NO338853B1 - Isolated specific binding member for human IL-13, isolated antibody VH and VL domain, composition containing them and methods for their preparation and use, and isolated nucleic acid and host cell. - Google Patents

Description

The present invention relates to specific binding members, especially human anti-IL-13 antibody molecules, and especially those that neutralize IL-13 activity. It further relates to anti-IL-13 antibody molecules for use in the diagnosis or treatment of IL-13 related disorders, including asthma, atopic dermatitis, allergic rhinitis, fibrosis, inflammatory bowel disease and Hodgkin's lymphoma.

Preferred embodiments according to the present invention use the antibody VH and/or VL domain of antibody molecules which are here called BAK502G9. Furthermore, preferred embodiments employ complementary determinant regions (CDRs) from BAK502G9, particularly the VH CDR3 in other antibody framework regions. Further aspects of the present invention provide compositions containing specific binding members of the invention.

The present invention provides antibody molecules of particular value in binding and neutralizing IL-13, and thus of use in any of a variety of therapeutic treatments, as indicated by the experimentation contained herein and further by the supporting technical literature.

Interleukin (IL)-13 is a 114 amino acid cytokine with an unmodified molecular mass of approximately 12 kDa [1,2]. IL-13 is most closely related to IL-4 with which it shares 30% sequence similarity at the amino acid level. The human IL-13 gene is located on chromosome 5q31 adjacent to the IL-4 gene. [1] [2]. This region of chromosome 5q contains gene sequences for other Th2 lymphocyte-derived cytokines including GM-CSF and IL-5 whose levels together with IL-4 have been shown to correlate with disease severity in asthmatics and rodent models of allergic inflammation [3] [4] [5] [6] [7] [8].

Although initially identified as a Th2 CD4+ lymphocyte-derived cytokine, IL-13 is also produced by Th1 CD4+ T cells, CD8+ T lymphocyte NK cells, and non-T cell populations such as mast cells, basophils, eosinophils, macrophages, monocytes and smooth muscle cells in the airways.

IL-13 has been reported to direct its effects through a receptor system that includes the IL-4 receptor oc chain (IL-4Roc), which itself can bind IL-4 but not IL-13, and at least two other cell surface proteins, IL- 13Ra2 [9] [10]. IL-13Rα1 can bind IL-13 with low affinity, and then recruit IL-4Rα to form a high-affinity functional receptor that signals [11] [12]. The Genbank database lists the amino acid sequence and nucleic acid sequence of IL-13Rocl as NP 001551 and Yl 0659, respectively. Studies in STAT6 (signal transducer and activator of transcription 6)-deficient mice have revealed that IL-13, in a similar manner to IL-4, signals by using the JAK-STAT6 reaction pathway [13] [14]. IL-13Roc2 shares 37% sequence identity with IL-13Ral at the amino acid level and binds IL-13 with high affinity [15] [16]. However, IL-13Roc2 has a shorter cytoplasmic tail that lacks known signaling motifs. Cells expressing IL-13Roc2 do not respond to IL-13 even in the presence of IL-4Rα

[17]. It is therefore postulated that IL-13Roc2 acts as a shutter receptor that regulates IL-13 but not IL-4 function. This is supported by studies in IL-13Roc2 deficient mice whose phenotype was consistent with an increased response to IL-13 [18] [19]. The Gengank database lists the amino acid sequence and nucleic acid sequence of IL-13Roc2 as NP_000631 and Y08768, respectively.

The signaling IL-13Rocl/IL-4Ra receptor complex is expressed on human B cells, mast cells, monocytes/macrophages, dendritic cells, eosinophils, basophils, fibroblasts, endothelial cells, airway epithelial cells and airway smooth muscle cells.

Bronchial asthma is a common persistent inflammatory disease of the lung that is characterized by airway hypersensitivity, mucus overproduction, fibrosis and increased serum IgE levels. Airway hypersensitivity (AHR) is the exaggerated contraction of the airways to non-specific stimuli such as cold air. Both AHR and mucus overproduction are thought to be responsible for the variable airway obstruction that leads to the shortness of breath that is characteristic of asthma attacks (exacerbations) and that is responsible for the mortality associated with this disease (around 2000 deaths/year in the UK).

The incidence of asthma, together with other allergic diseases, has increased significantly in recent years [20] [21]. For example, at the present time around 10% of the population in Great Britain (UK) has been diagnosed as asthmatic.

Current guidelines from the British Thoracic Society (BTS) and the Global Initiative for Asthma (GINA) suggest a stepwise approach to the management of asthma [22, 23]. Mild to moderate asthma can usually be controlled by the use of inhaled corticosteroids, in combination with beta-agonists or leukotriene inhibitors. However, due to the documented side effects of corticosteroids, patients tend not to adhere to treatment regimens that reduce the effectiveness of treatment [24-26]. There is a clear need for new treatments for individuals with more severe disease, who often derive very limited benefit from either higher doses of inhaled or oral corticosteroids recommended by asthma guidelines. Long-term treatment with oral corticosteroids is associated with side effects such as osteoporosis, reduced growth rate in children, diabetes and oral candidiasis [88]. As both beneficial and severe effects of corticosteroids are controlled via the same receptor, treatment is a balance between safety and effectiveness. Hospitalization of these patients, who represent around 6% of the UK asthma population, as a result of severe exacerbations account for the bulk of the significant economic burden of asthma on health authorities [89].

It is believed that the pathology of asthma is caused by ongoing Th2 lymphocyte-driven inflammation resulting from inappropriate responses of the immune system to innocuous antigens. Evidence has been mounting implicating IL-13 rather than the classical Th2-derived cytokine IL-4 as the key mediator in the pathogenesis of established airway disease.

Administration of recombinant IL-13 to the airways of naïve non-sensitized rodents caused many aspects of the asthma phenotype including airway inflammation, mucus production and AHR [27] [28] [29] [30] . A similar phenotype was observed in a transgenic mouse in which IL-13 was specifically overexpressed in the lung. In this model, more chronic exposure to IL-13 also resulted in fibrosis [31].

In rodent models with allergic disease, many aspects of the asthma phenotype have also been associated with IL-13. Soluble mouse IL-13Rα2, a potent IL-13 neutralizer, has been shown to inhibit AHR, mucus hypersecretion and the influx of inflammatory cells characteristic of this rodent model [27] [28] [30]. In complementary studies, mice in which the IL-13 gene had been deleted failed to develop allergen-induced AHR. AHR could be recovered in these IL-13 deficient mice by the administration of recombinant IL-13.1 in contrast, IL-4 deficient mice developed respiratory disease in this model [32] [33].

By using a long-term allergen-induced lung inflammation model, Taube et al. the effectiveness of soluble mouse IL-13Rα2 against established respiratory disease [34]. Soluble mouse IL-13Roc2 inhibited AHR, mucus overproduction and, to a lesser extent, airway inflammation. In contrast, soluble IL-4Rα, which binds and antagonizes IL-4, had little effect on AHR or airway inflammation in this system [35]. These findings were supported by Blease et al. who developed a chronic fungal model of asthma where polyclonal antibodies against IL-13 but not IL-4 were able to reduce mucus overproduction, AHR and subepithelial fibrosis [36].

A number of genetic polymorphisms in the IL-13 gene have also been linked to allergic disease. In particular, a variant of the IL-13 gene where the arginine residue at amino acid 130 is substituted with glutamine (R130Q) has been associated with bronchial asthma, atopic dermatitis and increased serum IgE levels [37] [38] [39] [40]. This particular IL-13 variant is also referred to as the R110Q variant (the arginine residue at amino acid 110 is substituted with glutamine) by some groups that exclude the 20 amino acid signal sequence from the amino acid count. Arima et al, [41] reported that this variant is associated with increased levels of IL-13 in serum. The IL-13 variant (R130Q) and antibodies against this variant are discussed in WO 01/62933. An IL-13 promoter polymorphism, which alters IL-13 production, has also been associated with allergic asthma [42].

Increased levels of IL-13 have also been measured in human subjects with asthma, atopic rhinitis (hay fever), allergic dermatitis (eczema) and chronic sinusitis. For example, levels of IL-13 were found to be higher in bronchial biopsies, sputum and broncho-alveolar secretion (BAL) cells from asthmatics compared to control individuals [43] [44]

[45] [46]. Furthermore, levels of IL-13 in BAL samples in asthmatic individuals were increased upon challenge with allergen [47] [48]. The IL-13 production capacity of CD4 (+) T cells has also been shown to be a useful marker for the risk of subsequent development of allergic disease in newborns [49].

Li et al [114] have recently reported effects of a neutralizing anti-mouse IL-13 antibody in a chronic mouse model of asthma. Chronic asthma-like response (such as AHR, severe airway inflammation, mucus overproduction) was induced in OVA-sensitized mice. Li et al reported that administration of an IL-13 antibody at the time of each OVA challenge suppresses AHR, eosinophilic infiltration, serum IgE levels, proinflammatory cytokine/chemokine levels, and airway remodeling [ 14 ].

In summary, these data provide an indication that IL-13 rather than IL-4 is a more attractive target for the treatment of human allergic disease. IL-13 may play a role in the pathogenesis of inflammatory bowel disease. Heller et al.

[116] reported that neutralization of IL-13 by administration of soluble IL-13Roc2 improved inflammation in: a mouse model of human ulcerative colitis [116]. Consistent with this, IL-13 expression was higher in rectal biopsy samples from ulcerative colitis patients compared to controls [117].

WO 9404680 A1 provides a mouse antibody, a monoclonal or a chimeric antibody that specifically binds to human IL-13. Antibodies can be produced against human IL-13 proteins both in naturally occurring form and in recombinant form. The antibodies are used therapeutically in that they bind to IL-13 and inhibit this protein's biological activity.

Apart from asthma, IL-13 has been associated with other fibrotic conditions. Increased levels of IL-13, up to 1000 times higher than IL-4, have been measured in the serum of patients with systemic sclerosis [50] and BAL samples from patients affected by other forms of pulmonary fibrosis [51]. In accordance with this, overexpression of IL-13 but not IL-4 in the mouse lung resulted in pronounced fibrosis [52] [53]. The contribution of IL-13 to fibrosis in tissues other than the lung has been carefully studied in a mouse model with parasite-induced liver fibrosis. Specific inhibition of IL-13 by administration of soluble IL-13oc2 or IL-13 knockdown, but not removal of IL-4 production, prevented fibrogenesis in the liver [54]

[55] [56].

Chronic obstructive pulmonary disease (COPD) includes patient populations with varying degrees of chronic bronchitis, small airway disease and emphysema and is characterized by progressive irreversible lung function impairment that responds poorly to current asthma-based therapy [90]. The incidence of COPD has increased dramatically in recent years to become the fourth leading cause of death in the world (World Health Organization). COPD therefore represents a large unmet medical need.

The underlying causes of COPD are still poorly understood. The "Dutch hypothesis" proposes that there is a common susceptibility to COPD and asthma and therefore that similar mechanisms may contribute to the pathogenesis of both disorders [57].

Zeng et al [58] have shown that overexpression of IL-13 in the mouse lung caused emphysema, increased mucus production and inflammation, which reflects the aspect of human COPD. Furthermore, AHR, an IL-13 dependent response in mouse models of allergic inflammation, has been shown to be predictive of lung function decline in smokers [59]. A link has also been established between an IL-13 promoter polymorphism and susceptibility to developing COPD [60].

The signs are therefore that IL-13 plays an important role in the pathogenesis of COPD, especially in patients with asthma-like features including AHR and eosinophilia. mRNA levels of IL-13 have been shown to be higher in autopsy tissue samples from individuals with a history of COPD compared to lung samples from individuals who had no reported lung disease (J. Elia, Oral communication at American Thoracic Society Annual Meeting 2002). In another study, increased levels of IL-13 were shown by immunohistochemistry in peripheral lung sections from COPD patients [91].

Hodgkin's disease is a common type of lymphoma, accounting for approximately 7,500 cases per year in the United States. Hodgkin's disease is unusual among malignancies in that the neoplastic Ree-Sternberg cell, often derived from B cells, constitutes only a small proportion of the clinically detectable mass. Hodgkin's disease derived cell lines and primary Ree-Sternberg cells often express IL-13 and its receptor [61]. As IL-13 promotes cell survival and proliferation in normal B cells, it was proposed that IL-13 could act as a growth factor for Reed-Sternberg cells. Skinnider et al. have shown that neutralizing antibodies against IL-13 can inhibit the growth of Hodgkin's disease-derived cell lines in vitro [62]. This finding suggested that Reed-Sternberg cells can increase their own survival by an IL-13 autocrine and paracrine cytokine loop. Consistent with this hypothesis, increased levels of IL-13 have been detected in the serum of some Hodgkin's disease patients compared to normal controls

[63]. IL-13 inhibitors can therefore prevent disease development by inhibiting the proliferation of malignant Reed-Sternberg cells.

Many human cancer cells express immunogenic tumor-specific antigens. Although many tumors spontaneously recur, a number evade the immune system (immunosurveillance) by suppressing T-cell-directed immunity. Terabe et al. [64] have shown a role for IL-13 in immunosuppression in a mouse model where the tumors spontaneously regress after initial growth and then return. Specific inhibition of IL-13, with soluble IL-13Roc2, protected these mice from tumor recurrence. Terabe et al. [64] went on to show that IL-13 suppresses the differentiation of tumor-specific CD8+ cytotoxic lymphocytes that direct anti-tumor immune responses. IL-13 inhibitors can therefore be used therapeutically to prevent tumor relapse or metastases. Inhibition of IL-13 has been shown to enhance antiviral vaccines in animal models and may be beneficial in the treatment of HTV and other infectious diseases [65].

WO 03035847 A2 describes an isolated mammalian mutated IL-13 protein and antibodies, and isolated nucleic acids encoding a mutated IL-13 protein or antibody, and their use.

It should be noted that reference generally herein to interleukin-13 or IL-13 is, except where the context indicates otherwise, a reference to human IL-13. This is also referred to in places as the "antigen". The present invention provides antibodies against human IL-13, particularly human antibodies, which cross-react with non-human primate IL-13, including cynomolgus monkey and rhesus monkey IL-13. Antibodies in conjunction with some antibodies according to the present invention do not recognize any variant of IL-13 where the arginine residue at amino acid position 130 is replaced by glutamine.

Brief description of the figures

Figure 1 shows neutralization potency (% inhibition) of BAKI67All (closed squares) and its derivative BAK615E3 (open squares) as scFv against 25 ng/ml human IL-13 in the TF-1 cell proliferation assay. The triangles represent an irrelevant scFv. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 2 shows the neutralizing potency (% inhibition) of BAK278D6 (closed squares) and its derivative BAK502G9 (open squares) as scFv against 25 ng/ml human IL-13 in the TF-1 cell proliferation assay. The triangles represent an irrelevant scFc. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 3 shows the neutralizing potency (% inhibition) of BAK209B11 (closed squares) as an scFv against 25 ng/ml mouse IL-13 in the TF-1 cell proliferation assay. The triangles represent an irrelevant scFv. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 4 shows the neutralizing potency (% inhibition) of BAK278D6 (closed squares) as an scFv against IL-13 in the TF-1 cell proliferation assay. The triangles represent an irrelevant scFv. Data represent the mean with standard error bars of triplicate determinations within the same experiment.

Figure 4A shows potency against 25 ng/ml human IL-13.

Figure 4B shows potency against 25 ng/ml human variant IL-13.

Figure 4C shows potency against 50 ng/ml non-human primate IL-13.

Figure 5 shows a comparison of the potency of anti-human IL-13 antibodies in the TF-1 proliferation assay. The data represent the average neutralization potency with standard error bars over 5-7 experiments against 25 ng/ml human IL-13. Performance relative to the commercially available antibody, B-B13, was evaluated statistically by performing a one-way ANOVA with Dunnett's test.<*>P<0.05,<**>P<0.01 compared to B- B13. Figure 6 shows the neutralizing potency (% inhibition) of BAK502G9 (closed squares), BAKI 167F2 (closed triangles) and BAKI 183H4 (closed inverted triangles) as human IgG4 against tagged IL-13 in the TF-1 cell proliferation assay. Open triangles represent an irrelevant IgG4. The data represent means with standard error bars of three separate experiments.

Figure 6A shows potency against 25 ng/ml human IL-13.

Figure 6B shows potency against 25 ng/ml human variant IL-13.

Figure 6C shows potency against 50 ng/ml non-human primate IL-13.

Figure 7 shows the neutralization potency (% inhibition) of BAK502G9 (closed squares), BAKI 167F2 (closed triangles), BAKI 183H4 (closed inverted triangles) as human IgG4 and commercial anti-human IL-13 antibodies (B-B13 - open squares; JES10 -5A2 - open inverted triangles) in the native IL-13 dependent HDLM-2 cell proliferation assay. Open triangles represent an irrelevant IgG4. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 8 shows a comparison of the potency of anti-human IL-13 antibody in the NHLF assay. Data represent the mean neutralizing potency (IC50 pM) with standard error bars over 4-5 experiments against 10 ng/ml human IL-13 in the NHLF eotaxin release assay. Performance relative to the commercially available antibody, B-B13, was evaluated statistically by performing a one-way ANOVA with Dunnett's test.<*>P<0.05,<**>P<0.01 compared to B- Bl 3. Figure 9 shows the neutralizing potency (% inhibition) of BAK502G9 (closed squares), BAKI 167F2 (closed triangles), BAKI 183H4 (closed inverted triangles) as human IgG4 against VCAM-1 upregulation of the surface of HUVEC in response to 10 ng/ ml human IL-13. Open triangles represent irrelevant IgG4. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 10 shows the neutralizing potency (% inhibition) of BAK502G9 (closed squares) BAKI 1867F2 (closed triangles), BAKI 183H4 (closed inverted triangles) as human IgG4 against eotaxin release from VCAM-1 upregulation on the surface of HUVEC in response to either 1 ng/ ml human IL-4 (Figure 10A) or 0.5 ng/ml human IL-1β (Figure 10B). Open triangles represent an irrelevant IgG4. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 11 shows the neutralizing potency (% inhibition) of BAK209B11 (squares) as a human IgG4 against 1 ng/ml mouse IL-13 in the factor dependent B9 cell proliferation assay. Open triangles represent an irrelevant IgG4. Data represent the mean with standard error bars of triplicate determinations within the same experiment. Figure 12 shows the relative level of IL-13 in lung homogenates from sensitized (s) (right-hand bar) and non-sensitized (ns) (left-hand bar) mice after challenge in a mouse model with acute lung allergic inflammation. The effect of sensitization was statistically evaluated by performing Student's t-test using quantity of IL-13 data.<*><0.5.<**><0.01 compared to non-sensitized control animals (n=5-6 mice) . Data represent the mean with standard error bars. Figure 13 illustrates the effects of i.v. administration of BAK209B11 as human IgG4 in different amounts compared to an isotype-matched IgG4 irrelevant control antibody on ovalbumin-induced leukocyte recruitment to the lung in ovalbumin-sensitized mice. The number of leukocytes is shown (x IO<4>). The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnetfs test using differential cell count data.<*><0.05.<**><0.01 compared to ovalbumin-challenged PBS control animals (=0% inhibition; n=5-8 mice). Data represent the mean with standard error bars. Figure 14 illustrates the effects of i.v. administration of BAK209B11 as human IgG4 in different amounts compared to an isotype-matched IgG4 irrelevant control antibody on ovalbumin-induced eosinophil recruitment to the lung in ovalbumin-sensitized mice. The number of eosinophils is shown (x IO<4>). The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnett's test using differential cell count data.<*><0.05.<**><0.01 compared to ovalbumin-challenged PBS control animals (=0% inhibition; n=5-8 mice). Data represent the mean with standard error bars. Figure 15 illustrates the effects of i.v. administration of BAK209B11 as human IgG4 in different amounts compared to an isotype-matched IgG4 irrelevant control antibody on ovalbumin-induced neutrophil recruitment to the lung in ovalbumin-sensitized mice. The number of neutrophils is shown (x IO<4>). The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnett's test using differential cell count data.<*><0.05.<**><0.01 compared to ovalbumin-challenged PBS control animals (=0% inhibition; n=5-8 mice). Data represent the mean with standard error bars. Figure 16 illustrates the effects of i.v. administration of BAK209B11 as human IgG4 in different amounts compared to an isotype-matched IgG4 irrelevant control antibody on ovalbumin-induced lymphocyte recruitment to the lung in ovalbumin-sensitized mice. The induction of lymphocytes was dose-dependently inhibited by BAK209B11 with maximal inhibition at 3 µg/ml of BAK209B11. The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnett's test using differential cell count data.<*><0.05.<**><0.01 compared to ovalbumin-challenged PBS control animals (=0% inhibition; n=5-8 mice). Data represent the mean with standard error bars. Figure 17 illustrates the effects of i.v. administration of BAK209B11 as human IgG4 in different amounts compared to an isotype-matched IgG4 irrelevant control antibody on ovalbumin-induced monocyte/macrophage recruitment to the lung in ovalbumin-sensitized mice. There was no significant increase in monocyte/macrophage levels in sensitized animals compared to control animals. However, such background levels of these cells were knocked down by >36 µg/ml BAK209B11 in sensitized animals. The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnetfs test using differential cell count data.<*><0.05.<**><0.01 compared to ovalbumin challenged PBS control animals (=0% inhibition ; n=5-8 mice). Data represent the mean with standard error bars. Figure 18 shows the effects of a commercial anti-IL-13 neutralizing antibody JES 10-5A2 on the entry of cells (number of leukocytes is shown (x 10<4>)) into the mouse "air-pouch" induced by the administration of bacterially derived recombinant human IL-13. The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnetfs test using differential cell count data.<*><0.05.<**><0.01 compared to CMC control animals (=0% inhibition; n =ll-13 mice). Data represent the mean with standard error bars. Figure 19 shows a sequence alignment of cynomolgus IL-13 against human IL-13 amino acid sequences. The seven amino acid residues that distinguish human and cynomolgus IL-13 are shaded. Rhesus and cynomolgus IL-13 have an identical amino acid sequence. Figure 20 illustrates the effect of a single 10 mg/kg i.v. bolus dose of BAK502G9 as human IgG4 on serum IgE levels in 4 allergic but unchallenged cynomolgus primates (2 male/2 female) over 29 days. Serum IgE concentration is significantly reduced from 100% (predose) to 66 + 10% of control values (p<0.05), at 4 and 5 days after dosing. This reduction in serum IgE concentration rises to 88 ± 8% of control levels at day 22.<*>= p<0.05 compared to predose IgE levels, repeated measures ANOVA followed by Dunnett's multiple comparison test (n=4 animals). Figure 20B shows relative serum IgE levels of male and female cynomolgus primates versus time after a single 10 mg/kg intravenous dose of BAK502G9. Relative serum IgE data are expressed as arithmetic mean + SEM percent of baseline value. Figure 21 illustrates the effects of intraperitoneal administration of BAK209B11 at various amounts (H=237 ug/day, M=23.7 ug/day and L=2.37 ug/day) compared to an isotype-matched IgG1 irrelevant control antibody on the lung function of ovalbumin-sensitized and challenged mice. In Figure 21A, lung function is represented by log PC50S (log methacholine concentration required to increase baseline PenH by 50%) before any treatment (day 0) and after sensitization, challenge and drug treatment (day 25). Figure 21A shows the raw data that was used to calculate the study endpoint, shown in Figure 21B (Delta log PC50). The data represent the mean with standard error bars of n=8.

