IVIG COMPOSITION AND METHOD OF TREATMENT OF ANTIBODY DEFICIENT PATIENTS

Abstract

The invention is in the field of therapy of antibody deficiencies such as immune diseases and inflammatory disorders. The inventors demonstrate for the first time the convergence of intestinal IgA and serum IgG responses toward the same microbial targets, under homeostatic conditions. Private anti-microbiota IgG specificities are induced in IgA-deficient patients, but are not found in IgG pools from healthy donors, partially explaining why substitutive IgG (IVIG) cannot regulate antibody deficiency-associated gut dysbiosis and intestinal translocation. Finally, in both controls and IgA-deficient patients, systemic anti-microbiota IgG responses correlate with reduced inflammation suggesting that systemic IgG responses contribute to the gut microbiota confinement. Accordingly, the invention relates to IVIGs (Intravenous immunoglobulins) composition containing at least 1% of immunoglobulins (Ig) from SIgAd (Selective IgA deficiency) patient and their use in the treatment of antibody deficiency disorders such as immune diseases, inflammatory disorders and autoimmune disease.

Claims

1. A composition of IVIGs (Intravenous immunoglobulins) containing at least 1% of immunoglobulin G (IgG) from SIgAd (Selective IgA deficiency) patients.

2. The composition of IVIGs according to claim 1, wherein said composition contains between 1% to 10% of immunoglobulin G (IgG) from SIgAd (Selective IgA deficiency) patients.

3. A method of preparation of the composition of IVIGs according to claim 1 comprising separating plasma proteins into individual stable fractions with different biological functions by Cohn's fractionation; or purifying immunoglobulins by ion-exchange chromatography.

4. A method of treating an antibody deficiency disorder selected from the group consisting of an immune disease, an inflammatory disorder and an autoimmune disease, in a subject in need thereof, comprising administering to the subject a therapeutically effective amount of a pharmaceutical composition comprising the composition of IVIGs of claim 1.

5. The method according to claim 4, wherein the immune disease is a Primary antibody deficiency or a Secondary antibody deficiency.

6. The method according to claim 5 wherein the Primary antibody deficiency is common variable immunodeficiency (CVID).

7. The method according to claim 4, wherein the inflammatory disorder is selected form the group consisting of a gut inflammatory disease, sepsis and graft versus host disease.

8. The method according to claim 7 wherein the gut inflammatory disease is inflammatory bowel disease.

9. The method according to claim 4, wherein the autoimmune disease is a neurological, haematological, nephrological, rheumatological and/or dermatological disease.

Description

FIGURES

[0052] FIG. 1: Systemic IgG and secretory IgA recognize a common spectrum of commensals. A. Representative flow cytometry dot plot showing from bottom to top isotype control, endogenous secretory IgA (without serum), human IgG anti-TNF (10 μg/ml; irrelevant IgG) and autologous systemic IgG (10 μg/ml) to fecal microbiota in a healthy donor. B. Flow cytometry analysis of the fraction of fecal microbiota bound by either secretory IgA, seric IgG or both in healthy donors (n=30). Median values are indicated and subgroups are compared with a non-parametric Mann-Whitney test.

[0053] FIG. 2: Systemic IgG bind a broad spectrum of commensals. A. Flow cytometry analysis of serum IgG binding to cultivated bacterial strains. Grey histograms represent isotype controls and dark lines anti-IgG staining. B. Flow cytometry analysis of serum IgG binding levels to 8 different bacterial strains in healthy donors (n=30). Blue strains (left) are typically poorly coated by secretory IgA from healthy individuals while pink strains (right) are representative of typical IgA targets.sup.15. Results are presented as A Median Fluorescence Intensity (MFI) i.e.: IgG=MFI IgG serum—MFI IgG negative control. Red bars show medians. Kruskal-wallis test was used to calculate p-value. C. Representative immunoblotting of Escherichia coli lysates probed with five different healthy human serums, with a normalized IgA and IgG levels. Ponceau staining indicates total amounts of bacteria lysates loaded. IgA and IgG binding were assessed by an HRP conjugated secondary antibody.

[0054] FIG. 3: IgA deficient patients harbour private anti-commensal IgG responses. A. Flow cytometry analysis of fecal microbiota bound by autologous seric IgG in healthy donors (n=30) and IgA deficient patients (n=15). Red bars represent medians. P-value was calculated by Mann-Whitney test. B. Representative flow cytometry analysis of autologous seric IgG binding (left) or polyclonal IgG derived from pooled serum of healthy donors binding (right) to fecal microbiota. In a healthy donor (top) and in an IgA deficient patient (bottom). C. Flow cytometry analysis of the IgG-bound fecal microbiota with IgG from autologous serum or polyvalent IgG in healthy donors (n=30) and IgA deficient patients (n=15). P-values were calculated by Wilcoxon-paired test. D. Flow cytometry detection of IgG on IgA deficient microbiota (n=9), following incubation with autologous serum or heterologous serum from another, randomly picked, IgA deficient individual. P-value was calculated by Wilcoxon-paired test.

[0055] FIG. 4: Private IgG anti-microbial signatures. A. Sorting strategy of IgG-bound and IgG-unbound microbiota in 10 healthy donors and 3 IgA deficient patients. Composition of sorted subsets was next analysed by 16S rRNA sequencing. B. Genera diversity in IgG+ and IgG− sorted fractions calculated by Shannon index. Dark symbols correspond to healthy donors, red symbols to IgA deficient patients. C. Median relative abundance of genera in IgG+ and IgG− sorted fractions. Dark symbols correspond to healthy donors, red symbols to IgA deficient patients.

