing hypothermia (*: p<0.05, #: p<0.01 vs. exitus). (h) Photograph of glass capillaries filled with urine illustrates hematuria in DKO mice expressing pCSF3. Genetic reconstitution by infection of an expression plasmid containing DNase1 (pDNase1) or DNase1l3 (pDNase1l3) restores the activity of DNase1 and DNase1l3 in serum as detected by (i) DPZ, (j) SRED, and (k) in vitro NET-degradation (Scale bar: 50 m). (l) Survival of DKO mice injected with a mixture of pCSF3 and pCtrl, pDNase1, or pDNase1l3. Co-expression of pDNase1 or pDNase1l3 prevents mortality CSF3-induced mortality (§: p<0.001 vs. all other groups).

[0119] FIG. 3 shows that extracellular DNases prevent intravascular DNA-clots and multi-organ damage in experimental neutrophila. (a) H&E stainings of lung sections of DKO mice expressing pCSF3 along with pCtrl, pDNase1, or pDNase1l3. Scale bar: 500 μm in overview, 50 μm in magnification. Co-expression of pCSF3 and pDNase1 or pDNase1l3 prevents the formation intravascular hematoxylin-rich clots. (b) Quantification hematoxylin-rich clots in lung, kidney, and liver (§: p<0.001 vs. all other groups). (c) Intravascular hematoxylin-rich clots stain for DNA (DAPI) and chromatin (Scale bar: 200 μm). Immunostaining for (d) chromatin and myeloperoxidase (MPO) shows abundant NETs as well as (e) traces of von Willebrand factor (vWF) and fibrin (Scale bar: 50 μm). Plasma analysis showed elevated levels of (e) LDH, (f) AST and ALT, (g) BUN and creatinine, and (h) anemia in DKO mice expressing pCSF3 along with pCtrl, but not in untreated WT and DKO mice (BL) or DKO mice expressing pCSF3 along with pDNase1 or pDNase1l3 (p<0.001 vs. all other groups).

[0120] FIG. 4 shows that mice lacking DNase1 and DNase1l3 and patients with thrombotic microangiopathies degrade NETs inefficiently. (a-g) DNase activity in human and murine plasma quantified by SRED. (a) Supplementation of plasma from normal healthy donors (NHD) with polyclonal antibodies against human DNase1 (α-hDNase1), but not with IgG, blocks the DNA degrading activity in a concentration-dependent manner (*: p<0.05 vs. IgG). (b) Plasma supplemented with 200 μg/ml α-hDNase1 and heparin, an inhibitor of DNase113, blocks residual DNase activity in plasma supplemented with 200 μg/ml α-hDNase1 (*: p<0.05 vs. NHD). (c) Plasma from patients with acute TMA supplemented with 500 μg/ml heparin shows a reduced activity of DNase1 (#: p<0.01 vs. NHD). (d) Plasma supplemented with 200 pig/ml α-hDNase1 shows a similar DNase1l3 activity in TMA patients and NHD (n/s, not significant vs. NHD). (e-f) Comparison of DNase activity in serial dilutions of plasma from mice and NHD. (e) Total DNase activity of WT mice is approximately 10-times higher than in NHD. (f) DNase1 activity in plasma from DNase1l3.sup.−/− mice is approximately 10-fold higher than human plasma supplemented with 500 μg/ml heparin. (g) DNase1l3 activity in plasma from DNase1.sup.−/− mice is approximately 10-fold higher than human plasma supplemented with 200 μg/ml α-hDNase1. (h-i) NET-degradation. (h) Activated neutrophils incubated with plasma from NHD alone or supplemented with 200 μg/ml α-hDNase1, plasma from TMA patients, and plasma from DKO mice (Scale bars: 200 μm). (i) Cell-free DNA in supernatants of activated neutrophils incubated with plasma from indicated sources (§: p<0.001 vs. NHD). NET-degradation by human plasma is dependent on DNase1. DNase1l3 in human plasma is not sufficient to degrade NETs. Plasma from patients with thrombotic microangiopathies and DKO mice cannot degrade NETs.