In Figure 21B, changing lung function is shown by a change in an individual mouse's log PC50 (delta log PC50). Delta log PC50 is defined as an individual's change in log PC50 at day 25 versus day 0. Data represent group mean delta log PC50

(average individual changes within the treatment groups) with standard error bars. The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnett's test using delta log PC50 data.<**>p<0.01 compared to ovalbumin-sensitized and challenged control mice (n=8 mice). Figure 22 illustrates the effects of local (i.po.) and systemic (i.v.) administration of BAK502G9 as human IgG4 at different amounts compared to an isotype-matched IgG4 irrelevant control antibody on total leukocyte recruitment (Figure 22A) and eosinophil recruitment (Figure 22B) to "air -pouch" to BALB/C mice. The data represent means with standard error bars to n=10. The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnetfs test using log-transformed data.<*>p<0.05.<**>p<0.01 compared to HuTL-13 challenged mice (n=10) . Figure 23: illustrates the effects of i.p. administration of BAK502G9 as human IgG4 compared with an isotype-matched IgG4 irrelevant control antibody on the development of AHR after intratracheal administration of human IL-13 to the airways of mice. The effect of antibody treatment was statistically evaluated by performing one-way ANOVA with Dunnetfs test using PC200Metacholine data<*><0.05.<**><0.01 compared to the human IL-13 positive control group (n= 6-8 mice). Data represent the mean with standard error bars. Figure 24 shows the neutralizing potency (% maximal response) of BAK502G9 (closed squares) as IgG4 against 30 ng/ml IL-13 in a human B-cell IgE production assay. Open squares represent an irrelevant IgG4. Data represent the mean with standard error bars of six donors from separate experiments. Figure 25 shows the effects of BAK502G9 on IL-13 induced potentiation of agonist-induced CA 2_|_ signaling in bronchial smooth muscle cells. The area under the curve (AUC) of the CA 2_|_ signaling response to histamine was determined for each antibody +/- IL-13 pre-treatment conditions. Combined data from three independent experiments are shown for irrelevant antibody CAT-001 (a) and BAK502G9 (b) as percentage difference versus untreated mice with AUC±SD (ns=not significant (p<0.05),<*>å< 0.05,<**>p<0.01). The results were statistically evaluated using a one-way analysis of variance (ANOVA) with Bonferronf's multiple comparison post-test.

Figure 26 shows the effects of phase II administered BAK502G9.

Figure 26A shows the effect on AHR as measured by change in area under the histamine dose-response curve (n=14).

Figure 26B shows the effect on AHR measured by changes in PC30 (n=18).

Figure 26C shows the effect on antigen priming (n=20).

Figure 26D shows effect on BAL inflammation (n=21).

Figure 27 shows the effect of BAK502G9 on IL-13-induced CD23 expression. Data are presented as a percentage of the response to IL-13 alone (100%) and expressed as mean ± SEM % control of 6 separate experiments from 6 individual donors (exposed in triplicate). Figure 18 shows the effect of BAK502G9 and irrelevant IgG4 on IL-13 and/or IL-4 induced PBMC CD23 expression. Data are presented as a percentage of the response to IL-4 alone (100%) and expressed as mean ± SEM % control of 4 separate experiments from 4 individual donors (performed in triplicate). Figure 29A shows the effect of BAK502G9 on NHLF eotaxin-1 production induced by 48 hours of cultivation with IL-13/TNF-a/TGF-fil containing medium. Data are shown as an arithmetic mean ±SEM from triplicate determinations of the medium used in this study to induce leukocyte for such change. Figure 29B shows effect of BAK502G9 on such change of human eosinophils induced by 1:16 dilution of conditioned medium. Data points represented are mean + SEM % blank medium for such change from separate experiments from four individual donors. Figure 30 shows the alignment of human IL-13 against mouse IL-13 which highlights the mutations that were introduced into human IL-13 chimeras. The four alpha helices are highlighted in boxes and loop 1 and loop 3 are labeled. Five chimeric proteins were produced in which helices B, C and D and loops 1 and loop 3 were replaced with the mouse sequence. Four further chimeric proteins were produced and numbered according to the amino acid of the human pre-protein (not to the numbering of the multiple array above) where arginine at residue 30 (position 34 above) was mutated, residues 33 and 34 (positions 37 and 38 above) were mutated, residues 37 and 38 (VH) were mutated (positions 41 and 42 above), and residues 40 and 41 (TQ) were mutated (positions 44 and 45 above). Figure 31 shows the alignment of human IL-13 against mouse IL-13 highlighting the mutations introduced into human IL-13 to produce the second panel of IL-13 chimeras. Six chimeras were produced in which the human residue or residues were substituted with the mouse residue or residues (highlighted by boxes). Four further chimeric proteins were produced (numbering is according to the amino acid position in the human pre-protein) where leucine at residue 58 (62 in the above figure) was mutated, leucine at residue 119 (residue 123 above) was mutated, lysine at position 123 (residue 127 above) was mutated and arginine at residue 127 (residue 132 above) was mutated. Figure 32 shows mutations made in human IL-13. Mutations in dark gray reduced binding to BAK502G9, mutations in light gray did not change binding. Linear sequence of pre-human IL 13 with mutated residues is indicated.

In various aspects and embodiments according to the invention, the inventive material included included in the claims below is provided.

The present invention relates to specific binding members for IL-13, particularly human, primate IL-13 and variant IL-13 (R130Q). Preferred embodiments of the present invention are antibody molecules, either whole antibody (eg, IgG, such as IgG4) or antibody fragments (eg, scFv, Fab, dAb). Antibody antigen binding regions were provided, there will also be antibody VH and VL domains. Within VH and VL domains are provided complementarity determining regions, CDRs, which may be provided within different framework regions, FRs, to form VH or VL domains as the case may be. An antigen-binding site can consist of an antibody VH domain and/or a VL domain.

An antigen binding site can be provided by arrangement of CDRs on non-antibody protein scaffold such as fibronectin or cytochrome B etc. [115,116]. Scaffolding for constructing new binding sites in proteins has been reviewed in detail by Nygren et al [116]. Protein scaffolds for antibody mimics are detailed in WO/0034784 where the inventors describe proteins (antibody mimics) which include a fibronectin type Ul domain which has at least one randomized loop. A suitable scaffold into which one or more CDRs can be inserted, e.g. a set of HCDRs can be provided by any domain member of the immunoglobulin gene superfamily.

Preferred embodiments according to the present invention are in what is here called the "BAK278D6 line". This is defined with reference to a set of six CDR sequences of BAK278D6 as follows: HCDR1 (SEW ID NO: 1), HCDR2 (SEC ID NO: 2), HCDR3 (SEC ID NO: 3), LCDR1 (SEQ ID NO: 4), LCDR2 (SEQ ID NO: 5) and LCDR3 (SEQ ID NO: 6). In one aspect, the present invention provides a specific binding member for human IL-13, which comprises an antibody antigen binding site which is composed of a human antibody VH domain and a human antibody VL domain and which comprises a set of CDRs where the VH domain comprises HCDR1, HCDR2 and HCDR3 and the VL domain comprises LCDR1, LCDR2 and LCDR3, where HCDR1 has the amino acid sequence of SEQ ID NO: 1, HCDR2 has the amino acid sequence of SEQ ID NO: 2, HCDR3 has the amino acid sequence of SEQ ID NO: 3, LCDR1 has the amino acid sequence of SEQ ID NO:4, LCDR2 has the amino acid sequence of SEQ ID NO: 5, and LCDR3 has the amino acid sequence of SEQ ID NO: 6.

The set of CDRs where HCDR1 has the amino acid sequence of SEQ ID NO: 1, HCDR2 has the amino acid sequence of SEQ ID NO: 2, HCDR3 has the amino acid sequence of SEQ ID NO: 3, LCDR1 has the amino acid sequence of SEQ ID NO: 4, LCDR2 has the amino acid sequence to SEQ ID NO: 5, and LCDR3 has the amino acid sequence of SEQ ID NO: 6, is referred to herein as the "BAK278D6 set of CDRs". HCDR1, HCDR2 and HCDR3 within the BAK278D6 set of CDRs are referred to as "BAK278D6 set of HCDRs" and LCDR1, LCDR2 and LCDR3 within the BAK278D6 set of CDRs are referred to as "BAK278D6 set of LCDRs". A set of CDRs with the BAK278D6 set of CDRs, BAK278D6 set of HCDRs, or BAK278D6 LCDRs, or one or two substitutions therein, is said to be of the BAK278D6 lineage.

As noted, in one aspect the invention provides a specific binding member for human IL-13, comprising an antibody antigen binding site composed of a human antibody VH domain and a human antibody VL domain and comprising a set of CDRs, wherein the set of CDRs 's is the BAK278D6 set of CDRs.

The present inventors have identified the BAK278D6 line as providing human antibody antigen binding domains against IL-13 which are of particular value. Within the line, the BAK502G9 has been identified as being of particular value. BAK278D6 and BAK502G9 sets of CDRs have already been identified above.

Following the lead of computational chemistry in applying multivariate data analysis techniques to structure/property activity relationships [94], quantitative activity property relationships of antibodies can be derived by applying well-known mathematical techniques such as statistical regression, pattern recognition and classification [95-100]. The properties of antibodies can be derived from empirical and theoretical models (for example) analysis of probable contact residues or calculated physicochemical properties) of antibody sequence, functional and three-dimensional structures and these properties can be considered individually or in combination.

An antibody-antigen binding site composed of a VH domain and a VL domain is formed by six loops of polypeptide; three from the light chain variable domain (VL) and three from the heavy chain variable domain (VH). Analyzes of antibodies with known atomic structure have clarified relationships between sequence and the three-dimensional structure of antibody by combining sites [101, 102]. These conditions suggest that, except for the third region (the loop) in VH domains, the binding site loops have one of a small number of main chain conformations: canonical structures. The canonical structure formed in a particular loop has been shown to be determined by its size and the presence of certain residues at key sites in both the loop and in framework regions [101, 102].

This study of sequence-structure relationships can be used to predict those residues in an antibody of known sequence but of unknown three-dimensional structure, which are important in maintaining the three-dimensional structure of its CDR loops and consequently maintaining binding specificity. These predictions can be backed up by comparing the yield predictions from the lead optimization tools.

In a structural approach, a model can be generated of the antibody molecule [103] by using a freely available or commercial package such as WAV [104]. A protein visualization and analysis software package such as Insight U [105] or Deppe View [106] can then be used to evaluate possible substitutions at any position in the CDR. This information can then be used to make substitutions that are likely to have a minimal or beneficial effect on activity.

The present inventors analyzed sequence data of the panel of clones whose sets of CDRs are shown in Table 1.

The analysis tested the hypothesis that binary combinations of listed amino acid variations in the CDRs from the presented set of scFv variants led to an scFv variant with at least the starting potency of the parent scFv BAK278D6.

All scFv variants in the panel shown in Table 1 have been selected for improved affinity and have been confirmed to exhibit higher potency.

The observed amino acid variations can be either beneficial, non-beneficial or neutral in their effect on the starting potency of scFv BAK278D6 in the TF-1 assay at 44 nM.

No linkage was observed between any of the two amino acid variations confirming that there was no synergy, either "positive" or "negative", between any of the two selected amino acid variations.

There are four scenarios where such a binary combination will fulfill the hypothesis and three scenarios where the hypothesis will not be of value. Synergistic amino acid variants are not considered as no linkage was observed.

The hypothesis is applicable where:

Al: mutation 1 is favorable and mutation 2 is favorable

A2: mutation 1 is favorable and mutation 2 is neutral

A3: mutation 1 is neutral and mutation 2 is neutral

A4: mutation 1 is beneficial and mutation 2 is non-beneficial

(with the effect of 1 offsetting the effect of 2).

The hypothesis is not valid where:

Bl: mutation 1 is non-beneficial and mutation 2 is neutral

B2: mutation 1 is unfavorable and mutation 2 is unfavorable B3: mutation 1 is favorable and mutation 2 is unfavorable

(with the effect of 2 offsetting the effect of 1).

For A4 to be possible, mutation 1 must be very favorable to counterbalance the negative effect of mutation 2 on potency. Because such a highly favorable mutation would be present in the library of variants used for selection, it would select for and would therefore appear frequently in the panel of variants. Because synergy can be excluded, such mutation would be beneficial in any type of sequence context and should therefore recur in different scFv variants. An example of such a frequent amino acid change is the change in light chain CDR1 Asn27Ile. However, this mutation on its own (in clone BAK531E2) has only a moderate 2-fold effect on potency (final IC50 of 23.2 nM). On its own, this mutation would not allow the scenario outlined in A4, as it is not a strong beneficial mutation. This suggests that each clone in the presented set of IL-13 binding clones (table 1) that has a light chain CDR1 Asn27Ile variant together with one or more further mutations is at least as potent as the variant that has the single light chain CDR1 Asn27Ile mutation. The other mutations are either neutral or positive, but do not have a negative or harmful effect.

A further example is in the heavy chain CDR3 Asn99Ser (see Table 1). As a clone carrying this particular single amino acid variation has not been observed, the potency of such a clone has been estimated to be about 12.0 nM by the following rationale: BAK278D6 potency is 44 nM. Changes in VL CDR1 N27I + VH CDR3 N99S lead to BAK502G9 with a potency of 8 nM, i.e. a 5.5-fold improvement.

BAK278D6 potency is 44 nM. Alteration of VL CDR1 N27I leads to BAK531E2 with a potency of 23 nM, i.e. a 1.9-fold improvement.

BAK278D6 potency is 44 nM. Change VH CDR3 N99S to provide a possible clone with a potency of 12.2 nM, ie a 2.9-fold improvement (5.5/1.9 = 2.9).

The binary combination of heavy chain CDR3 Asn99Ser with light chain CDR1 Asn27Ile gives a scFv BAK0502G9 with a potency of 8 nM. As synergy is excluded, the contribution to the heavy chain CDR3 Asn99Ser change in BAK502G9 is therefore additive.

Therefore, each clone in the presented set of IL-13 binding clones (Table 1) that has a heavy chain CDR3 AsnH99Ser change along with one or more further mutations would have a potency of at least 12 nM or greater, within a permitted assay window of 2.5 times for n=l-2.

The inventors thus note that a highly advantageous amino acid variation that would advantageously be selected has not been observed. As discussed above, two variations that were mainly represented in Table 1 with ScFv variants were analyzed in more detail. Any scFv variant in Table 1 with any of these mutations along with one or more further mutations showed a potency at least as improved as a clone containing either of these two single amino acid variations in parent BAK278D6. There is therefore no evidence that a highly advantageous amino acid variation that would allow scenario A4 is present in the panel.

This observation led the inventors to conclude that there were no non-beneficial mutations present in this set of scFv variants. This means that scenarios A4 and Bl to B3 are not relevant and that the hypothesis is valid.

Accordingly, as already noted, the present invention provides specific binding members comprising CDRs of the BAK278D6 or BAK502G9 set of CDRs.

The relevant set of CDRs were provided within antibody framework regions or other protein scaffolds, e.g. fibronectin or cytochrome B [115, 116]. Ideally, antibody framework regions are used, and where they are used, they are preferably germline, more desirably, the antibody framework region for the heavy chain can be DP14 from the VH1 family. The preferred framework region for the light chain may be X3-3H. For the BAK502G9 set of CDRs, it is preferred that the antibody framework regions are for VH FRI, SEQ ID NO: 27, for VH FR2, SEQ ID NO: 28, for VH FR3, SEQ ID NO: 29, for light chain FRI, SEQID NO: 30, for light chain FR2, SEQID NO: 31, for light chain FR3, SEQ ID NO: 32.1 a very preferred embodiment then a VH domain is provided with the amino acid sequence according to SEQ ID NO: 15, this is called "BAK502G9 VH domain". In an even more preferred embodiment, a VL domain is provided with the amino acid sequence of SEQ ID NO: 16, this being termed "BAK502G9 VL domain". A highly preferred antibody antigen binding site provided according to the present invention is composed of the BAK502G9 VH domain, SEQ ID NO: 15 and the BAK502G9 VL domain, SEQ ID NO: 17. This antibody antigen binding site can be provided within any desired antibody molecule format, e.g. scFv, Fab, IgG, IgG4, dAb, etc., which are discussed further elsewhere here.

In a further highly preferred embodiment, the present invention provides an IgG4 antibody molecule comprising the BAK502G9 Vh domain, SEQ ID NO: 15, and the BAK502G9 VL domain, SEQ ID NO: 16. This is called here "BAK502G9 IgG4".

Other IgG4 or other antibody molecules comprising the BAK502G9 VH domain, SEQ ID NO: 15, and the BAK502G9 VL domain, SEQ ID NO: 16, are provided according to the present invention, as are other antibody molecules comprising BAK502G9 set to HCDRs (SEQ ID NO: 7, 8 and 9) within an antibody VH domain, and BAK502G9 set to LCDRs (SEQ ID NO: 10, 11 and 12) within an antibody VL domain.

It is appropriate to point out here that "and/or" where used here must be taken as a specific explanation of each of the two specified features or components with or without the other. For example, "A and/or B" must be viewed as specifically defining each of (i) A, (ii) B and (iii) A and B, just as if each were set forth individually herein.

As noted, the present invention provides a specific binding member that binds human IL-13 and that comprises the BAK502G9 VH domain (SEQ ID NO: 15) and the BAK502G9 VL domain (SEQ ID NO: 16).

Generally, a VD domain is paired with a VL domain to provide an antibody antigen binding site, although as discussed further below, a VH domain alone can be used to bind antigen. In a preferred embodiment, the BAK502G9 VH domain (SEQ ID NO: 15) is paired with the BAK502G9 VL domain (SEQ ID NO: 16), so that an antibody antigen-binding site is formed that includes both the BAK502G9 VH and VL domains.

One or more CDRs can be taken from the BAK502G9 VH or VL domain and incorporated into an appropriate frame. This is discussed further here. The BAK502G9 HCDRs 1, 2, and 3 are shown in SEQ ID NO: 7, 8, and 9, respectively. The BAK502G9 LCDRs 1, 2, and 3 are shown in SEQ ID NO: 10, 11, and 12, respectively.

Variants of the VH and VL domains and CDRs of the present invention, including those whose amino acid sequences are set forth herein, and which can be used in specific binding members for IL-13 can be provided by methods of sequence change or mutation and screening. Such methods were also provided according to the present invention.

Variable domain amino acid sequence variants of any of the VH and VL domains whose sequences are specifically derived herein may be used in accordance with the present invention, as discussed. Special variants may include one or more amino acid sequence changes (addition, deletion, substitution and/or insertion of an amino acid residue), may be less than about 20 changes, less than about 15 changes, less than about 10 changes or less than about 5 changes, 4 , 3, 2 or 1. Changes can be made in one or more frame regions and/or one or more CDRs.

Also described is a specific binding member that competes for binding to antigen with any specific binding member that both binds the antigen and that includes a specific binding member, the VH and/or VL domain as disclosed herein, or HCDR3 as disclosed herein, or variants of any of these. Competition between binding members can be readily analyzed in vitro, for example using ELISA and/or by tagging a specific reporter molecule to a binding member that can be detected in the presence of other untagged binding member or members, to enable identification of specific binding members that bind to the same epitope or an overlapping epitope.