[0056] FIG. 5: Microbiota specific IgG and inflammation A. Percentage of serum IgG-bound microbiota correlated with sCD14 levels in autologous serum of healthy donors (triangles) and SIgAd patients (dark points). Spearman coefficient (r) and p-value (p) are indicated. B. Flow cytometry analysis of IgG-bound microbiota following IVIG exposure in healthy donors and CVID patients. C. sCD14 levels measured by ELISA in plasmas of healthy donors and CVID patients. D. Seric IL-6 levels measured by Simoa technology in plasmas of healthy donors and CVID patients. E. Flow cytometry analysis of CD4+CD45RA-PD-1+ lymphocytes in peripheral blood mononuclear cells of healthy donors and CVID patients. Percentage among CD4+ T cells is presented. For all dot plots, black lines represent medians. Mann-Whitney test was used to calculate p-values (*p<0.05, ***p<0.001)

[0057] FIG. 6: In vivo intestinal IgG binding to gut microbiota. Flow cytometry analysis of the fraction of fecal microbiota bound by intestinal IgG in healthy donors (HD; n=30) and selective IgA deficient patients (SIgAd; n=15). Bars represent medians.

[0058] FIG. 7: Anti-commensals IgG react mostly in a Fab-dependent manner A-B. Flow cytometry analysis of 30 healthy (A) and 15 IgA deficient (B) fecal microbiota samples incubated with seric IgG or human IgG anti-TNF. C. Flow cytometry analysis of 10 IgA deficient fecal microbiota samples incubated with heterologous seric IgG or human IgG anti-TNF. Wilcoxon-paired test was used to calculate p-values. **p<0.01;***p<0.001; ****p<0.0001

EXAMPLE 1

[0059] Material & Methods

[0060] Human Samples

[0061] Fresh stool and blood samples were simultaneously collected from n=30 healthy donors, n=15 selective IgA deficiency and n=10 common variable immunodeficiency patients.

[0062] Healthy donors were recruited among laboratory staff and relatives. Patients followed for clinical manifestations associated with antibody deficiencies were recruited from two French clinical immunology referral centers (Department of Clinical Immunology at Saint Louis hospital and Department of Internal Medecine at Pitrié-Salpêtrière hospital, Paris). Patient's inclusion criteria were (i) undetectable seric IgA levels (<0,07 mg/mL) in at least three previous samples in the past year (ii) either selective IgA deficiency (n=15 selective IgA deficient patients), or associated with IgG and/or IgM deficiency integrating a global antibody production defect (n=10 CVID patients). Clinical and biological data were collected at inclusion time.

[0063] Surgical samples from histologically normal intestine were obtained from twelve donors undergoing gastric bypass or tumorectomy at Pitié-Salpêtrière hospital, Paris.

[0064] Oral and written consent were obtained from patients and healthy donors before inclusion in the study.

[0065] PBMC and Plasma

[0066] 30 mL of blood were collected in ACD tubes (BD Vacutainer®) and PBMC were isolated by density gradient procedure (Ficoll 400, Eurobio, Les Ulis, France) and then stored in liquid nitrogen after soft freezing in isopropanol. Supernatants were collected as plasma and immediately stored at −80° C.

[0067] Stool Collection and Whole Microbiota Purification

[0068] Stool were collected immediately after emission in a container allowing anaerobic bacteria preservation (Anaerocult band, Merck, Darmstadt, Germany), aliquoted in a CO2-rich 02-low atmosphere and stored at −80° C. Fecal microbiota were extracted by gradient purification in anaerobic conditions (Freter chamber) as previously described.sup.37. Briefly, thawed feces were diluted in 1×-PBS (Eurobio), 0,03% w/v sodium deoxycholate (NaDC), 60% w/v Nycodenz (Sigma-aldrich, St Louis, USA) and loaded on a continuous density gradient obtained by a freezing-thawing cycle of a Nycodenz solution. Fecal bacteria were obtained after ultracentrifugation (14567×g, 45 min, +4° C.) (Beckman Coulter ultracentrifuge, swinging rotor SW28) and washed three times in 1×-PBS (Eurobio), 0,03% w/v sodium NaDC. The final pellet was diluted in 1×PBS-10%Glycerol, immediately frozen in liquid nitrogen and then stored at −80° C.

[0069] Bacterial Flow Cytometry

[0070] Specific seric antibodies levels against purified microbiota or cultivable strains were assessed by a flow cytometry assay as previously described.sup.11. Briefly, 10.sup.7 bacteria (purified microbiota or cultivable strains) were fixed in a solution of 4% paraformaldehyde and simultaneously stained with a cell proliferation dye (eFluor 450, eBiosciences, Calif., USA). After washing with 1 mL of a 1×-PBS solution, cells were resuspended to a final concentration of 4.10.sup.8 bacteria/mL in a 1×-PBS, 2% w/v BSA, 0.02% w/v Sodium azide solution. Then 10.sup.7 bacteria were incubated in a 96-V bottom well plate with a 10 μg/mL IgG solution (from either human serum or pooled human IgG Hizentra®—CSL Behring France or human anti-TNF Remicade®—MSD France) per condition. Immune complexes were washed twice with a 1×-PBS, 2% w/v BSA, 0.02% w/v Sodium azide (200 μL/well, 4000×g, 10 minutes, +4° C.) and then incubated with secondary conjugated antibodies, either isotype controls mix or goat anti-human IgA-FITC and goat anti-human IgG-A647 (Jackson Immunoresearch Laboratories, West Grove, USA). Acquisition of the cells events was performed on a FACS CANTO II flow cytometer (Becton Dickinson) after washing and analysis was performed with Flow-Jo software (Treestar, Ashland, USA). Medians of fluorescence were used to measure the seric IgG response levels against the cultivable strains. Intestinal IgA binding was quantified by the same assay without incubation with seric immunoglobulins. Results are expressed as median, minimum and maximum percentages throughout the manuscript.