[0121] FIG. 5 shows that DNase1 and DNase1L3 protect against host injury in septicemia. WT mice (N=5) and mice with a combined deficiency in DNase1 and DNase1L3 (D1/D1l3.sup.−/−) expressing Dnase1 (D1, N=7), Dnase1l3 (D1l3, N=8), or a control plasmid (Ctrl, N=11) were treated with lipopolysaccharide and heat-killed E. coli to induce septicemia. (A) Survival time of septic mice. (B) Concentration of hemoglobin in blood. (C) Representative photographs of plasma and urine. (D) LDH concentration in plasma. (E) Quantification of schistocytes in blood smears per field of view (FOV). (F) Quantification of occluded blood vessels in lungs per FOV. (G) Representative hematoxylin and eosin stainings (H & E) of lungs of WT mice and D1/D13.sup.−/− mice expressing D1 or D1l3. (H) Representative H & E staining of lungs of D/D1l3.sup.−/− mice expressing the control plasmid. Arrowheads point to occluded blood vessels. Scale bars: 500 μm. (I) Representative H & E staining of partially and fully occluded blood vessel. Arrows point to NETs covering the intercellular space. Inserts are overviews. Scale bars: 50 μm. Statistics: (A) log-rank test; ** P<0.01 vs. all other groups, (B to F) one-way ANOVA followed by Bonferroni's multiple comparisons post hoc test; *** P<0.001, ** P<0.01.

[0122] FIG. 6 shows that neutrophilic WT mice accumulate NETs in blood and prone to septicemia. (A) NET-like structures (arrows) in DNA-stainings of blood smears from WT mice expressing CSF3 for indicated times or Ctrl for 14 days (N=5). Scale bar: 50 μm. (B) Susceptibility of WT mice expressing Ctrl or CFS3 for 2 weeks to a single injection of a low dose of LPS. ALT, AST, and LDF levels in plasma 4 hours post LPS injection. Statistics: (A, C) one-way ANOVA followed by Bonferroni's multiple comparisons post test, (B) Student's t-test; * P<0.01, *** P<0.001 vs. (A) BL.

[0123] FIG. 7 shows the results of the histological analysis revealing a robust and systemic deposition of fibrin in vital organs.

[0124] FIG. 8 shows the effect of hepatic expression of CSF3 in lupus-prone MRLlpr mice. (A) Analysis of proteinuria indicated that CSF3-expression induced an early and severe disease onset of SLE. (B) Continuous monitoring of the mice after CSF3-gene delivery indicated rheumatoid arthritis (RA)-like symptoms, namely swollen paws within 4-5 weeks. (C) In depth analysis of the bone structure by micro-CT showed that CSF3-expression precipitated robust bone degeneration, thus confirming the development of RA.

EXAMPLES

[0125] The experiments referred to below were approved by the Hamburg State Administration for animal research. All environmental parameters within the animal facility were in compliance with the German Law for the Care and Use of Laboratory animals.

Example 1: Analysis of DNase Activity in DNase-Deficient Mice

[0126] We used DNase1- and DNase1l3-deficient mice with a C57B16-genetic background (Napirei et al. (2000), see above; Mizuta et al. (2013) PloS one, 8:e80223) to generate mice lacking both DNases. To characterize the DNase activity in these animals, we applied the denaturing PAGE zymography (DPZ) method, as previously described with modifications (Napirei, et al. (2009) FEBS J 276 (4): 1059-73). In brief, SDS-PAGE gels were prepared with 4% (v/v) stacking gels without DNA and 10% (v/v) resolving gels containing 200 μg/mL of salmon testes DNA. For DNase1 detection, 0.5 μl of murine serum were mixed with 14.5 μL of water and 5 μl SDS gel-loading buffer (BioRad), boiled for 5 min, and loaded onto the gels. The PageRuler Prestained Protein Ladder (MBI Fermentas, St. Leon-Rot, Germany) was used as molecular marker. Electrophoresis was carried out at 120 V using Tris/glycine electrophoresis buffer (25 mM Tris, 192 mM glycine, 0.1% (w/v) SDS, pH 8.7). After electrophoresis, proteins were refolded by incubating the gels overnight at 37° C. in a solution containing 5% (w/v) milk powder, 10 mM Tris/HCl pH 7.8, 3 mM CaCl.sub.2, 3 mM MgCl.sub.2, 100 U/mL penicillin and 100 μg/mL streptomycin. Next, the gels were transferred to a buffer containing 10 mM Tris/HCl pH 7.8, 3 mM CaCl.sub.2, 3 mM MgCl.sub.2, 100 U/mL penicillin, 100 μg/mL streptomycin and 1×SYBR Safe. Gels were incubated for 24 hours at 37° C. The fluorescence was recorded by a fluorescence scanner.

[0127] For DNase1l3 detection, 2 μl of serum were mixed with 12 μL of water, 1 μl of beta-mercaptoethanol (BME) and 5 μl SDS gel-loading buffer, boiled for 5 min, and loaded onto the gels. Electrophoresis was carried out as before. SDS was removed by washing the gels with 10 mM Tris/HCl pH 7.8 for 30 min at 50° C. twice, and the proteins were refolded by incubating the gels 48 hours at 37° C. in a solution containing 10 mM Tris/HCl pH 7.8, 1 mM BME, 100 U/mL penicillin and 100 μg/mL streptomycin. Next, the gels were incubated for 48 hours at 37° C. in a buffer containing 10 mM Tris/HCl pH 7.8, 3 mM CaCl.sub.2, 3 mM MnCl.sub.2, 1 mM BME, 100 U/mL penicillin, 100 μg/mL streptomycin. SYBR Safe was then added to a concentration of 1× and fluorescence was recorded by a fluorescence scanner.