Furthermore, a specific binding member comprising a human antibody antigen binding site is described which competes with a BAK502G9 antibody molecule, particularly BAK502G9 scFv and/or IgG4, for binding to IL-13. Also described is a specific binding member comprising a human antibody antigen binding site that competes with an antibody antigen binding site for binding to IL-13, wherein the antibody antigen binding site is composed of a VH domain and a VL domain, and wherein the VH and VL domains comprise a set of CDRs from the BAK278D6 line.

Various methods are available in the art to provide antibodies against IL-13 that can compete with a BAK502G9 antibody molecule, an antibody molecule with a BAK502G9 set of CDRs, or that antibody molecule with a set of CDRs from the BAK278D6 line, for binding to IL-13.

Furthermore, a method is described for providing one or more specific binding members capable of binding the antigen, the method includes contacting a library of specific binding members in accordance with the invention and said antigen, and selecting one or more specific binding members in the library capable of binding said antigen.

The library may be displayed on the surface of bacteriophage particles, each particle containing nucleic acids encoding the antibody VH variable domain displayed on its surface, and optionally also a displayed VL domain if present.

After selection of specific binding members capable of binding the antigen and displayed on bacteriophage particles, nucleic acid can be taken from a bacteriophage particle displaying said selected specific binding member. Such nucleic acid can be used in the subsequent production of a specific binding member or an antibody VH variable domain (optionally an antibody VL variable domain) by expression from nucleic acid with the sequence of nucleic acid taken from a bacteriophage particle that displays an aforesaid selected specific binding member.

An antibody VH variable domain with the amino acid sequence of an antibody VH variable domain of a said selected specific binding member can be provided in isolated form, so can a specific binding member comprising such a VH domain. The ability to bind IL-13 can be further tested, also the ability to compete with BAK502G9 (e.g. in scFv format and/or IgG format, e.g. IgG4) for binding to IL-13. The ability to neutralize IL-13 can be tested, as discussed further below.

Binding affinity and neutralization potency of different specific binding members can be compared under appropriate conditions.

The antibodies of the present invention have a number of advantages over existing commercial anti-IL-13 antibodies, particularly three commercial rodent anti-human IL-13 antibodies, named JES10-5A2 (SioSource), B-B13 (Euroclone) and clone 321166 (R&D Systems). The potency of the antibodies of the present invention was compared with commercial antibodies JES10-A2 and B-B13. Clone 321166 was not evaluated as previous experiments revealed that this clone was considerably less potent than other known commercial antibodies.

The effectiveness and application of the rodent commercial IL-13 antibodies in humans is probably limited, due to their increased potential to induce immunogenic responses and therefore more rapid clearance from the body. Kinetic analysis of the antibodies of the present invention in non-human primates indicates that these antibodies have a clearance rate similar to that of other known human or humanized antibodies.

Antibodies provided according to the present invention do not recognize human primate IL-13, including rhesus and cynomolgus IL-13. Determining the efficacy and safety profiles of an antibody in non-human primates is very valuable as it provides a means of predicting the antibody's safety, pharmacokinetics and pharmacodynamic profile in humans.

Furthermore, antibodies according to the present invention recognize the human IL-13 variant, R130Q, which is associated with asthma. Cross-reactivity with variant IL-13 allows antibodies of the present invention and compositions comprising antibodies of the present invention to be used for the treatment of patients with wild-type and variant IL-13.

The inventors have demonstrated that BAK502G9, 1167F2 and 1183H4 are significantly more potent against naturally occurring IL-13 than known commercial antibodies (Figure 7).

In addition to antibody sequences, a specific binding member according to the present invention may comprise other amino acids, for example to form a peptide or polypeptide, such as a folded domain, or to give the molecule another functional characteristic in addition to the ability to bind antigen. Specific binding members of the invention may carry a detectable label, or may be conjugated to a toxin or a targeting moiety or enzyme (eg via a peptidyl bond or linker).

In a further aspect, the invention provides an isolated nucleic acid comprising a sequence that codes for a specific binding member, the VH domain and/or VL domains according to the present invention, and methods for making a specific binding member, a VH domain and/or a VL domain according to the invention, which comprises expressing said nucleic acid under conditions to obtain production of said specific binding member, VH domain and/or VL domain to recover it. Specific binding members of the invention may be used to treat or diagnose the human or animal body, such as to treat (which may include prophylactic treatment) of a disease or disorder in a human patient comprising administering to said patient an effective amount of a specific binding member according to the invention. The conditions treatable in accordance with the present invention include any in which IL-13 plays a role, particularly asthma, atopic dermatitis, allergic rhinitis, fibrosis, chronic obstructive pulmonary disease, scleroderma, inflammatory bowel disease and Hodgkin's lymphoma. Furthermore, the antibodies according to the present invention are also used to treat tumors and viral infections, as these antibodies will inhibit IL-13 controlled immunosuppression [64, 65].

A further aspect according to the present invention provides nucleic acid, generally isolated, which encodes an antibody VH variable domain and/or VL variable domain according to the invention.

It also describes nucleic acid, generally isolated, which encodes a VH CDR or VL CDR sequence which is disclosed herein, in particular a VH CDR selected from SEQ ID NO: 7, 8 and 9 or a VL CDR selected from SEQ ID NO: 10, 11 and 12, most preferably BAK502G9 VH CDR3 (SEQ ID NO: 9). Nucleic acid encoding the BAK502G9 set of CDRs, nucleic acid encoding the BAK502G9 set of HCDRs and nucleic acid encoding the BAK502G9 set of LCDRs are also disclosed, as are nucleic acids encoding individual CDRs, HCDRs , LCDRs and sets of CDRs, HCDRs, LCDRs in the BAK278D6 line.

A further aspect provides a host cell that has been transformed with nucleic acid according to the invention.

A still further aspect provides a method of producing an antibody VH variable domain, the method including causing expression from the encoded nucleic acid. Such a method may comprise cultivating host cells under conditions for the production of said antibody VH variable domain.

Analogous methods for the production of VL variable domains and specific binding members comprising a VH and/or VL domain are provided as further aspects of the present invention.

A method of production may include a step of isolation and/or purification of the product.

A method of manufacture may comprise and formulate the product into a composition that includes at least one additional component, such as a pharmaceutically acceptable excipient.

These and other aspects according to the invention are described in further detail below.

TERMINOLOGY

Specific binding member

This describes a member of a pair of molecules that have binding specificity for each other. The members of a specific bond pair may be naturally derived or wholly or partially synthetically produced. One member of the pair of molecules has an area on its surface or cavity that specifically binds to and is therefore complementary to a particular spatial and polar organization of the other member of the pair of molecules. Thus, the members of the pair have the property of binding specifically to each other. Examples of types with specific binding pairs are antigen-antibody, biotin-avidin, hormone-hormone receptor, resptor ligand, enzyme substrate. The present invention relates to antigen-antibody type reactions.

Antibody molecule

This describes an immunoglobulin either naturally or partially or completely synthetically produced. The term also covers any polypeptide or protein comprising an antibody-binding domain. Antibody fragments comprising an antigen-binding domain are molecules such as Fab, scFv, Fv, dAb, Fd; and diastoff.

It is possible to take monoclonal and other antibodies and use techniques with recombinant DNA technology to produce other antibodies or chimeric molecules that have retained the specificity of the original antibody. Such techniques may involve introducing DNA encoding the immunoglobulin variable region, or the complementarity determining regions (CDRs), of an antibody into the constant regions, or constant regions plus framework regions, of a different immunoglobulin. See for example EP-A-184187, GB 2188638A or EP-A-239400, and a large amount of subsequent literature. A hybridoma or other cell that produces an antibody may be subjected to genetic mutation or other changes, which may or may not alter the binding specificity of the antibodies that are produced.

Since antibodies can be modified in a number of ways, the term "antibody molecule" should be understood to cover any specific binding member or substance having an antibody antigen binding domain with the required specificity. This term thus covers antibody fragments and derivatives, including any polypeptide comprising an immunoglobulin binding domain, whether natural or wholly or partially synthetic. Chimeric molecules comprising an immunoglobulin binding domain, or equivalent, fused to other polypeptides are therefore included. Cloning and expression of chimeric antibodies is described in EP-A-0120694 and eP-A-0125023 and a large body of subsequent literature.

Furthermore, techniques available in the field for constructing antibodies have made it possible to isolate human and humanized antibodies. For example, human hybridomas can be made as described by Kontermann et al [107]. Phage display, another established technique for generating specific binding members has also been described in detail in many publications such as Kontermann et al [107] and WO92/01047 (discussed further below). Transgenic mice in which the mouse antibody genes are inactivated and functionally replaced with human antibody genes while other components of the mouse immune system are intact, can be used to isolate human antibodies to human antigens [108].

Synthetic antibody molecules can be made by expression from genes made using oligonucleotides synthesized and assembled within suitable expression vectors, for example as described by Knappik et al. J. Mol. Biol. (2000) 296, 57-86 or Krebs et al. Journal of Immunological Methods 254 2001 67-84.

It has been shown that fragments of an entire antibody can perform the function of binding antigens. Examples of binding fragments are (i) the Fab fragment consisting of VL, vH, CL and CH1 domains; (ii) the Fd fragment consisting of the VH and CH1 domains; (iii) the Fv fragment consisting of the VL and VH domains from a single antibody; (iv) the dAb fragment (Ward, E.S. et al., Nature 341, 544-546 (1989), McCafferty et al (1990) Nature, 348, 552-554) consisting of a VH domain; (v) isolated CDR regions: (vi) F(ab')2 fragments, a bivalent fragment comprising two linked Fab fragments (vii) single-chain Fv molecules (scFv), where a VH domain and a VL domain are linked by a peptide linkers that allow the two domains to associate or form an antigen-binding site

(Bird et al, Science, 242 423-426,1988; Huston et al, PNAS USA, 85, 5879-5883, 1988); (viii) bispecific single-chain Fv dimers (PCT/US92/09965) and (ix) "diabodies" multivalent or multispecific fragments constructed by gene fusion (WO94/13804; P. Holliger et al, Proe. Nati. Acad. Sei. USA 90 6444-6448, 1993). Fv, scFv or diasubstance molecules can be stabilized by the incorporation of disulfide bridges linking the VH and VL domains (Y. Reiter et al, Nature Biotech, 14, 1239-1245, 1996). Minidrugs comprising an scFv linked to a CH3 domain can also be made (S, Hu et al, Cancer Res., 56, 3055-3061, 1996).

Where bispecific antibodies are to be used, these can be conventional bispecific antibodies, which can be produced in a number of ways (Holliger, P, and Winter G. Current Opinion Biotechnol, 4, 446-449 (1993)), e.g. prepared chemically or from hybrid hybridoma, or they can be any of the bispecific antibody fragments mentioned above. Examples of bispecific antibodies include those with the BiTE™ technology where the binding domains of two antibodies with different specificities can be used and directly linked via short flexible peptides. This combines two antibodies on a short single polypeptide chain. Diastoffers and scFv can be constructed without an Fc region, using only variable domains, potentially reducing the effects of anti-idiotypic reaction.

Bispecific diabodies, as opposed to bispecific whole antibodies, may also be particularly useful because they can be easily constructed and expressed in E. coli. Diasubstances (and many other polypeptides such as antibody fragments) with appropriate binding specificity can be easily selected by using phage display (WO94/13804) from libraries. If one arm of the diasubstance is to be kept constant, for example, with a specificity directed against IL-13, then a library can be created where the other arm varies and an antibody with appropriate specificity is selected. Bispecific whole antibodies can be made by "knobs-into-holes" engineering (J.B.B. Ridgeway et al, Protein Eng-, 9, 616-621, 1996).

Antigen-binding domain

This describes the part of an antibody molecule that includes the region that specifically binds to and is complementary to part or all of an antigen. Where an antigen is large, an antibody can only bind to a specific part of the antigen. This part is called an epitope. An antigen-binding domain can be provided by one or more antibody variable domains (e.g. a so-called Fd antibody fragment consisting of a VH domain). Preferably, an antigen-binding domain comprises an antibody light chain variable region (VL) and an antibody heavy chain variable region (VH).

Specific

This can be used to refer to the situation where one member of a specific binding pair will not show any significant binding to molecules other than its specific binding partner or partners. The expression is also usable where, e.g. an antigen binding domain is specific for a particular epitope carried by a variety of antigens, in this case the specific binding member carrying the antigen binding domain will be able to bind to the different antigens carrying the epitope.

Include

This is generally used in the sense of including, ie allowing the presence of one or more features or components.

Isolated

This refers to the state that the specific binding members according to the invention, or nucleic acid encoding such binding members, will generally be in according to the present invention. Isolated members and isolated nucleic acid will be free or substantially free of material with which they are naturally associated such as other polypeptides or nucleic acids with which they co-exist in their natural environment, or the environment in which they are made (eg, cell culture) when such preparation is by recombinant DNA technology practiced in vitro or in vivo. Members and nucleic acid can be formulated with diluents or auxiliaries and still for practical purposes isolated, for example the members will normally be mixed with gelatin or other carriers if they are used to coat microtiter plates for use in immunoassays, or they will be mixed with pharmaceutically acceptable carriers or diluents when used in diagnosis or therapy. Specific binding members may be glycosylated, either naturally or by systems in heterologous eukaryotic cells (eg, CHO or NSO (ECACC 85110503) cells), or they may be (eg, if produced by expression in a prokaryotic cell) non-glycosylated .

Naturally occurring IL-13

This generally refers to a condition where the IL-13 protein or fragments thereof can occur. Naturally occurring IL-13 means IL-13 protein that is naturally produced by a cell without prior introduction of coding nucleic acid by using recombinant technology. Thus, naturally occurring IL-13 can be as produced naturally by, for example, CD4+ T cells and/or as isolated from a mammal, e.g. human, non-human primate, rodent such as rat or mouse.

Recombinant IL-13

This refers to a condition where the IL-13 protein or fragments of it can occur. Recombinant IL-13 means IL-13 protein or fragments thereof produced by recombinant DNA in a heterologous host. Recombinant IL-13 may differ from naturally occurring IL-13 by glycosylation.

Recombinant proteins expressed in prokaryotic bacterial expression systems are not glycosylated while those expressed in eukaryotic systems such as mammalian or insect cells are glycosylated. However, proteins expressed in insect cells have a different glycosylation than proteins expressed in mammalian cells.

By "substantially as shown" is meant that the relevant CDR or VH or VL domain will be either identical or very similar to the specified regions from which the sequence is set forth herein. By "very similar", it is contemplated that from 1 to 5, preferably from 1 to 4 such as 1 to 3 or 1 or 2, or 3 or 4, amino substitutions can be made in the CDR and/or VH or VL domain.

The structure for carrying a CDR or a set of CDRs according to the invention will usually be an antibody heavy or light chain sequence or substantial part thereof where the CDR or set of the CDR is located at a location corresponding to the CDR or the set of CDRs of naturally occurring VH and VL antibody variable domains encoded by rearranged immunoglobulin genes. The structures and locations of immunoglobulin variable domains can be determined by reference to (Kabat, E. A. etal, Sequences of Proteins of Immunological Interest. 4th Edition. US Department of Health and Human Services. 1987, and updates thereof now available on the Internet.

(http://immuno.bme.nwu.edu or find "Kabat" using any search engine).

CDRs can also be carried by other scaffolds such as fibronectin or cytochrome B [115, 116].

Preferably, a CDR amino acid sequence is carried as a CDR in a human variable domain or a substantial part thereof.

Variable domains used according to the invention can be obtained from any germline or rearranged human variable domain, or can be a synthetic variable domain based on consensus sequences of known human variable domains. A CDR sequence according to the invention (e.g. CDR3) can be introduced into a repertoire of variable domains that lack a CDR (e.g. CDR3), by using recombinant DNA technology.

For example, Marks et al (Bio/Technology,\ 992, 10:779-783) describe methods for producing antibody variable domain repertoires where the consensus primers directed at or flanking the 5' end of the variable domain region are used together with consensus primers to the third framework region of human VH genes to provide a repertoire of VH variable domains lacking a CDR3. Marks et al further describe how this repertoire can be combined with a CDR3 of a specific antibody. Using analogous techniques, CDR3-derived sequences of the present invention can be shuffled with the repertoires of VH or VL domains lacking a CDR3, and the shuffled complete VH or VL domains combined with a cognate VL or VH domain to provide specific binding members according to the invention. The repertoire can then be displayed in a suitable host system such as the subject display system of WO92/01047 or any of the subsequent large bodies of literature, including Kay, B. K, Winter, J.; and Mc Cafferty, J. (1996) Phage Display of Peptides and Proteins: A Laboratory Manual, San Diego: Academic Press, so that appropriate specific binding members can be selected. A repertoire can consist of anything from IO4 individual members upwards, for example from IO<6> to 108 or 10<10> members. Other suitable host systems include yeast display, bacterial display, T7 display, ribosome display and more. For an overview of ribosome display see Lowe D and Jermutus L, 2004, Curr. Pharm. Biotech, 517-27, also WO92/01047.

Analogous sliding or combinatorial techniques are also explained by Stemmer {Nature, 1994, 370:389-391), who describes the technique in relation to a pMaktamase gene but observes that the approach can be used for the generation of antibodies.

A further alternative is to generate new VH or VL regions bearing CDR-derived sequences by using random mutagenesis of one or more selected VH and/or VL genes to generate mutations within the entire variable domain. One such technique is described by Gram et al (1992, Proe. Nati. Acad. Sci., USA, 89:3576-3580), who used lean PCR. In preferred embodiments, one or two amino acid substitutions are made within a set of HCDRs and/or LCDRs.

Another method that can be used is to direct mutagenesis to CDR regions in VH or VL genes. Such techniques are elaborated by Barbas et al. (1994, Proe. Nati. Acad. Sci., USA, 91:3809-3813) and Schier et al (1996, J. Mol. Biol. 263:551-567). All the techniques described above are known in the field and do not in themselves form part of the present invention. A person skilled in the art will be able to use such techniques to provide specific binding members of the invention using routine methodology in the art.

The invention further relates to a method for providing an antibody antigen-binding domain that is specific for IL-13 antigen, the method comprises providing by means of addition, deletion, substitution or insertion of one or more amino acids in the amino acid sequence of a VH domain set forth here a VH domain which is an amino acid sequence variant of the VH domain, optionally combining the VH domain thus provided with one or more VL domains, and testing the VH domain or the VH/VL combination or combinations to identify a specific binding member or an antibody antigen binding domain that is specific from the IL-13 antigen and optionally with one or more preferred properties, preferably the ability to neutralize IL-13 activity. Said VL domain may have an amino acid sequence which is substantially as set forth herein.

An analogous method can be used where one or more sequence variants of the VL domain which is detailed here is combined with one or more VH domains.

In a preferred embodiment, BAK502G9 VH domain (SEQ ID NO: 15) can be subjected to a mutation to provide one or more VH domain amino acid sequence variants, and/or BAK502G9 VL (SEQ ID NO: 16).

The invention further relates to a method for preparing a specific binding member that is specific for IL-13 antigen, the method comprising: (a) providing a starting repertoire of nucleic acid encoding a VH domain that either includes a CDR3 to be replaced or lacks a CDR3 coding region; (b) combining said repertoire with a donor nucleic acid encoding an amino acid sequence substantially as set forth herein for a VH CDR3 such that said donor nucleic acid is inserted into the CDR3 region of the repertoire, to provide a product repertoire of nucleic acids encoding a VH domain; (c) expressing the nucleic acids of said product repertoire;

(d) selecting a specific binding member specific for IL-13; and

(e) recovering said specific binding member or encoding nucleic acid

the.

Again, an analogous method can be used where a VL CDR3 according to the invention is combined with a repertoire of nucleic acids that encode a VL domain that either includes a CDR3 to be replaced or lacks a CDR3 coding region.

Likewise, one or more or all three CDRs can be inserted into a repertoire of VH or VL domains which are then screened for a specific binding member or specific binding members that are specific for IL-13.

In a preferred embodiment, one or more of BAK502G9 and CDR1 (SEQ ID NO: 7), HCDR2 (SEQ ID NO: 8) and HCDR3 (SEQ ID NO: 9), or the BAK502G9 set of HCDRs, can be used and/ or one or more of BAK502G9 LCDR1 (SEQ ID NO: 10), LCDR2 (SEQ ID NO: 11), or the BAK502G9 set of LCDRs.

A substantial part of an immunoglobulin variable domain will comprise at least the three CDR regions, together with their intervening framework regions. Preferably, the portion will also include at least about 50% of either or both of the first and fourth framework regions, the 50% being the C-terminal 50% of the first framework region and the N-terminal 50% being the fourth framework region. Additional residues at the N-terminal or the C-terminal end of the essential part of the variable domain may be those not normally associated with naturally occurring variable domain regions. For example, construction of specific binding members of the present invention made by recombinant DNA techniques may result in the introduction of N- or C-terminal residues as encoded by linkers introduced to promote cloning or other manipulation steps. Other manipulation steps include the introduction of linkers such as linking variable domains of the invention to further protein sequences including immunoglobulin heavy chains, other variable domains (for example in the production of diastoff) or protein tags as discussed in more detail elsewhere herein.