[0071] Cytokines Quantification

[0072] IL-6 and IL-10 were measured in the serum using a 3-step digital assay relying on Single Molecule Array (Simoa) technology HD-1 Analyzer (Quanterix Corporation, Lexington, USA). Working dilutions were ¼ for all sera in working volumes of 25 μL. Lower limit of quantification for IL-6 and IL-10 are respectively of 0.01, 0.021 pg/mL.

[0073] Soluble CD14 Quantification

[0074] Soluble CD14 was quantified in plasma (400-fold dilution) by ELISA (Quantikine® ELISA kit, R&D, Minneapolis, USA). Experimental procedure followed the manufacturer's recommendations. Lower limit of quantification for soluble CD14 is of 6 pg/mL.

[0075] Peripheral Blood Mononuclear Cell Phenotyping

[0076] T cell phenotyping was performed using a combination of the following antibodies : CD3-H500, CCR7-PE-Cy7, CD4-APC-Cy7 (BD Biosciences), CD45RA-PercP Cy5.5 (e-Bioscience), CD8-A405 (Invitrogen), CD279-APC (BioLegend). Acquisition of cells events was performed using a FACS CANTO II flow cytometer (Becton Dickinson) and analysis was performed using the Flow-Jo software (Treestar).

[0077] Intestinal B Cells Phenotyping

[0078] Lamina propria was digested by collagenase A (Roche) in RPMI (Life Technologies) for 30 minutes at 37° C. Lymphocytes were purified by centrifugation over Ficoll 400 (Eurobio) and stained with the following antibodies: anti-CD45 APC-H7, anti-CD19 BV421, anti-IgD FITC, anti-CD27 PE-Cy7 (all purchased from BD Biosciences), and anti-IgA PE (Jackson Immunoresearch), or anti-IgG1 PE, anti-IgG2 AF488, anti-IgG3 A647 (Southern Biotech). Dead cells were excluded with LIVE/DEAD™ Fixable Aqua Dead Cell Stain Kit (Invitrogen). Acquisition of cells events was performed using a FACS CANTO II flow cytometer (Becton Dickinson) and analysis was performed using the Flow-Jo software (Treestar).

[0079] Analysis of IgG-Coated Bacteria

[0080] Purified microbiota (10.sup.9/condition) was washed in 1×-PBS and stained with isotype control (A647-conjugated Goat IgG, Jackson Immunoresearch Laboratories) as a negative control or anti-human IgG-A647 (Jackson Immunoresearch Laboratories). Acquisition and sorting were performed on a 2 lasers-2 ways Fluorescent-activated cell sorter (S3 cell sorter, Bio-Rad Laboratories, California, USA). 10.sup.6 bacteria per fraction were collected and immediately stored at −80° C. as dry pellets. Purity for both fractions was systematically verified after sorting with a minimum rate of 80%. Genomic DNA was extracted and the V3-V4 region of the 16S rRNA gene was amplified by semi-nested PCR. Primers V3fwd (+357): 5′ TACGGRAGGCAGCAG 3′ (SEQ ID N° 1) and V4rev (+857): 5′ ATCTTACCAGGGTATCTAATCCT 3′ (SEQ ID N° 2) were used during the first round of PCR (10 cycles). Primers V3fwd and X926_Rev (+926) 5′ CCGTCAATTCMTTTRAGT 3′ (SEQ ID N° 3) were used in the second PCR round (40 cycles). Polymerase chain reaction amplicon libraries were sequenced using a MiSeq Illumina platform (Genotoul, Toulouse, France). The open source software package Quantitative Insights Into Microbial Ecology (QIIME).sup.38 was used to analysed sequences with the following criteria: (i) minimum and maximum read length of 250 bp and 500 bp respectively, (ii) no ambiguous base calls, (iii) no homopolymeric runs longer than 8 bp and (iv) minimum average Phred score >27 within a sliding window of 50 bp. Sequences were aligned with NAST against the GreenGenes reference core alignment set (available in QIIME as core_set_aligned.fasta.imputed) using the ‘align_seqs.py’ script in QIIME. Sequences that did not cover this region at a percent identity >75% were removed. Operational taxonomic units were picked at a threshold of 97% similarity using cd-hit from ‘pick_otus.py’ script in QUIIME. Picking workflow in QUIIME with the cd-hit clustering method currently involves collapsing identical reads using the longest sequence-first list removal algorithm, picking OTU and subsequently inflating the identical reads to recapture abundance information about the initial sequences. Singletons were removed, as only OTU that were present at the level of at least two reads in more than one sample were retained (9413±5253 sequences per sample). The most abundant member of each OTU was selected through the ‘pick_rep_set.py’ script as the representative sequence. The resulting OTU representative sequences were assigned to different taxonomic levels (from phylum to genus) using the GreenGenes database (release August 2012), with consensus annotation from the Ribosomal Database Project naïve Bayesian classifier [RDP 10 database, version 6.sup.39. To confirm the annotation, OTU representative sequences were then searched against the RDP database, using the online program seqmatch (http://rdp.cme.msu.edu/segmatch/segmatch_intro.jsp) and a threshold setting of 90% to assign a genus to each sequence.