[0128] To measure total DNase activity, we dissolved 55 μg/ml DNA from salmon testes (Sigma-Aldrich) in buffer with 20 mM MES pH 6.5, 10 mM MnCl.sub.2, 2 mM CaCl.sub.2, and 2×SYBR Safe (Life Technologies) in distilled water. The DNA solution was heated at 50° C. for 10 min and mixed with an equal volume of 2% agarose GP-36 (Nacalai Tesque). The mixture was poured into plastic trays and stored at room temperature until solidification. Two μl of sample (e.g. murine serum) were applied to wells of 1.0 mm radius. Gels were incubated for up to 24 hours at 37° C. in a humid chamber, the fluorescence of the gels was recorded with a fluorescence scanner (Molecular Imager FX, Bio-Rad). Image J (NIH) was used for the quantification of the area and intensity of the circles reflecting DNase activity. Absolute units were obtained upon extrapolation against a standard curve obtained by diluting recombinant human DNase1 (Pulmozyme, Roche) in HBSS+ containing 0.1% BSA.

[0129] Results:

[0130] Murine and human serum degrades the DNA-backbone of NETs (FIG. 1a). We therefore speculated that circulating DNases disarm NETs during inflammation. We fractionated murine serum and detected two DNA-degrading enzymes corresponding to DNase1 and DNase1l3 (FIG. 1b) (Napirei, et al. (2000) Nat. Genet. 25 (2): 177-81. Sisirak, et al. (2016) Cell 166 (1): 88-101). In mice, DNase1 and DNase1l3 are present in serum at steady-state-conditions, but independently expressed by non-hematopoietic tissues (Napirei, et al. (2004) Biochem. J. 380 (Pt 3): 929-37) or macrophages and dendritic cells (Sisirak, et al. (2016) Cell 166 (1): 88-101), respectively. We generated mice lacking both DNases (DKO) and detected DNA-degrading activity in WT, DNase1.sup.−/−, and DNase1l3.sup.−/− sera, but not in DKO sera (FIG. 1c, d).

Example 2: In Vitro NET Degradation Assay

[0131] NET-degradation was analyzed according to the protocol of Hakkim et al (Hakkim, et al. (2010) Proceedings of the National Academy of Sciences of the United States of America 107 (21): 9813-8) with modifications. Purified neutrophils in serum-free Dulbecco's Modified Eagle's Medium (DMEM; Life Technologies) were seeded to sterile 96-well plates (Falcon, BD Technologies, Heidelberg, Germany) at a concentration of 5×10.sup.4 cells per well. To induce NET-formation, neutrophils were activated with 100 nM phorbol 12-myristate 13-acetate (PMA; Sigma-Aldrich) for 4 h at 37° C. with 5% CO.sub.2 and humidity. We added adding phosphate buffered saline (PBS, naïve NETs) or 2% paraformaldehyde (PFA; Sigma Aldrich, fixed NETs) in PBS to the NETs and stored (PFA; Sigma Aldrich, fixed NETs) and stored the 96-well plates overnight at 4° C. We considered PBS- and PFA-treated NETs as naïve NETs and fixed-NETs, respectively. Thereafter, NETs were washed with PBS, treated for 5 minutes with PBS containing 0.5% Triton X-100, and washed with PBS again. NETs were incubated with 10% murine serum or plasma, human citrated plasma, or culture supernatants of HEK cells expressing murine DNase1 or DNase1l3. All samples were diluted 10-fold in HBSS with divalent cations (HBSS+; Life Technologies) and supplemented with 10 μM PPACK (Santa Cruz) in the case of citrated plasma. NETs were allowed to be degraded for different time at 37° C. with 5% CO.sub.2 and humidity as indicated in the figures or figure legends. We collected supernatant and stopped NET-degradation by adding 2% PFA in PBS for 1 hours at room temperature. For analysis of NET-degradation, (non-degraded) NETs were then labeled fluorescently by adding 2 μM of the fluorescent DNA dye SytoxGreen (Life Technologies. Images of fluorescently stained nuclei and NETs were acquired with an inverted fluorescence microscope (Zeiss Axiovert 200M, Oberkochen, Germany). In addition, we quantified the fragments of NET-DNA in the supernatants (cell-free DNA), which are generated during NET-degradation. In brief, culture supernatants were diluted with DNAq buffer (0.1% BSA and 2 mM EDTA in PBS) and supplemented with 1 μM SytoxGreen to label DNA. Fluorescence was recorded using a MTP reader (Excitation: 485 nm; Emission: 535 nm). DNA concentrations were calculated based on a standard curve of known concentrations of DNA (lambda DNA, ThermoScientific). In parallel, auto-fluorescence and the endogenous DNA-concentrations in human and murine samples and culture supernatant, which were not exposed to NETs, were considered as background.

