The present invention relates to the use of antibody non-agonist ligands to IL-13Ra1 (UniProt P78552) in the treatment of atopic (allergic) inflammatory diseases, sepsis, and / or neutropenia.

Description

The canonical response of the immune system to tissue damage and invasion by pathogens involves the activation of tissue-resident stromal and immune cells, as well as the fast mobilization of neutrophil granulocytes (neutrophils) from the bone marrow (BM) to the periphery. In fact, neutrophils are the first innate immune cells to migrate to the site of action and rapidly exert several effector functions, including the secretion of cytokines and chemokines, thereby guiding and attracting additional innate and adaptive immune cells. The efficacy of neutrophil recruitment hinges on a multistep process, including vigorous de novo generation, proliferation and maturation of neutrophil precursors in the BM, followed by their release as mature neutrophils to the bloodstream. Once neutrophils are in the circulation, they patrol the body in search of molecular cues of tissue inflammation, which induce neutrophils to leave the blood vessel and infiltrate the affected tissue. All these steps of neutrophil mobilization are driven by cytokines and chemokines, most notably granulocyte colony-stimulating factor (G-CSF) and the C-X-C chemokine receptor 2 (CXCR2)-binding chemokines CXCL1 and CXCL2, also termed keratinocyte chemoattractant (KC) and macrophage inflammatory protein-2 (MIP-2), respectively.

G-CSF is synthesized by stromal and immune cells upon acute inflammation and infection and leads to expansion and mobilization of myeloid cells in the BM, most prominently neutrophils. While G-CSF directly stimulates neutrophil precursors to proliferate, G-CSF's influence on neutrophil mobilization relies on CXCR2 and CXCR4. Notably, neutrophil release from the BM is controlled by signals from CXCR4 and its ligand CXCL12, which cause neutrophils to remain in their BM niche, while CXCR2 and its ligands CXCL1 and CXCL2 mediate neutrophil egress from the BM. G-CSF signals tip this balance towards egress by favoring signals via CXCR2 over CXCR4. Thus, provision of recombinant G-CSF to humans or mice leads - via the above-mentioned effects - to prominent blood neutrophilia and subsequent neutrophil migration to tissues whereby neutrophils follow CXCR2-binding chemokines and ultimately so-called end-target chemoattractants.

Strikingly, type-2 cell-mediated inflammation appears to be an exception to the above- mentioned pattern. Type-2 cell immune responses are characterized by the cytokines interleukin-4 (IL-4), IL-5, IL-9, IL-13, IL-25, IL-33, and thymic stromal lymphopoietin, which in turn are produced by, stimulate, and recruit type-2 immune cells, such as T-helper-2 cells as well as different innate immune cells, including type-2 innate lymphoid cells, eosinophils, basophils, mast cells, and IL-4- and/or IL-13-activated macrophages. However, while these innate immune cells predominate during type-2 cell immune responses, neutrophils are conspicuously absent in type-2 cell inflammation. Moreover, chronic type-2 cell-driven inflammatory disorders, such as atopic dermatitis, are often associated with recurrent bacterial infections that are usually contained by neutrophils. Notably, marked neutropenia in target organs is a common finding in many allergic diseases. Thus, patients suffering from atopic dermatitis contain normal counts of blood neutrophils but show a paucity of neutrophils in skin lesions, even though their skin is colonized with bacteria known to induce neutrophil recruitment. Not surprisingly, a dominant type-2 cell immune signature characterizes the skin of atopic dermatitis patients, with IL-4 being prominently expressed.

Based on the above-mentioned state of the art, the objective of the present invention is to provide means and methods for the treatment of conditions or diseases impacted by neutrophil dysregulation, particularly for the treatment of neutropenia and atopic inflammatory disease. This objective is attained by the subject matter of the independent claims of the present specification.

The finding that IL-4 is instrumental in initiating, polarizing and maintaining type-2 immunity motivated the inventors to inquire whether IL-4 signals directly affect neutrophil expansion, migration or function.

Neutropenia is a life-threatening condition as it is accompanied by an increased risk of infections that can rapidly lead to death. The most severe form of neutropenia is called agranulocytosis. Causes of neutropenia are manifold, most commonly including medications (chemotherapy, indomethacin and other drugs), cancer (especially leukemias), radiation, autoimmune diseases, genetic diseases, hemodialysis, and vitamin deficiency (vitamin B12 or folic acid).

The results presented herein demonstrate that IL-4 receptor (IL-4R) stimulation on

neutrophils reduces their expansion and egress from BM, and dampens their recruitment to peripheral tissues. Together, these data demonstrate that type II IL-4R signalling can directly impair neutrophil recruitment during infection and inflammation, and that a non-agonist ligand specifically reactive to IL-13Ra1 is of use in the treatment of use in the treatment of a condition selected from neutropenia, allergic inflammation and sepsis. Terms and definitions

Amino acid sequences are given from amino to carboxyl terminus. Capital letters for sequence positions refer to L-amino acids in the one-letter code (Stryer, Biochemistry, 3 ^rd ed. p. 21 ).

"Capable of forming a hybrid" in the context of the present invention relates to sequences that under the conditions existing in the extracellular bodily compartments or within the cytosol of a mammalian cell, are able to bind selectively to their target sequence. Such hybridizing sequences may be contiguously reverse-complimentary to the target sequence, or may comprise gaps, mismatches or additional non-matching nucleotides. The minimal length for a sequence to be capable of forming a hybrid depends on its composition, with C or G nucleotides contributing more to the energy of binding than A or T/U nucleotides, and the backbone chemistry.

In certain embodiments, the hybridizing sequence is at least 80% identical, particularly 85%, 90%, 92%, 94%, 95%, 96%, 97%, 98% or 99% identical to the reverse complimentary sequence of a DNA encoding SEQ ID 1 or SEQ ID 2. In certain embodiments, the hybridizing sequence comprises deoxynucleotides, phosphothioate deoxynucleotides, LNA and/or PNA nucleotides or mixtures thereof.

In the context of the present specification, the term antibody is used in its meaning known in the art of cell biology and immunology; it refers to whole antibodies including but not limited to immunoglobulin type G (IgG), type A (IgA), type D (IgD), type E (IgE) or type M (IgM), any antigen binding fragment or single chains thereof and related or derived constructs. A whole antibody is a glycoprotein comprising at least two heavy (H) chains and two light (L) chains inter-connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable region (V _H ) and a heavy chain constant region (C _H ). The heavy chain constant region is comprised of three domains, C _H 1 , C _H 2 and C _H 3. Each light chain is comprised of a light chain variable region (abbreviated herein as V _L ) and a light chain constant region (C _L ). The light chain constant region is comprised of one domain, C _L . The variable regions of the heavy and light chains contain a binding domain that interacts with an antigen. The constant regions of the antibodies may mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component of the classical complement system.

The term antibody-like molecule in the context of the present specification refers to a molecule capable of specific binding to another molecule (its "target") with high affinity, particularly with a coefficient of dissociation Kd < 10E-8 mol/l. An antibody-like molecule binds to its target in a similar fashion as an antibody binds specifically to its target. The term antibody-like molecule encompasses a repeat protein, such as a designed ankyrin repeat protein, a polypeptide derived from armadillo repeat proteins, a polypeptide derived from leucine-rich repeat proteins and a polypeptide derived from tetratricopeptide repeat proteins.

The term antibody-like molecule further encompasses a polypeptide derived from protein A domains, a polypeptide derived from fibronectin domain FN3, a polypeptide derived from consensus fibronectin domains, a polypeptide derived from lipocalins, a polypeptide derived from Zinc fingers, a polypeptide derived from Src homology domain 2 (SH2), a polypeptide derived from Src homology domain 3 (SH3), a polypeptide derived from PDZ domains, a polypeptide derived from gamma-crystallin, a polypeptide derived from ubiquitin, a polypeptide derived from a cysteine knot polypeptide and a polypeptide derived from a knottin.

The term protein A domains derived polypeptide refers to a molecule that is a derivative of protein A and is capable of specifically binding the Fc region and the Fab region of immunoglobulins.

The term armadillo repeat protein refers to a polypeptide comprising at least one armadillo repeat, wherein an armadillo repeat is characterized by a pair of alpha helices that form a hairpin structure.

The term humanized camelid antibody in the context of the present specification refers to an antibody consisting of only the heavy chain or the variable domain of the heavy chain (VHH domain) and whose amino acid sequence has been modified to increase their similarity to antibodies naturally produced in humans and, thus show a reduced immunogenicity when administered to a human being.

A general strategy to humanize camelid antibodies is shown in Vincke et al. "General strategy to humanize a camelid single-domain antibody and identification of a universal humanized nanobody scaffold", J Biol Chem. 2009 Jan 30;284(5):3273-3284, and US201 1 165621 A1 .