Although in a preferred aspect according to the invention specific binding members comprising a pair of VH and VL domains are preferred, single binding domains based on either VH or VL domain sequences form further aspects according to the invention. It is known that simple immunoglobulin domains, especially VH domains, are able to bind target antigens in a specific manner.

In the case of both single specific binding domains, these domains can be used to screen for complementary domains capable of forming a two-domain specific binding member capable of binding IL-13.

This can be achieved by phage display screening methods using the so-called hierarchical two-way combinatorial approach described in WO92/01047, where an individual colony containing either an H or L chain clone is used to infect a complete library of clones as clones for the other chain (L or H) and the resulting two-chain specific binding member are selected in accordance with phage display techniques such as those described in that reference. This technique is also detailed in Marks et al, ibid.

Specific binding members according to the present invention may further comprise antibody constant regions or parts thereof. For example, a VL domain can be attached at its C-terminus to an antibody chain constant domain including Ck or CA. chains, preferably CX chains. Similarly, a specific binding member based on a VH domain can be attached at its C-terminal end to all or part (e.g. a CH1 domain) of an immunoglobulin thin chain derived from any antibody isotype, e.g. IgG, lg A, IgE and IgM and any of the isotype subclasses, especially IgGl and IgG4. IgG4 is preferred. IgG4 is preferred because it does not bind complement and does not create effector functions. Any synthetic or other constant region variant that has these properties and stabilizes variable regions is also preferred for use in embodiments of the present invention.

Specific binding members according to the invention may be labeled with a detectable or functional label. Detectable labels include radioactive labels such as<131>I or "Tc, which can be attached to antibodies of the invention using conventional chemistry known in the art of antibody imaging. Labels also include enzyme labels such as horseradish peroxidase. Labels further include chemical entities such as biotin which can be detected via binding to a specific cognate detectable entity, eg labeled avidin.

Specific binding members of the present invention are designed to be used for diagnosis or treatment in human or animal subjects, preferably human.

Accordingly, further aspects of the invention provide pharmaceutical compositions comprising such a specific binding member, and the use of such a specific binding member in the production of a drug for administration, for example in a method of making a drug or pharmaceutical composition comprising formulating the specific binding member with a pharmaceutically acceptable excipient.

Clinical indications where an anti-IL-13 antibody can be used to provide therapeutic benefit include asthma, atopic dermatitis, allergic rhinitis, fibrosis, chronic obstructive pulmonary disease, inflammatory bowel disease, scleroderma and Hodgkin's lymphoma. As already explained, anti-IL-13 treatment is effective for all these diseases.

Anti-IL-13 treatment can be given orally, by injection (for example, subcutaneously, intravenously, intraperitoneally or intramuscularly), by inhalation or topically (for example, intraocularly, intranasally, rectally, in wounds, on skin). The administration route can be determined by the physicochemical characteristics of the treatment, by special considerations for the disease or by the requirement to optimize effect or to minimize side effects.

It is taken into account that anti-IL-13 treatment will not be limited to use in clinics. Therefore, subcutaneous injection using a target-free device is also preferable.

Combination therapies can be used to provide significant synergistic effects, particularly the combination of an anti-IL-13 specific binding member with one or more other drugs. A specific binding member according to the present invention can be provided in combination or in addition to short- or long-acting beta-agonists, corticosteroids, cromoglycate, leukotriene (receptor) antagonists, methylxanthines and their derivatives, IL-4 inhibitors, muscarinic receptor antagonists, IgE inhibitors , histamine-like inhibitors, IL-5 inhibitors, eotaxin/CCR3 inhibitors, PDE4 inhibitors, TGF-beta antagonists, interferon-gamma, perfenidone, chemotherapeutic agents and immunotherapeutic agents.

Combination therapy with one or more short- or long-acting beta-agonists, corticosteroids, cromoglycate, leukotriene (receptor) antagonists, xanthines, IgE inhibitors, IL-4 inhibitors, IL-5 inhibitors, eotaxin/CCR3 inhibitors, PDE4 inhibitors can be used to treat asthma . Antibodies according to the present invention can also be used in combination with corticosteroids, anti-metabolites, antagonists of TGF-beta and its downstream signaling reaction pathway, for the treatment of fibrosis. Combination therapy with these antibodies with PDE4 inhibitors, xanthines and their derivatives, muscarinic receptor antagonists, short and long beta antagonists may be useful to treat chronic obstructive pulmonary disease. Similar consideration of combinations applies to the use of anti-IL-13 treatment for atopic dermatitis, allergic rhinitis, chronic obstructive pulmonary disease, inflammatory bowel disease, scleroderma and Hodgkin's lymphoma.

The compositions described herein are administered to individuals. Administration is preferably in a "therapeutically effective amount", this being sufficient to show benefit to a patient. Such gain can be at least an improvement of at least one symptom. The actual amount given, and the rate and time course of administration, will depend on the nature and severity of what is being treated. The prescription for treatment, i.e. stipulations regarding dose etc., is within the responsibility of general practitioners and other medical doctors. Appropriate doses of antibody are well known in the art; see Ledermann J.A. et al. (1991) Int. J. Cancer 47: 659-664; Bagshawe KD. et al. (1991) Antibodies, Immunoconjugates and Radiopharmaceuticals 4: 915-922.

The exact dose will depend on a number of factors, including whether antibodies are for diagnosis or for treatment, the size and location of the area to be treated, the exact nature of the antibody (eg, whole antibody, fragment, or diabody), and the nature of any detectable label or other molecule attached to the antibody. A typical antibody dose will be in the range of 100 µg to 1 g for systemic use, and 1 µg to 1 mg for topical use. Usually the antibody will be a complete antibody, preferably the IgG4 isotype. This is a dose for a simple treatment of an adult patient, which can be proportionally adjusted for children and newborns, and also adjusted for other antibody formats in relation to molecular weight. Treatments can be repeated at daily, twice a week, weekly or monthly intervals, with care from the doctor. In preferred embodiments according to the present invention, treatment is periodic, and the period between administrations is about two weeks or more, preferably about three weeks or more, more desirably about four weeks or more, or about once a month.

Specific binding members described herein will typically be administered in the form of a pharmaceutical composition, which may comprise at least one component in addition to the specific binding member.

Thus, pharmaceutical compositions according to the present invention, and can be used according to the present invention, can include, in addition to the active ingredient, a pharmaceutically acceptable excipient, carrier, buffer, stabilizer or other materials that are well known to those skilled in the art. Such materials should be non-toxic and should not interfere with the effectiveness of the active ingredient. The exact nature of the carrier and other material will depend on the route of administration, which may be oral or by injection, e.g. intravenous.

Pharmaceutical compositions for oral administration may be in tablet, capsule, powder or liquid form. A tablet may comprise a solid carrier such as gelatin or an excipient. Liquid pharmaceutical compositions generally comprise a liquid carrier such as water, petroleum, animal or vegetable oils, mineral oil or synthetic oil. Physiological saline, dextrose or other saccharide solution or glycols such as ethylene glycol, propylene glycol or polyethylene glycol may be included.

For intravenous injection or injection into the affected site, the active ingredient will be in the form of a parenterally acceptable aqueous solution which is pyrogen-free and has appropriate pH, isotonicity and stability. Those skilled in the art are able to make suitable solutions by, for example, using an isotonic vehicle such as sodium chloride injection, Ringer's injection, lactated Ringer's injection. Preservatives, stabilizers, buffers, antioxidants and/or other additives may be included where necessary.

A composition may be administered alone or in combination with other treatments, either simultaneously or sequentially depending on the condition to be treated.

Specific binding members of the present invention can be formulated in liquid or solid forms depending on the physicochemical properties of the molecule and the route of delivery. Formulations may include excipients, or combinations of excipients, for example: sugars, amino acids and surfactants. Liquid formulations can include a wide range of antibody concentrations and pH. Solid formulations can be produced by lyophilization, spray drying, or drying by supercritical fluid technology, for example. Formulations with anti-IL-13 will depend on the intended route of delivery: for example, formulations for pulmonary delivery may consist of particles with physical properties that ensure penetration into the deep lung by inhalation; topical formulations may include viscosity-modifying agents, which prolong the time the drug remains at the site of action.

The present invention provides a method comprising, causing or allowing binding of a specific binding member as provided herein to IL-13. in vitro, for example in ELISA, Western blotting, immunocytochemistry, immunoprecipitation, affinity chromatography or cell-based assays such as a TF-1 assay.

The amount of binding of specific binding member to IL-13 can be determined. Quantitation can be related to the amount of the antigen in a test sample that may be of diagnostic interest.

A kit comprising a specific binding member or antibody molecule according to any aspect or embodiment of the present invention was also described. In a kit, the specific binding member or antibody molecule can be labeled to allow its reactivity in a sample to be determined, e.g. as described further below. Components in a kit are generally sterile and in sealed tubes or other containers. The kit can be used in diagnostic analysis or other procedures for which antibody molecules are useful. A kit may contain instructions for using the components in a method, e.g. a method in accordance with the present invention. Associated materials to assist in or cause such a method to be performed may be included within a kit.

The reactivities of antibodies in a sample can be determined by any suitable means. Radioimmunoassay (RIA) is a possibility. Radioactively labeled antigen was mixed with unlabeled antigen (the test sample) and allowed to bind to the antibody. Bound antigen is physically separated from unbound antigen and the amount of radioactive antigen bound to the antibody is determined. The more antigen there is in the test sample, the less radioactive antigen will bind to the antibody. A competitive binding assay can also be used with non-radioactive antigen, by using antigen or an analogue coupled to a reporter molecule. The reporter molecule can be a fluorochrome, phosphor or laser dye with spectrally isolated absorption or emission characteristics. Suitable fluorochromes include fluorescein, rhodamine, phycoerythrin and Texas red. Suitable chromogenic stains include diaminobenzidine.

Other reporters include macromolecular colloidal particles or particulate matter such as latex beads that are colored, magnetic, or paramagnetic, and biological or chemically active agents that can directly or indirectly cause detectable signals that can be observed visually, detected electronically, or otherwise recorded. These molecules can be enzymes that catalyze reactions that develop or change colors or cause changes in electrical properties, for example. They can be molecularly excitable, so that electronic transitions between energy states result in characteristic spectral absorptions or emissions. They may include chemical objects used in conjunction with biosensors. Biotin/avidin or biotin/triptavidin and alkaline phosphatase detection systems can be used.

The signals generated by individual antibody reporter conjugates can be used to derive quantifiable absolute or relative data of the relevant antibody binding in samples (normal and test).

The present invention also provides the use of a specific binding member as above to measure antigen levels in a competition assay, i.e. a method for measuring the level of antigen in a sample by using a specific binding member as provided according to the present invention in a competition assay. This may be where the physical separation of bound from unbound antigen is not necessary. Coupling a reporter molecule to the specific binding member such that a physical or optical change takes place upon binding is one possibility. The reporter molecule can directly or indirectly generate detectable and preferably measurable signals. The connection of reporter molecules can be direct or indirect, covalent, e.g. via a peptide bond or non-covalently. Coupling via a peptide bond may result from recombinant expression of a gene fusion encoding antibody and reporter molecule.

The present invention also provides for measuring levels of antigen directly, by using a specific binding member according to the invention for example in a biosensor system.

The method of determining binding is not a feature of the present invention and those skilled in the art are able to select a suitable method according to their preference and general knowledge.

Also described herein is a specific binding member that competes for binding to IL-13 with any specific binding member defined herein, e.g. BAK502G9 IgG4. Competition between binding members can be easily analyzed in vitro, for example by tagging a specific reporter molecule to one binding member that can be detected in the presence of other untagged binding member or members, to enable the identification of specific binding members that bind the same epitope or an overlapping epitope .

Competition can be determined for example by using ELISA where IL-13 is immobilized to a plate and a first tagged binding member together with one or more other non-tagged binding members was added to the plate. The presence of an untagged binding member competing with the tagged binding member is observed by a decrease in the signal emitted by the tagged binding member.

To test for competition, a peptide fragment of the antigen can be used, especially a peptide that includes an epitope of interest. A peptide having the epitope sequence plus one or more amino acids at one or the other end can be used. Such a peptide may be said to "consist substantially" of the specified sequence. Specific binding members according to the present invention may be such that their binding for antigen is inhibited by a peptide with or including the sequence provided. To test for this, a peptide with both sequences plus one or more amino acids can be used.

Specific binding members that can bind a specific peptide can be isolated, for example, from a phage display library by clamping with the peptide or peptides.

The present invention further provides an isolated nucleic acid encoding a specific binding member of the present invention. Nucleic acid can include DNA and/or RNA. In a preferred aspect, the present invention provides a nucleic acid which codes for VH domain or VL domain or antibody molecule, e.g. scFv or IgG4, according to the invention as defined above.

The present invention also provides constructs in the form of plasmids, vectors, transcription or expression cassettes comprising at least one polynucleotide as above.

The present invention also provides a recombinant host cell comprising one or more of the constructs above. A nucleic acid encoding any of the VH domain or VL domain or antibody molecule, e.g. scFv or IgG4 as provided, in itself forms an aspect according to the present invention, so does a method for the production of the coded product, this method includes expression from coding nucleic acid. Expression can be readily achieved by culturing under appropriate conditions recombinant host cells containing nucleic acid. After production by expression of the VH or VL domain, a specific binding member can be isolated and/or purified using any suitable technique, and then used as appropriate.

Specific binding members, VH and/or VL domains, and encoding nucleic acid molecules and vectors according to the present invention can be provided isolated and/or purified, for example from their natural environment, in substantially pure or homogeneous form, or in the case of nucleic acid , free or substantially free of nucleic acid or genes of other origin than the sequence that codes for a polypeptide with the required function. Nucleic acid according to the present invention may comprise DNA or RNA and may be wholly or partially synthetic. Reference to a nucleotide sequence set forth here includes a DNA molecule with the specified sequence, and includes an RNA molecule with the specified sequence where U is substituted for T, unless context requires otherwise.

Systems for cloning and expression of a polypeptide in a variety of different host cells are well known. Suitable host cells include bacteria, mammalian cells, plant cells, yeast and baculovirus systems, and transgenic plants and animals. Mammalian cell lines available in the art for heterologous polypeptide expression include Chinese hamster ovary (CHO) cells, HeLa cells, baby hamster kidney cells, NSO mouse melanoma cells, YB2/0 rat myeloma cells, human embryonic kidney cells, human embryonic retinal cells, and many others. A common preferred bacterial host is E. coli.

The expression of antibodies and antibody fragments in prokaryotic cells such as E. coli is well established in the field. For an overview see, for example, Pluckthun, A. Bio/Technology 9: 545-551 (1992). Expression in eukaryotic cells in culture is also available to those skilled in the art as an option for production of a specific binding member eg Chadd HE and Chamow SM (2001) 110 Current Opinion in Biotechnology 12: 188-194, Andersen DC and Krummen L (2002) Current Opinion in Biotechnology 13: 117, Larrick JW and Thomas DW (2001) Current Opinion in Biotechnology 12: 411-418.

Suitable vectors can be selected or constructed, containing appropriate regulatory sequences, including promoter sequences, terminator sequences, polyadenylation sequences, enhancer sequences, marker genes and other sequences as appropriate. Vectors can be plasmids, viral e.g. subject, or mid subject as appropriate. For further details see, for example, Molecular Cloning: a Laboratory Manual. 3rd edition, Sambrook and Russel, 2001, Cold Spring Harbor Laboratory Press. Many known techniques and protocols for manipulating nucleic acid, for example in the preparation of nucleic acid constructs, mutagenesis, sequencing, introduction of DNA into cells and gene expression, and analysis of proteins, are described in detail in Current Protocols in Molecular Biology, Second Edition, Ausubel et al., ed., John Wiley & Sons, 1988, Short Protocols in Molecular Biology: A Compendium of Methods from Current Protocols in Molecular Biology, Ausubel et al., ed., John Wiley & Sons, 4th ed. 1999.

A further aspect according to the present invention thus provides a host cell containing nucleic acid according to the invention. Such a host cell may be in vitro and may be in culture. Such a host cell may be in vivo. In vivo presence of the host cell may allow intracellular expression of the specific binding members of the present invention as "intrabodies" or intracellular antibodies. Intrastoffer can be used for gene therapy [112].

A method is also described which comprises introducing such nucleic acid into a host cell. The introduction can use any available technique. For eukaryotic cells, suitable techniques may include calcium phosphate transfection, DEAE-dextran, electroporation, liposome-directed transfection and transduction using retroviruses or other viruses, e.g. vaccinia or for insect cells, baculovirus. To introduce nucleic acid into the host cell, a eukaryotic cell in particular can be used as a viral or a plasmid-based system. The plasmid system can be maintained episomally or it can be incorporated into the host cell or into an artificial chromosome [110, 111]. Incorporation can be either by random or targeted integration of one or more copies at a single or at several loci. For bacterial cells, suitable techniques may include calcium chloride transformation, electroporation and transfection using bacteriophage.

The introduction can be followed by causing or allowing expression from the nucleic acid, e.g. by culturing the host cells under conditions for expression of the gene.

In one embodiment, the nucleic acid according to the invention is integrated into the genome (eg chromosome) of the host cell. Integration can be promoted by the inclusion of sequences that promote recombination with the genome, according to standard techniques.

The present invention also provides a method comprising using a construct as mentioned above in an expression system to express a specific binding member or polypeptide as above.

Aspects and embodiments of the present invention will now be illustrated by way of examples with reference to the following experimentation.

EXAMPLE 1

Isolation of anti-IL-13 scFv

ScFv antistorage repertoire

A large single-chain Fv (scFv) human antibody library derived from splenic lymphocytes from 20 donors and cloned into a phagemid vector was used for selections [66].

Selection of scFv

ScFv recognizing IL-13 were isolated from phage display libraries in a series of repeated selection cycles for recombinant bacterially derived human or mouse IL-13 (Peprotech) essentially as described in [67]. Briefly, after incubation with the library, the immobilized antigen that had been previously coupled to paramagnetic beads, and bound phage were recovered by magnetic separation while unbound phage was washed away. Bound phage was then captured as described by Vaughan et al

[67] and the selection process was repeated. Different solid surfaces and capture procedures were used in different rounds of selection to reduce non-specific binding. Antigen was either covalently coupled to beads (Dynabeads M-270 carboxylic acid) or modified by biotinylation prior to the second capture by streptavidin-coated beads (Dynabeads M-280) according to the manufacturer's protocols (Dynal). A representative part of the clones from the yield of the selection rounds was DNA sequenced as described in Vaughan et al [67] and Osbourn et al [70]. Unique clones were examined for their ability to neutralize IL-13 as purified scFv preparations in IL-13 dependent cell proliferation assays.

Ribosome display libraries were created and screened for scFv that specifically recognize recombinant, bacterially derived human or mouse IL-13 (Peprotech), essentially as described in Hanes et al [113]. Initially, the BAK278D6 leader clone from the initial selections was converted to ribosome display format, and this template was then used for library generation. At the DNA level, a T7 promoter was added to the 5' end for efficient transcription into mRNA. At the mRNA level, the construct contained a prokaryotic ribosome-binding site (Shine-Dalgarno sequence). At the 3' end of the single chain, the stop codon was removed and part of glll (gene HI) was added to act as a spacer [113].

Ribosome display libraries derived from BAK278D6 were generated by mutagenesis of antibody complementarity determining regions (CDRs) where PCR reactions were performed with non-proofreading Taq polymerase. Affinity-based selections were performed where, after incubation with the library, the biotinylated human IL-13 was captured by streptavidin-coated paramagnetic beads (Dynal M280) and bound tertiary complexes (mRNA-ribosome-scFv-IL-13) were recovered by magnetic separation while unbound complexes were washed away. The mRNA encoding the bound scFvs was then removed by RT-PCR as described in Hanes et al [113] and the selection process was repeated with decreasing concentrations (loo nM - 100 pM over 5 rounds) with biotinylated human IL-13 present. during the selection.

"Error-prone" PCR was also used to further increase library size. Three false intensities were used (2.0, 3.5 and 7.2 mutations per 1,000 bp following a standard Per reaction, as described in the manufacturer's protocol (Clontech)) during the selection regimen. Initial error-prone PCR reactions took place before round one selections were initiated at 100 nM. A subsequent round of error-prone PCR was performed before round three selections at 10 nM biotinylated human IL-13. As above, a representative part of the clones from the yield of the selection rounds was DNA sequenced as described in Vaughan et al [67] and Osbourn et al [70]. Unique clones were examined for their ability to neutralize IL-13 as purified scFv preparations in IL-13 dependent cell proliferation assays.

EXAMPLE 2

Neutralizing potency of anti-IL-13 scFv in the IL-13 dependent TF-1 cell proliferation assay

The neutralizing potency of purified scFv preparations against human and mouse IL-13 bioactivity was investigated using the TF-1 cell proliferation assay. Purified scFv preparations were made as described in example 3 in WO01/66754. Protein concentrations of purified scFv preparations were determined using the BCA method (Pierce). TF-1 is a human premyeloid cell line established from a patient with erythroleukemia [68]. The TF-1 cell line is factor dependent for survival and proliferation. Accordingly, TF-1 cells responded to either human or mouse IL-13 [69] and were maintained in medium containing human GM-CSF (4 ng/ml, R&D Systems). Inhibition of IL-13 dependent proliferation was determined by measuring the reduction in incorporation of tritiated thymidine into the newly synthesized DNA of dividing cells.

