Use of IGG1 immunoglobulins and/or ligands of the CD32 receptor for treating inflammatory diseases and incidents via the mucosa

Abstract

The present invention concerns the use of immunoglobulins of IgG.sub.1 type, and more generally of ligands of the CD32 receptor, for the treatment of inflammatory diseases and manifestations, in particular of allergies and auto-immune diseases, more particularly the treatment of allergic asthma, the immunoglobulins and ligands being administered via mucosal route, in particular via sublingual route.

Claims

1. A method of treatment of a type 1 hypersensitivity in a subject, comprising administering to a subject in need thereof a therapeutically effective amount of anti-IgE antibody having the capacity of binding the IgEs fixed to the receptors on the surface of a mastocyte or basophile without dissociating said IgEs from their receptors, the anti-IgE antibody being administered via mucosal route.

2. The method according to claim 1, the antibody having the capacity of bridging IgEs fixed to the receptors on the surface of a mastocyte or basophile.

3. The method according to claim 1, the antibody recognizing a part of the IgE not involved in the fixing of the IgE to its receptor on the mastocyte or basophile.

4. The method according claim 1, wherein in the sign of the type 1 hypersensitivity is systemic anaphylaxis, localized anaphylaxis, allergic rhinitis, asthma, atopical dermatitis, conjunctivitis, eczema, mastocytosis induced by anaphylactic shock, an allergic sign associated with parasitosis.

Description

FIGURES

(1) FIG. 1 shows the effect on bronchial hyper-reactivity (measured in PenH value or <<enhanced pause>>), of stimulation with metacholine on a group of three BALB/c mice non-sensitized to ovalbumin (group 1) and groups of five mice after sensitization to ovalbumin and desensitization with: group 2: PBS; group 3: ovalbumin at a dose of 500 μg via sublingual administration, twice a week for two months; group 4: the rat control isotype of IgG1κ type at a dose of 25 μg via sublingual administration, twice a week for two months; group 5: rat anti-mouse IgE antibody of clone R35-72 at a dose of 25 μg via sublingual administration twice a week for two months; group 6: rat anti-mouse IgE antibody of clone R35-72, at a dose of 10 μg via sublingual administration twice a week for two months.

(2) FIG. 2 shows the reactivity of the airways by measuring the Penh value in response to administration of metacholine (100 mg/ml). Eight mice were analyzed in each group. The horizontal bars represent the mean response in each group, each dot representing the Penh value obtained for a given animal. The results represent two independent experiments.

(3) FIG. 3 shows the count of macrophages and eosinophils in bronchial alveolar washings of mice given various treatments. The results are given as the mean±standard deviation of the mean N=8 mice per group. * p<0.05 compared with mice desensitized with PBS (placebo). The data were compared using the non-parametric test (Kruskal-Wallis).

EXAMPLE 1

(4) Material and Methods

(5) Sensitization and Desensitization of Mice

(6) BALB/c mice were sensitized with ovalbumin (OVA) as described in Razafindratsita et al. (2007, J. Allergy Clin. Immunol. 120: 278-285).

(7) Groups of 5 mice were then treated via sublingual route twice a week for two months with: 500 μg OVA per administration; 10 or 25 μg of rat anti-mouse IgE IgG1 κ (clone R35-72, BD Biosciences Pharmingen, San Diego, Calif., USA) per administration; 25 μg of non-specific rat IgG1κ (control isotype; eBioscience, San Diego, Calif., USA) per administration or PBS for the group of control mice.

(8) The mice were then subjected to allergenic challenge with aerosols of OVA (1% weight/volume), twice within two consecutive days.

(9) In parallel, a group of three healthy mice not subjected to ovalbumin was used as control group.

(10) The anti-IgE used, obtained from the R35-72 clone, was a rat antibody of IgG1κ type which is known for its ability to bind with IgEs on the surface of mastocytes and/or basophiles and of thereby inducing the release by these cells of pro-inflammatory mediators (Kubo et al. (2003) J. Immunol. 170: 775-780, Zhou et al. (2007) J. Exp. Med. 204: 2797-2802).

(11) The non-specific rat IgG1κ monoclonal antibody has no specificity for murine IgEs.