The term "specifically reactive" when used in the context of describing the binding of a ligand to its target refers to non-covalent binding of the ligand to the target characterized by a Kd < 10 ^"7 mol/l, typically even < 10 ^"8 mol/l or < 10 ^"9 mol/l, and a binding characteristic of the ligand to other biomolecules present in the biological context of the target characterized by a Kd > 10 ^"4 , even > 10 ^"3 . In other words, specific binding of the ligand refers to the ability of the ligand to selectively bind to the target but no other structure.

Any and all US patent documents mentioned in the present specification are to be deemed incorporated herein by reference. A first aspect of the invention relates to the use of a non-agonist polypeptide ligand specifically reactive to IL-13Ra1 (UniProt ID P78552) in the treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis. The non-agonist polypeptide ligand of IL-13Ra1 is capable of binding to IL-13Ra1 , thereby inhibiting the biological effect of IL-4 and IL-13 on cells on which the IL-13Ra1 is expressed.

UniProt ID numbers in this document refer to entries in the Universal Protein Resource Knowledgebase.

The protein sequence of the IL-13 receptor alpha 1 canonical isoform is

SEQ ID No 1

MEWPARLCGL WALLLCAGGG GGGGGAAPTE TQPPVTNLSV SVENLCTVIW TWNPPEGASS NCSLWYFSHF GDKQDKKIAP ETRRSIEVPL NERICLQVGS QCSTNESEKP SILVEKCISP PEGDPESAVT ELQCIWHNLS YMKCSWLPGR NTSPDTNYTL YYWHRSLEKI HQCENIFREG QYFGCSFDLT KVKDSSFEQH SVQIMVKDNA GKIKPSFNIV PLTSRVKPDP PHIKNLSFHN DDLYVQWENP QNFISRCLFY EVEVNNSQTE THNVFYVQEA KCENPEFERN VENTSCFMVP GVLPDTLNTV RIRVKTNKLC YEDDKLWSNW SQEMSIGKKR NSTLYITMLL IVPVIVAGAI IVLLLYLKRL KIIIFPPIPD PGKIFKEMFG DQNDDTLHWK KYDIYEKQTK EETDSWLIE NLKKASQ

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is specifically reactive to the extracellular part of the receptor molecule, which is described by the sequence

SEQ ID NO 2 P78552[22 - 427]

GGGGAAPTET QPPVTNLSVS VENLCTVIWT WNPPEGASSN CSLWYFSHFG DKQDKKIAPE TRRSIEVPLN ERICLQVGSQ CSTNESEKPS ILVEKCISPP EGDPESAVTE LQCIWHNLSY MKCSWLPGRN TSPDTNYTLY YWHRSLEKIH QCENIFREGQ YFGCSFDLTK VKDSSFEQHS VQIMVKDNAG KIKPSFNIVP LTSRVKPDPP HIKNLSFHND DLYVQWENPQ NFISRCLFYE VEVNNSQTET HNVFYVQEAK CENPEFERNV ENTSCFMVPG VLPDTLNTVR IRVKTNKLCY EDDKLWSNWS QEMSIGKKRN ST

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of chronic neutropenia.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of chronic neutropenia caused by aplastic anemia, particularly aplastic anemia caused by viral infection, exposure to toxic chemicals or radiation.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of acute neutropenia, particularly neutropenia caused or associated with infection such as infection with Salmonella typhi, Mycobacterium (tuberculosis), or cytomegalovirus. In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of neutropenia caused or associated with administration of a pharmaceutical drug, particularly administration of indomethacin, propylthiouracil, levamisole, clozapine, valproate, penicillamine, or trimethoprim / sulfamethoxazole combination therapy.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of agranulocytosis. In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of neutropenia caused or associated with deficiency of vitamin B12 or folic acid.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of allergic inflammation, particularly allergic asthma, atopic dermatitis, allergic rhinitis or allergic inflammation of the eye or conjunctivitis. Further allergic disease conditions amenable to such treatment are eosinophilic disorders, food allergy, urticarial and allergic bronchopulmonary aspergillosis.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of bacteremia and sepsis, particularly sepsis caused by bacterial infection. In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is used for the treatment of sepsis caused by or associated with a bacterial infection of the skin, a bacterial infection of the lung, a bacterial infection of the urinary tract, a bacterial infection of a joint, a bacterial infection of the gut, a bacterial infection caused by intravenal catheterization or injection, or an infection of a prosthesis.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 for use in treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis is selected from the group consisting of an antibody, an antibody fragment and an antibody-like molecule.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 for use in treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis is an antibody.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 for use in treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis is a human or humanized gamma immunoglobulin.

In certain embodiments, the antibody fragment is a Fab domain or an Fv domain of an antibody, or a single-chain antibody fragment, which is a fusion protein consisting of the variable regions of light and heavy chains of an antibody connected by a peptide linker. In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is a single domain antibody, consisting of an isolated variable domain from a heavy or light chain. In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL- 13Ra1 is a heavy-chain antibody consisting of only heavy chains such as antibodies found in camelids. In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 antibody-like molecule is a repeat protein, such as a designed ankyrin repeat protein.

In certain embodiments, the non-agonist polypeptide ligand specifically reactive to IL-13Ra1 for use in treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis does not specifically bind to IL-4Ra.

In certain embodiments, the non-agonist anti-IL-13Ra1 polypeptide ligand is selected from the group consisting of an antibody, an IL-13Ra1 -specific antibody fragment, an antibody-like molecule and a protein A domains derived polypeptide. Non-antibody scaffolds targeting IL-13Rct1 (antibody-like molecules)

Alternative targeting proteins have been proposed recently, which are more diverse in molecular structure than human immunoglobulin-derived antibody fragments and antibody- derived constructs and formats, and thus allow additional molecular formats by creating heterodimeric and multimeric assemblies, leading to new biological functions. A number of such targeting proteins have been described (reviewed in (Binz et al., Nat. Biotech 2005, Vol 23: 1257-1268)). Non-limiting examples of such targeting proteins are camelid antibodies, protein scaffolds derived from protein A domains (termed "Affibodies", Affibody AB), tendamistat (an alpha-amylase inhibitor, a 74 amino acid beta-sheet protein from Streptomyces tendae), fibronectin, lipocalin ("Anticalins", Pieris), T-cell receptors, ankyrins (designed ankyrin repeat proteins termed "DARPins", Univ. Zurich and Molecular Partners; see US2012014261 1 (A1 )), A-domains of several receptors ("Avimers", Avidia) and PDZ domains, fibronectin domains (FN3) ("Adnectins", Adnexus), consensus fibronectin domains ("Centyrins", Centyrex/Johnson&Johnson) and Ubiquitin ("Affilins", SCIL Proteins) and knottins (Moore and Cochrane, Methods in Enzymology 503 (2012), 223-251 and references cited therein).

From these proteins, multimeric and multispecific assemblies can be constructed (Caravella and Lugovskoy, Current Opinions in Chemical Biology 2010, 14:520-528; Vanlandschoot et al. Antiviral Research 201 1 92:389-407; Lofblom et al. 201 1 Current Opinion in Biotechnology 201 1 22:843-848, Boersma et al. 201 1 Curr. Opin. Biotechnol. 22:849-857). It is also possible to fuse these and other peptidic domains to antibodies to create so-called Zybodies (Zyngenia Inc., Gaithersburg, MD).

All of these scaffolds, with different inherent properties, have in common that they can be directed to bind specific epitopes, by using selection technologies well known to practitioners in the field (Binz et al., Nat. Biotech 2005, 23: 1257-1268). For example, IL-13Ra1 can be expressed in insect cells, using a baculovirus expression system (Frei et al., Nat Biotechnol. 2012 30:997-1001 ). The IL-13Ra1 can then be biotinylated and thus be immobilized on streptavidin-coated magnetic beads or on microtiter plates coated with streptavidin or neutravidin (Steiner et al. (2008) J. Mol. Biol. 382, 121 1- 1227); (Zahnd et al. (2007) J. Mol. Biol. 369, 1015-1028.)). The IL-13Ra1 so immobilized can then serve as target for diverse protein libraries in either phage display or ribosome display format. A large variety of different antibody libraries has been published (Mondon et al., Frontiers in Bioscience. 13:1 1 17-1 129, 2008.) and the technology of selecting binding antibodies is well known to the practitioners of the field. Phage display is a suitable format for antibody fragments (Fab fragments, scFv fragments or single domain antibodies s) (Hoogenboom Nature Biotechnology. 23(9):1 105-1 1 16, 2005 Sep) and any other scaffold that contain disulfide bonds, but it can also be used for scaffolds not containing disulfide bonds (e.g., Steiner et al. (2008) J. Mol. Biol. 382, 121 1-1227) (Rentero et al. Chimia. 65(1 1 ):843-5, 201 1., Skerra A. Current Opinion in Biotechnology. 18(4):295-304, 2007 Aug). Similarly, ribosome display can be used for antibody fragments (Hanes et al. (2000), Nat. Biotechnol. 18, 1287-1292) and for other scaffolds (Zahnd et al. (2007) Nat. Methods 4, 269- 279; Zahnd et al. (2007) J. Mol. Biol. 369, 1015-28.). A third powerful technology is yeast display (Pepper et al., Combinatorial Chemistry & High Throughput Screening. 1 1 (2):127- 134, 2008 Feb.). In this case a library of the binding protein of interest is displayed on the surface of yeast, and the target epitope is either directly labeled with a fluorescent dye or its polyhistidine (his ₆ )-tag is detected with an anti-his-tag antibody, which is in turn detected with a secondary antibody. Such methods are well known to the practitioners in the field (Boder et al., Methods in Enzymology. 328:430-44, 2000.).