[0132] In selected experiments human plasma was supplemented with rabbit-polyclonal antibodies against recombinant human DNase1 (Pulmozyme, Roche), generated by Exbio (Praha, Czech Republic), and with heparin to block DNase1 and DNase113 activity, respectively.

[0133] In addition, plasmids containing the cDNA for DNase1 or DNase1l3, described elsewhere (Napirei, et al. (2009) FEBS J 276 (4): 1059-73), were transfected into Human Embryonic Kidney (HEK) cells with Lipofectamine 3000 (Thermo Scientific, Darmstadt, Germany) in serum-free conditions with DMEM media (Gibco) supplemented with 10% KnockOut serum replacement (Gibco). After 72 hours, supernatants were collected, centrifuged at 500×g for 10 minutes, sterile filtered, and concentrated by ultracentrifugation with 3K columns (Amicon Ultra, Millipore). Supernatants were used to study NET-degradation of recombinant murine DNase1 and/or DNase1l3 in vitro.

[0134] Results:

[0135] Analysis of NET-degradation in vitro showed that NETs are stable in DKO-sera, whereas sera from WT, DNase1.sup.−/−, and DNase1l3.sup.−/− mice degraded NETs (FIG. 1e, f). Supplementation of DKO-serum with recombinant murine DNase1 or DNase1l3 restored the NET-degrading activity (FIG. 1g, h). In vitro, DNase1 preferentially cleaves protein-free DNA, while DNase1l3 targets DNA associated with proteins, such as polynucleosomes (Napirei, et al. (2005) Biochem. J. 389 (Pt 2): 355-64). To test whether DNase1 and DNase1l3 degrade NETs by distinct mechanisms, we cross-linked proteins in NETs using fixative. Fixed NETs were efficiently degraded by DNase1l3.sup.−/− sera, but were resistant to degradation by DNase.sup.−/− sera (not shown). In line with these results, exposure of NETs to recombinant DNase1 or DNase1l3 showed that both DNases degrade naïve NETs (not shown), whereas fixed NETs are resistant to DNase1l3 activity (not shown). Collectively, these in vitro data identify extracellular DNase1 and DNase1l3 in murine serum as redundant NET-degrading enzymes.

Example 3: In Vivo NET Degradation

[0136] For in vivo expression, we used the pLIVE plasmid (Mirus Bio), a vector, which allows a strong and maintained hepatocyte-specific expression of the protein of interest upon delivery. PCR of DNase1 cDNA (Genbank Accession Number NM010061) was performed using the pair of primers DNase1-F 5′-GTCGACATGCGGTACACAGG-3′ (SEQ ID NO:2) and DNase1-R 5′-CTCGAGTCAGATTTTTCTGAGTGTCA-3′ (SEQ ID NO:3) containing SalI and XhoI restriction sites. PCR of DNase1l3 cDNA (Genbank Accession Number AF047355) was performed using the pair of primers DNase1l3-F 5′-GAAGTCCCAGGAATTCAAAGATGT-3′ (SEQ ID NO:4) and DNase1l3-R 5′-GCGTGATACCCGGGAGCGATTG-3′ (SEQ ID NO:6) containing BamHI and SacI restriction sites. Both cDNAs were cloned using the T4 ligase (New England Biolabs, Frankfurt, Germany) into the MCS site of the vector pLIVE previously digested with the appropriate enzymes. The DNase1l3 pLIVE vector was subjected to site-directed mutagenesis with the pair of primers mutD1l3-F 5′-AGTCGACTCCCGGCCACCATGTCCCTGCA-3′ (SEQ ID NO:7) and mutD1l3-R 5′-TGCAGGGACATGGTGGCCGGGAGTCGACT-3′ (SEQ ID NO:5) in order to match the consensus Kozak sequence. The synthesis of the pLIVE-CSF3 plasmid (pCSF3) was outsourced to Eurofins Genomics. The cDNA of the CSF3 gene (Genbank Accession Number BC120761) was inserted in the MCS between restriction sites SalI and XhoI. Sequence of all the generated vectors was confirmed by double stranded sequencing. As a control we used the parental pLIVE plasmid (pCtrl). Plasmids were amplified and purified from potential contaminations with endotoxin.