(12) Determination of Bronchial Hyper-Reactivity

(13) The measurement of bronchial hyper-reactivity was performed 24 hours after the last challenge, by whole body plethysmography (Buxco Europe Ltd, Winchester, UK) as described by Hamelmann et al. (1997, Am. J. Respir. Crit. Care Med. 156: 766-775) and bronchial resistance was estimated by measuring PenH (enhanced pause). The PenH index was obtained by determining the ratio between the PenH values measured after exposure to inhalation of USA metacholine and after exposure to atomized PBS.

(14) Results

(15) To determine the effect on allergenic response of an anti-IgE antibody having the capacity to bind and bridge murine IgEs on the surface of mastocytes and basophiles, mice were sensitized with ovalbumin and then desensitized with this anti-IgE antibody. These mice were compared with mice sensitized to ovalbumin and then desensitized with the same ovalbumin, or desensitized with a control isotype, or non-desensitized. All these mice were compared with mice non-sensitized with ovalbumin.

(16) Bronchial hyper-reactivity with metacholine was increased in mice sensitized to ovalbumin and treated with PBS (FIG. 1, group 2), as shown by the high value of the “PenH” parameter compared with the value obtained for mice non-sensitized with ovalbumin (FIG. 1, group 1).

(17) An improvement in bronchial hyper-reactivity was observed in mice sensitized then desensitized with ovalbumin (FIG. 1, group 3).

(18) A greater improvement in bronchial hyper-reactivity was observed in mice sensitized with ovalbumin then desensitized in increasing order of improvement, with: anti-IgE at a dose of 10 μg per administration (FIG. 1, group 6); the control isotype at a dose of 25 μg per administration (FIG. 1, group 4); anti-IgE at a dose of 25 μg per administration (FIG. 1, group 5).

(19) This example shows in most surprising manner that the administration of an anti-IgE antibody having the capacity to bind with IgEs on the surface of mastocytes and basophiles without dissociating said IgEs from their receptors, allows a very substantial reduction to be induced in the bronchial hyper-reactivity of mice sensitized to ovalbumin.

(20) This decrease is dependent upon the quantity of administered anti-IgE and is much more efficient than desensitization with ovalbumin and at molar concentrations at least 60 times less (bearing in mind that the molecular weights of ovalbumin and of an immunoglobulin are respectively 45 kDa (Nisbet et al. (1981) Eur. J. Biochem. 115: 335-345) and about 150 kDa.

(21) At least part of the effect is related to the specificity of the anti-IgE, as indicated by the difference in PenH value observed with 25 μg per administration between the anti-IgE and the control isotype, which is of the same order as the difference observed between the allergen (ovalbumin) and the placebo (PBS). The effect obtained with the control isotype may possibly be explained by the inducing of non-specific signals reducing the anaphylactic reaction at the time of fixation of this antibody onto the receptor with IgG inhibitor: RFcγIIb (Kang et al. (2007) Immune Network 7: 141-148).

EXAMPLE 2

(22) This study shows the effect of bridging or non-bridging anti-IgE antibodies and of IgG, type antibodies administered via sublingual route in immunotherapy using an in vivo murine model of allergic asthma.

(23) Material and Methods

(24) Mice, Reagents and Antibodies

(25) Female mice aged 6 to 8 weeks were obtained from Charles River (L'Arbresle, France). The phosphate buffer (PBS) was obtained from Invitrogen (Carlsbad, Calif.). Level V ovalbumin (OVA) with low endotoxin content was obtained from Sigma (St. Louis, Mo.) and was also purified on an endotoxin-removing gel (Pierce, Rockford, Ill.). The concentrations of residual endotoxin determined by the Endochrome K test (R1708K, Charles River, Wilmington, Mass.), were always less than 0.1 enzymatic unit (EU)/μg of protein. A polymerized form of corn maltodextrin (capsular polysaccharide or CPS) was used as system for administering a muco-adhesive particle antigen (Baudner et al. (2002) Infect. Immun. 70:4785-4790; Razafindratsita et al. (2007) J. Allergy Clin. Immunol. 120:278-285).

(26) The monoclonal antibodies (mAb) cited in Table 1 were used as purified antibodies for sublingual administration.