In certain embodiments, the non-agonist anti-IL-13Ra1 polypeptide ligand is an immunoglobulin consisting of two heavy chains and two light chains. In some embodiments, the non-agonist anti-IL-13Ra1 polypeptide ligand is a single domain antibody, consisting of an isolated variable domain from a heavy or light chain. In some embodiments, the non- agonist anti-IL-13Ra1 polypeptide ligand is a heavy-chain antibody consisting of only heavy chains such as antibodies found in camelids.

According to an alternative aspect of the invention, a pharmaceutical composition comprising a non-agonist polypeptide ligand specifically reactive to IL-13Ra1 is provided for use in treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis.

Similarly, the non-agonist anti-IL-13Ra1 polypeptide ligand according to the above described aspects and embodiments is used in a method of manufacture of a medicament for the treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis. Medicaments according to the invention are manufactured by methods known in the art, especially by conventional mixing, coating, granulating, dissolving or lyophilizing.

Also within the scope of the present invention is a method or treating or preventing a condition selected from neutropenia, allergic inflammation and sepsis in a patient in need thereof, comprising administering to the patient a non-agonist polypeptide ligand specifically reactive to IL-13Ra1 as specified in any of the aspects or embodiments of the invention above.

Such treatment according to the invention may be for prophylactic or therapeutic purposes. For the administration, the inhibitor or compound is preferably in the form of a pharmaceutical preparation comprising the inhibitor or compound in chemically pure form and, optionally, a pharmaceutically acceptable carrier or adjuvants. The dosage of the inhibitor or compound depends upon the species, its age, weight, individual condition, the individual pharmacokinetic data, the mode of administration, and whether the administration is for prophylactic or therapeutic purposes. The daily dose administered may range from approximately 0.1 mg/kg to approximately 1000 mg/kg, preferably from approximately 0.5 mg/kg to approximately 100 mg/kg, of an inhibitor or compound according to the above aspects or embodiments of the invention.

According to an alternative aspect of the invention, the effect underlying the instant invention, namely the suppression of IL-13Ra1 signaling on neutrophils and related cells in the treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis, is achieved by inhibiting of IL-13Ra1 gene expression mediated by a single-stranded or double-stranded interfering ribonucleic acid oligomer or a precursor thereof, comprising a sequence tract complementary to, in other word capable of forming a hybrid with, IL-13Ra1 mRNA.

The art of silencing or "knocking down" genes, by degradation of mRNA or other effects, is well known. Examples of technologies developed for this purpose are siRNA, miRNA, shRNA, shmiRNA, or dsRNA. A comprehensive overview of this field can be found in Perrimon et al, Cold Spring Harbour Perspectives in Biology, 2010, 2, a003640.

Alternatively, IL-13Ra1 gene expression is effected through a single-stranded or double- stranded antisense ribonucleic or deoxyribonucleic acid, comprising sequences complementary to a sequence comprised in an operon that expresses the IL-13Ra1 encoding gene. Such an operon sequence may include, without being restricted to, an intron, an exon, an operator, a ribosome binding site or an enhancer sequence. Such antisense molecules may for example be 12-50 nucleotides in length. According to yet another alternative aspect of the invention, suppression of IL-13Ra1 signaling on neutrophils and related cells in the treatment or prevention of a condition selected from neutropenia, allergic inflammation and sepsis, is achieved by an expression vector, comprising a sequence encoding an interfering ribonucleic acid oligomer or precursor thereof, as is described in the preceding paragraphs. Optionally, the sequence is under the control of an RNA-polymerase promoter sequence operable in a mammalian cell. Such an expression vector allows for the production of an interfering RNA within the cell. Methods for making and using such expression vectors are known in the art.

Similarly, the invention in another aspect relates to a dosage form for the prevention or treatment of neutropenia, allergic inflammation or sepsis. The dosage form comprises a non- agonist polypeptide ligand specifically reactive to IL-13Ra1 according to one of the above aspects of the invention. In certain embodiments, the active agent is applied by parenteral administration, such as by subcutaneous, intravenous, intrahepatic or intramuscular injection forms. Optionally, a pharmaceutically acceptable carrier and/or excipient is present. The pharmaceutical compositions comprise from approximately 1 % to approximately 95% active ingredient, preferably from approximately 20% to approximately 90% active ingredient.

A non-agonist polypeptide ligand specifically reactive to IL-13Ra1 according to the above aspects or embodiments of the invention can be administered alone or in combination with one or more other therapeutic agents. Possible combination therapies can take the form of fixed combinations of the inhibitor, or compound with one or more other therapeutic agents known in the prevention or treatment of neutropenia, allergic inflammation or sepsis. The administration can be staggered or the combined agents can be given independently of one another or in the form of a fixed combination.

In certain embodiments, a pharmaceutical composition comprises an non-agonist polypeptide ligand specifically reactive to IL-13Ra1 according to the above aspects or embodiments of the invention, in combination with a therapeutically active amount of the granulocyte colony-stimulating factor (G-CSF, UniProt ID P09919).

The skilled artisan is aware of a broad range of possible recipes for providing topical formulations, as exemplified by the content of Benson and Watkinson (Eds.), Topical and Transdermal Drug Delivery: Principles and Practice (1 st Edition, Wiley 201 1 , ISBN-13: 978- 0470450291 ); and Guy and Handcraft: Transdermal Drug Delivery Systems: Revised and Expanded (2 ^nd Ed., CRC Press 2002, ISBN-13: 978-0824708610); Osborne and Amann (Eds.): Topical Drug Delivery Formulations (1 ^st Ed. CRC Press 1989; ISBN-13: 978- 0824781835). Wherever alternatives for single separable features are laid out herein as "embodiments", it is to be understood that such alternatives may be combined freely to form discrete embodiments of the invention disclosed herein.

The invention is further illustrated by the following examples and Figures, from which further embodiments and advantages can be drawn. These examples are meant to illustrate the invention but not to limit its scope.

Brief description of the Fig.

Fig. 1 shows that neutrophil influx during skin infection is modulated by IL-4 signals: (A- E) 3x10 ⁷ colony-forming units (cfu) of Group A Streptococcus (GAS) M1 were injected subcutaneously into the shaved flank of C57BL/6 mice pretreated with

PBS or IL-4— anti-IL-4 mAb complexes (IL-4cx) or given a neutralizing anti-IL-4 mAb throughout the experiment, starting one day prior to infection.

(A) Skin lesion size was measured 48 hours (left panel) and 72 hours (right panel) postinfection.

(B) Leg swelling of the infected and uninfected site was determined over 72 hours postinfection. (C) Quantification of flow cytometric analysis of CD1 1 b ⁺ Ly6G ⁺ cells per 0.1 g skin 6 hours postinfection.

(D) Immunohistochemistry and quantification of Ly6G ⁺ cells in skin 6 hours postinfection. Scale bars represent 100 μηη. (E) Cfu in skin 72 hours postinfection normalized to PBS-treated mice.

Data are pooled from 2-3 independent experiments with a total of 4 (D) and 8-12 mice per condition (A-C, and E; 2-3 mice for uninfected) and represented as mean ± SEM. *P<0.05; **P<0.01 ; ***P<0.001.

Fig. 2 shows that flood neutrophilia upon systemic infection is suppressed by IL-4: (A

and B) C57BL/6 mice received either PBS or 10 ⁵ cfu LM intravenously (i.v.) with or without a neutralizing anti-G-CSF mAb (anti-G-CSF) on days -1 and 0 of infection. Shown are flow cytometric analysis 24 hours postinfection of CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophil frequencies in blood (A) and neutrophil counts in blood and spleen (B).