TF-1 cell assay protocol

TF-1 cells were obtained from R&D Systems and maintained according to the attached protocols. Assay medium comprised RPMI-1640 with GLUTAMAXI (Invitrogen) containing 5% fetal bovine serum (JFH) and 1% sodium pyruvate (Sigma). Before each analysis, TF-1 cells were pelleted by centrifugation at 300 x g for 5 minutes, the medium was removed by aspiration and the cells resuspended in analysis medium. This process was repeated twice with cells resuspended to a final concentration of 10<5> cells/ml in assay medium. Test solutions with antibody (in triplicate) were diluted to the desired concentration in analysis medium. An irrelevant antibody not directed against IL-13 was used as a negative control. Recombinant bacterially derived human or mouse IL-13 (Peprotech) was added to a final concentration of 50 ng/ml when mixed with the appropriate test antibody in a total volume of 100 µl/well in a 96-well assay plate. The concentration of IL -13 used in the assay was selected as the dose which, at the final assay concentration, produced approximately 80% of the maximal proliferative response. All samples were incubated for 30 minutes at room temperature. 100 µl of resuspended cells were then added to each assay point to give a total assay volume of 200 (xl/well. Assay plates were incubated for 72 hours at 37°C under 5% CO2. 25 µl of tritiated thymidine (10 (xCi/ml, NEN) was then added to each assay point and the assay plates were returned to the incubator for another 4 hours. Cells were harvested on glass fiber filter plates (Perkin Eimer) using a cell harvester. Thymidine incorporation was determined using a Packard TopCount microplate liquid scintillation counter. Data were analyzed using end Graphpad Prism software.

Results

Despite alternating selection cycles between human and mouse antigen, no cross-reactive neutralizing antibodies were obtained. Two distinct anti-human and one anti-mouse IL-13 neutralizing scFvs were obtained from selections. BAK278D6 (VH SEQ ID NO: 13; VL SEQ ID NO: 14) and BAK167A11 (VH SEQ ID NO: 23; VL SEQ ID NO: 24) recognized human IL-13 most BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 24) ID NO: 26 recognized mouse IL-13. BAK278D6 (Figure 2) and BAK167A11 (Figure 1) as scFv neutralized 25 ng/ml human IL-13 with an IC50 of 44 nM and 111 nM, respectively. BAK209B11 (Figure 3) as a The scFv neutralized 25 ng/ml mouse IL-13 with an IC50 of 185 nM.

EXAMPLE 3

Neutralizing potency of leader clones from targeted optimization of heavy chain CDR3 from parental clones in IL-13 dependent TF-1 cell proliferation assay

Osbourn et al. [70] have shown that targeted mutagenesis of residues within the heavy chain CDR3 can significantly improve the affinity of antibodies. Selections were performed as described in Example 1, on scFv repertoires where residues within the heavy chain CDR3 to BAK278D6 (SEQ ID NO: 6) BAK167A11 (SEQ ID NO: 57) had been randomized by mutagenesis. Unique clones from selection yields were identified by DNA sequencing and their neutralizing potency examined as scFv in the TF-1 cell proliferation assay, as described in Example 2.

Results

Significant gain in potency was achieved for both lines. The most potent clones from the BAK167A11 line were BAK615E3, BAK612B5 and BAK582F7 which had scFv IC50 of 3 nM (figure 1), 6.6 nM, 6.65 nM respectively against 25 ng/ml human IL-13 in the TF-1 cell proliferation assay. From the BAK278D6 line, the most potent clone was BAK502G9, which had as scFv IC50 of 8 nM against 25 ng/ml human IL-13 in the TF-1 cell proliferation assay (figure 2).

EXAMPLE 4

Neutralizing potency of BAK167A11 and BAK278D6 lines against non-human primate IL-13 and an IL-13 variant associated with asthma in the TF-1 factor-dependent cell proliferation assay

None of the BAK167A11 and BAK278D6 human IL-13 neutralizing lines cross-reacted with mice. The inventors therefore determined the following criteria for the line selected for further optimization and clinical development: should preferably cross-react with non-human primate IL-13 and should recognize a variant of IL-13 where arginine at amino acid position 130 is substituted with glutamine (R130Q). This variant has been genetically associated with asthma and other allergic diseases [37, 39, 41, 71]. Cross-reactivity was determined by the ability of purified scFv preparations to bind to non-human primate IL-13 and IL-13 variant by surface plasmon resonance (BIAcore) analysis. Functional activity was determined using the TF-1 cell proliferation assay.

Production of wild-type, variant and non-human primate IL-13

A cDNA for wild-type human IL-13 was obtained from InvivoGen and modified by site-directed mutagenesis (Stratagene Quikchange<®>kit) to yield a cDNA encoding a variant IL-13. The coding sequence for both rhesus and cynomolgus monkey IL-13 was obtained by PCR on a genomic DNA template using degenerate primers based on the human IL-13 sequence. Both non-human primate (rhesus and cynomolgus) sequences were identical to each other, but differed from human IL-13 by seven amino acids (Figure 19). Recombinant wild-type, variant and non-human primate IL-13 was then expressed using the baculovirus expression system (Invitrogen). Expression constructs had a carboxyl-terminal affinity tag added to the expressed protein which allowed purification from insect cell-conditioned medium to near homogeneity.

Qualitative binding analysis using BIAcore

The binding affinity of purified scFv preparations to non-human primate, variant and wild-type IL-13 was determined by surface plasmon resonance measurements using a

BIAcore 2000 Biosensor (BIAcore AB) as described in Karlsson et al [72]. Briefly, IL-13 was coupled to the CM5 sensor chip by using an amine coupling kit (BIAcore) at a surface density of about 200 Ru and three test concentrations of scFv (about 350 nM, 175 nM and 88 nM) in HBS-EP buffer passed over the sensor chip surface. The resulting sensorgrams were evaluated using BIA evaluation 3.1 software to provide relative binding data.

TF-1 analysis protocol

The assay was performed essentially as described in Example 2 with the following modifications: non-human primate IL-13, human variant IL-13 (R130Q) and wild-type human IL-13 were used at concentrations of 50 ng/ml, 25 ng/ml, respectively. ml and 25 ng/ml.

Results

BIAcore binding assay data indicated that the BAK278D6 but not the BAK167A11 line had the necessary cross-reactivity profile for further therapeutic development (Table 2). This finding was supported by bioassay data showing that BAK278D6 (Figure 4) and BAK501G9 (Figure 6) were able to neutralize human IL-13, the human IL-13 (R130Q) variant and non-human primate IL-13 in TF -1 the cell proliferation assay with near equivalent potency. Although BAK615E3 (VH SEQ ID NO: 33, VL SEQ ID NO: 34) had a significantly increased potency against human IL-13 relative to its parent BAK167A11 (VH SEQ ID NO: 23; VL SEQ ID NO: 24) in the TF-1 cell proliferative assay (Figure 1), then no clone bound non-human primate or variant IL-13 in the BIAcore binding assay.

Germline framework regions of BAK278D6 and BAK502G9

The deduced amino acid sequence of BAK278D6 VH (SEQ ID NO: 13) and VL (SEQ ID NO: 14) was aligned against known human germline sequences in the VBASE database

[73] and the closest germline identified by sequence similarity. The closest germline for the VH domain of BAK278D6 (SEQ ID NO: 149 and its derivatives, was identified as DP14, a member of the VH1 family. The BAK278D6 VH has 9 changes from the DP14 germline in-frame regions. The closest germline to the VL in BAK278D6 was identified as Vx3 3h. The BAK278D6 VL domain (SEQ ID NO: 14) has only 5 changes from the germline within framework regions. Framework regions of BAK278D6 and its derivatives were returned to the germline by site-directed mutagenesis (Stratagene Quikchange kit) to identically match native human antibodies.

EXAMPLE 5

Neutralization potency of leader clones from targeted optimization of heavy chain CDR1 and heavy chain CDR2 sequences of BAK502G9 in the human IL-13 dependent TF-1 cell proliferation assay

A second phase of optimization was performed using the BAK502G9 sequence, with germline framework regions as a template. Selections were performed essentially as described in Example 1 on scFv repertoires where both residues within the heavy chain CDR1 or heavy chain CDR2 in BAK502G9 had been randomized by mutagenesis. Unique clones from the selection yield were identified by DNA sequencing and their neutralizing potency tested as purified scFv preparations in the TF-1 cell proliferation assay as described in Example 2. Vectors were constructed for the most potent scFv clones to allow re-expression as fully human IgG4 antibody as described by Persic et al. (1997) Genes 187; 9-18) with a few modifications. An oriP fragment was included in the vectors to facilitate the applications in HEK-EBNA 292 cells and to allow episomal replication. The VH variable domain was cloned into the polylinker between the secretion leader sequence and the human gamma 4 constant domain in the expression vector pEU8.1(+). The VL variable domain was cloned into the polylinker between the secretion leader sequence and the human lambda constant domain in the expression vector pEU4.1(-).

Whole antibody was purified from conditioned medium of EBNA-293 cells co-transfected with constructs expressing heavy and light chains by protein A affinity chromatography (Amersham Pharmacia). The purified antibody preparations were sterile filtered and stored at 4°C in phosphate buffered saline (PBS) before evaluation. Protein concentration was determined by measuring absorbance at 280 nM using the BCA method (Pierce). Reformatted human IgG4 whole antibodies were compared with commercially available anti-human IL-13 antibodies in the TF-1 proliferation assay described in Example 2.

Results

As shown in figure 5, the commercial antibody B-B13 (mouse IgGl - Euroclone 5) was shown to be significantly more potent against human IL-13 than the commercial antibody JES10-5A1 (rat IgGl - Biosource) with an IC5o of 1021 pM respectively and 471 pM. Eight clones, namely BAKI 111D10, BAKI 166G02, BAKI 167F02, BAKI 167F04, BAKI 183H4, BAKI 184C8, BAKI 185E1, BAKI 185F8, derived from BAK502G9 (hence "BAK502G9 line"), in which the heavy chain CDR1 or CDR2 had been targeted , showed improved potency as scFv compared to the commercial antibodies. These improvements were maintained upon conversion to whole antibody human IgG4. Derivatives of BAK502G9 as whole antibodies (IgG4) had an IC50i ranging from 244 pM to 283 pM. BAK502G9 as a whole antibody IgG4 had an IC50 of 384 pM. In summary, large improvements in potency could be provided by targeting heavy chain CDR1 (SEQ ID NO: 7) or CDR2 (SEQ ID NO: 8) of BAK502G9. Statistical comparisons with B-B13 were made using an ANOVA followed by a Dunnet's post test analysis (InStat software).

Further characterization

Selected anti-human antibodies from the BAK278D6 line underwent further characterization to determine their specificity. These included BAK502G9 (VH SEQ ID NO: 15; VL SEQ ID NO: 16) and its derivatives BAKI 167F1 (VH SEQ ID NO: 35; VL SEQ ID NO: 36) and BAKI 183H4 (VH SEQ ID NO: 37; VL SEQ ID NO: 38), which are representative examples of clones with modifications to heavy chain CDR1 and heavy chain CDR2 respectively in BAK502G9.

EXAMPLE 6

Neutralizing potency of leader clones from targeted optimization of heavy chain CDR1 and heavy chain CDR2 sequences of BAK502G9 against non-human primate IL-13 and an IL-13 variant associated with asthma in the TF factor-dependent cell proliferation assay.

Cross-reactivity of anti-human IL-13 antibodies was determined by their ability to inhibit non-human primate IL-13 and IL-13 variant-driven TF-1 cell proliferation as described in Example 4.

Results

Optimized anti-human IL-13 antibodies BAKI 167F2 (VH SEQ ID NO: 35; VL SEQ ID NO: 36) and BAKI 183H4 (VH SEQ ID NO: 37; VL SEQ ID NO: 38) maintained the specificity of their parent BAK502G9 ( VH SEQ ID NO: 15; VL SEQ ID NO: 16) (Figure 6). Potency gains against wild-type IL-13 were reflected in their ability to neutralize non-human primate IL-13 and IL-13 variants with substantially equivalent potency. The IC 50 for BAK502G9 against human, human variant and non-human primate IL-13 was 1.4 nM, 1.9 nM and 2.0 nM, respectively. The IC50 of BAKI 167F2 against human, human variant and non-human primate IL-13 was 1.0 nM and 1.1 nM and 1.3 nM, respectively. The IC 50 of BAKI 183H4 against human, human variant and non-human primate IL-13 was 0.9 nM, 1.0 nM and 1.6 nM, respectively. These clones are suitable for therapeutic use.

EXAMPLE 7

Neutralizing potency of lead anti-human IL-13 antibodies against native human IL-13 in the HDIM-2 cell proliferation assay

The human IL-13 sequence has four potential N-glycosylation sites. The inventors have demonstrated the ability of BAK278D6 and its derivatives to neutralize recombinant IL-13 expressed either in bacterial or baculovirus expression systems. Although there is evidence that many processing events known in mammalian systerns also occur in insects, key differences are in protein glycosylation, particularly N-glycosylation [74].

The inventors investigated the ability of BAK278D6 derivatives to neutralize native IL-13 released from human cells.

HDLM-2 cells were isolated by Drexler et al [75] from a patient with Hodgkin's disease. Skinnider et al [76] showed that HDLM-2 cell proliferation was partially dependent on autocrine and paracrine release of IL-13. Lead anti-human IL-13 antibodies were examined for their ability to inhibit HDLM-2 cell proliferation driven by the release of native (or naturally occurring) IL-13.

HDLM- 2 cell analysis protocol

HDLM-2 cells were obtained from the Deutsche Sammlung von Mikroorganismenund Zellkulturen (DSMZ) and maintained according to provided protocols. The assay medium comprised RPI-1640 with Glutamax I (Invitrogen) containing 20% fetal bovine serum. Before each analysis, the cells were pelleted by centrifugation at 300 x g for 5 minutes, the medium was removed by aspiration and the cells were resuspended in fresh medium. This process was repeated three times and the cells were finally resuspended in a final concentration of 2 x 10<5> cells/ml in analysis medium. 50 µl of resuspended cells were added to each assay point in a 96 well assay plate. Test solutions with antibodies (in triplicate) were diluted to the desired concentration in analysis medium. An irrelevant isotype antibody not directed against IL-13 was used as a negative control. The appropriate test antibody in a total volume of 50 µl/well was added to the cells, each assay point giving a total assay volume of 100 µl/well. Assay plates were incubated for 72 hours at 37°C under 5% CO 2 . 25^1 tritiated thymidine (10^Ci/ml, NEN) was then added to each assay point and assay plates were placed back into the incubator for another 4 hours. Cells were harvested on glass fiber filter plates (Perkin Eimer) using a cell harvester. Thymidine incorporation was determined by using a Packard TopCount microplate liquid scintillation counter Data were analyzed using Graphpad Prism software.

Results

As shown in Figure 7, BAK502G9 (VH SEQ ID NO: 15; VL SEQ ID NO: 16), and its derivatives BAKI 183H4 (VH SEQ ID NO: 37; VL SEQ ID NO: 38) and BAKI 167F2 (VH SEQ ID NO: 35; VL SEQ ID NO: 36) capable of causing a dose-dependent inhibition of cell proliferation with relative potencies similar to those observed in other bioassays. IC50 of BAK502G9, BAKI 183H4, BAKI 167F2 as human IgG4 were 4.6 nM, 3.5 nM and 1.1 nM respectively. The IC50 for the commercial antibodies JES10-5A2 and B-B13 were 10.7 nM and 16.7 nM, respectively.

EXAMPLE 8

Neutralizing potency of lead anti-human IL-13 antibodies against IL-13-dependent responses in disease-relevant primary cells

Secondary bioassays were performed using primary cells and readouts more relevant to respiratory disease. These included eotaxin release from normal human lung fibroblasts (NHLF) and vascular adhesion molecule 1 (VCAM-1) upregulation on the surface of human umbilical cord endothelial cells (HUVEC). Both IL13-dependent responses could contribute to eosinophil recruitment, a feature of asthma phenotypes [92].

NHLF assay protocol IL-13 has been shown to cause eotaxin release from lung fibroblasts [77] [78] [79]. Factor-dependent eotaxin release from NHLF was determined by ELISA.

NHLF was obtained from Biowhittaker and maintained according to provided protocols. The assay medium was FGM-2 (Biowhittaker). Test solutions with antibody (in triplicate) were diluted to the desired concentration in assay medium. An irrelevant antibody not directed against IL-13 was used as a negative control. Recombinant bacterially derived human IL-13 (Peprotech) was then added to a final concentration of 10 ng/ml and mixed with the appropriate test antibody in a total volume of 200 µl. The concentration of IL-13 used in the assay was selected as the dose that produced approximately 80% of the maximal response. All samples were incubated for 30 minutes at room temperature. Analysis samples were then added to NHLF that had previously been seeded at a density of 1 x 10<4> cells per well in 96-well analysis plates. Assay plates were incubated at 37°C for 16-24 hours at 37°C under 5% CO2. Assay plates were centrifuged at 300 x g for 5 min to pellet detached cells. Eotaxin levels in the supernatant were determined by ELISA using reagents and methods described by the manufacturer (R&D Systems). Data were analyzed using Graphpad Prism software.

Results

BAK278D6 line clones were able to inhibit human IL-13 dependent eotaxin release from NHLF. Relative potency was similar to that observed in the TF-1 cell proliferation assay (Figure 8). BAK502G9 (VH SEQ ID NO: 15; VL SEQ ID NO: 16), BAK1183H4 (VH SEQ ID NO: 37; VL SEQ ID NO: 38), BAK1167F2 (VH SEQ ID NO: 35; VL SEQ ID NO: 36) had IC50 of 207 pM, 118 pM and 69 pM, respectively, against 10 ng/ml human IL-13. Commercial antibodies JES10-5A2 and B-B13 had IC50 of 623 pM and 219 pM, respectively.

HUVEC assay protocol

IL-13 has been shown to upregulate expression of VCAM-1 on the cell surface of HUVEC [80, 81]. Factor-dependent VCAM-1 expression was determined by detection of upregulation of VCAM-1 receptor cellular expression using a time-resolved fluorescence readout.

HUVEC were obtained from Biowhittaker and maintained according to provided protocols. The assay medium was EGM-2 (Biowhittaker). Test solutions with antibody (in triplicate) were diluted to the desired concentration in analysis medium. An irrelevant antibody not directed against IL-13 was used as a negative control. Recombinant bacterially derived human IL-13 (Peprotech) was added to a final concentration of 10 ng/ml and it was mixed with the appropriate test antibody in a total volume of 200 µl The concentration of IL-13 used in the assay was chosen as the dose which gave about 80% of the maximum response. All samples were incubated for 30 minutes at room temperature. Analysis samples were then added to HUVEC that had previously been seeded at 4 x 10 cells per well in 96-well analysis plates. Assay plates were incubated at 37°C for 16-20 hours under 5% CO2. Assay medium was then removed by aspiration and replaced with blocking solution (PBS containing 4% dry Marvel milk powder). Assay plates were incubated at room temperature for 1 hour. Wells were washed three times with PBST Tween before 100 µl (1:500 dilution in PBST/1% Marvel<®>) of biotinylated anti-VCAM-1 antibody (Serotec) was added to each well. Assay plates were incubated at room temperature for 1 hour. Wells were washed three times with Delfia wash buffer (Perkin Eimer) before 100 µl Eurpium-labeled streptavidin or anti-mouse IgG1 (1:1000 dilution in Delfia assay buffer, Perkin Eimer) was added to each well. Assay plates were then incubated at RT for 1 hour. Wells were washed 7 times with Delfia wash buffer

(Perkin Eimer). Finally, 100 µl of enhancer solution (Perkin Eimer) was added to each well and fluorescence intensity was determined using a Wallac 1420 VTKTOR2 plate reader (Standard Europium protocol). Data were analyzed using Graphpad Prism software.

Results

Typical data for BAK502G9 (VH SEQ ID NO: 15; VL SEQ ID NO: 16), BAKI 183H4 (VH SEQ ID NO: 37; VL SEQ ID NO: 38), BAK1167F2 (VH SEQ ID NO: 35; VL SEQ ID NO: 36) as complete antibody human IgG4 is shown in Figure 9. Relative potency was similar to that observed in the TF-1 cell proliferation assay. The IC50 for BAK502G9, BAKI 183H4 and BAKI 167F2 were 235 pM, 58 pM and 55 pM, respectively, against 10 ng/ml human IL-13.

EXAMPLE 9

Neutralizing potency of anti-IL-13 antibodies against IL- i<p>and IL-4 dependent on VCAM-1 upregulation

The specificity of the BAK278D6 line of clones was investigated in a modification of the HUVEC bioassay. Together with IL-13, both IL-4 and IL-ip have been shown to upregulate expression of VCAM-1 on the cell surface of HUVEC [80, 81].

HUVEC assay protocol

The assay was performed essentially as described in Example 5 with the following modifications. Recombinant human IL-1β and IL-4 (R&D Systems) were used in place of human IL-13 at 0.5 ng/ml and 1 ng/ml, respectively, and represented the dose that produced approximately 80% of the maximal response.