(27) TABLE-US-00001 TABLE 1 List of antibodies used. Antibody Antigen Species/Isotype Manufacturer LO-ME-2 IgE Rat IgG.sub.2a/κ Invitrogen, (bridging) ref: 04-7000 Control Rat IgG.sub.2a e-biosciences, isotype ref: 14-4321 R35-92 IgE Rat IgG.sub.1/κ BD Biosciences, (non-bridging) ref: 553416 R35-72 IgE Rat IgG.sub.1/κ BD Biosciences, (bridging) ref: 553413 Control Rat IgG.sub.1/κ e-biosciences, isotype ref: 14-4301
Purification of Antibodies

(28) To obtain all the antibodies in an identical buffer, the samples of anti-IgE were dialysed against 10 volumes of PBS using 30 kD membranes (Amicon Ultra-4, Millipore Corp.). The desalted samples were also filtered (Millex 0.22 μm, Millipore Corp.) to prevent any bacterial growth in the absence of sodium azide. Finally, the protein concentrations were determined using optical density of 280 nm (OD280) (Secoman XL, Uvikon) before and after filtration.

(29) Sublingual Immunotherapy in BALB/c Mice

(30) For sensitization, the mice were immunized via intraperitoneal route (i.p.) at days 0 and 14 with 10 μg OVA adsorbed on 2 mg Al(OH).sub.3, administered in 100 μl PBS. On day 21, a challenge test of 20 min using an aerosol was performed with 1% (weight/volume) OVA over 4 consecutive days using an aerosol administration system (Buxco Europe Ltd, Winchester, UK). To induce tolerance, the anti-IgE antibodies (10 μg and 25 μg per dose) were applied via sublingual route to groups of 8 mice, twice a week for 2 months. The control mice were treated via sublingual route with PBS or isotypic antibodies (IgG.sub.1/κ and IgG.sub.2a). As positive efficacy controls, some mice were treated via sublingual route with CPS-OVA (500 μg). Airway hyper-reactivity measurements (AHR) were performed by whole body plethysmography (Puxco Europe Ltd, Winchester, UK) and the results were expressed by lengthening the pause (“enhanced pause” or Penh). The Penh index, expressed as an increase relative to the basal resistance of the airways, was obtained by dividing the Penh value measured after exposure by inhalation to increasing doses of metacholine (0 to 100 mg), by the Penh value measured after inhalation of atomized PBS, as described in Razafindratsita et al. (2007).

(31) For the analysis of inflammatory cells in bronchial-alveolar washings (BAL), the mice were anaesthetized by intraperitoneal injection of pentobarbital (50 mg/kg body weight), and BALs were performed with 3×400 μl PBS. The BAL fluid was centrifuged at 800 g for 10 min at 4° C. The cell pellets were re-suspended in PBS and left to rotate on glass slides by cytocentrifugation, fixed and labelled with May/Grunwald Giema (RAL Reagents, Martillac, France). The eosinophils and macrophages were counted under optical microscopy using 200 times magnification.

(32) Antibody Response

(33) Samples of blood were collected at the retro-orbital sinus to evaluate the levels of antibodies specific to OVA using ELISA. The sera were collected after centrifugation at 10000 rpm for 10 min. For the detection of the IgG.sub.1 and IgG.sub.2a antibodies, purified OVA (0.2 μg) was plated overnight at 4° C. on ELISA plates (Nunc, Roskilde, Denmark). After the steps of washing and saturation, the mouse sera dilutions (1/100 to 1/12800 for IgG.sub.1 and 120 to 1/12560 for IgG.sub.2a) were incubated for 1 h at 37° C. The plates were washed and anti-mouse biotinylated rat IgG.sub.1 (dilution 1/100, BD Pharmingen, San Jose, Calif.) or IgG.sub.2a antibodies (dilution 1/200, BD Pharmingen) were added for 1 h at 37° C. Anti-mouse rat IgG antibodies conjugated with streptavidine-peroxidase (dilution 1/400, BD Pharmingen) were used for detection, using orthophenylenediamine (OPD) as substrate (Sigma Chemicals Aldrich). The reaction was stopped with 3N HCl and the optical densities were determined using an ELISA plate reader at 492 nm (Labsystems, Helsinki, Finland).