(C and D) C57BL/6 mice received either PBS or 10 ⁵ cfu LM i.v. without or with pretreatment using IL-4cx on days -3 to -1 prior to infection, followed by analysis 24 hours postinfection. Shown are flow cytometric analysis of CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophil frequencies in blood (C) and their quantification (D). Data are representative of 1 out of 2 independent experiments with a total of 4-5 animals per condition and are represented as mean ± SD. ns = not significant; *P<0.05.

shows that IL-4 antagonizes G-CSF effects on neutrophils: (A and B) C57BL/6 mice were treated with PBS, G-CSF-anti-G-CSF mAb complexes (G-CSFcx), IL- 4cx, or G-CSFcx plus IL-4cx for 3 consecutive days. Bone marrow (BM), blood and spleen were analyzed 16 hours after last injection. (A) Expression of Ly6G versus CD1 1 b in CD3 ^" BM, blood and spleen cells. (B) Quantification of CD1 1 b ⁺ Ly6G ⁺ neutrophils in BM, blood and spleen. (C) C57BL/6 mice were treated with PBS, G-CSF, IL-4, or G-CSF plus IL-4 for 3 consecutive days. Shown is quantification of CD1 1 b ⁺ Ly6G ⁺ neutrophils in BM, blood and spleen 16 hours after last injection. (D-F) C57BL/6 mice were pretreated for 3 days with PBS, G- CSFcx, IL-4cx, or G-CSFcx plus IL-4cx, followed by systemic (i.v.) infection with 10 ⁵ cfu LM the next day: (D) Quantification of LM cfu in spleen and liver 24 and 72 hours postinfection. (E) Mice were assessed for weight change. (F) Mice were monitored for survival. Plots (A) are representative of one out of three independent experiments with 2-3 mice per group each; quantifications (B and C) are pooled from 3 independent experiments with a total of 7 mice per group and are displayed as mean ± SEM. Data shown as mean ± SD are representative of 1 out of 2 independent experiments with a total of 4 animals per condition (D) or pooled from 2 experiments with a total of 6 mice per condition (E and F). ns = not significant; *P<0.05; **P<0.01 ; ***P<0.001 ; ****P<0.0001 .

shows that IL-4 acts directly on neutrophils via type I I IL-4R: (A and B) Wild-type (WT) mice received 3 injections of PBS, G-CSFcx, or G-CSFcx plus IL-4cx. Histograms show IL-4 receptor (IL-4R) subunit expression compared to isotype control (gray shaded area) of mice receiving PBS (black line) or G-CSFcx (blue line) and quantification by mean fluorescence intensity (MFI) of IL-4Ra, common γ-chain (y _c ) and IL-13Ra1 expression on CD1 1 b ⁺ Ly6G ⁺ neutrophils isolated from BM (A) or spleen (B). (C) Stimulation of splenocytes from WT, IL-4Ra-deficient (Il4ra ^~ ' ^~ ), y _c -deficient (//2rg ^"/_ ), and IL-13Ra1 -deficient (Il13ra1 ^~ ' ^~ ) mice with PBS or IL-4 (500 ng/ml) for 15 minutes, followed by quantification of phosphorylated STAT6 (pSTAT6) in CD1 1 b ⁺ Ly6G ⁺ neutrophils. Shown is change in percentage of MFI of pSTAT6 compared to PBS. (D) Stimulation of WT splenocytes with titrated concentrations of IL-4 and IL-13, followed by quantification of pSTAT6 in CD1 1 b ⁺ Ly6G ⁺ neutrophils. Shown is change in percentage of MFI of pSTAT6 compared to PBS. (E) WT and H4ra ^~ ' ^~ mice were treated with PBS, G-CSFcx, or G-CSFcx plus IL-4cx for 3 consecutive days. Spleen and blood were analyzed 16 hours after last injection. Shown is expression of CD1 1 b versus Ly6G in blood CD3 ^" cells (left panel) and quantification of CD1 1 b ⁺ Ly6G ⁺ neutrophils in indicated organs (middle and right panels). (F) WT and H2rg ^~ ' ^~ were treated and assessed as in (E). (G) Immune cell-lineage-depleted BM cells of WT (CD45.1 ⁺ ) and H4ra ^~ ' ^~ (CD45.2 ⁺ ) mice were mixed at a 1 :1 ratio and adoptively transferred to irradiated CD45.2 ⁺ H4ra ^~ ' ^~ hosts. After reconstitution, BM chimeric mice were injected for 3 consecutive days with PBS, G-CSFcx, or G-CSFcx plus IL-4cx and change in ratios of CD45.2 ⁺ to CD45.1 ⁺ cells within CD3 ^" CD1 1 b ⁺ Ly6G ⁺ blood neutrophils were determined by flow cytometry 16 hours after last injection. (H) WT and H4ra ^~ ' ^~ mice were infected i.v. with 10 ⁵ cfu LM and monitored for survival. Data (A, B, D-G) are representative of 1 out of 2 independent experiments and shown as mean ± SD or are pooled from 2-3 independent experiments and given as mean ± SEM (C and H), with 2-3 animals per condition each, ns = not significant; *P<0.05; **P<0.01 ; ***P<0.001 ; ****P<0.0001 .

shows that IL-4 induces a BM-resident phenotype in neutrophils. (A-D) WT mice were treated with PBS, G-CSFcx or IL-4cx for 3 consecutive days. BM and spleen were analyzed 16 hours after last injection. (A) Quantification of geometric MFI of CXCR2 and CXCR4 in neutrophils from BM. (B) Histogram showing CXCR4 expression on neutrophils in BM. (C) Quantification of geometric MFI of CXCR2 and CXCR4 in neutrophils from spleen. (D) Histogram of CXCR2 expression on neutrophils in spleen. Data are representative of 1 out of 2 experiments with 2-3 mice per group each and are presented as mean ± SD. ns = not significant; *P<0.05; **P<0.01 ; ****P<0.0001 .

shows that IL-4 inhibits CXCR2-mediated migration in vitro and in vivo. (A) Purified BM-derived CD1 1 b ⁺ Ly6G ⁺ neutrophils were pretreated with either PBS or IL-4 (30 ng/ml), followed by migration towards CXCL2 (100 ng/ml) over 240 minutes. (B) Purified BM neutrophils were pretreated with either PBS or IL-4 (30 ng/ml), followed by migration towards titrated concentrations of CXCL2 for 2 hours. (C and D) Purified BM neutrophils were pretreated titrated amounts of IL- 4, followed by migration towards a fixed concentration of CXCL1 (100 ng/ml) (C) or CXCL2 (100 ng/ml) (D) for 2 hours. (E) Purified BM neutrophils were pretreated in vitro with PBS, IL-2, IL-4, IL-7, IL-13, or IL-15 followed by migration towards CXCL1 (100 ng/ml) for 2 hours. (F) Purified BM neutrophils from WT or H4ra ^~ ' ^~ mice were pretreated with either PBS or IL-4 (30 ng/ml), followed by migration towards to CXCL2 (100 ng/ml) for 2 hours. (G) WT and Cxcr2 ^~ ' ^~ mice harboring an airpouch received either PBS or monosodium urate crystals (MSU) into their airpouch and were analyzed the next day by flow cytometry for CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils in their airpouch (top and middle panels) and blood (bottom panels). (H) WT mice harboring an airpouch were treated i.v. with either PBS or IL-4cx, followed 15 minutes later by injection of PBS (control), MSU or IL- 1 β into the airpouch. Shown are total CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils that migrated into the airpouch. Data are representative of at least 2 independent experiments and are displayed as mean ± SD (A-D), or are pooled from 2-3 independent experiments (E, F and H) and shown as mean ± SEM, with 2-3 animals per group each, ns = not significant; *P<0.05; **P<0.01 ; ***P<0.001. shows IL-4 signaling interferes with p38 MAPK-PI3K crosstalk. (A) CXCL1- induced migration of purified neutrophils was analyzed upon treatment in vitro with PBS, phosphoinositide 3-kinase (PI3K) inhibitor LY294002, or p38 mitogen- activated protein kinase (MAPK) inhibitor SB203580. (B) WT splenocytes were stimulated with either PBS (gray shaded area) or IL-4 (30 ng/ml) for 5 (orange line) or 15 minutes (red line), followed by assessment by flow cytometry of p38 MAPK phosphorylation in CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils. (C) WT splenocytes were stimulated with either PBS, IL-4 plus DMSO, or IL-4 plus p38 inhibitor SB203580, followed by assessment by flow cytometry of p38 MAPK phosphorylation in CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils. Shown are histograms of phosphorylated p38 MAPK expression in CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils (left) and quantification of MFI values of phosphorylated p38 MAPK (right). (D) CXCL1-induced migration of purified neutrophils incubated with PBS, IL-4 plus DMSO, IL-4 plus SB203580, or IL-4 plus LY294002. (E) Flow cytometric quantification of phosphorylated p38 MAPK in CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils 15 minutes after in vivo treatment with either PBS, IL-4, or IL-4 plus p38 inhibitor SB203580. (F) WT mice were treated with PBS, G-CSFcx, G-CSFcx plus IL-4cx, or G-CSFcx plus IL-4cx plus SB203580 for 3 days. Shown are dot plots (left) and frequencies (right) of CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils from blood 16 hours after last injection. Data are representative of 2-3 independent experiments with 2-5 mice per condition each and are represented as mean ± SD. ns = not significant; *P<0.05; **P<0.01 ; ***P<0.001.