Results

None of the clones evaluated from the BAK278D6 line neutralized VCAM-1 upregulation in response to either human IL-1β or IL-4, thus demonstrating specificity for IL-13 (Figure 10). IL-4 is the most closely related to IL-13, sharing 30% sequence identity at the amino acid level [82].

COMPARISON EXAMPLE 10

Neutralizing potency of BAK209B11 as a human IgG4 in a mouse IL-13 dependent mouse B9 cell proliferation assay

BAK206B11, identified as an anti-mouse IL-13 neutralizing clone as an scFv as described in Example 1, was reformatted as a whole antibody human IgG4 as described in Example 5 and its potency evaluated in the mouse IL-13 dependent B9 cell proliferation assay . B9 is a mouse B-cell hybridoma cell line [83]. B9 is factor dependent for survival and proliferation. In this regard, B cells respond to mouse IL-13 and are maintained in medium containing human IL-6 (50 pg/ml, R&C Systems). Inhibition of mouse IL-13 dependent proliferation was determined by measuring the reduction in incorporation of tritiated thymidine into the newly synthesized DNA of dividing cells.

B9 cell assay protocol

B9 cells were obtained from the European Collection of Animal Cell Culture ECACC and maintained according to the provided protocols. The assay was performed essentially as described for the TF-1 assay in Example 2, but with the following modifications. The assay medium comprised RPMI-1640 with GLUTAMAX 1 (Invitrogen) containing 5% fetal bovine serum (Hyclone) and 50 µM 2-mercaptoethanol (Invitrogen). Recombinant bacterially derived mouse IL-13 (Peprotech) replaced human IL-13 at a final assay concentration of 1 ng/ml.

Results

BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 26) as a human IgG4 neutralized 1 ng/ml mouse IL-13 with an IC50 of 776 pM in the B9 assay (Figure 11). BAK209B11 therefore represented a useful tool for investigating the role of IL-13 in mouse models of disease. This is clearly demonstrated in Example 12, which shows the efficacy of BAK209B11 in a mouse model of acute lung inflammation.

EXAMPLE 11

Determination of affinity of anti-IL-13 antibodies by BIAcore analysis

The affinity of BAK502G9 (VH SEQ ID NO: 15; VL SEQ ID NO: 16), BAKI 167F2

(VH SEQ ID NO: 35; VL SEQ ID NO: 36) and BAKI 183H4 (VH SEQ ID NO: 37; VL SEQ ID NO: 38) for human IL-13 and BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 26) for mouse IL-13 as human IgG4 was determined by surface plasmon resonance

measurements using a BIAcore 2000 Biosensor (BIAcore AB) essentially as described in [72]. Briefly, antibodies were coupled to the CM5 sensor chip using an amine coupling kit (BIAcore) at a surface density of approximately 500 Ru and a serial dilution of IL-13 (between 50 nM and 0.78 nM) in HBS-EP buffer passed over the sensor tip surface. The resulting sensorgrams were evaluated using BIA evaluation 3.1 software to provide kinetic data.

Results

BAK502G9, BAKI 167F2 and BAKI 183H4 in IgG4 bound human IL-13 with high affinity with Kd of 178 pM, 136 pM and 81 pM, respectively, corresponding to their relative potency in cell-based assays. BAK209B11 bound mouse IL-13 with an affinity of 5.1 nM (Table 3).

COMPARISON EXAMPLE 12

Efficacy of BAK209B11 in a mouse model of acute allergic lung inflammation

Mouse model with acute allergic lung inflammation

The efficacy of BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 26), an anti-mouse IL-13 neutralizing human IgG4 antibody, was investigated in a mouse model of acute allergic lung inflammation. This model was performed essentially as described by Riffo-Vasquez et al [84] and is characterized by its end point by increased bronchial alveolar secretion (BAL) IL-13 (Figure 12), cellular infiltration in the lung and BAL (Figure 13), increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

Female Balb/C mice (Charles River UK) were treated with either anti-mouse IL-13 antibody BAK209B11 (at 12, 36, 119 or 357 µg doses) or an isotype-matched control antibody (357 µg dose). On days 0 and 7, mice in each group were sensitized by intraperitoneal injection of 10 µg ovalbumin (Ova) in 0.2 ml of vehicle (saline containing 2% AI2O3 (Rehydragel) as an excipient). A separate control group of non-sensitized mice received an equal volume of the vehicle. Mice were challenged with ovalbumin on days 14, 15, and 16. Ovalbumin was diluted to 1% (w/v) in sterile saline prior to nebulization. All inhalation challenges were administered in a Plexiglas exposure chamber. Ova was aerosolized using a deVilbiss Ultraneb 2000 nebulizer (Sunrise Medical) in a series of three 20-minute exposures separated by 1-hour intervals.

BAK209B11 or an irrelevant human IgG4 was administered intravenously, 1 day before the first challenge and then 2 hours before each subsequent challenge (4 doses in total). The model ended at day 17, 24 hours after the last challenge. Blood (serum) and BAL were collected. Serum was analyzed from total IgE. BAL was obtained by injecting 3 aliquots of saline (0.3 ml, 0.3 ml and 0.4 ml) and pooling the samples. Total leukocyte and differential cell count were obtained from BAL cells.

Results

Ovalbumin challenge of sensitized mice caused a significant (p<0.05) increase in total BAL cell recruitment compared to non-sensitized but challenged mice. This recruitment was dose-dependently inhibited by BAK209B11; significant (p<0.05) inhibition was seen with >36 µg BAK209B11 but no control antibody (Figure 13). Similar effects were also seen on eosinophils (Figure 14) and neutrophils (Figure 15) with significant (p<0.05) inhibition of cellular influx at a minimal BAK209B11 dose of 36 µg. This inhibition was not seen with the control antibody. Lymphocytes were also induced in sensitized but not in non-sensitized mice upon challenge. This induction was dose-dependently inhibited by BAK209B11, with maximal inhibition seen with 36 µg BAK209B11. Control antibody had no effect (Figure 16). Although monocyte/macrophages were not induced in sensitized animals compared to non-sensitized animals, background levels were reduced by >36 µg BAK209B11, but not by control antibody (Figure 17). Serum IgE levels were significantly increased in sensitized animals compared to non-sensitized animals after challenge (p<0.05). This increase was reduced after treatment with 36 µg BAK209B11, but not by the control antibody.

In summary, systemic administration of BAK209B11, a mouse IL-13 neutralizing antibody, but not control antibody, inhibited inflammatory cell influx and the upregulation of serum IgE levels caused by sensitization and subsequent challenge with ovalbumin in a mouse model of allergic inflammation.

Examples 13 to 20 are prophetic.

COMPARISON EXAMPLE 13

The effect of BAK209B11 in the Lloy mouse model of acute lung inflammation Mouse model of acute allergic lung inflammation

The efficacy of BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 26), an anti-mouse IL-13 neutralizing antibody, was examined in a second mouse model of acute allergic lung inflammation. This model was performed essentially as described by McMillan et al. [85] and is characterized by its endpoint by increased BAL and lung tissue IL-13, cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

Female Balb/C mice (Charles River UK) were administered different doses of anti-mouse IL-13 antibody BAK209B11 or an isotype-matched control antibody as follows. On days 0 and 12, mice in each group were sensitized (SN) by intraperitoneal injection of 10 (xg ovalbumin (Ova) in 0.2 ml of the vehicle (saline containing 2 mg Al (OH)3 as an excipient [calculated as described in Example 12]).A separate control group of non-sensitized mice (NS) received an equal volume of the vehicle. Mice were challenged with ovalbumin for 20 minutes on days 19, 20, 21, 22, 23 and 24. Ovalbumin was diluted to 5% (w/v) in saline prior to nebulization. All inhalation challenges were administered in a Plexiglas exposure chamber. Ova was aerosolized using a deVilbiss Ultraneb 2000 nebulizer (Sunrise Medical). On days 18, 19, 20, 21, 22, 23 and 24, mice were administered different intraperitoneal doses (237 µg, 23.7 µg or 2.37 µg; indicated in Figures 21 at H, M and L) of anti-mouse IL-13 antibody BAK209B11 mulgGI or an isotype-matched control antibody ( 237 u.g). Respiratory function was examined on days 0 and 25 during increasing methacholine challenges r and measured by using conscious plethysmography (Buxco). PC50 (concentration of methacholine required to increase baseline PenH by 50%) was estimated for individual mice at both day 0 and day 25 from 4-parameter free curve fitting methacholine dose-response curves.

The model ended at day 25, 24 hours after the last challenge. Blood, serum, BAL and lung tissue were collected.

Results

Lung function was evaluated for individual animals at day 0 (pre-treatment) and vday 25 (post-challenge) and was quantified by calculating PC50 values (concentration of methacholine required to increase baseline PenH by 50%) (Figure 21 A). An individual's airway hyperresponsiveness (AHT) was determined by the change in log PC50 at day 25 versus day 0 (log day 25 PC50 - log day 0 PC50). This delta logPCsovar at primary endpoints of the study; PC50 data were log transformed due to the requirements for endpoint ANOVA. Individual changes were averaged within group to generate group mean delta log PC50 (as shown in Figure 21B).

Ovalbumin challenge of sensitized mice caused a significant AHR compared to non-sensitized and challenged mice (p<0.01). BAK 209B11 caused a clear and dose-dependent decrease in AHR while the control antibody had no effect.

COMPARISON EXAMPLE 14

Efficacy of BAK209B11 in the Gerard mouse model of acute lung inflammation

Mouse model for acute allergic lung inflammation

The efficacy of BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 26), an anti-mouse IL-13 neutralizing human IgG4 antibody, was investigated in a third mouse model of acute allergic lung inflammation. This model was performed essentially as described by Humbles et al. [86] and is characterized by its endpoint by increased BAL and lung tissue IL-13, cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

Female Balb/C mice (Charles River UK) were administered different doses of anti-mouse IL-13 antibody BAK209B11 or an isotype-matched control antibody. On days 0, 7 and 14, mice in each group were sensitized (SN) by intraperitoneal injection of 10 µg ovalbumin (Ova) in 0.2 ml of the vehicle (saline containing 1.125 mg Al(OH)3 as an excipient [calculated as described in example 12]). A separate control group of non-sensitized mice (NS) received an equal volume of the vehicle. Mice were challenged with ovalbumin for 20 min on days 21, 22, 23, and 24. Ovalbumin was diluted to 5% (w/v) in saline prior to nebulization. All inhalation challenges were administered in a Plexiglas exposure chamber. Ova was aerosolized using a deVilbiss Ultraneb 2000 nebulizer (Sunrise Medical).

The model ended at day 25, 24 hours after challenge. Blood, serum, BAL and lung tissue were collected.

COMPARISON EXAMPLE 15

Efficacy of BAK209B11 in the Lloyd chronic model of lung inflammation

Mouse model for chronic allergic lung inflammation

The efficacy of BAK209B11 (VH SEQ ID NO: 25; VL SEQ ID NO: 26), an anti-mouse IL-13 neutralizing human IgG4 antibody, was investigated in a model of chronic allergic lung inflammation. This model was performed essentially as described by Temelkovski et al. [87] and is characterized by its end point of cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

Female Balb/C mice (Charles River UK) were dosed with different doses of anti-mouse IL-13 antibody BAK209B11 or an isotype-matched control antibody. On days 0 and 11, mice in each group were sensitized (SN) by intraperitoneal injection of 10 (xg ovalbumin (Ova) in 0.2 ml of the vehicle (saline containing 2 mg of Al (OH)3 as an excipient [calculated as described in Example 12]).A separate control group of non-sensitized (NS) mice received an equal volume of the vehicle. Mice were challenged with ovalbumin for 20 minutes on days 18, 19, 20, 21, 22, 23, 28, 30 , 32, 35, 37, 39, 42, 44, 46, 49, and 51. Ovalbumin was diluted to 5% (w/v) in saline prior to nebulization. All inhalation challenges were administered in a Plexiglas exposure chamber. Ova was aerosolized using a deVilbiss Utraneb 2000 nebulizer (Sunrise medical).

The model ended at day 52, 24 hours after challenge. Blood, serum, BAL and lung tissue were collected.

EXAMPLE 16

Efficacy of anti-human IL-13 antibodies against exogenous human IL-13 given to the mouse air pocket model

The effect of anti-human IL-13 antibodies on the pro-inflammatory action of human IL-13 was investigated in a basal mouse model. This model was carried out essentially as described by Edwards et al [93] and was characterized by the end point of cellular infiltration into the air pocket.

Model protocol

An air pocket was created on the back of female Balb/C mice by subcutaneous injection of 2.5 ml of sterile air at day 0. The air pocket was reinflated with another 2.5 ml of sterile air at day 3. 2 µg of huIL-13 in 0, 75% CMC was injected directly into the lung at day 6. 24 hours later, the mice were killed and the air pocket was flushed with 1 ml of heparinized saline. Antibody treatments were either given with huIL-13 (into the pocket) or given systemically.

Results

Human IL-13, injected into the air pocket (i.po.), caused a significantly increased infiltration of total leukocytes (p<0.01) and eosinophils (p<0.01) at 24 hours after challenge compared to vehicle ( 0.75% carboxymethylcellulose (CMC) in saline i.po.) treated mice.

Locally administered BAK502G9 (200 mg, 20 mg or 2 mg into the pocket) significantly and dose-dependently inhibited the total leukocyte (p<0.01) and eosinophil (p<0.01) infiltration into the air pocket caused by 2 µg huIL-13 in 0.75% CMC.

Systemic administration of BAK209B11 (30 mg/kg, 10 mg/kg and 1 mg/kg) also significantly and dose-dependently inhibited the total leukocyte (p<0.01) and eosinophil (p<0.01) infiltration into the air pocket caused by 2 µg huIL-13 in 0.75% CMC.

EXAMPLE 17

Generation of human IL-13 knock-in/mouse IL-13 knock out transgenic mice for the purposes of evaluating the efficacy of anti-human IL-13 antibodies in models of lung allergic inflammation

The present inventors have generated mice that express human instead of mouse IL-13 by gene targeting. The mouse IL-13 gene has been replaced from start to stop codon with the relevant part of the human IL-13 gene. This mouse strain expresses human IL-13, rather than mouse IL-13, in response to the same stimuli as wild-type mice, as the endogenous IL-13 promoter and IL-13 pA tail remained unchanged. It has been shown that human IL-13 can bind to signal through mouse IL-13 receptors to generate the same physiological consequences as signaling caused by mouse IL-13 ligated mouse IL-13 receptors. For example, exogenous human IL-13 caused inflammatory cell recruitment into the mouse air pocket (Figure 18). These transgenic mice allow us to evaluate non-mouse cross-reactive anti-human IL-13 antibodies in establishing mouse models of disease.

This mouse has been used in the acute allergic airway inflammation models (as described in Examples 18 and 19) and chronic allergic airway inflammation models (as described in Example 20) and allows the evaluation of anti-human IL-13 antibody pharmacology in allergic airway disease.

EXAMPLE 18

Efficacy of anti-human IL-13 antibodies in the huIL-13 transgenic Lloyd mouse model of acute lung inflammation

Mouse model for acute allergic lung inflammation

The efficacy of anti-human IL-13 neutralizing human IgG4 antibodies was examined in a mouse model of acute allergic lung inflammation using the transgenic mice prepared in Example 17. This model was performed essentially as described by McMillan et al. [85] and example 13. The model was characterized by the end point of increased BAL and lung tissue IL-13, cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

The protocol for this model was as described in Example 13 except that anti-human IL-13 antibodies were dosed instead of BAK209B11.

EXAMPLE 19

Efficacy of anti-human IL-13 antibodies in the huIL-13 transgenic Gerard mouse model of acute lung inflammation

Mouse model for acute allergic lung inflammation

The efficacy of anti-human IL-13 neutralizing human IgG4 antibodies was examined in another mouse model of acute allergic lung inflammation using the transgenic mouse generated in Example 17. This model was performed essentially as described by Humbles et al, [86] and in example 14. The model is characterized by its endpoint by increased BAL and lung tissue IL-13, cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

The protocol for this model was as described in example 14 except that anti-human IL-13 antibodies were dosed instead of BAK209B11.

EXAMPLE 20

Efficacy of anti-human IL-13 antibodies in the huIL-13 transgenic Lloyd chronic model of lung inflammation

The effect of anti-human IL-13 neutralizing human IgG4 antibodies was examined in a model of chronic allergic lung inflammation using the transgenic mice made in Example 17. This model was performed essentially as described by Temelovski et al. [87] and in example 15 and is characterized by the end point of cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness (AHR).

Model protocol

The protocol for this model was as described in Example 15 except that anti-human IL-13 antibodies were dosed instead of BAK209B11.

EXAMPLE 21

Pharmacokinetics and pharmacodynamics of anti-human IL-13 antibodies in Ascaris. suum-allergic cynomolgus monkeys.

Pharmacokinetics and pharmacodynamics of 502G9 were evaluated in 4 allergic but unchallenged cynomolgus primates (2 male/2 female) following a single 10 mg/kg i.v. bolus dose. The experiment ran for 29 days. The antibody's pharmacokinetic parameters were determined from a geo-averaged mean

serum drug concentration curve and is given in detail below in table 4.

In the same study, serum IgE concentrations were also monitored using a human IgE ELISA kit (Bethyl laboratories, USA).

Results

Serum IgE concentrations were significantly reduced after a single 10 mg/kg i.v. bolus dose of BAK502G9, from 100% control levels (predose) to 66 ± 10% of control values (p<0.05), at 4 and 5 days post-dose. This decrease in serum IgE concentration went up again to 88 ± 8% of control levels by day 22 (see Figure 20). Again, these data were derived by normalizing each animal's serum IgE concentration to predose levels, where predose concentrations were 100%, and then making average curves from the 4 animals tested.

The two male monkeys had relatively low predose total serum IgE (60 ng/mL and 67 ng/mL). These IgE levels do not change in a manner that indicated a trend after treatment with BAK502G9 (Figure 20B). The two female monkeys had relatively high predose total serum IgE (1209 ng/mL and 449 ng/mL). These IgE levels were reduced after treatment with BAK502G9, maximally by 60% at 7 days, and returned to approximately predose levels at 28 days post administration (Figure 20B). These data provide indication that BAK502G9 lowers serum IgE concentrations in animals with relatively high circulating IgE of IgE.

EXAMPLE 22

Effectiveness of anti-human IL-13 antibodies in cynomolgus models for skin allergy

The effects of anti-human IL-13 neutralizing human IgG4 antibodies were investigated in a primate model of acute allergic skin inflammation. This model was performed by injecting human IL-13 and A. suum antigen intradermally into cynomolgus monkeys. 24-96 hours later, dermal biopsies and serum samples were taken. The model was characterized by its endpoint by cellular infiltration into the skin.

EXAMPLE 23

Effectiveness of anti-human IL-13 antibodies in cynomolgus models for lung allergy

The effect of anti-human IL-13 neutralizing human IgG4 antibodies was investigated in a primate model of acute allergic lung inflammation. This model was performed by exposing asuum allergic cynomolgus primates to nebulized asuum antigen, thereby generating an allergic reaction. This allergy was characterized by the end point of cellular infiltration into the lung and BAL, increased serum IgE levels and airway hyperresponsiveness.

Pharmacodynamics were additionally evaluated ex vivo using a flow cytometric method. CD23 is the high-affinity IgE receptor and can be expressed on peripheral human blood mononuclear cells. CD23 expression can be induced, with respect to the number of cells expressing CD23 and also in how much CD23 each cell expresses by both IL-13 and IL-4. IL-13, but not IL-4 driven response can be inhibited by anti-human IL-13 antibodies.

The animals were previously selected to participate in this 2-phase study on the basis of previously established AHR after nebulization antigen (ascaris suum extract) challenge. In phase I, airway function was examined during intravenous histamine challenge on days 1 and 11. PC30, the dose of histamine required to generate a 30% increase in pulmonary resistance (Rl) above baseline, was determined from each histamine dose-response curve. On days 9 and 10, animals were challenged with individually tailored doses of nebulized antigen previously shown to generate a 40% increase in Rl as well as a 35% decrease in dynamic compliance (Cdyn). Historically, in this model, a greater Rlblitt was observed after the second challenge with a given allergen dose than the first; this is antigen priming. The two antigen challenges caused AHR, as measured by an increased area under the histomine dose-response curve and/or a fall in PC30, and BAL, as well as eosinophilia at day 11 compared to day 1. Animals showing an AHR phenotype were selected to be with in phase H

Phase II was run exactly like Phase I except that all animals received a 30 mg/kg BAK502G9 infusion on days 1, 5, and 9. The effects of BAK502G9 were examined by comparing the changes seen in Phase II to those seen in Phase I for individual animals.

Blood, serum, BAL and lung tissue were collected. Serum IgE levels were measured by ELISA. Serum from BAK502G9 treated cynomolgus monkeys was shown to inhibit the expression of CD23 on human peripheral blood mononuclear cells induced by IL-13, but not IL-4. The magnitude of this inhibition was consistent with the serum BAK502G9 levels predicted by the PK ELISA.