[0081] Immunoblotting

[0082] 10.sup.8 CFU of wild type Escherichia coli were freezed (−80° C.) and thawed (37° C.) three times in 30 μL of lysis buffer (50mM Tris-HCL, 8M urea). Lysis efficiency was verified by Gram staining. Proteins were separated using 4%-20% polyacrylamide gel electrophoresis (Mini-PROTEAN TGX Stain-Free Precast Gels; Bio-Rad) in reducing conditions (dithiothreitol DTT and sodium dodecyl sulfate SDS, Bio-Rad) and transferred to nitrocellulose. Membranes were incubated with 10 μg/ml of human seric IgG or IgA of different healthy donors. Human IgG were detected with horseradish peroxidase-conjugated goat anti-human IgG used at 1:50,000 or goat anti-human IgG used at 1:20,000 followed by enhanced chemi-luminescence revealing reaction (Clarity™ Western ECL, Bio-Rad). Human IgA were detected with horseradish peroxidase-conjugated goat anti-human IgA used at 1:20 000 (Bethyl Laboratories). All incubations were in 1×-PBS with 5% non fat milk and washing steps in 1×-PBS with 0.1% Tween.

[0083] IgG Gene Expression Analysis

[0084] Total RNA of jejunal lamina propria fraction and PBMC were extracted with the RNeasy Mini kit (QIAGEN). cDNAs were synthesized from and prepared with M-MLV reverse transcriptase (Promega). SYBR green primers were designed by manufacturer (Roche) and used for qRT-PCR using the 7300 real time PCR system (Applied Biosystem). Data were normalized to ribosomal 18S RNA.

[0085] Results

[0086] 1/Convergence of Intestinal IgA and Serum IgG Toward the Same Bacterial Cells

[0087] To determine the level of humoral systemic response against fecal microbiota, we have elaborated a flow cytometric assay derived from a previously reported technology.sup.11. This protocol allows to probe concomitantly IgA and IgG microbiota coating. We found that approximately 8% of the fecal microbiota is targeted by secretory IgA (median[min-max]%; 8[0.8-26.7]%; n=30) in healthy donors, in concordance with previous reports.sup.12. As shown, the proportion of bacteria in vivo bound by secretory IgA in human feces is highly variable between healthy individuals (FIG. 1B). IgG-bound bacteria are virtually absent from healthy human feces (median [min-max]%; 0.03[0-0.16]%; n=30 ; FIG. S1 and 1A), in agreement with the lack of IgG transport to the intestinal lumen. In healthy donors, seric IgG bound a median rate of 1.1% of fecal bacteria (median [min-max]%; 1.1[0.2-3.2]%; FIG. 1B). Surprisingly, seric IgG targeted exclusively secretory IgA bound bacteria (FIG. 1A). Conversely, all IgA-coated bacteria (IgA.sup.+ bacteria) were not targeted by seric IgG. Of note, an irrelevant human monoclonal IgG (chimeric anti-human TNF containing a human Fc IgG fraction) exhibits markedly reduced binding to IgA+bacteria, compared to serum IgG (FIG. 1A, S2), demonstrating that IgG binding to IgA-coated bacteria is mostly Fab-mediated.

[0088] To confirm that systemic IgG binding is directed against IgA-bound bacteria, we evaluated in vitro serum IgG binding to cultivable bacterial strains. We selected four bacterial strains that were not preferentially bound by IgA in human feces and four others that were previously defined as classical IgA targets in vivo.sup.12-14. As shown in FIG. 2, IgG from healthy individuals (n=30) bind much more significantly Bifidobacterium longum, Bifidobacterium adolescentis, Faecalibacterium prausnitzii and Escherichia coli, known to be particularly enriched in the IgA-coated fraction of healthy individuals, than three different strains of Bacteroides sp. and Parabacteroides distasonis, known to be particularly enriched in the IgA-uncoated fraction of the fecal microbiota (FIG. 2A-B). The majority of anti-commensal IgG antibodies are of the IgG2b and IgG3 isotypes in mice. Using isotype-specific secondary antibodies we detected minimal IgG1 binding, but high seric IgG2 reactivity, to Bifidobacterium adolescentis, Bifidobacterium longum and Escherichia coli, suggesting that IgG2 is involved in commensals targetting in humans (FIG. S3).

[0089] Since anti-commensal IgG might possibly be triggered during mucosal immune responses, we characterized lamina propria B cells and detected the presence of IgG2+ B cells throughout the intestine (FIG. S4). Of note, IgG transcripts are more abundant in LP tissue that in PBMCs, as measured by qPCR (FIG. S4).

[0090] These results demonstrate that human IgG recognize a wide range of commensal under homeostatic conditions. Systemic humoral immunity (notably IgG2) converges with mucosal immunity to bind the surface of commensals.

[0091] 2/Inter-Individual Variability and Non Overlapping Anti-Commensal IgA and IgG Molecular Targets.