(34) For the detection of IgE antibody titres, anti-mouse IgE antibodies (1 μg/well, Bethyl Laboratories, Montgomery, Tex.) were plated on ELISA plates. After the steps of washing and saturation, dilutions of mouse sera (1/10 to 1/320) were incubated for 1 h at 37° C. Digoxygenin-OVA was incubated (at 1/10 dilution) for 1 h at 37° C. and Fab fragments of mouse anti-digoxygenin antibodies conjugated with horseradish peroxidase (HRP) (Roche) were used for detection at a dilution of 1/1000. A 2,2′-azino-bis(3-ethylbenzthiazoline-6 sulfonic) acid substrate (ABTS) was added (Roche). The optical densities were measured using an ELISA plate reader at 405 nm.

(35) The antibody titres were defined as the reverse of the last dilution at which the value of the optical density was twice higher the background noise.

(36) To determine the levels of antibodies specific to the allergen at the mucosal surfaces, samples of saliva were collected for determining the IgA level by ELISA. In brief, microplates were coated with goat anti-mouse IgA (0.1 μg/well, Bethyl Laboratories), washed and dilutions of the supernatant (1/20 to 1/320) were incubated for 1 h at 37° C., followed by digoxygenin-OVA (dilution 1/100). The plates were washed and rabbit anti-digoxygenin antibodies conjugated with HRP (Roche) were used for detection as described above.

(37) Statistical Analysis

(38) The data were compared using a non-parametric test (Kruskal-Wallis). The results were considered to have statistical significance at a p value of less than 0.05.

(39) Results

(40) The Anti-IgEs of Clones R35-72 and R35-92 and the Corresponding IgG.sub.1/κ Isotypes Improve Clinical Efficacy in Mice Specifically Sensitized with OVA and in Pulmonary Inflammation.

(41) As described above, the mice sensitized with OVA develop airway hyper-reactivity (AHR) associated with high Penh values detectable by whole body plethysmography, and signs of lung inflammation with cell infiltrations. Anti-IgE antibodies (clones LO-ME-2, R35-72 and R35-92) were tested as candidates for immunotherapy in this in vivo murine model of established asthma. Antibodies of the corresponding isotype and CSP-OVA were used as controls in these experiments.

(42) As shown in FIG. 2, and as expected, the healthy mice (i.e. non-sensitized) displayed low Penh values whilst the mice sensitized with OVA and treated via sublingual route with PBS (placebo) displayed strong AHR. The mice sensitized with OVA and treated via sublingual route with CSP-OVA were used as positive control and exhibited low Penh values. Sublingual treatment with the IgG.sub.2a/κ isotype control or with anti-IgE antibodies (isotype IgG.sub.2a/κ) of the LO-ME-2 clone (10 or 25 μg) did not have any impact on AHR. On the contrary, sublingual treatment with anti-IgEs (isotype IgG.sub.1) of the clones R35-72 and R35-92 induced a reduction in AHR in most of the animals compared with the mice treated with PBS. Surprisingly, the IgG.sub.1 isotype control induced a similar reduction in AHR suggesting that immune mechanisms non-specific of the antigen are involved in the induction of tolerance in this model.

(43) The reduction in AHR observed after treatment with the anti-IgEs corresponding to the clones R35-72 and R35-92 (10 or 25 μg) and the corresponding isotype control IgG.sub.1 was associated with a significant decrease (p<0.05) in the number of eosinophils in the BALs as observed with the positive control CSP-OVA (FIG. 3).

(44) The Therapeutic Sublingual Administration of Anti-IgE Antibodies does not Alter the IgG, IgE or IgA Responses.

(45) Sublingual immunotherapy (SLIT) with anti-IgE antibodies of the LO-ME-2 clone (isotype IgG.sub.2a/κ) increased the IgG.sub.1 and IgE specific to OVA only at the dose of 10 μg. In all the other experimental groups there was no detectable change in the IgE or IgG sera antibodies specific to OVA. The levels of OVA-specific salivary IgA antibodies were increased in mice treated with CSP-OVA compared with mice which had been given PBS. On the other hand, none of anti-IgE antibodies or control antibodies with the corresponding isotypes altered the level of IgA antibodies specific to OVA.