[0137] The pLIVE-expression vectors described above were administered to mice via the method of hydrodynamic tail vein injection, as described elsewhere. In brief, 50 μg of plasmid were diluted in 0.9% saline in a volume equivalent to 10% of the body mass of the mouse. Mice were anaesthetized with isoflurane (2.5% of Isoflurane, Abbvie) and the plasmid solution was then injected intravenously over 5 to 8 seconds via the tail vein. In rare cases, mice did not fully recover from the injection within the first 24 hours and were excluded from the study.

[0138] Mice were injected with 50 μg of pCSF3 or CSF3 to induce G-CSF expression and neutrophilia. For co-expression studies, 50 μg pCSF3 was mixed with 50 μg pCtrl or 50 μg pDNase1 or 50 μg pDNase1l3. A solution containing both plasmids was administered via hydrodynamic tail vein injection. Weight and temperature were monitored every 8 hours during the first week after injection, and daily afterwards. Temperature was measured in the perianal area by a contactless infrared-thermometer (Etekcity). Severe hypothermia was defined as decrease in body temperature of >4° C., compared to the body temperature before the plasmid injection. In all cases of hypothermia was accompanied with symptoms of distress. Therefore hypothermic mice were sacrificed and blood, urine, and organs were collected. Non-hypothermic mice did not show any signs of distress and were euthanized at the end of the experiments to collect blood, urine, and organs.

[0139] Results:

[0140] To test the requirement of DNase1 or DNase1l3 for NET-degradation in vivo, we overexpressed murine CSF3 in WT, DNase1.sup.−/−, DNase1l3.sup.−/−, and DKO mice. CSF3 encodes the granulocyte-colony stimulating factor (G-CSF), which mobilizes neutrophils from the bone marrow to circulation (Semerad, et al. (2002) Immunity 17 (4): 413-23) and stimulates a subpopulation of neutrophils to release NETs ex vivo and in vivo (Demers, et al. (2012) Proceedings of the National Academy of Sciences of the United States of America 109 (32): 13076-81). We induced G-CSF expression by injecting a liver-specific expression plasmid containing the CSF3 coding sequence (pCSF3) into 4-week old WT mice. The pLIVE plasmid (Mirus Bio, USA) was used to express G-CSF in mice. The vector enables a long-lasting and hepatocyte-specific expression of proteins. Murine CSF3 (Genbank Accession Number BC120761) was inserted into the MCS between restriction sites SalI and XhoI. Sequence of the vector was confirmed by double-stranded DNA sequencing. As a control, we used the parental pLIVE plasmid without additional inserts. All plasmids were purified using PureLink HiPure Plasmid Maxiprep Kit and potential contaminations of endotoxin were removed using High Capacity Endotoxin Removal Spin Columns (both Thermo Scientific). The pLIVE-plasmid containing CSF3 or empty control plasmids were administered to mice via hydrodynamic tail vein injection. In brief, 50 μg of plasmid were diluted in 0.9% saline in a volume equivalent to 10% of the body mass of the mouse. Mice were anaesthetized with isoflurane and the plasmid solution was then injected intravenously over 5-8 seconds via the tail vein. In rare cases, mice did not fully recover from the injection within the first 24 hours and these animals were excluded from the study. For co-expression studies, 50 μg of the CSF3-plasmid were mixed with 50 μg of the empty control plasmid. The solution containing both plasmids was administered via hydrodynamic tail vein injection. High levels of G-CSF were detectable in plasma 3 days after pCSF3 administration (FIG. 2a) and the concentration of circulating neutrophils increased steadily henceforward (FIG. 2b). After 2 weeks, we detected approximately 40-fold higher neutrophil counts in G-CSF-overexpressing mice, when compared to animals injected with the control plasmid lacking CSF3 (pCtrl, FIG. 2b). Neutrophilic mice developed increased numbers of resident neutrophils in vital organs (not shown), splenomegaly (FIG. 2c), but grew normally (FIG. 2d), did not show signs of organ injury (FIG. 2e), and were macroscopically indistinguishable from untreated animals (not shown).