shows that migration of human neutrophils towards CXCL8 is impaired by stimulation with IL-4 or IL-13. (A and B) Human neutrophils were isolated from peripheral blood of a healthy donor and pre-treated in vitro for 1 hour with different concentrations (ng/mL) of (A) recombinant human interleukin-4 (IL-4) or (B) recombinant human IL-13, followed by migration towards a fixed concentration of recombinant human CXCL8 (IL-8; 100ng/mL) for 90 minutes. Shown are the counts of migrated neutrophils. (C-F) Representative histograms (left panels) of expression levels of the indicated IL-4 receptor (IL-4R) subunits (C) CD124 (IL-4Ra), (D) CD132 (common γ chain, yc), (E) CD213a1 (IL-13Ra1 ), and (F) CD213a2 (IL-13Ra2) on human neutrophils isolated from peripheral blood of a healthy donor at basal level (black filled histograms), indicated as to for time-point 0, and upon in vitro stimulation with recombinant human G-CSF for 24 hours (black lines). Bars (right panels) show the mean fluorescence intensity (MFI) ± standard deviation of either basal level (indicated as to for time-point 0; open bars) or upon stimulation with G-CSF (grey filled bars). Statistical significance was calculated using Student's f-test.

Examples

Neutrophil influx during bacterial skin infection is modulated by IL-4 signals

The skin of patients with atopic dermatitis usually expresses higher IL-4 concentrations, shows some degree of neutropenia, and is the target of recurrent bacterial infections. In order to assess whether a similar correlation could be observed in mice, we took advantage of the well-established murine skin infection model using Group A Streptococcus (GAS) (Nizet et al., 2001 , Nature 414, 454-457; Zinkernagel et al., 2008, Cell Host Microbe 4, 170- 178). Subcutaneous inoculation of GAS led to a large skin lesion within 48 hours, with further progression after 72 hours (Fig. 1A). This was accompanied by a prominent inflammatory swelling of the infected leg (Fig. 1 B), which was paralleled by a significant influx of CD1 1 b ⁺ Ly6G ⁺ cells (Fig. 1 C), with Ly6G being specific for neutrophils.

Of note, increasing systemic IL-4 concentrations exacerbated the skin lesion. Thus, provision of recombinant mouse IL-4 in the form of IL-4— anti-IL-4 monoclonal antibody (mAb) complexes (IL-4cx) in order to prolong IL-4's in vivo biological half-life (Boyman et al., 2006, Science 311, 1924-1927; Finkelman et al., 1993, J Immunol 151, 1235-1244) increased skin lesion size to 165% of PBS at 48 hours (Fig. 1A) and caused enhanced leg swelling at 24 hours throughout 72 hours after infection Fig. 1 B). Despite this prominent inflammation, animals receiving IL-4cx demonstrated an attenuated neutrophil response to GAS skin infection by showing lower neutrophil influx into the skin (Fig. 1C and 1 D). Conversely, bacterial loads of GAS were 4-5-times higher in the skin of mice receiving IL-4cx (Fig. 1 E).

Consistent with the effects of elevating IL-4 concentrations, decreasing systemic IL-4 by the use of a neutralizing anti-IL-4 mAb reduced cutaneous bacterial loads and skin lesion size by half at 72 hours after infection (Fig. 1A and 1 E), and decreased leg swelling from as early as

6 hours after infection (Fig. 1 B), while influx of CD1 1 b ⁺ Ly6G ⁺ neutrophils into the skin was almost doubled upon IL-4 neutralization (Fig. 1C and 1 D). Altogether, these results indicate that neutrophil skin infiltration upon bacterial cutanenous infection is affected by systemic IL- 4 concentration.

IL-4 suppresses blood neutrophilia following systemic infection

Similar to skin infection by GAS, which also led to systemic effects (Fig. S1 ), intravenous injection of Listeria monocytogenes (LM) at a high dose of 10 ⁵ colony-forming units (cfu) led to systemic infection causing prominent blood neutrophilia (Fig. 2A and 2B). LM infection- induced blood neutrophilia was dependent on G-CSF secretion. Thus, LM infection caused a significant surge in G-CSF concentration, as measured using serum of LM-infected mice to stimulate a G-CSF-sensitive cell line, the proliferation of which was abrogated by using a neutralizing anti-G-CSF mAb in vivo or in vitro. Moreover, administration of a neutralizing anti-G-CSF mAb prior and during LM infection of mice significantly reduced blood and spleen neutrophilia to almost initial counts (Fig. 2A and 2B). Pretreatment of LM-infected mice with IL-4cx prevented blood neutrophilia and led to blood neutrophil frequencies as seen in uninfected mice (Fig. 2C and 2D). We observed similar findings in the blood of GAS-infected animals. These results suggest that IL-4 signals oppose G-CSF action on neutrophils, although a role of other cytokines, including granulocyte macrophage colony-stimulating factor, cannot be excluded.

IL-4 antagonizes G-CSF effects on neutrophils

To determine whether IL-4 mediated its effects on neutrophils during GAS and LM infection by affecting G-CSF, or rather other factors produced upon infection, we induced 'sterile' neutrophilia by injection of recombinant mouse G-CSF. As for most other soluble cytokines, G-CSF has a short in vivo half-life, unless it is coupled to polyethylene glycol or a particular anti-G-CSF mAb, thus forming G-CSFcx with extended biological half-life (Rubinstein et al., 2013, J Hematol Oncol 6, 75), similar to IL-4cx. Administration of G-CSFcx caused a prominent increase in CD1 1 b ⁺ Ly6G ⁺ neutrophils in BM, blood, and spleen (Fig. 3A and 3B). Notably, G-CSFcx-mediated neutrophilia was abrogated by concomitant injection of IL-4cx Fig. 3A and 3B). Moreover, administration of IL-4cx alone led to a slight decrease of neutrophil percentages and counts in the assessed compartments (Fig. 3A and 3B). Similar results were obtained using unmodified cytokines, that is G-CSF and IL-4 not complexed with anti-G-CSF and anti-IL-4 mAbs, respectively, although these unmodified cytokines had to be injected at higher doses to account for the cytokines' short in vivo half-life (Fig. 3C).

We next assessed whether IL-4 was able to antagonize G-CSF during systemic infection. To this end, mice received a lethal dose of 10 ⁵ cfu LM intravenously, leading to high bacterial titers in the spleen already 24 hours later and in the liver at 72 hours after infection (Fig. 3D). This was accompanied by rapid weight loss and death of animals around day 4 after infection (Fig. 3E and 3F). Notably, injection of G-CSFcx rapidly and significantly lowered bacterial loads in spleen and liver of infected mice (Fig. 3D), reversed the weight loss and prevented death of animals (Fig. 3E and 3F). However, co-administration of IL-4cx completely abrogated the beneficial effects G-CSFcx, and these mice succumbed to lethal LM infection with a similar kinetic as phosphate-buffered saline (PBS)-treated control mice (Fig. 3D-3F). Altogether, these data demonstrate that IL-4 signals dominantly suppress G-CSF-mediated neutrophil expansion in BM, blood, and spleen, both during homeostasis and systemic infection.

IL-4 acts directly on neutrophils via type II IL-4R

IL-4 exerts its pleotropic effects by binding to two types of IL-4Rs, both of which subsequently lead to phosphorylation of signal transducer and activator of transcription 6 (STAT6). Heterodimerization of IL-4Ra (also termed CD124) with the common γ-chain (y _c , also known as CD132) forms the type I IL-4R, which is expressed typically by hematopoietic cells, including B and T cells. Conversely, type II IL-4Rs are thought to be predominant on non-hematopoietic cells as well as macrophages and consist of heterodimers of IL-4Ra and IL-13Ra1 . Notably, also IL-13 can signal via type II IL-4Rs by binding to IL-13Ra1 and subsequently IL-4Ra.