[0092] It was previously suggested that murine IgG would target a restricted number of bacterial proteins and favored highly conserved outer membrane proteins.sup.8. Reactivity of human serum IgG against bacterial lysates from a Gram-negative strains was evaluated by immunoblotting. We observed that IgG labeled several E. coli bands (FIG. 2C), suggesting that multiple bacterial products are involved in the induction of systemic antibodies.

[0093] Interestingly, this analysis reveals a great deal of inter-individual variability, as it is not always the same bacterial products that react with the tested serums. We then compared the overlap between bacterial products labeled by IgG and IgA and found distinct binding profiles (FIG. 2C). Finally, in the 5 individuals tested, although some bacterial products (notably a 15 Kd antigen) are frequently targeted in most subjects and without isotype restriction, it clearly appears that IgA and IgG never share exactly the same binding pattern at a molecular level.

[0094] Taken together, these results demonstrate although IgG converges with IgA to bind the surface of commensals, it appears that IgA and IgG do not systematically target the same bacterial antigens, even at the individual level.

[0095] 3/Private Anti-Microbiota IgG Specificities are Induced in IgA-Deficient Patients

[0096] The existence of seric IgG able to bind IgA-coated bacteria could equally suggest that some gut bacteria (or bacterial antigens) might cross the intestinal barrier: (i) in spite of IgA, or (ii) because of IgA. In order to explore these two putatively opposing roles for IgA, we studied the systemic anti-commensal IgG response in SIgAd. These patients had undetectable seric and digestive IgA levels while seric IgG were in the normal range.sup.15. Anti-microbiota IgG levels were significantly higher in SIgAd compared to controls (median [min-max]%;

[0097] 3.3[0.2-20.2]% versus 1.1%[0.2-3.2]%; FIG. 3A). Using irrelevant human IgG, we confirmed that, like in healthy donors, IgG interact with fecal bacteria in a Fab-dependent manner (Figure S2B). These data support an enhanced triggering of systemic IgG immunity against fecal microbiota when lacking secretory IgA, as shown in the murine model of polymeric immunoglobulin receptor deficiency.sup.6.

[0098] Considering this high level of anti-microbiota IgG in SIgAd, and the similarity of SIgAd and healthy microbiota composition.sup.15, we investigated how anti-microbiota IgG repertoires from healthy donors and IgA deficient patients were overlapping. Using polyclonal IgG from pooled serum of healthy donors, we assessed IgG-bound microbiota using either healthy or SIgAd purified microbiota. We showed that pooled polyclonal IgG and autologous healthy sera recognized a similar percentage of fecal bacteria (median [min-max]%;1[0-3.7] % vs 1.1[0.2-3.2]%, respectively, FIG. 3B-C). In contrast, pooled polyclonal IgG bound a smaller bacterial fraction of IgA deficient-microbiota compared to autologous patient serum (median [min-max]%;0.4[0-3.6] % vs 3.3[0.2-20.2] %, FIG. 3B-C). In order to test whether similar specificities are induced in all or most IgA deficient individuals, we compared their IgG reactivity to autologous or heterologous gut microbiota. In this experiment (FIG. 3D), each IgA-deficient microbiota was incubated either with autologous serum (i.e.: autologous condition), or with serum from an unrelated IgA deficient individual (i.e.: heterologous condition). As shown in FIG. 3D, no significant difference was seen between autologous or heterologous conditions (median autologous IgG+ microbiota 1.2% versus median heterologous IgG+ microbiota 1.4%). Of note, heterologous seric IgG also predominantly interact with fecal microbiota in a Fab-dependent manner (FIG. S2C).

[0099] This set of data suggests that peculiar anti-microbiota IgG specificities are induced in IgA-deficient patients, but not in healthy individuals.

[0100] 4/IgG Specifically Recognize a Broad Spectrum of Bacteria

[0101] To more deeply decipher anti-commensal IgG specificities in both healthy donors and IgA deficient patients, we next performed a stringent flow-sorting to isolate IgG-bound bacteria and identified their taxonomy by 16S rRNA sequencing (FIG. 4A). We observed extensive inter-individual variability at genus level irrespective of immunological status (healthy donors vs IgA deficient patients). Microbial diversity calculated by Shannon index varied between donors, but on average bacterial diversity of IgG.sup.+ and IgG.sup.− bacteria was not significantly different (FIG. 4B). We postulated that IgG might preferentially interact with dominant taxa, and therefore compared relative abundance of IgG-bound and IgG-unbound genera. Both fractions exhibited equal distributions of rare and abundant genera (FIG. 4C), thus IgG target commensals irrespectively of their frequency. Interestingly, we found that individual IgG.sup.+ and IgG.sup.− fecal bacterial profiles were remarkably different, supporting a strong IgG bias against peculiar taxa that cannot be explained by an expansion of the latter. Besides, anti-commensals IgG were not restricted to pathobionts, but also targeted symbiotic genera such as Faecalibacterium, whose the most common species (i.e.: F.prausnitzii) has been assigned anti-inflammatory properties in both healthy donors and IgA deficient patients.sup.16. From this part we conclude that anti-commensal IgG recognize a diverse array of both pathobionts and commensal bacteria. Importantly, each individual harbored a private IgG antimicrobial signature.