[0141] We then compared the course of G-CSF-induced neutrophilia in WT mice to DNase1.sup.−/−, DNase1l3.sup.−/−, and DKO mice. Single KO mice were macroscopically indistinguishable from neutrophilic WT mice (FIG. 2f) and untreated controls (not shown). All DKO mice died within 7 days after induction of G-CSF overexpression and developed a rapidly progressing and severe hypothermia with concomitant hematuria (FIG. 2f, g). No such phenotype was observed in DKO mice injected with the control plasmid (FIG. 2f). In order to conclude that the lack of both circulating DNases is responsible for this phenotype, we performed several control experiments. An off-target mutation has been recently reported for the DNase1.sup.−/− mice used in this study (Rossaint, et al. (2014) Blood 123 (16): 2573-84). We therefore reproduced the phenotype in single and DKO mice derived from an independently generated DNase1.sup.−/− strain (Bradley, et al. (2012) Mammalian genome 23 (9-10): 580-6.) (not shown). DNase1- and DNase1l3-deficient mice spontaneously develop autoimmunity with features of lupus nephritis.sup.10,11. Furthermore, DNase1 and/or DNase1l3 degrade nuclear material generated during cell death (Sisirak, et al. (2016) Cell 166 (1): 88-101), and DNase1l3 has been implicated in B-cell development (Shiokawa, et al. (2007) Cell Death Differ. 14 (5): 992-1000). To rule out actions of DNase1 and DNase1l3 other than degrading extracellular DNA, we treated DKO mice with a liver-specific expression plasmid containing the cDNA of DNase1 (pDNase1) or DNase1l3 (pDNase1l3). Both enzymes contain a secretion sequence (Napirei, et al. (2009) FEBS J 276 (4): 1059-73), and consequently, hepatic expression of DNase1 or DNase1l3 restored their enzymatic activity in circulation (FIG. 2h, i) as well as the NET-degrading activity of serum (FIG. 2j). Importantly, DKO mice genetically reconstituted with DNase1 or DNase1l3 in circulation survived G-CSF-induced neutrophilia, did not develop severe hypothermia or hematuria (FIG. 2k), and were macroscopically indistinguishable from untreated DKO mice (not shown).

Example 4: Analysis of Pathology

[0142] LDH, AST, ALT, Bilirubin, Creatinine and BUN in plasma were analyzed by using standardized kits (Biotron Diagnostics) following the manufacturer instructions. Mouse G-CSF was quantified with a Quantikine ELISA Kit (R&D), following manufacturer's indications. Hemoglobin in EDTA-blood was quantified by an automated hemocytometer (Idexx ProCyte Dx Hematology Analyzer). To quantify neutrophils, whole blood was incubated on ice for 15 minutes with 0.2 μg of PE-labelled anti-mouse CD11b (M1/70, Biolegend) and 0.5 μg of Alexa Fluor 488 anti-mouse Ly6G (1AB, Biolegend). Sample was then diluted with 0.5 ml PBS and analyzed with the BD FACS Calibur. Data collection and sorting was performed using the CellQuest™ Software (BD immunocytometry systems, CA, USA).

[0143] For immunohistochemistry, paraffin-embedded sections were de-paraffinized, rehydrated, and subjected to antigen retrieval for 25 minutes at 100° C. in citrate buffer (10 mM sodium citrate, 0.1% Tween, pH 6). Thereafter, sections were blocked for 30 minutes with 2.5% normal goat serum (Vector Labs) followed by incubation with a mouse-on-mouse blocking kit (Vector) for one hour. The sections were then incubated over night at 4° C. with 2 μg/ml of the primary antibody against CRAMP (PA-CRLP-100, Innovagen), citrullinated histone 3 (ab5103, Abeam), fibrin-specific (clone 59D8), the complex of histone H2A, H2B, and DNA to detect chromatin, respectively. Sections were incubated with anti-rabbit and anti-mouse conjugated with AlexaFluor488 or AlexaFluor555 (Life Technologies) for 1 hour. After washing, DAPI staining (1 μg/ml) was applied for 2 minutes and washed. Autofluorescence was quenched by 25 minutes incubation with Sudan Black (0.1% in 70% EtOH), and sections were mounted with Fluoromount G (Southern Biotech). Images of fluorescently labeled sections were acquired with an inverted fluorescence microscope (Zeiss Axiovert 200M, Oberkochen, Germany) or a confocal microscope (Leica TCS SP5).

[0144] Results:

[0145] Histological analysis of vital organs from hypothermic DKO mice revealed intravascular hematoxylin-rich clots, which fully or partially occluded blood vessels in lungs, liver, and kidneys (FIG. 3a, b). Untreated DKO mice, neutrophilic WT mice, and neutrophilic DKO mice expressing pDNase1 or pDNase1l3 did not show occluded blood vessels (FIG. 3b). We observed two distinct staining patterns in intravascular hematoxylin-rich clots: a dotted pattern illustrating nuclei of leukocytes and an abundant diffuse staining pattern covering the space between nuclei (FIG. 3a). Positive staining with intercalating DNA dyes and antibodies against DNA-histone-complexes (FIG. 3c) identified extracellular chromatin as a major clot component. Extracellular chromatin was mainly derived from NETs, as evidenced by co-localization with markers of neutrophil granules such myeloperoxidase (FIG. 3d), cathelin-related antimicrobial peptide (CRAMP, not shown), and by the NET-surrogate marker citrullinated histones (not shown). Additionally, the DNA-clots contained von Willebrand factor (vWF) and fibrin (not shown), supporting the previously described pro-thrombotic activity of NETs (Engelmann, et al. (2013) Nat. Rev. Immunol. 13 (1): 34-45). The formation of DNA-clots was associated with organ damage, as evidenced by increased levels of plasma LDH (FIG. 3f). Elevated plasma levels of ALT and AST, indicated liver damage (FIG. 3g), while high levels of BUN and creatinine in plasma, and hematuria indicated renal failure (FIG. 3h). In addition, DKO mice developed anemia (FIG. 3i). In summary, these data suggest that mice lacking DNase1 and DNase1l3 died of multi-organ damage induced by systemic intravascular DNA-clots comprising NETs.