CD1 1 b ⁺ Ly6G ⁺ neutrophils isolated from the BM of control mice showed low expression of IL- 4Ra, background expression of y _c and IL-13Ra1 (Fig. 4A). Upon administration of G-CSFcx to mice, IL-4Ra significantly increased by 8.5-fold along with a marked 3.8-fold increase in expression of IL-13Ra1 , both as measured by mean fluorescence intensity (MFI), whereas y _c remained unchanged (Fig. 4A). The addition of IL-4cx to G-CSFcx treatment of mice did not alter the effects of G-CSFcx on upregulation of these receptor subunits (Fig. 4A). In comparison to BM neutrophils, their splenic counterparts showed similar expression of IL- 4Ra and y _c during steady-state, whereas IL-13Ra1 expression was more prominent already on resting neutrophils from spleen compared to BM (Fig. 4B). Injection of G-CSFcx to animals led to the appearance of IL-4Ra ^h ' ^9h CD1 1 b ⁺ Ly6G ⁺ neutrophils in the spleen and further increased IL-13Ra1 on splenic neutrophils, whereas a change in y _c was not evident (Fig. 4B). As with BM neutrophils, co-administration of G-CSFcx with IL-4cx did not lead to decreased expression of these receptor subunits but, if anything, addition of IL-4cx slightly enhanced IL-4Ra and IL-13Ra1 expression on splenic neutrophils (Fig. 4B). These results suggest that G-CSF signals, as occurring during inflammation and infection, dynamically upregulate type II IL-4Rs on BM and spleen neutrophils.

Further evidence for the involvement of the type II IL-4R came from studies of STAT6 phosphorylation in mature neutrophils from spleen. Thus, we stimulated splenocytes of wild- type (WT), IL-4Ra-deficient (H4ra ^~ ' ^~ ), y _c -deficient (//2rg ^"/_ ), and IL-13Ra1-deficient (H13ra1 ^~ ' ^~ ) mice for 15 minutes (corresponding to the timepoint of maximum STAT6 phosphorylation; Fig. S3) with IL-4 to assess phospho-STAT6 expression. Phosphorylation of STAT6 in CD1 1 b ⁺ Ly6G ⁺ neutrophils occurred efficiently in WT and H2rg ^~ ' ^~ neutrophils, whereas H4ra ^~ ' ^~ and H13ra ' ^~ neutrophils failed to phosphorylate STAT6 (Fig. 4C). Although IL-13 is also able to bind the type II IL-4R, 10-100-fold higher doses of IL-13 were necessary to reach phospho-STAT6 expression in neutrophils seen with 5 ng/ml IL-4 (Fig. 4D).

The above-mentioned data demonstrate that functional type II IL-4Rs consisting of IL-4Ra and IL-13Ra1 are prominently expressed on BM and splenic neutrophils upon G-CSF signals, whereas the expression of type I IL-4Rs on neutrophils is negligible in this context. This suggests that IL-4 might - by triggering type II IL-4Rs directly, rather than indirectly via lymphocytes - act on neutrophils during inflammation and infection. Along these lines, IL-4- mediated inhibition of G-CSFcx-induced neutrophilia was independent of B and T cells, as demonstrated by using animals deficient in the recombination-activating gene 1 (Rag1). Moreover, while the vigorous expansion of CD1 1 b ⁺ Ly6G ⁺ neutrophils in blood and spleen by treatment with G-CSFcx was antagonized by co-injection of IL-4cx in WT mice (Fig. 4E and 4F, and Fig. 3), use of H4ra ^~ ' ^~ mice completely abrogated the effects of IL-4cx (Fig. 4E). Conversely, IL-4cx were able to fully counteract the effects of G-CSFcx even in H2rg ^~ ' ^~ mice, similar to the extent seen in WT (Fig. 4F).

To obtain further evidence of direct IL-4R signaling on neutrophils in an IL-4R-proficient environment, we generated BM chimeras using a 1 :1 mix of Ly5.2 (CD45.2)-congenic H4ra ^~ ' ^~ and Ly5.1 (CD45.1 )-congenic WT BM adoptively transferred to lethally irradiated H4ra ^~ ' ^~ mice (Fig. 4G). Upon reconstitution at week 3, BM-chimeric mice were treated using PBS, G- CSFcx, or G-CSFcx plus IL-4cx. Whereas the ratio of Ly5.2 H4ra ^~ ' ^~ to Ly5.1 WT CD1 1 b ⁺ Ly6G ⁺ neutrophils in blood was 1 .25 in PBS-treated animals and changed only minimally in mice receiving G-CSFcx, co-administration of G-CSFcx plus IL-4cx caused a marked change and led to a ratio of Ly5.2 H4ra ^~ ' ^~ to Ly5.1 WT neutrophils of 3 (Fig. 4G), showing that IL-4 acted directly on neutrophils to inhibit their expansion.

To explore the functional consequence of IL-4Ra deficiency during LM infection, we intravenously injected 10 ⁵ cfu LM to mice, which was lethal for WT, whereas H4ra ^~ ' ^~ survived this dose of infection (Fig. 4H). Altogether, these data demonstrate that IL-4 directly acts on neutrophils via type II IL-4Rs to induce STAT6 signaling and inhibit G-CSF-mediated neutrophil expansion during inflammation and infection.

IL-4 induces a BM-resident, non-migratory phenotype of neutrophils

Our results so far suggest that, in the presence of cell-intrinsic type II IL-4R-mediated signals, neutrophil accumulation in the blood, spleen and peripheral sites such as the skin was reduced. Neutrophil migration is controlled by the chemokine receptors CXCR2 and CXCR4, with CXCR2 mediating BM egress and migration to inflamed sites, whereas CXCR4 serves to keep neutrophils in the BM. Surface staining for CXCR2 and CXCR4 on BM neutrophils revealed that injection of G-CSFcx to mice nearly doubled expression of CXCR2, while CXCR4 decreased to 28% of PBS (Fig. 5A and 5B), thus favoring BM egress of neutrophils. In splenic neutrophils G-CSFcx treatment did not significantly affect neither of these two chemokine receptors (Fig. 5C and 5D).

Conversely, administration of IL-4cx increased CXCR4 in BM neutrophils with CXCR2 remaining low or slightly decreasing (Fig. 5A and 5B), thus promoting BM retention of neutrophils. Moreover, in splenic neutrophils, IL-4cx significantly reduced CXCR2 expression to 8% of PBS (Fig. 5C and 5D), which disfavors migration of neutrophils towards the CXCR2-binding chemokines CXCL1 and CXCL2.

IL-4 inhibits CXCR2-mediated migration in vitro and in vivo

We next investigated the effects of IL-4 during neutrophil migration by purifying CD1 1 b ⁺ Ly6G ⁺ BM neutrophils, followed by assaying their migration towards CXCL1 and CXCL2 in vitro. In the presence of IL-4, a significant decrease of neutrophil migration towards these chemokines was evident, which could not be overcome by shorter or longer times of migration or by increasing the dose of the chemokine (Fig. 6A and 6B, and data not shown). IL-4 decreased migration towards these chemokines in a dose-dependent manner (Fig. 6C and 6D). In contrast to IL-4, other y _c or T helper 2 cell-type cytokines, including IL-2, IL-7, IL- 13 and IL-15, did not reduce CXCR2-induced neutrophil migration (Fig. 6E). Consistent with our IL-4R expression data (Fig. 4), the inhibitory effect of IL-4 on neutrophil migration was abrogated in neutrophils from H4ra ^~ ' ^~ animals (Fig. 6F).

To assess the effects of IL-4 in a model of neutrophil migration in vivo, we used the well- established airpouch model (Perretti et al. (1995), Br J Pharmacol 116, 2251-2257; Ryckman et al. (2003), Arthritis Rheum 48, 2310-2320). Neutrophil migration to an airpouch containing monosodium urate crystals (MSU) was crucially dependent on CXCR2. Thus, in contrast to their WT counterparts, Cxcr2 ^~ ' ^~ neutrophils were unable to migrate and accumulate in the MSU airpouch, although Cxcr2 ^~ ' ^~ mice contained even above-normal percentages of blood neutrophils compared to WT animals (Fig. 6G). Having established that neutrophil migration to an MSU-containing airpouch relied on CXCR2, we treated animals with either PBS or IL- 4cx, prior to assessing neutrophil migration to MSU or IL-Ι β. In both experimental conditions, increased systemic IL-4 concentration significantly inhibited the accumulation of neutrophils in the airpouch (Fig. 6H). Hence, IL-4 signals potently dampen migration of neutrophils towards CXCR2-binding chemokines. IL-4 signaling interferes with p38 MAPK-PI3K crosstalk

Neutrophil chemotaxis is intracellular^ regulated by two antagonistic signaling pathways, involving phosphoinositide 3-kinase (PI3K) and p38 mitogen-activated protein kinase (MAPK). CXCR2 binding activates PI3K to cause migration of neutrophils. Conversely, once neutrophils arrive to a site of inflammation and infection, so-called end-target chemoattractants, such as N-formyl-Met-Leu-Phe (fMLP) from bacteria or complement factor C5a, stimulate the p38 MAPK pathway, which overrides PI3K-mediated signals. This in turn allows neutrophils to be guided away from CXCL1 and CXCL2 gradients towards end-target chemoattractants. Consistent with this notion, PI3K was crucial for in vitro chemotaxis of purified CD1 1 b ⁺ Ly6G ⁺ neutrophils towards CXCR2-binding chemokines, as shown by using PI3K inhibitor LY294002 (Fig. 7A). In contrast, addition of the selective ρ38αβ inhibitor SB203580 did not hamper neutrophil migration towards CXCR2-binding chemokines (Fig. 7A).