[0102] 5/High Anti-Microbiota IgG Levels Correlate with Reduced Systemic Inflammation

[0103] Microbiota-specific serum IgG responses contribute to symbiotic bacteria clearance in periphery and maintain mutualism in mice.sup.2. We thus hypothesized that anti-commensals IgG might influence the balance of systemic inflammatory versus regulatory responses in humans. Hence, we measured plasma levels of sCD14 (a marker of monocyte activation,.sup.17) and observed that seric IgG-coated bacteria inversely correlated with soluble CD14 (r=-0.42, p<0.005; FIG. 5A) in both healthy donors and SIgAd patients. These results are in line with the finding that IgG replacement therapy reduced endotoxemia.sup.18. To further explore the potential link between anti-microbiota IgG and systemic inflammation, we explored CVID patients (characterized by both IgG and IgA defects). These patients benefit from IVIG treatment. Yet, we show that IVIG do not efficiently bind CVID microbiota. As shown in FIG. 5B, IVIG bound a reduced fraction of CVID microbiota compared to control microbiota (median [min-max]%; 0.37[0.00-1.14]% vs 1.06[0.00-3.7]%). We then determined plasma levels of sCD14 and IL-6 (an inflammatory cytokine reflecting T-cell activation) and evaluated the expression of PD-1 (a T-cell co-inhibitory molecule induced after activation) on CD4+ T cells. IL-6 as well as sCD14 levels were consistently higher in CVID patients than in healthy donors (IL-6, median [min-max]%, 1.8(0.7-60.1) pg/ml versus 0.6(0.33-2.4) pg/ml; sCD14, median [min-max]%; 2063 (590-5493) pg/ml versus median 2696(1147-4283) pg/ml; FIG. 5C-D). Moreover, CD45RA-PD1+CD4+T cells tended to increase in CVID patients, as compared with healthy donors (median [min-max]%; 20.3(4.26-59.6)% versus 10(2.09-41.9)%, FIG. 5E).

[0104] Altogether, in both controls and IgA-deficient patients, systemic anti-microbiota IgG responses correlate with reduced inflammation.

[0105] Discussion

[0106] Anti-commensal IgG have been described in patients with inflammatory diseases .sup.5,19,20. Here, we characterize for the first time a broad anti-commensal IgG response under homeostatic conditions in humans. Previous work demonstrated that symbiotic Gram-negative bacteria disseminate spontaneously and drive systemic IgG responses.sup.8. We show here that a diverse array of commensal bacteria, including Gram-positive and Gram-negative species, can induce systemic IgG. We show that a pathobiont like E. coli induce less systemic IgG responses than a presumably beneficial symbiont like B. adolescentis (FIG. 2B). Therefore the systemic IgG response in healthy humans does not appear preferentially driven by pathobionts, but also by commensals. In mice it has been shown that commensal microbes induce serum IgA responses that protect against sepsis.sup.21, illustrating the consequence of systemic anti-microbial IgA binding to both pathogenic strains and commensals. We postulate that systemic anti-microbiota IgG, also mainly induced by commensals, could have the same protective role.

[0107] Strikingly, systemic IgG and secretory IgA converge towards the same autologous microbiota subset. Yet, it seems unlikely that secretory IgA enhances systemic IgG responses, since IgA deficiency is associated with high proportions of IgG+ microbiota, as detected using bacterial flow cytometry on SIgAd microbiota labeled with autologous serum. In addition, induction of anti-commensal IgG has been shown to be microbiota-dependent, but IgA-independent in mice.sup.2,6. Systemic IgG could reflect asymptomatic gut microbiota translocation episodes in healthy individuals. Repeated bacterial translocations might occur more frequently in the absence of secretory IgA, accounting for elevated anti-microbiota IgG levels in these patients.

[0108] IgA do not activate complement via the classical pathway.sup.22. Interestingly, the anti-Bifidobacterium adolescentis IgG response is primarily restricted to the IgG2 isotype (FIG. S3), which less efficiently triggers the classical route of complement than IgG1 and IgG3.sup.23. Furthermore, IgG2 poorly interact with type I FcγRs, while IgG1 and IgG3 demonstrate affinity for most FcγRs.sup.24. These distinct binding patterns have functional consequences. IgG1 antibodies mediate phagocytosis and induce potent pro-inflammatory pathways while IgG2 are rather involved in dendritic cell or B cell activation.sup.25,26. Besides its specific Fc domain interaction, IgG2 is usually, but not exclusively, associated with anti-carbohydrate responses.sup.27. IgA was also recently shown to bind multiple microbial glycans.sup.28. Thus, IgA and IgG2 could be viewed as playing similar roles, but in different compartments. Much effort has been recently expended to develop bacterial glycan or protein microarray. Glycomics could represent a new option in order to better decipher anti-microbiota antibody targets.sup.27,29.

[0109] Importantly, we show that IgA and IgG do not systematically target the same bacterial antigens at an individual level (FIG. 2C). Therefore IgG and IgA epitopes are not strictly overlapping. This result could further illustrate antibacterial IgA/IgG synergy, and explain the absence of isotype competition allowing the observed IgA/IgG co-staining of bacteria (FIG. 1).