Example 5: Comparison of DNase Activity in Healthy Donors, Patients, and Mice

[0146] Plasma from normal healthy donors (NHD) and patients with thrombotic miroangiopathies (Jimenez-Alcazar, et al. (2015) J. Thromb. Haemost. 13 (5): 732-42), as well as mice was analyzed by SRED and in-vitro NET-degradation as outlined in Example 1.

[0147] Results:

[0148] The importance of circulating DNases to maintain homeostasis during inflammation is supported by our previous clinical observations. We identified elevated markers of neutrophil activation (Fuchs, et al. (2012) Blood 120 (6): 1157-64) along with a deficiency in circulating DNase1 in plasma from patients with acute thrombotic microangiopathies (TMA), a rare heterogeneous disease associated with infection and autoimmunity (Jimenez-Alcazar, et al. (2015) J. Thromb. Haemost. 13 (5): 732-42). DKO mice subjected to experimental neutrophilia and endotoxemia develop several features of patients with acute TMA, including strongly elevated levels of LDH, thrombocytopenia, anemia, hematuria, and organ damage (Kremer Hovinga, et al. (2010) Blood 115 (8): 1500-11; quiz 662). We therefore questioned whether TMA patients have a dual-deficiency in circulating DNase1 and in circulating DNase1l3. To discriminate DNase1 and DNase1l3 activity, we generated antibodies against human DNase1 (α-hDNase1), which block the enzymatic activity in a concentration-dependent manner in plasma from normal healthy donors (NHD) (FIG. 4a). Heparin is a known inhibitor of DNase1l3 (Napirei, et al. (2009) FEBS J 276 (4): 1059-73) and was used to block the DNA-degrading activity by DNase1l3 (FIG. 4b). Analysis of heparinized plasma confirmed the deficiency in DNase1 activity in TMA patients compared to healthy individuals (FIG. 4c). However, plasma spiked with α-hDNase1 showed a similar DNA-degrading activity in between patients and healthy donors (FIG. 4d), indicating normal levels of DNase1l3 activity in acute TMA and suggesting that DNase1 and DNase1l3 are independently regulated in diseased humans.

[0149] These data furthermore illustrate a discrepancy between humans and mice. Whereas TMA patients show normal levels of DNase1l3 activity, mice with normal levels of DNase1l3 do not develop vascular occlusions and organ damage in our experimental models. We therefore compared circulating DNase1 and DNase1l3 in mice to humans. Murine plasma showed approximately 10-fold higher DNA-degrading activity than human plasma (FIG. 4e). Comparison of plasma from DNase1.sup.−/− and DNase1l3.sup.−/− mice to plasma from NHD supplemented with α-hDNase1 and heparin, respectively, showed that activity of both DNases is approximately 10-fold higher in mice than in healthy donors (FIG. 4f, g). Finally, we speculated the DNase1l3 activity observed in NHD and TMA patients is not sufficient to degrade NETs efficiently. Indeed, NETs were stable in plasma from NHD supplemented with α-hDNase1 and in plasma from TMA patients. (FIG. 4h, i). Consequently, TMA patients and mice lacking DNase1 and DNase1l3, both cannot degrade NETs efficiently.

Example 6: Analysis of Septicemia

[0150] Escherichia coli (XEN 14, Perkin Elmer) was grown overnight in lysogeny broth media containing 50 μg/ml kanamycin. Bacteria were pelleted by centrifuging at 4000×g for 10 minutes, washed with and resuspended in PBS. Aliquots of 1.5×10.sup.9 bacteria/ml were incubated at 70° C. for 15 minutes to heat-kill the bacteria. Aliquots were stored at −20° C. until further use.