IL-4 signaling has previously been shown to activate p38 MAPK in human polymorpho- nuclear cells, including neutrophils ( Ratthe et al. (2007), J Leukoc Biol 81, 1287-1296). We thus hypothesized that the inhibitory effect of IL-4 on neutrophil migration might rely on p38 MAPK-mediated interference with PI3K signaling. Stimulation of purified CD1 1 b ⁺ Ly6G ⁺ neutrophils with IL-4 in vitro caused phosphorylation of p38 MAPK within 5 minutes, which further increased after 15 minutes (Fig. 7B). Concomitant use of the selective ρ38αβ inhibitor SB203580 completely abrogated the appearance of phospho-p38 MAPK in neutrophils (Fig. 7C). This indicates that IL-4 induces phosphorylation of the p38a MAPK family member in neutrophils, as only p38a and ρ38δ have been found in neutrophils, whereas ρ38β was absent in these cells (Hale et al. (1999), J Immunol 162, 4246-4252).

In line with these findings, neutrophil migration in vitro towards CXCL1 was dampened by IL- 4, and IL-4-mediated inhibition was abrogated by addition of ρ38αβ inhibitor SB203580 (Fig. 7D). Conversely, PI3K inhibitor LY294002 did not affect IL-4-induced inhibition of neutrophil migration (Fig. 7D).

Injection of IL-4 to mice also caused p38a MAPK phosphorylation of neutrophils 15 minutes after injection, which was inhibited by co-administration of ρ38αβ inhibitor SB203580 (Fig. 7E). Treatment of mice with G-CSFcx induced blood neutrophilia, which was antagonized by co-injection of IL-4cx (Fig. 7F and 7G), as also shown in Fig. 3. Notably, simultaneous administration of SB203580 was able to block IL-4's inhibitory effect on neutrophils, thus allowing neutrophils to accumulate in the blood in response to G-CSFcx (Fig. 7F and 7G). These data demonstrate that IL-4 signaling in neutrophils activates p38a MAPK thus overriding G-CSF-CXCR2-PI3K-mediated signals and inhibiting neutrophil recruitment and migration. DISCUSSION

The data of the present specification show that IL-4 inhibited neutrophil recruitment and migration during bacterial infection and inflammation via directly interacting with type II IL- 4Rs on neutrophils. Thus, increased G-CSF production upon local and systemic bacterial infection or 'sterile' inflammation as mimicked by injection of long-lasting G-CSF usually expands BM neutrophils. Moreover, G-CSF is known to induce neutrophil egress from the BM by weakening CXCF -mediated retention signals and augmenting the neutrophils' sensitivity towards CXCR2-binding chemokines. As shown here, G-CSF signals also increased the expression of type II IL-4Rs, made of IL-4Ra and IL-13Ra1 , rendering neutrophils more sensitive to IL-4. Accordingly, increasing IL-4 concentration, either physiologically during infection and inflammation or by administration of IL-4, directly bound to and activated in neutrophils STAT6 and p38 MAPK and led to up-regulation of CXCR4 as well as downregulation of CXCR2. Hence, IL-4 signalling in neutrophils resulted in decreased G-CSF-mediated expansion and increased retention of neutrophils in the BM. Moreover, in addition to decreasing CXCR2 expression, IL-4 signals also inhibited neutrophil migration by opposing CXCR2-PI3K-mediated signals via the activation of the PI3K antagonist p38 MAPK. Thus, type II IL-4R-mediated signaling in neutrophils efficiently interferes with several central and peripheral mechanisms of neutrophil recruitment.

In placing these data in the context of previous findings, the literature on the effects of IL-4 on neutrophils is rather controversial. In freshly-isolated human neutrophils in vitro, IL-4 prolongs survival, increases phagocytosis and killing, facilitates lysozyme production, β- glucuronidase secretion and fMLP-mediated respiratory burst, and enhances migration towards zymosan-activated serum and IL-5 ( Bober et al., (1995). Clin Exp Immunol 99, 129- 136; Ratthe et al. (2007), J Leukoc Biol 81, 1287-1296). The results pertaining to fMLP are actually in line with the model proposed herein and could possibly be explained by the finding that fMLP (and other end-target chemoattractants) activate p38 MAPK; thus, IL-4 could synergize with fMLP to stimulate p38 MAPK in neutrophils in this context. As for survival, in our hands, IL-4 did not alter the viability of freshly-isolated murine BM neutrophils (data not shown). Conversely, another study shows that upon stimulation with interferon-γ (IFN-γ) or tumor necrosis factor-a, in vitro phagocytic activity of human neutrophils is inhibited by IL-4 (Bober et al. (2000), Arthritis Rheum 43, 2660-2667).

In light of our data, these findings raise the following questions. Do resting and cytokine- activated or human and mouse neutrophils express different types of IL-4Rs? And do type I versus type II IL-4Rs on neutrophils mediate different effects? Notably, one of the studies shows that resting human neutrophils express type I IL-4Rs, but lack IL-13Ra1 Ratthe et al. (2007), J Leukoc Biol 81, 1287-1296). Our findings in mice demonstrate that IL-13Ra1 is found on BM and particularly splenic neutrophils and its expression is further enhanced by G- CSF, which also upregulates IL-4Ra on neutrophils.

The above-mentioned data refer to the in vitro use of IL-4 on human neutrophils. The effects of IL-4 on neutrophils in vivo reported in the literature are difficult to interpret in terms of direct and indirect actions. In an a-galactosylceramide-induced mouse hepatitis model, early IL-4 production by natural killer T cells increases neutrophil infiltrates and inflammation in the liver, whereas hepatic neutrophil counts are curtailed by later IFN-γ secretion by natural killer T cells (Wang et al. (2013), Hepatology 58, 1474-1485), indicating that IL-4 signals enhanced neutrophil accumulation in the liver. Conversely, other findings suggest that IL-4 signals dampen neutrophil or myeloid cell recruitment, which is in line with our herein described mechanism. Thus, in mice infected with Schistosoma japonicum, neutrophil recruitment to the liver is increased in animals lacking both IL-4 and IL-13. Also, in animals sensitized with house dust mite extract, inhibition of IL-4 by using a neutralizing anti-IL-4 mAb leads to increased neutrophil counts in the lungs Choy et al. (2015), Sci Transl Med 7, 301 ra129). Moreover, intravenous injection of IL-4 is able to hamper IL-i p-mediated recruitment of neutrophils to an airpouch, although the mechanism by which IL-4 exerted its effect has been unclear and IL-4 is unable to inhibit CXCL8-mediated neutrophil chemotaxis in this publication (Perretti et al. (1995), Br J Pharmacol 116, 2251-2257). Furthermore, systemic administration of IL-4 improves joint inflammation in three different arthritis models, including the rat adjuvant arthritis model, collagen-induced arthritis model in mice, and K/BxN-mediated joint inflammation in mice (Bober et al., 2000 (ibid.); Hemmerle et al. (2014), Proc Natl Acad Sci U S A 111, 12008-12012; Wermeling et al. (2013), Proc Natl Acad Sci U S A 110, 13487-13491 ), as well as curtailed delayed-type pleuritis in mice (Fine et al. (2003), Inflammation 27, 161-174). However, as some of these models rely on a late readout between days 17 and 22 and a sensitization and challenge procedure similar to delayed-type hypersensitivity, it is conceivable that in the reported models IL-4 mediates its effects via dampening T helper 1 cell responses. Hence, the previously reported in vivo findings show a rather heterogeneous picture of the effects of IL-4 on neutrophils and leave the question unanswered as to whether IL-4 affected neutrophils directly or via its action on other immune cells.