[0110] Recent studies suggested that murine secretory IgA are polyreactive and bind a broad but defined subset of microbiota.sup.30,31. Similarly, up to 25% of intestinal IgG.sup.+ plasmablasts could produce polyreactive antibodies.sup.9. We therefore hypothesized that the cross-reactive potential of anti-commensal IgG may act as a first line of defense against potentially harmful bacteria. In line with this idea, it can be noted that homeostatic anti-commensal IgG confer protection against pathogens such as Salmonella.sup.8. Conversely, IgG directed against Klebsiella pneumoniae, an opportunistic pathogen, cross-react with commensal microbes.sup.32. Clonally related memory B cells expressing cross-specific anti-K. pneumoniae antibodies were found in both lamina propria and peripheral blood in humans suggesting that generation of anti-commensal antibodies might be triggered in the mucosal compartment. At the same time, anti-commensal memory B cells might recirculate in periphery.sup.32. Altogether, it appears possible that bacteria-specific IgG would arise from the gut, as all bacteria-specific IgG isotypes we characterized in human sera are also present in the gut (FIG. S4), and also because a large proportion of gut IgG+ B cells are expected to be commensal-specific.sup.9. However, it remains presently unknown whether serum IgG responses mainly originate from the gut and/or are induced the periphery following bacterial translocation.

[0111] We report that each individual harbors a private set of anti-commensal IgG in both healthy donors and IgA deficient patients. Since our analysis was limited to 3 IgA deficient patients, further study might precisely reveal how SIgAd anti-commensal IgG bind a distinct set of commensals. While IVIG preparations contain an extended set of anti-commensal IgG, we observe that IVIG less efficiently bind CVID microbiota. These observations are consistent with reported alterations of gut microbiota in CVID patients.sup.33. Microbiota perturbations are also associated with selective IgA deficiency. The latter perturbations are less pronounced than in CVID, since the presence of IgM appears to preserve SIgAd microbiota diversity.sup.15. Nevertheless, IgA deficiency condition is also associated in severe cases with bacterial translocation, colitis and dysbiosis. These complications are not accessible to substitutive Ig replacement therapy.sup.34. Indeed, IVIG do not appear to contain high-enough concentrations as well as appropriate specificities of anti-commensal IgG. As shown in FIG. 3, healthy control serum usually less efficiently binds IgA deficient microbiota than autologous serum. Similarly, IVIG poorly targets CVID gut microbiota (FIG. 5B). In addition, local mucosal antibody responses might be important in regulating microbiota composition in a way that cannot be substituted by IVIG. These findings expand our understanding of how IVIG fail to treat gastro-intestinal symptoms in CVID and IgA deficient patients. Dysbiosis and gastro-intestinal complications might not accessible to substitutive Ig replacement therapy, since, as we show, healthy IgG repertoire does not contain adequate “dysbiotic-specific” antibodies.

[0112] It was recently shown in mice that maternally-derived anti-commensal IgG dampen aberrant mucosal immune responses and strengthen epithelial barrier.sup.7,35. The contribution of systemic anti-commensal IgG to the regulation of microbiota/immune homeostasis was not explored in the latter studies. Here, we show that anti-commensal IgG are negatively associated with sCD14, suggesting they might quell inflammation. In support of this, we measured higher levels of sCD14 and IL-6 in plasma of patients lacking both IgA and IgG compared to controls (FIG. 5).

[0113] Altogether, these data suggest that systemic IgG and intestinal IgA cooperate in different body compartments to limit systemic pro-inflammatory pathways. While selective IgA deficient patients harbour elevated seric anti commensal IgG levels, CVID patients can not mount an appropriate IgG response. These findings suggest that : in selective IgA deficiency, microbiota confinement is obtained at the price of a strong inflammatory response, and in CVID, confinement is lost and Ig replacement therapy do not substitute for a specific autologuous IgG response. We therefore propose that IgA supplementation might have beneficial effects on gut dysbiosis and systemic inflammatory disorders associated with antibody deficiencies. IgA might be orally delivered through a carrier system allowing colon delivery. Polymers such as gellan gum or pectin, are degraded specifically by the colonic microbiota and could thus release polymer-bound IgA locally.sup.36.

[0114] In summary, we report for the first time a systemic anti-commensal IgG response that is restricted to intestinal IgA-coated bacteria in humans. We demonstrate that in the absence of IgA, anti-commensal IgG responses are amplified and associated with reduced systemic inflammation. Finally, the present study provides new therapeutic perspectives based on IgA supplementation in patients with CVID or SIgAd, while SIgAd -derived IgG supplementation might be considered in CVID.

REFERENCES

[0115] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

[0116] 1. Honda K, Littman D R. The microbiota in adaptive immune homeostasis and disease. Nature. 2016;535:75.

[0117] 2. Slack E, Hapfelmeier S, Stecher B, Velykoredko Y, Stoel M, Lawson M A E, et al. Innate and adaptive immunity cooperate flexibly to maintain host-microbiota mutualism. Science. 2009;325:617-20.

[0118] 3. Donskey C J. The role of the intestinal tract as a reservoir and source for transmission of nosocomial pathogens. Clin Infect Dis Off Publ Infect Dis Soc Am. 2004;39:219-26.

[0119] 4. MacFie J. Current status of bacterial translocation as a cause of surgical sepsis. Br Med Bull. 2004;71:1-11.

[0120] 5. Beaugerie L, Sokol H. Clinical, serological and genetic predictors of inflammatory bowel disease course. World J Gastroenterol. 2012;18:3806-13.

[0121] 6. Johansen F E, Pekna M, Norderhaug IN, Haneberg B, Hietala M A, Krajci P, et al. Absence of epithelial immunoglobulin A transport, with increased mucosal leakiness, in polymeric immunoglobulin receptor/secretory component-deficient mice. J Exp Med. 1999;190:915-22.

[0122] 7. Koch M A, Reiner G L, Lugo K A, Kreuk L S M, Stanbery A G, Ansaldo E, et al. Maternal IgG and IgA Antibodies Dampen Mucosal T Helper Cell Responses in Early Life. Cell. 2016;165:827-41.