[0151] In preliminary experiments, we tested daily intraperitoneal injections of 1 μg/g of LPS from Salmonella enterica serotype thyphimurium (product number L6511, lot number 025M4042V, Sigma-Aldrich) in 0.9% saline. We observed that by day 3, clots of NETs occluded the vasculature in approximately 20% of DKO mice, but not in WT mice, Dnase1.sup.−/− mice, or Dnase1l3.sup.−/− mice. To improve the outcome, mice received an intravenous injection of 1.5×10.sup.7 heat-killed E. coli/g along with the third LPS injection. The shown survival time indicates the time after the injection of E. coli. Blood and organs were collected at the time of euthanasia. Insufficient biosamples were obtained from two animals (1×DKO+Ctrl, 1×DKO+Dnase1l3) to perform the complete analysis shown in FIG. 4. Mice were euthanized and scored as “non-surviving” if the animals showed signs of severe distress (irresponsiveness to touch). All non-surviving mice showed hematuria and paleness of extremities. All surviving mice were euthanized and scored as “surviving” 24 hours after the intravenous injection of heat-killed E. coli.

[0152] Results:

[0153] Septicemia is a potent and rapid trigger of intravascular NET formation in mice. Thus, we hypothesized that a defect in NET degradation may aggravate the disease. Indeed, mice with a combined deficiency in DNase1 and DNase1L3, but not wild-type mice, were highly susceptible to low doses of lipopolysaccharide and heat-killed E. coli (FIG. 5A). Similar to neutrophilic DKO mice, blood analysis of septic DKO mice showed hemolytic anemia and hematuria (FIGS. 5, B and C), along with increased levels of plasma LDH and schistocytes in blood smears (FIGS. 5, D and E). Furthermore, we detected abundant partially or fully occluded blood vessels in the lung (FIG. 5, F to H). A detailed analysis of partially occluded vessels revealed clots of NETs within the vascular lumen (FIG. 5I). In fully occluded vessels the NET clots were congested with entrapped erythrocytes and leukocytes (FIG. 5I). Hepatic expression of Dnase1 or Dnase1l3 in DKO mice prevented vascular occlusion and restored the wild-type phenotype. Thus, circulating DNase1 or DNase1L3 prevent the formation of NET clots and host injury in septicemia.

Example 7: Analysis of Septicemia in Neutrophilic WT Mice

[0154] WT mice were injected with 50 μg of CSF3-plasmid (CSF3) to induce G-CSF expression and neutrophilia. Blood smears of EDTA-anticoagulated blood were prepared on polylysine-coated slides (Hecht Assistant, Germany). After air-drying, they were incubated for 1 minute in methanol supplemented with 1 μM SytoxGreen on dry ice to stain NETs. WT mice expressing CSF3 or Ctrl plasmid for 2 weeks received a single intraperitoneal injection of 1 μg/g of LPS from Salmonella enterica serotype thyphimurium (product number L6511, lot number 025M4042V, Sigma-Aldrich) in 0.9% saline. After 4 hours mice are scarified and blood was collected. LDH, AST, and ALT were quantified by using standardized kits (Biotron Diagnostics, CA, USA) following the manufacturer instructions. Paraffin section of lung and liver were stained for with a fibrin-specific antibody (clone 59D8).

[0155] Results:

[0156] Hepatic expression of CSF3, which encodes G-CSF, resulted in a steadily increasing concentration of spontaneously formed NETs in blood smears (FIG. 6A). Interestingly, neutrophilic mice were highly susceptible to low doses of lipopolysaccharide, when compared to mice with wild-type neutrophil blood counts (FIG. 6B). Furthermore, histological analysis revealed a robust and systemic deposition of fibrin in vital organs (FIG. 7), indicating disseminated intravascular coagulation, a severe complication of endotoxemia and sepsis in patients. Thus, the overexpression of G-CSF in wild-type mice resulted neutrophilia, which is tolerated, but represents a severe pro-inflammatory and pro-thrombotic state.

Example 8: Precipitation of Early-Onset SLE and RA in Neutrophilic MRLlpr Mice

[0157] MRLlpr mice were injected with 50 μg of CSF3-plasmid (CSF3) to induce G-CSF expression and neutrophilia at the age of 4 weeks. After 5 weeks (35 days) urine was collected.

[0158] Results:

[0159] The effect of hepatic expression of CSF3, which encodes G-CSF, was analyzed in lupus-prone MRLlpr mice. Analysis of proteinuria indicated that CSF3-expression induced an early and severe disease onset of SLE (FIG. 8A). Furthermore, continuous monitoring of the mice after CSF3-gene delivery indicated rheumatoid arthritis (RA)-like symptoms, namely swollen paws within 4-5 weeks (FIG. 8B). In depth analysis of the bone structure by micro-CT showed that CSF3-expression precipitated robust bone degeneration, thus confirming the development of RA (FIG. 8C).