Of note, our results indicate that type II IL-4R signaling antagonizes some of the actions of G-CSF on BM neutrophils, particularly those effects of G-CSF pertaining to expansion and egress of BM neutrophils. Previous reports showed that G-CSF exerted both STAT3- dependent and STAT3-independent effects in neutrophils and also stimulated suppressor of cytokine signaling 3 (SOCS3), which in turn acts as an important negative feedback regulator of G-CSF receptor signaling by inhibiting STAT3 (Carow and Rottenberg, 2014, Front Immunol 5, 58; Nguyen-Jackson et al., 2010, Blood 115, 3354-3363). Thus it is conceivable that type II IL-4R signaling might interfere directly, or indirectly via SOCS3, with one of the pathways downstream of the G-CSF receptor.

The therapeutic implications of the herein described type II IL-4R-p38 MAPK-CXCR2 axis are manifold. As mentioned previously, individuals suffering from allergic disorders show a relative paucity of neutrophils in the affected organs. For example, the skin of patients with atopic dermatitis contains more IL-4 and lower counts of neutrophils and is more susceptible to bacterial infections that are usually contained by neutrophils. Thus, therapeutic approaches targeting IL-4Ra, IL-13Ra1 or p38 MAPK might lower the risk of recurrent bacterial infections.

Likewise, patients undergoing myeloablative chemotherapy go through a phase of critical neutropenia and severely increased risk of systemic infections. Currently, long-lasting G-CSF formulations, such as PEGylated G-CSF, are given to patients in order to stimulate production and recruitment of neutrophils. Based on our data, concomitant inhibition of type II IL-4R-p38 MAPK signalling might be considered for severe cases.

IL-4 is well known to directly affect many immune cells, including macrophages, dendritic cells, B cells and T cells, thereby driving type 2 cell-mediated immunity. By showing a direct action of IL-4 on neutrophils via type II IL-4Rs we extend the range of target cells of IL-4 and also demonstrate that type II IL-4Rs can become very prominent on immune cells and potently affect their responses during immunity and immunopathology.

EXPERIMENTAL PROCEDURES

Animals

C57BL/6, CD45.1 (Ly5.1 )-congenic, Cxcr2 ^~ ' ^~ , l\2rg ^~ ' ^~ , Rag1 ^~ ' ^~ (all on a C57BL/6 background), and Balb/c and H4ra ^~ ' ^~ (on a Balb/c background) were purchased from The Jackson Laboratory (Bar Harbor). H13ra1 ^~ ' ^~ (on a Balb/c background) were provided by Regeneron Pharmaceuticals (Ramalingam et al., 2008, Nat Immunol 9, 25-33). Experiments were approved by the Cantonal Veterinary Office and performed in accordance with Swiss law.

Infections and in vivo treatments

Mice were infected either systemically with 10 ⁵ cfu Listeria monocytogenes (LM) or subcutaneously with 3x10 ⁷ cfu Group A Streptococcus (GAS) M 1 as previously described (Nizet et al., 2001 (ibid.); Zinkernagel et al., 2008 (ibid.)). Where indicated, animals received over three consecutive days daily injections of PBS, free cytokines (5 pg human G-CSF [Neupogen]; 7.5 pg mouse IL-4 [mlL-4, eBioscience]), or cytokine-anti-cytokine monoclonal antibody (mAb) complexes (1 pg human G-CSF complexed with 6 pg anti-human G-CSF mAb clone BVD1 1-37G10, SouthernBiotech; 1 .5 pg mlL-4 complexed with 7.5 pg anti-mlL-4 mAb clone 1 1 B1 1 , BioXcell) prior to infection, as previously published (Boyman et al., 2006 (ibid.); Finkelman et al., 1993 (ibid.); Rubinstein et al., 2013 (ibid.)). Also, where indicated, mice received daily intraperitoneal injections of 100 μ9 neutralizing mAb against mG-CSF (MAB414; R&D) or mlL-4 (1 1 B1 1 ; BioXcell). p38 MAPK activity was blocked in vivo by administering mice three injections of 300 μg of SB203580 (Calbiochem) (Heit et al., 2002 Nat Immunol 9, 743-752).

Bacterial load

To determine LM bacterial load, liver was flushed with cold PBS, and processed liver and spleen were incubated for 20 minutes with 0.05% Triton X-100 (Sigma). Serial dilutions were plated on brain heart infusion agar (Oxoid) plates and cfu were counted after 24 hours of incubation at 37°C. To determine GAS load, skin homogenates were centrifuged and serial dilutions of supernatants were plated on THY agar (Oxoid) plates.

Immunohistochemistry

Murine skin samples were embedded in O.C.T. Compound (Sakura) and stained using anti- Ly6G mAb (1A8; BioXcell). Sections were analyzed using ImageScope for image acquisition (Aperio Technologies, Inc.).

Flow cytometry

Single cell suspensions of organs were prepared according to standard protocols and stained for analysis by flow cytometry, as previously published (Bouchaud et al. O. (2013), J Exp Med 210, 2105-21 17) using fluorochrome-labeled mAbs directed against the following mouse antigens (from eBioscience, unless stated otherwise): Annexin V (BD Biosciences), CD3 (145-2C1 1 ; BD Biosciences), CD4 (GK1.5), CD1 1 b (M1/70), CD45.1 (A20), CD45.2 (104), Ly6G (1A8), CD132 (TUGm2; BioLegend), CXCR2 (TAB2164P; R&D), CXCR4 (2B1 1 ; BD Biosciences), IL-4Ra (mll_4R-M1 ; BD Biosciences), IL-13Ra1 (13MOKA), phospho- STAT6 (pY641 ; BD Biosciences), and phospho-p38 (pT180, pY182; BD Biosciences). For in vitro phosphostainings splenocytes were stimulated with cytokines for 15 minutes and subsequently fixed by addition of paraformaldehyde and ice-cold methanol. Skin was cut into small pieces and incubated for 1 hour at 37°C with an enzymatic cocktail consisting of 5 pg/ml Liberase™ (Roche), 1 pg/ml DNAase I (Sigma) and 5 pg/ml Dispase II (Roche) in RPMI media. Subsequently, cells were liberated by extensive pipetting and filtered. Cells were acquired on a BD FACSCanto™ II or BD LSR Fortessa flow cytometer and analyzed using FlowJo software (Tristar Inc.).

Bone marrow chimeras

Immune lineage-negative (Lin ^" ) bone marrow (BM) cells of WT CD45.1 -congenic and H4ra ^~ ' ^~ CD45.2-congenic mice were purified by negative selection using magnetic beads (StemCell Technologies) and biotinylated mAbs against CD19, CD3, MHC class II, NK1.1 , and Ter1 19. Lin ^" BM cells from WT and H4ra ^~ ' ^~ mice were mixed at a 1 :1 ratio and injected intravenously into irradiated (950 rad) H4ra ^~ ' ^~ CD45.2-congenic host mice. BM chimeric mice received 1 mg/ml sulfamethoxazol and 0.2 mg/ml trimethoprim in their drinking water for two weeks and were left for three weeks in order to allow for reconstitution of neutrophils before use.

Neutrophil migration assay

CD3 ^" CD1 1 b ⁺ Ly6G ⁺ neutrophils were obtained by positive selection using Ly6G microbeads (Miltenyi Biotec), yielding a purity of 92-95%. 10 ⁵ purified neutrophils were pretreated for 20 minutes with PBS or cytokines (30 ng/ml), including IL-2, IL-4, IL-7, IL-13, or IL-15 (from eBioscience and Peprotech), before seeding into the upper chamber of a 5 μm-transwell (Corning Costar). Subsequently, migration of neutrophils was determined for 2 hours towards CXCL1 or CXCL2 (both 100 ng/ml; PeproTech) given to the lower chamber. Where indicated, SB203580 or Ly294002 (both 30 μΐηοΙ; Calbiochem) was added 15 minutes before pretreatment. Airpouch model

An airpouch was generated in the back of mice by subcutaneous injection of 4 ml sterile air on days 0 and 3, as previously described (Perretti et al. (1995), Br J Pharmacol 116, 2251- 2257; Ryckman et al. (2003), Arthritis Rheum 48, 2310-2320). On day 6, we administered intravenously PBS or IL-4cx, followed by injection into the airpouch of either 1 mg monosodium urate crystals (MSU) or 10 ng IL-1 β (Peprotech) in 1 ml sterile PBS. Mice were left overnight and subsequently euthanized, followed by flushing of the airpouch with PBS plus 2 mM EDTA to collect cells within. The content of the airpouch was counted and analyzed by flow cytometry for leukocytes.

Statistical analysis

Differences between groups were examined for statistical significance by using unpaired Student's f-test or one-way or two-way analysis of variance (ANOVA) with Bonferroni's post- test correction.