Results

BAK502G9 significantly inhibited AHR as measured by RLAUC (p<0.05) (Figure 26A, Table 7). An inhibitory effect of BAK502G9 on AHR, as measured by PC30, was observed but did not reach statistical significance (Figure 26B; Table 7). BAK502G9 also significantly inhibited both antigen priming (p<0.01) (Figure 26C; Table 7) and BAL inflammation. BAK502G9 significantly inhibited total cell (p<0.05) and eosinophil (p<0.05) but not macrophage, lymphocyte or mast cell influx into BAL (Figure 26D; Table 7).

EXAMPLE 24

Efficacy of anti-human IL-13 antibodies against the asthmatic phenotype developed when human IL-13 was administered to the mouse lung

Mouse model for airway hyperresponsiveness

The efficacy of the anti-human IL-13 neutralizing antibody BAK502G9 against the development of airway hyperresponsiveness (AHR) following administration of human IL-13 to the mouse lung was investigated. This model was performed essentially as described by Yang et al [119] with the exception that human IL-13 was used instead of mouse IL-13.

Model protocol

To develop the phenotype, male Balb/C mice were exposed to two doses of human IL-13 separated by a 48-hour interval. Briefly, mice were anesthetized with an intravenous injection of 1100 ul saffan solution (1:4 diluted in water). Mice were intubated with a 22-gauge catheter needle, through which human recombinant IL-13 (25 µg dissolved in 20 µl phosphate-buffered saline (PBS)) or vehicle control (PBS) was instilled. Airway function was examined 24 hours after the last administration of IL-13 during increased methacholine challenges and measured using conscious plethysmography (Buxco). PC200 (concentration of methacholine required to increase baseline penH by 200%) was determined from 4 parameter non-fixed curve fitting of methacholine dose response curves. Antibody treatments were given by intraperitoneal injection 24 hours before each dose of IL-13.

Results

Intratracheal installation of human IL-13 into naïve wild-type mice resulted in the development of significant (p<0.05) airway hyperresponsiveness relative to control animals as determined by PC200 methacholine concentrations. Systemically administered BAK502G9 (lmg/kg) significantly inhibited (p<0.01) the development of AHR while the null control antibody had no effect (Figure 23).

EXAMPLE 25

Neutralizing potency of BAK502G9 as a human IgG4 against human IL-13 dependent IgE release from human B cells

B-cell switching assay protocol

IL-13 has been shown to induce IgE synthesis in human B cells in vitro [120]. Factor-dependent IgE release from human B cells was determined by ELISA. The neutralizing potency of BAK502G9 as a human IgG4 was investigated against human IL-13 dependent IgE release from human B cells.

Peripheral blood mononuclear cells (PBMC) were purified from human buffy coat (Blood Transfusion Service) by centrifugation through a 1.077 g/l density gradient. B cells were purified from PBMC with a B cell isolation kit II (Miltenyi Biotec), using reagents and methods described by the manufacturer. Assay medium included Iscove's modified Dulbecco's medium (Life Technologies) containing 10% fetal bovine serum and 20 µg/ml human transferrin (Serologicals Proteins Inc). After purification, B cells were resuspended to a final concentration of 10<6>/ml in analysis medium. 50 µl of resuspended cells were added to each assay point in a 96-well assay plate. 50 µl of 4 µg/ml of the anti-CD40 antibody EA5 (Biosource) was added to the assay wells as appropriate. Test solutions of antibodies (six replicates) were diluted to the desired concentration in assay medium. An irrelevant antibody that did not target IL-13 was used as a negative control. 50 µl/well of the appropriate test antibody was added to the cells. Recombinant bacterially derived human IL-13 (Peprotech) was then added to a final concentration of 30 ng/ml to give a total assay volume of 200 µl/well. The concentration of IL-13 used in the assay was chosen to give a maximal response. Assay plates were incubated for 14 days at 37°C under 5% CO2. IgE levels in the supernatant were determined by ELISA using reagents and protocols supplied by the manufacturer BD Biosciences, Bethyl Laboratories). Data were analyzed using Graphpad prism software.

Results

As shown in Figure 24, BAK502G9 (VH SEQ ID NO: 15; VL SEQ ID NO: 16) was able to inhibit human IL-13 dependent IgE production by human B cells. BAK502G9 as human IgG4 had an IC50 of 1.8 nM against 30 ng/ml human IL-13.

EXAMPLE 26

The effect of BAK502G9 against IL-13-directed potentiation of histamine-induced Ca<2+>signaling in primary human bronchial smooth muscle cells

IL-13 has been shown to directly modulate the contractility of airway smooth muscle [121, 122]. Intracellular calcium mobilization is a necessity for smooth muscle contraction. Recent studies have shown that IL-13's ability to alter smooth muscle contractility is controlled in part through modulation of contractile agonist-induced Ca2_|_ signaling [123, 124]. The effect of BAK502G9, an anti-human IL-13 antibody formatted as an IgG4, against IL-13 directed changes in shape change measurement. Samples were taken at high flow rate and access was terminated after 1000 eosinophilic events or 1 minute, whichever was earlier. Shape change was calculated as a percent FSC caused by shape change buffer alone (100% blank shape change). Data are expressed as the mean % blank shape change + SEM from 4 separate experiments. Each experiment used cells from an individual buffycoat (and thus individual donor), performed in duplicate for each point.

Results

NHLF cells simultaneously stimulated with 9.6 nM IL-13, 285.7 pM TNF-α and 160 pM TGF-P1 and cultured for 48 hours secreted 9.6 nM eotaxin-1 into the culture medium. In contrast, NHLF cells cultured with maintenance medium alone secreted 0.1 nM eotaxin-1 into the culture medium. This eotaxin-1 production was IL-13 dependent as IL-13/TNF-a/TGF-pi co-stimulated NHLF cells' production of eotaxin-1 was dose-dependently inhibited by BAK502G9 with an IC50 of 32.4 nM (Figure 29A).

The primary purpose of this part of the study was to investigate eosinophilic shape change. The magnitude of eosinophilic shape change in response to 3 nM eotaxin (positive control) was 122.2±2.1% n=4). Eotaxin-1 induced shape change was completely inhibited by 100 nM of an anti-eotaxin antibody CAT-213, mean shape change 101.0±1.0% (n=4).

Medium from NHLF cells co-stimulated with 9.6 nM IL-13, 285.7 pM TNF-oc and 160 pM TGF-pi and cultured for 48 hours (conditioned medium) induced a clear eosinophilic shape change (Figure 29B). In contrast, medium from NHLF cultured for 48 hours in NHLF maintenance medium alone did not induce eosinophilic shape change (Figure 29B).

Addition of anti-IL-13 antibody BAK502G9 to co-stimulated medium prior to NHLF cultivation resulted in a dose-dependent inhibition of eosinophil shape change, with a geometric mean IC50 of 16.8 nM when examined at 1:16 dilution (Figure 29B).

The abilities of stimulants (IL-13, TNF-oc and TGF-pl) that were not cultured with NHLF cells to induce eosinophil and neutrophil shape change were also investigated. 9.6 nM IL-13, 285.7 pM TNF-oc and 160 pM TGF-pl did not induce a clear eosinophilic shape change. This suggests that the eosinophilic shape change ability of conditioned medium that develops during NHLF cell culture with the stimulants is not due to either stimulant alone or in combination (Figure 29B).

EXAMPLE 29

Mapping of anti-IL-13 antibodies to human IL-13

The epitope mapping of a representative IL-13 antibody BAK502G9 was performed using a molecular approach and standard peptide removal.

Molecular approach

IL-13 chimeras were constructed, where parts of the human IL-13 sequence were replaced with mouse sequence. These chimeras were used in binding studies with representative IL-13 antibodies to help identify the specific epitope.

Two panels of IL-13 chimeras were produced. The first panel contained nine chimeras (Figure 30) and was used to locate the general position of the epitope. The second panel contained ten chimeras (Figure 31) and was used to fine map the epitope.

The chimeric IL-13 sequences were assembled using PCR and cloned into the Gateway<®> entry vector, which was then recombined with a destination vector pDEST8 (modified to encode a detection and affinity tag at the C-terminal end of the recombinant protein). These expression vectors were used to transform DHlOBac™ chemically competent E. coli that allowed site-specific transposition of tagged chimeric IL-13 into the baculovirus shuttle vector (bacmid). Recombinant bacmid DNA was isolated for each IL-13 chimera and transfected into SJ9 (Spodoptera frugiperda) insect cells using Cellfectin® reagent. Recombinant baculovirus was harvested 72 hours after transfection and passed through S/9 insect cells two more times.

Insect 2000-500 ml culture supernatant was purified on an affinity column and eluted material was concentrated from 16 to 1 ml and loaded onto a size exclusion Superdex 200 HR10/300GL column for final purification and buffer recovery.

A homogeneous competition assay using biotinylated human IL-13, streptavidin-anthophyocynate, and Europium-labeled BAK502G9 was developed. The analysis is as follows: Eu-BAK502G9 binds biotinylated-human IL-13, the complex was then recognized by the streptavidin APC conjugate and when a flash of light is used, the energy is transferred from the APC labeling to the Europium in close proximity, and time-resolved fluorescence can be measured. Competition for this binding is introduced using unlabeled human IL-13 (as a control) and the chimeric constructs. This competition is quantified to calculate the relative affinities of the IL-13 mutants for IL-13 antibodies and allows mutations that alter binding to be identified.

Results

Chimeric construct IL-13-Helix D (Table 5) was found to be the weakest competitor against biotinylated human IL-13 for binding BAK502G9, which indicated that helixD within the IL-13 molecule was involved in BAK502G9 epitope binding (Table 5). Reduced activity was also seen with the 4041 and 3334 mutants where residues 40, 41 and 33, 34 of the parent sequence were altered, respectively, indicating the potential involvement of helixA in the recognition of BAK502G9. The reduced activities of loop3 were disregarded as this loop had a reduced number of amino acids in the mutant compared to the human molecule and probably changes the overall structure of the protein. Other reductions in the ability of the chimeric IL-13 molecules to compete with BAK502G9 binding were not considered significant for

such amino acid changes.

A more targeted set of mutations within helix D (Figure 26) was then tested. Results obtained are shown in Table 6 and are as follows: Results show that chimeric constructs 116117TK 8where lys in vd position 116 was replaced with threonine and the aspartate at position 117 was replaced with lysine), 123KA (where lysine at position 123 was replaced) and 127RA (where the arginine at position 127 was substituted) was least able to compete for binding to BAK502G (123KA and 127RA did not compete at 1 µM). Other residues implicated in binding to BAK502G9 due to their reduced efficiency in the competition assay include helixD residues 124Q (where lysine has been replaced by glutamine) and 120121 SY (a leucine histidine pair has been changed to a serine tyrosine pair). Mutations of leucine at position 58L also reduce binding and analysis of the 3-dimensional structure revealed that this residue packs against helixD and can either be directly contacted by BAK502G9 or affect the alignment of helixD.

These experiments show that residues within helixD are critical for the binding of BAK509G9 to IL-13. Especially the lysine at position 123 and the arginine at position 127 are critical for this binding as mutation of both destroys binding to BAK502G9.

Epitope removal

The epitope mapping of BAK502G9 was also performed using the standard peptide removal procedure. Here, IgG is immobilized on solid phase and allows capture of the IL-13 ligand. The complex formed is then subjected to specific proteolytic cleavage, during which accessible peptide bonds will be cleaved, however, those protected by the IgG:ligand interphase will remain intact. A peptide containing the epitope thus remains bound to IgG. This can then be emitted, collected and identified by mass spectrometry (ms).

Two complementary techniques are used, the first making use of the Ciphergen ProteinChip Reader MALDI-TOF mass spectrometer, where it was possible to covalently link IgG to a mass spectrometer chip and then perform the cutting and extraction in situ. The second technique used biotinylated BAK502G9 coupled to streptavidin-coated beads and allowed the collection of sufficient peptide for sequence confirmation by tandem mass spectrometry (ms/ms).

Although the two procedures differ in absolute detail and scale, they essentially involved the same steps, coupling of IgG, blocking of unreacted binding sites, washing, ligand capture, removal of unbound ligand, cutting and a final washing step.

The MALDI-TOF ms approach made use of proprietary ms chips that were activated with caronyldiimdazole that covalently binds to free primary amine groups to which IgG at 1-2 mg/ml in PBS was coupled overnight at 4°C. The chip was then blocked with an ethanolamine solution at room temperature for 1 h and then washed thoroughly with PBS or HBS plus an appropriate detergent. A one picomole aliquot of IL-13 was then applied to the chip in either PBS or HBS and allowed to bind to the chemically immobilized IgG for 2 hours at room temperature. This was followed by further washes in PBS or HBS with or without detergent to remove any non-specifically bound IL-13. A solution of trypsin ranging from 200 to 3.1 µg/ml in PBS or HBS was then used for IgG: the ligand complex and cutting was allowed to proceed for 30 min at room temperature, after which the chip was washed in PBS or HBS plus detergent, PBS or HBS and finally water. After applying a suitable MALDI-TOF ms matrix, the chip was then placed directly into the mass spectrometer and analyzed.

The bead-based approach started with the biotinylation of IgG, using an NHS biotin compound, at a molar ratio of 1 IgG to 4 biotin molecules. Removal of unbound biotin and the byproducts from the reaction using gel filtration followed. The biotinylated IgG was then allowed to bind to neutravidin-coated agarose beads, where an attempt was made to maximize the IgG capture. Aliquots of IgG-coated beads were then distributed in a concentrator spin column and washed with Dulbecco's PBS + 0.05% Tween 20 followed by resuspension in Dulbecco's PBS + 0.05% Tween 20. A pulse of IL-13 was then applied to the resuspended the IgG beads and binding was allowed to occur for 10 min after which the liquid phase was removed by centrifugation and the beads were washed with Dulbecco's PBS + 0.05% Tween 20 followed by resuspension in Dulbecco's PBS + 0.05% Tween 20.

The bullet: IgG: ligand complex was then subjected to proteolysis with either trypsin or chymotrypsin and incubated at room temperature or 37°C. The beads were then again washed in Dulbecco's PBS + 0.05% Tween 20 followed by a further wash in Dulbecco's PBS without detergent. The beads were then resuspended in water, acetonitrile, trifluoroacetic acid mixture and the supernatant was recovered. This was then either analyzed by MALDI-TOF ms or by reverse phase HPLC mass spectrometry, including tandem (ms/ms) fragmentation using a ThermoQuest LCQ ESI ion-capture mass spectrometer. An attempt was then made to match the resulting fragmentation coin with the human IL-13 sequence and the separate heavy and light chain sequence of BAK502G9 IgG.

During the experimental sequence, a number of controls, primarily blank surfaces, IgG alone and isotype controls were used to demonstrate that the identified peptides were derived specifically from IgG captured IL-13 and not a product of BAK502G9 or non-specifically bound IL- 13 cutting.

Results

The experimental series consistently yielded single IL-13 specific peptides for each cut. Data from the LCQ ion capture instrument revealed that the tryptic fragment had a monoisotopic mass of 3258Da (MH+) and the chymotrypsin fragment a monoisotopic mass of 3937Da (MH+).

A search of these masses against the appropriate in silico cut of human IL-13 yielded close hits to related peptides in the C-terminal portion of the molecule.

Hit for trypsin peptide mass: 3258Da

At a tolerance of 1000 ppm, the 3258Da sequence matches from aspartic acid at position 106 to the C-terminal asparagine at position 132. There are no other hits with this tolerance. This region is highlighted on the sequence of the precursor form of human IL-13 below.

Hit for chymotrypsin peptide mass: 393 7Da

At a tolerance of 1000 ppm, 3937Da matches the sequence from serine at position 99 to the C-terminal asparagine at position 132. This region is highlighted on the sequence of the precursor form of human IL-13 below.

Both of these hits show that BAK502G9 IgG preserves the C-terminal part of the IL-13 molecule during proteolysis of the antibody:ligand complex.

The identity of both peptides was successfully confirmed by ms/ms, none showing any significant sequence parallels to BAK502G9. Ms/ms fragment mapping tailored to identify either Y or B ions hit 26 of 104 possible ions in a charged state for the trypsin peptide and 19 of 128 possible ions for the chymotrypsin peptide. An overview of all charged states shows identification of 23 of the 27 amino acid residues for the trypsin fragment and 29 of the 33 residues for the chymotrypsin fragment. This is sufficient to confirm identity.

The experimental sequence as a whole has identified the portion of the BAK502G9 epitope on human IL-13 to lie within the 27 C-terminal amino acid residues. These findings support the findings from the molecular approach described in detail above.

REFERENCES

1. McKenzie, A.N., et al. J Immunol, 1993. 150(12): pp. 5436-44.

2. Minty, A., et al. Nature, 1993. 362(6417): pp. 248-50.

3. Nakamura, Y., et al. Am J Reapir Cell Mol Biol, 1996. 15(5): p 680-7. 4. Robinson, D.S., et al. N Engl J Med, 1992. 326(5); pp. 298-304. 5. Walker, C, et al. Att J Respir Crit Care Med, 1994. 150(4): p 1038-48. 6. Humbert, M., et al. Am JR Respir Crit Care Med, 1996. 154(5)r p. 1497-504. 7. Corrigan, C.J. and A.B. Kay Int Arch Allergy Appl

Immunol, 1991. 94(1-4): pp. 270-1. 8. Bentleyr A ,M-., -et- al— Am J Respir Geil- Mol- B±oi; 1993". 8(1): p. 35-42. 9. Murata, T., et al. Int J Hematol, 1999. 69(1): p. 13-20. 10. Andrews ;' A.L., et al J Biol Chem, 2002. 277(48): p 46073-8 11. Miloux, B., et al FEBS Lett, 1997. 401(2-3): p 163-6. 12. Hilton, D.J., et al Proe Nati Acad Sei USA, 1996. 93(1): p 497-501 13. Kup

: pp. 939-48. 14. Nelms, K., et al. Annu Rev Immunol, 1999. 17: p. 701-38. 15. Zhang, J.G., et al. J Biol Chem, 1997. 272(14): p 9474-80. 16. Caput, D., et al. JBiolChem, 1996. 271(28): p. 16921-6.

17. Kawakami, K., et al. Blood, 2001. 97(9); p. 2673-9.

18. Wood, N., et al. J Exp Med, 2003. 197(6): p 703-709. 19. Chiaramonte, M.G., et al. J Exp Med, 2003. 197(6): p 687-701. 20. Beasley, R., et al. J Allergy Clin Immunol, 2000. 105(2

Pt 2): p. S466-72.

21. Peat, J.K. and J. Li J Allergy Clin Immunol, 1999. 103(1

Pt 1): p. 1-10. 22. Society, B.T., British guideline on the management of asthma. Thorax, 2003. 58 Soppi 1: p. il-94. 23. GINA, Global Strategy for Asthma Management and

Prevention. 2002, National Institute of Health.

24. Milgrom, H., B. Bender, and F. Wamboldt. Ann Allergy

A3thma Immunol, 2002. 88(5): p 429-31.

25. Fish, L, and CL. Lung, Adherence to asthma therapy. Ann

Allergy Asthma Immunol, 2001. 86(6 Suppl 1): p. 24-30. 26. Bender, B.G. J Allergy Clin Immunol, 2002. 109(6 Suppl): p, S554-9. 27. Wills-Karp, M., et al. Science, 1998. 282(5397): pp. 2258-61. 28. Grunig, 6,, et al. Science, 1998. 282(5397): pp. 2261-3. 29. Venkayya, R. et al. Year J Respir CéH Mol 3x61, 207)2. 26(2): pp. 202-8. .30, Morse, B., et al. Am J Physiol Lung Cell Mol Physiol, 2002. 282(1): p L44-9. 31. Zhu, 2., et al. J Clin Invest, 1999. 103(6): p 779-88. 32. Walter, D.M., et al. 0 Immunol, 2001. 167(8): p. 4668-75. 33. Cohn, L., J.S. Tepper, and K. Bottomly. J Immunol, 1998. 161(8): p 3813-6. 34. Taube, C, et al. JImmunol, 2002. 169(11): p. 6482-9. 35. Yang, E.S., et al. J. Allergy Immunol., 2002. 109; p.

A168. 36. Blease, K., et al. J Immunol, 2001. 166(8}: p. 5219-24. 37. Heinzmann, A., et al. Hum Mol Genet, 2000. 9(4}: p. 549-59. 38. Howard, T.D., et al. al. Am JHum Genet, 2002. 70(1): p. 230-6.