(46) This study therefore shows that anti-IgE antibodies of isotype IgG.sub.1/κ of the R35-72 and R35-92 clones allow an increase in tolerance induction. In addition, the control isotype antibodies IgG.sub.1/κ also promote the induction of tolerance. On the contrary, anti-IgE antibodies of isotype IgG.sub.2a/κ of the LO-ME-2 clone and the IgG.sub.2a/κ isotype control antibodies do not alter lung function in mice sensitized to OVA. Therefore, these data suggest that the tolerance-inducing effect of the anti-IgE antibodies of the R35-72 or R35-92 clones could involve the Fc region of the IgG.sub.1/κ isotypes. This raises the possibility that said anti-IgE antibodies could mediate their tolerance-inducing effect both by binding with the IgEs linked to FcεRI via their Fab regions, and by binding with the regulator receptor FcγRIIb, via their Fc region.

EXAMPLE 3

(47) This study shows the distribution of the Fc receptors of IgGs in the tongue tissues of mice, and the characterization of N-glycosylation of the Fc regions of the antibodies used in Example 2.

(48) Material and Methods

(49) Immuno-Histology

(50) For immunohistology, tissues taken from the spleen and tongue were sampled from naive mice frozen at −80° C. Tissue sections (4-6 μm in width) were cut in series, air-dried for at least 30 min, fixed in acetone for 1-2 min, and incubated for 10 min in 3% hydrogen peroxide (Sigma) to block endogenous peroxidase activity. After washing in Tris buffer (TBS: 0.05 M Tris, 0.15 M NaCl, pH 7.4), the primary antibodies i.e. anti-CD16/32 (clone 2.4G2, BD Biosciences) or anti-SIGN-R1 (clone ER-TRP9, Abcam) (dilution 1/100 in TBS) were added to the samples and incubated for 1 h at room temperature. The tissue sections were washed in TBS and incubated with biotinylated rabbit anti-goat secondary IgG antibodies (Sigma, 1/400) for 30 min before adding biotin-streptavidine horseradish peroxidase (SA-HRP, Sigma). After 30 min, the samples were washed and the specific labelling was visualized using diaminobenzidine (DAB, Sigma) as substrate. Tissue sections taken in the absence of primary antibody were included as negative controls.

(51) Analysis of N-Glycosylation of the Antibodies Using LC-ESIMS

(52) Anti-IgE rat IgG antibodies (R35-72 and R35-92 by BD Biosciences, eBRG1 and eBR2 by eBiosciences, LO-ME2 by Invitrogen) were denatured (6 M urea) and reduced (75 mM DTT, 50° C., 15 min) before analysis by LC-ESI MS. Analysis of the reduced heavy chains by ESI MS was performed after chromatographic separation of the light chains and heavy chains. The IgGs were also subjected to LC-ESIMS after deglycosylation (PNGase F treatment, Glycoprofile II, Sigma) or enzymatic removal of sialic acid (sialidase Au, QA Bio) following the manufacturer's instructions. In brief, about 2 μg of reduced antibody were injected into an Acquity C.sub.4, 10 cm×2.1 mm, 1.7 μm (BEH300, Waters) thermostated at 80° C. and connected to an RS-HPLC system (Dionex). A gradient of CH.sub.3CN (with 0.1% v/v formic acid) was set up with a flow rate of 400 μl/min to ensure proper UV detection at 210 nm. A Qq-TOF mass spectrometer (Maxis, Bruker) was connected to the RS-HPLC for correct mass measurement and was operated under positive ionization mode. Deconvolution of the mass spectrum was conducted using the MaxEnt algorithm (Waters Corp.) and the following parameters: 45000-55000 Da, auto-spacing of data points, resolution 40000. Therefore the masses of the intact desialylated and deglycosylated heavy chains were obtained so as to identify the glycosylation profiles.

(53) Results

(54) Identification of the Cells Carrying the Fc Receptors of IgGs in the Tongue Tissues of Naive and OVA-Sensitized BALB/c Mice

(55) In order to determine the tissue distribution of the Fc receptor of the IgGs, the inventors analyzed the tongue tissues of naive and OVA-sensitized BALB/c mice by immunohistology using specific antibodies, i.e. CD16/CD32 and SIGN-R1. CD16/CD32 was detected at the mucosal/sub-mucosal interface both at the ventral and dorsal tissue sites of the tongue and in the muscular region, both in the naive mice and the OVA-sensitized mice. On the other hand, SIGN-R1 was only detected in the muscular tissue (both in naive mice and OVA-sensitized mice), far from the sublingual administration site.

(56) Analysis of N-Glycosylation of the Antibodies by LC-ESIMS

(57) The IgGs were denatured and reduced before analysis by LC-ESIMS of the intact, desialylated and deglycosylated heavy chains, in order to identify their respective glycosylation profiles.

(58) The different oligosaccharidic structures were determined by LC-ESIMS, both before and after enzymatic digestion (i.e. deglycosylation and removal of sialic acid). The R35-92 antibody displayed structural heterogeneity of the glycoforms of heavy chains, with the successive addition of different monosaccharides to the N-linked oligosaccharide. The mass spectrum of the deglycosylated R35-92 antibody shows the weight (49075 Da) of the polypeptide of the heavy chain, whilst the mass spectrum of the native heavy chain shows a weight increase of 1298 Da corresponding to the addition of a non-fucosylated, non-galactosylated, non-sialylated, biantennary structure (observed weight 50373 Da). Biantennary, sialylated, galactosylated and fucosylated glycans were also evidenced. The treatment with sialidase confirmed the presence of both mono and bisialylated, biantennary, fucosylated structures. Similar experiments were performed on each antibody. The results are summarized in Table 2 below.

(59) TABLE-US-00002 TABLE 2 Summary of the characterization of N-glycosylation of the heavy chains Clone Isotype Activity Sialylation Fucosylation R35-72 IgG.sub.1 + ND 40% R35-92 IgG.sub.1 + 30% 50% eBRG1 IgG.sub.1 + ND ND eBRG2 IgG.sub.2a − ND 100%  LO-ME* IgG.sub.2a − ND 50% ND: non-detected *the LO-ME antibody shows an unusual mass spectrum profile

(60) According to RS-HPLC followed by high resolution and high precision ESIMS analysis, all the tested anti-IgE monoclonal rat IgG antibodies (R35-72, R35-92, eBRG1, eBRG2 and LO-ME2) display N-glycosylated heavy chains. Regular oligosaccharide structures, of which the majority are biantennary and non-galactosylated, were observed. Glycosylation differed between antibodies' since significant sialylation (about 30%) was only observed in sample R35-92. In addition, the absence of core fucose (sample eBRG1) and partial or total fucosylation (samples R35-92 and eBRG2 respectively) were also evidenced. Therefore, neither sialylation nor fucosylation play a critical role in tolerance-inducing activity.

EXAMPLE 4

(61) This study shows the identification of the receptors of the Fc part of IgG.sub.1 in human buccal tissues.

(62) The anti-CD32 antibodies (clone AT10, ref ab41899, Abcam) recognize the family CD32 inhibitor receptors.

(63) The presence of cells expressing CD32 was detected by immuno-histochemistry at the mucosal/sub-mucosal junction of human gum tissue.

(64) The inventors observed moderate to pronounced intensity labelling of immune cells in very high quantities in the papillary corium (buccal side).

(65) The anti-CD16 antibodies (clone 2H7, ref ab74512, Abcam) recognize the family of CD16 activator receptors.

(66) The presence of cells expressing CD16 was detected by immuno-histochemistry at the mucosal/sub-mucosal junction of human gum tissue.

(67) The inventors observed moderate intensity labelling of immune cells in small quantities in the papillary corium (buccal side).

(68) The inventors have therefore shown that at human gums level, the balance between the CD16 activator receptors and CD32 receptors (which comprise inhibitor receptors) leaned in favour of the latter. They are detected in greater quantity at the papillary corium. As a result, the presence of a larger number of CD32 receptors is in favour of the use of CD32 ligands, and in particular of IgG.sub.1 immunoglobulin, via mucosal route, in particular via sublingual route, in human for the treatment of inflammatory diseases and manifestations.