39. Kauppi, F., et al. Genomics, 2001. 77(1-2): p. 35-42.

40. Graves, P.E., et al. J Allergy Clin Immunol, 2000. 105(3): pp. 506-13. 41. Arima, K., et al. J Allergy Clin Immunol, 2002. 109(6}: p. 980-7. 42. van der Pouw Kraan, T.C., et al. GenesImmun, 1999. 1(1): p. 61-5. 43. Humbért, M ., et al. J Allergy Clin Immunol, 1997. 99(5): p 657-65. 44. Kotsimbos, T. C., P. Ernst, and Q. A. Hamid, Proe Assoc Am

Physicians, 1996. 108(5): pp. 368-73. 45. Komai-Koma, M., F.Y. Liew, and P.C. Wilkinson, J Immunol, 1995, 155(3): pp. 1110-6. 46. Haseer, T., et al. Am 0 Respir Crit Care Med,.1997. 155(3): pp. 845-51. 47. Huang, S.K., et al. J Immunol, 1995. 155(5): pp. 2688-94. 48. Kroegel, C, et al. Eur Respir J, 1996. 9(5): p 899-904. 49. Ohshima, Y., et al. Pedlatr Res, 2002. 51(2): p. 195-200. 50. Hasegawa, M., et al. J Rheumatol, 1997. 24(2): p. 328-32. 51. Hancock, A,, et al. Am J Respir Cell Mol Biol, 1998. 18(1): p 60-5. 52. Lee, C.G., et al. JExp Med, 2001. 194(6): pp. 809-21. 53. Jain-Vora, S., et al. Am J Respir Cell Mol Biol, 1997. 17(5): p 541-51. 54. Fallon, P.G., et al: J Immunol, 2000. 164(5): pp. 2585-91. 55. Chiaramonte, M.G., et al. JClin Invest, 1999. 104(6): p. 777-85. 56. Chiaramonte, M.G., et al. Hepatology, 2001. 34(2): p. 273-82. 57. Sluiter, H.J., et al. Eur Respir J, 1991. 4(4): p 479-89. 59. Zheng, T., et al. J Clin Invest, 2000. 106(9); pp. 1081-93. 59- Tashkin, D.P., et al., Methacholine reactivity predicts changes in lung function over time in smokers with early chronic obstructive pulmonary disease. The Lung Health Study Research Group. Am J Respir Crit Care Med, 1996. 153(6 Pt 1): p 1802-11. 60- Van Der Pouw Kraan, T.C., et al. Genes Immun, 2002. 3(7): p. 436-9. 61. Cinnider, B.F., et al. Blood, 2001. 97(1): p. 250-5. 62. Kapp, U., et al. J Exp Med, 1999. 189(12): p 1939-46. 63. Fiumara, P., F. Cabanillas, and A. Younee, Blood, 2001. 98(9): p. 2877-8. 64. Terabe, M., et al. Nat Immunol, 2000. 1(6): pp. 515-20. 65- Ahlers, J.D., et al. Proe Nati Aoad Sei USA, 2002. 99(20} : p. 13020-5. 66. Hutchings, C, Generation of Naive Human Antibody Libraries, in Antibody Engineering, R. Kontermann and S. Dubel, Editors. 2001, Springer Laboratory Manuals,

Berlin, pp. 93-108.

67. Vaughan, T.J., et al. Nat Biotechnol, 1996. 14(3): pp. 309-14.

68. Kitamura, T., et al. Blood, 1989. 73(2): pp. 375-80.

69. Lefort, S., et al. FEBS Lett, 1995. 366(2-3): p. 122-6. 70. Osbourn, J.K., et al. Iramunotechnology, 1996. 2(3): p. 181-96. 71. Howard, T.D., et al. Am J Respir Cell Mol Biol, 2001. 25(3): p. 377-84. 72. Karlsson, R,, A. Michaelsson, and L. Mattsson, J Immunol Methods, 1991. 145(1-2): p 229-40. 73. Tomlinson, VBASE. 1997, MRC Center for Protein

Engineering, Cambridge, OK.

74. Altmann, F., et al. Glycoconj J, 1999. 16(2): p 109-23, 75. Drexler, H.G., et al. Leuk Res, 1986. 10(5): p. 487-500. 76. Skinnider, B.F., 0. Kapp, and T.W. Mak, Leuk Lymphoma, 2002. 43(6); p. 1203-10. 77. Terada, N., et al. Clin Exp Allergy, 2000. 30(3): pp. 348-55. 78. Wenzel, S.E., et al. J Immunol, 2002. 169(8): p 4613-9. 79. Richter, A., et al. Am J Respir Cell Mol Biol, 2001. 25(3): p. 385-91. 80. Bochner, B.S., et al. J Immunol, 1995. 154(2): p 799-803. 81. Kotowicz, K., et al. Int Immunol, 1996. 8(12): p. 1915-25. 82. McKenzie, A.N., et al. Journal of Immunology, 1993. 150(12) : p. 5436-44. 83. Bouteiller, C.L., et al. J Immunol Methods, 1995. 181(1): p 29-36. 84. Riffo-Vasquez, Y., et al. ClinExp Allergy, 2000. 30(5): p. 728-38. 85. McMillan, S.J., et al. J Exp Med, 2002. 195(1): p, 51-7. 86. Humbles, A.A., et al. Pro Hat! Acad Sei USA, 2002. 99(3): p. 1479-84. 87. Temelkovski, J., et al. Thorax, 1998. 53(10): pp. 849-56. 88. Belvisi, M.G., et al., JPulm Pharraacol Ther, 2001. 14(3): p, 221-7. 89. Barnes, P.J., et al. Eur Respir J, 1996. 9(4): p 636-42.

90. Barnes, P.J., Pharmacol Ther, 2003. 97(1): pp. 87-94.

91. Wardlaw, A.J., Clin Med, 2001. 1(3): pp. 214-8.

92. Edwards, J.C., et al. J Pathol, 1981. 134(2): p 147-56. 93. McDonough, J.E., et al. W.M. BUI ot, and J.C. Hog. TGF~beta Isofons, and IL- 13 Immvnostainxng on Lvng Tiasue from Patlents with COPD. in ATS 99th International Conference. 2003. Seattle. 94. Wold et al. Multivariate data analysis in cbemistry. Chemometrics - Mathematics and Statistics in Chemistry (Ed.: B. Kowalski), D. Reidel Publishing Company, Dordrecht, Holland, 1984 (ISBN 90-277-1846-6). 95- Norman et al, Applied Regression Analysis. Wiley-Interscience; 3rd edition (April 1998) ISBN: 0471170828 96. Abraham Kandel, Eric Backer. Computer-Assisted Reasoning in Cluster Analysis.Prentice Hall PTR?(May 11, 1995), ISBN: 0133418847 97. Wojtek Krzanowski. Principles of Multivariate Analysis: A Dser's Perspective (Oxford Statistical Science Series, No 22 (Paper)). Oxford University Press; (December 2000), ISBN: 0198507089 98 . lan H. KTtténVETBé" Frank. Dåta MlhXng: Practical Måchine Learning Toola and Techniques with Java Implementations. Morgan Kaufmann; (October 11, 1999), ISBN: 1558605525

99- David G, T. Denison (Editor), Christopher C. Holmes, Bani K. Mallick, Adrian F, M. Smith. Bayesian Methods for Nonlinear Classification and Regression (Wiley Series in Probability and Statistics), John Wiley&Sons; (July 2002), ISBN: 0471490369

100. Arup K. Ghose, Veilarkad K. Viswanadhan. Combinatorial Library Design and Evaluation Principles, Software, Tools, and Applications in Drug Discovery. ISBN; 0-8247-04B7-8

101. Chothia C. et al. Journal Molecular Biology (1992) 227,

799-817.

102. Al-Lazikani, et al. Journal Molecular Biology (1997)

273(4), 927-948.

103. Chothia, et al. Science, 233,755-758 (1986).

104. Whitelegg, N.R.J, and Rees, A.R (2000). Prot. Eng., 13,

819-824.

105. Available from Accelerys Inc.

106. Guex, N. and Peitsch, M. C, (1997). Slectrophoresis

(1997) 18, 2714-2723.

107. Kabat E A et al (1991): Sequences of Proteins of Immunological Interest, 5<1*>Edition. US Department of Health and Human Services, Public Service, NIH, Washington.

108. Kontermann R and Dubel Stefan; (2001) Antibody

Engineering, Springer Laboratory Manuals.

109. Mendez et al (1997); Nature Genetics Vol. 2: 146-156.

110. Csonka E et al (2000) Journal of Cell Science, 113: 3207-3216.

Ul. Vanderbyl S et al (2002) Molecular Therapy, 5(5): 10.

112. Marasco WA (1997) Gene Therapy, 4(1): 11.

in

113. Hanes J et al (2000). Methods in Enzymology, Vol 328:24.

114. Li et al (2003). Abstract for poster [605] submitted at The American Thoracic Society Annual Meeting, 2003, Seattle.

115. Koide et al (1998). Journal of Molecular Biology, Vol

284:1141-1151.

116. Hygren et al (1997). Current Opinion in Structural

Biology, Vol 7:4 63-469.

117. Heller, F., et al. (2002) Immunity, 17(5):629-38.

118. Inoue, S-, et al. (1999) Am J Gastroenterol, 94(9}:2441-6.

119. Yang, M., et al. Am J Respir Cell Mol Biol. 2001. 25(4): p. 522-30

120. Punnonen J., et al 1993. Proe Nati Acad Sei. 90(8):3730-4.

121. Grunstein, M., et al. Am J Physiol Lung Cell Mol Physiol 2002. 282: p L520-L528.

122. Laporte, J., et al. Am J Respir Crit Care Med 2001. 164: p. 141-148,

123. TllJba 0., et al. Br J Pharmacol 2Q03. 140(7): pp. 1159-62.

124. Deshpande, D., et al. Am J Respir Cell Mol Biol 2004.

31(1): pp. 36-42; Bpub Feb 5 as doi:10.1165/rcmb.2003-0313OC.

125. Humbert et al. 1997. J. Allergy Clin. Immunol., 99:657.

126. Berry, M.A., Parker, D., Neale, N., Woodman, L., Morgan, A. Monk, P.D.. Submitted to J. Allergy Clin Immunol.

127. Obase et al. Ann Allergy Asthma Immunol. 2001; 86(3):304-10.

128. Chu et al. 2000; J. Allergy Clin. Immunol. 106:1115

129. Terada et al. 2000. Clin. Exp. Allergy., 30: 348-55.

130. Wenzel et al. 2000. J. Immunol. 169: 4613-19.

Vdinf = volume of distribution during time 0 - infinity, calculated from it

extrapolated AUC

Clinf= clearance during time 0 to infinity, calculated from the extrapolated AUC. AUCinf = the area under the curve (measured as total drug exposure) during to 0 to infinity, including an extrapolated term based on the elimination rate constant (k) and the last observed

the serum drug concentration.

AUCgxt= percentage of the total AUC extrapolated.

Xo.5= Drug half-life in the terminal elimination phase.

Second set of chimeric constructs 21 animals showing AHR (PC30) in phase I and one additional animal with an antigen-priming phenotype were drawn for testing in phase II (22 total). Each animal did not have AHR measured at both AUC and PC30. Only animals that showed AHR in phase I and whose AHR was examined in both phase I and phase II were included in the AHR results. Statistical testing was performed using InStat. Testing was a 2-way Student's t-test against the null hypothesis that the endpoint did not include the number 0 (ie, there was no change in phase II compared to phase I);

<*>p<0.05,<**>p<0.01. Data are shown as arithmetic mean + SEM (n=14-21).

<a>5 animals were excluded from the AUC analysis as they did not show AHR (increased AUC) in phase I. A further 3 animals were excluded due to a technical error in phase II airway function data collection.

<b>3 animals were excluded from the PC3o analysis due to a technical error in phase II airway function data collection (same animals as in a). The additional animal with antigen priming phenotype was excluded as it did not show PC30AHR in phase I.

<c>2 animals were excluded from the antigen priming assay because there was a technical error in phase I airway function data collection.

dl animals were excluded from the BAL analysis due to marked BAL inflammation at the start of the study.

Claims (37)

1. Isolated specific binding member for human IL-13, characterized in that it comprises an antibody-antigen binding site which is composed of a human antibody VH domain and a human antibody VL domain and which comprises a set of CDRs, HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, wherein the VH domain comprises HCDR1, HCDR2 and HCDR3 and the VL domain comprises LCDR1, LCDR2 and LCDR3, wherein the set of CDRs consists of a set of CDRs selected from the group consisting of: BAK278D6 the set of CDRs, defined in which HCDR1 has the amino acid sequence of SEQ ID NO: 1, HCDR2 has the amino acid sequence of SEQ ID NO: 2, HCDR3 has the amino acid sequence of SEQ ID NO: 3, LCDR1 has the amino acid sequence of SEQ ID NO: 4, LCDR2 has the amino acid sequence of SEQ ID NO: 5, and LCDR3 has the amino acid sequence according to SEQ ID NO: 6, BAK502G9 the set of CDRs, defined in which HCDR1 has the amino acid sequence according to SEQ ID NO: 7, HCDR2 has the amino acid sequence according to SEQ ID NO: 8, HCDR3 has the amino acid sequence according to SEQ ID NO: 9, LCDR1 h ar the amino acid sequence according to SEQ ID NO: 10, LCDR2 has the amino acid sequence according to SEQ ID NO: 11, LCDR3 has the amino acid sequence according to SEQ ID NO: 12, where the specific binding member binds human IL-13 and rhesus or cynomolgus IL-13. 2. Isolated specific binding member according to claim 1, characterized in that HCDR1, HCDR2 and HCDR3 in the VH domain are within a germline framework and/or LCDR1, LCDR2 and LCDR3 in the VL domain are within a germline framework. 3. Isolated specific binding member according to claim 2, characterized in that HCDR1, HCDR2 and HCDR3 in the VH domain are within the germline framework VH1 DPI 4. 4. Isolated specific binding member according to claim 2 or claim 3, characterized in that LCDR1, LCDR2 and CLDR3 in the VL domain are within the germline framework VL VX3 3H. 5. Isolated specific binding member according to any one of claims 1 to 4, characterized in that it binds a human IL-13 variant where arginine at position 130 is replaced by glutamine. 6. Specific binding member according to any one of claims 3 to 5, characterized in that it comprises the BAK502G9 VH domain (SQ ID NO: 15). 7. Specific binding member according to any one of claims 3 to 6, characterized in that it comprises the BAK502G9 VL domain (SEQ ID NO: 16). 8. Specific binding member according to any one of claims 1 to 7, characterized in that it neutralizes human IL-13. 9. Specific binding member according to any one of claims 1 to 8, characterized in that it comprises an scFv antibody molecule, or an antibody constant region or a complete antibody, optionally where the complete antibody is IgG4. 10. Isolated antibody VH domain, characterized in that it is a specific binding member according to any one of claims 1 to 9. 11. Isolated antibody VL domain, characterized in that it is a specific binding member according to any one of claims 1 to 9. 12. Composition, characterized in that it comprises a specific binding member, antibody VH domain or antibody VL domain according to any one of claims 1 to 11 and at least one additional component. 13. Composition according to claim 12, characterized in that it comprises a pharmaceutically acceptable excipient, vehicle or carrier. 14. Isolated nucleic acid, characterized in that it comprises a nucleotide sequence that codes for a specific binding member or antibody VH or VL domain of a specific binding member according to any of claims 1 to 9. 15. Host cell, characterized in that it has been transformed with the nucleic acid according to claim 14 in vitro. 16. Method for producing a specific binding member or antibody VH or VL domain, characterized in that the method comprises growing host cells according to claim 15 under conditions for the production of said specific binding member or antibody VH or VL domain. 17. Method according to claim 16, characterized in that it further comprises isolating and/or purifying said specific binding member or antibody VH or VL variable domain. 18. Method according to claim 16 or claim 17, characterized in that it further comprises formulating the specific binding member or antibody VH or VL variable domain into a composition that includes at least one additional component. 19. Method for producing an antibody-antigen binding domain specific for human IL-13, characterized in that the method comprises providing by means of addition, deletion, substitution or insertion of one or more amino acids in the amino acid sequence of a parent VH domain comprising HCDR1, HCDR2 and HCDR3, where the parent VH domain HCDR1, HCDR2 and HCDR3 is the BAK278D6 set of HCDRs, defined in which HCDR1 has the amino acid sequence according to SEQ ID NO: 1, HCDR2 has the amino acid sequence according to SEQ ID NO: 2, HCDR3 has the amino acid sequence according to SEQ ID NO: 3, or BAK502G9 the set of HCDRs, defined in which HCDR1 has the amino acid sequence according to SEQ ID NO: 7, HCDR2 has the amino acid sequence according to SEQ ID NO: 8, HCDR3 has the amino acid sequence according to SEQ ID NO: 9, a VH domain which is an amino acid sequence variant of the parent VH domain, and optionally combining the VH domain thus provided with one or more VL domains to provide one or more VH/VL combinations; and testing said VH domain which is an amino acid sequence variant of the parent VH domain or the VH/VL combination or combinations to identify an antibody-antigen binding domain specific for human IL-13. 20. Method according to claim 19, characterized in that the parent VH domain amino acid sequence is selected from the group consisting of SEQ ID NO: 13 and SEQ ID NO: 15. 21. Method according to claim 19 or claim 20, characterized in that said one or more VL domains are provided by means of addition, deletion, substitution or insertion of one or more amino acids in the amino acid sequence of a mother VL domain comprising LCDR1, LCDR2 and LCDR3, in which mother The VH domain of LCDR1, LCDR2 and LCDR3 is the BAK278D6 set of LCDRs, defined in which LCDR1 has the amino acid sequence of SEQ ID NO: 4, LCDR2 has the amino acid sequence of SEQ ID NO: 5, LCDR3 has the amino acid sequence of SEQ ID NO: 6, or BAK502G9 the set of LCDRs , defined in which LCDR1 has the amino acid sequence according to SEQ ID NO: 10, LCDR2 has the amino acid sequence according to SEQ ID NO: 11, LCDR3 has the amino acid sequence according to SEQ ID NO: 12, which produces one or more VL domains each of which is an amino acid sequence variant of the parent VL domain. 22. Method according to claim 21, characterized in that the parent VL domain amino acid sequence is selected from the group consisting of SEQ ID NO: 14 and SEQ ID NO: 16. 23. Method according to any one of claims 19 to 22, characterized in that said VH domain which is an amino acid sequence variant of the mode VH domain is provided by CDR mutagenesis. 24. Method according to any one of claims 19 to 23, characterized in that it further comprises providing the antibody-antigen binding site within an IgG, scFv or Fab antibody molecule. 25. Method for producing a specific binding member that binds human IL-13, characterized in that the method comprises: providing the starting nucleic acid encoding a VH domain or a starting repertoire of nucleic acids each encoding a VH domain, wherein the VH domain or VH domains either comprise a HCDR1, HCDR2 and/or HCDR3 to be replaced or lacking a HCDR1, HCDR2 and/or HCDR3 coding region; to combine said starting nucleic acid or starting repertoire with a donor nucleic acid or donor nucleic acids which encode or which are produced by mutation of the amino acid sequence of HCDR1 (SEQ ID NO: 1) or HCDR1 (SEQ ID NO: 7), HCDR2 (SEQ ID NO: 2) or HCDR2 (SEQ ID NO: 8) and/or HCDR3 (SEQ ID NO: 3) or HCDR3 (SEQ ID NO: 9) so that said donor nucleic acid or donor nucleic acids are inserted into the CDR1, CDR2 and/or CDR3 region of the starting nucleic acid or the starting repertoire, to provide a product repertoire of nucleic acids encoding VH domains: to express the nucleic acids of said product repertoire to produce product VH domains; optionally combining said product VH domains with one or more VL domains; and selecting a specific binding member that is specific for human IL-13, wherein the specific binding member comprises a product VH domain and optionally a VL domain; and recovering said specific binding member or nucleic acid encoding it. 26. Method according to claim 25, characterized in that the donor nucleic acids are produced by mutation of said HCDR1 and/or HCDR2. 27. Method according to claim 25, characterized in that the donor nucleic acid is produced by mutation of HCDR3. 28. Method according to claim 27, characterized in that it comprises providing the donor nucleic acid by mutation of nucleic acid which codes for the amino acid sequence according to HCDR3 (SEQ ID NO: 3) or HCDR3 (SEQ ID NO: 9). 29. Method according to claim 25, characterized in that it comprises providing the donor nucleic acid by random mutation of nulleic acid. 30. Method according to any one of claims 25 to 29, characterized in that it further comprises anchoring a product VH domain that is comprised within the recovered specific binding member to an antibody constant region. 31. Method according to any one of claims 25 to 29, characterized in that it comprises providing an IgG, scFv or Fab antibody molecule comprising the product VH domain and a VL domain. 32. Method according to any one of claims 19 to 31, characterized in that it further comprises testing the antibody-antigen binding domain or the specific binding member that binds human IL-13 for the ability to neutralize human IL-13 to identify an antibody-antigen binding domain or specific binding member that binds to and neutralizes human IL-13. 33. Method according to claim 32, characterized in that the antibody fragment is an scFv antibody molecule, or a Fab antibody molecule, possibly where the method further comprises providing the VH domain and/or the VL domain to the antibody fragment in a complete antibody. 34. A method according to any one of claims 19 to 33, characterized in that it further comprises formulating the specific binding member that binds to IL-13, the antibody-antigen binding site or an antibody VH or VL variable domain to the specific binding member or antibody-antigen binding site that binds IL-13, to a composition that includes at least one additional component. 35. Method according to any one of claims 19 to 34, characterized in that it further comprises binding a specific binding member that binds human IL-13 to IL-13 or a fragment of IL-13, where said binding takes place in vitro. 36. Method, characterized in that it comprises binding a specific binding member that binds IL-13 according to any one of claims 1 to 9 to human IL-13 or a fragment of human IL-13, where said binding takes place in vitro. 37. Method according to any one of claims 35 or 36, characterized in that it comprises determining the amount of binding of specific binding member to IL-13 or a fragment of IL-13.