[0123] 8. Zeng M Y, Cisalpino D, Varadarajan S, Hellman J, Warren H S, Cascalho M, et al. Gut Microbiota-Induced Immunoglobulin G Controls Systemic Infection by Symbiotic Bacteria and Pathogens. Immunity. 2016;44:647-58.

[0124] 9. Benckert J, Schmolka N, Kreschel C, Zoller M J, Sturm A, Wiedenmann B, et al. The majority of intestinal IgA+ and IgG+ plasmablasts in the human gut are antigen-specific. J Clin Invest. 2011;121:1946-55.

[0125] 10. Iversen R, Snir O, Stensland M, Kroll J E, Steinsbo Ø, Korponay-Szabó I R, et al. Strong Clonal Relatedness between Serum and Gut IgA despite Different Plasma Cell Origins. Cell Rep. 2017;20:2357-67.

[0126] 11. Moor K, Fadlallah J, Toska A, Sterlin D, Balmer M L, Macpherson A J, et al. Analysis of bacterial-surface-specific antibodies in body fluids using bacterial flow cytometry. Nat Protoc. 2016;11:1531-53.

[0127] 12. Palm N W, de Zoete M R, Cullen T W, Barry N A, Stefanowski J, Hao L, et al. Immunoglobulin A coating identifies colitogenic bacteria in inflammatory bowel disease. Cell. 2014;158:1000-10.

[0128] 13. D′Auria G, Peris-Bondia F, D{circumflex over (z)}nková M, Mira A, Collado M C, Latorre A, et al. Active and secreted IgA-coated bacterial fractions from the human gut reveal an under-represented microbiota core. Sci Rep. 2013;3:3515.

[0129] 14. Kau AL, Planer J D, Liu J, Rao S, Yatsunenko T, Trehan I, et al. Functional characterization of IgA-targeted bacterial taxa from undernourished Malawian children that produce diet-dependent enteropathy. Sci Transl Med. 2015;7:276ra24.

[0130] 15. Fadlallah J, El Kafsi H, Sterlin D, Juste C, Parizot C, Dorgham K, et al. Microbial ecology perturbation in human IgA deficiency. Sci Transl Med. 2018;10.

[0131] 16. Sokol H, Pigneur B, Watterlot L, Lakhdari O, Bermúdez-Humarán L G, Gratadoux J-J, et al. Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identified by gut microbiota analysis of Crohn disease patients. Proc Natl Acad Sci USA. 2008;105:16731-6.

[0132] 17. Bazil V, Strominger JL. Shedding as a mechanism of down-modulation of CD14 on stimulated human monocytes. J Immunol Baltim Md 1950. 1991;147:1567-74.

[0133] 18. Perreau M, Vigano S, Bellanger F, Pellaton C, Buss G, Comte D, et al. Exhaustion of bacteria-specific CD4 T cells and microbial translocation in common variable immunodeficiency disorders. J Exp Med. 2014;211:2033-45.

[0134] 19. Landers C J, Cohavy O, Misra R, Yang H, Lin Y-C, Braun J, et al. Selected loss of tolerance evidenced by Crohn's disease-associated immune responses to auto- and microbial antigens. Gastroenterology. 2002;123:689-99.

[0135] 20. Macpherson A, Khoo U Y, Forgacs I, Philpott-Howard J, Bjarnason I. Mucosal antibodies in inflammatory bowel disease are directed against intestinal bacteria. Gut. 1996;38:365-75.

[0136] 21. Wilmore J R, Gaudette B T, Gomez Atria D, Hashemi T, Jones D D, Gardner C A, et al. Commensal Microbes Induce Serum IgA Responses that Protect against Polymicrobial Sepsis. Cell Host Microbe. 2018;23:302-311.e3.

[0137] 22. Russell M W, Mansa B. Complement-fixing properties of human IgA antibodies. Alternative pathway complement activation by plastic-bound, but not specific antigen-bound, IgA. Scand J Immunol. 1989;30:175-83.

[0138] 23. Bindon C I, Hale G, Brüggemann M, Waldmann H. Human monoclonal IgG isotypes differ in complement activating function at the level of C4 as well as C1q. J Exp Med. 1988;168:127-42.

[0139] 24. Bruhns P, Iannascoli B, England P, Mancardi D A, Fernandez N, Jorieux S, et al. Specificity and affinity of human Fcgamma receptors and their polymorphic variants for human IgG subclasses. Blood. 2009;113:3716-25.

[0140] 25. Nimmerjahn F, Gordan S, Lux A. FcγR dependent mechanisms of cytotoxic, agonistic, and neutralizing antibody activities. Trends Immunol. 2015;36:325-36.

[0141] 26. White A L, Chan H T C, French R R, Willoughby J, Mockridge C I, Roghanian A, et al. Conformation of the human immunoglobulin G2 hinge imparts superagonistic properties to immunostimulatory anticancer antibodies. Cancer Cell. 2015;27:138-48.

[0142] 27. Schneider C, Smith D F, Cummings R D, Boligan K F, Hamilton R G, Bochner B S, et al. The human IgG anti-carbohydrate repertoire exhibits a universal architecture and contains specificity for microbial attachment sites. Sci Transl Med. 2015;7:269ra1.

[0143] 28. Bunker J J, Erickson S A, Flynn T M, Henry C, Koval J C, Meisel M, et al. Natural polyreactive IgA antibodies coat the intestinal microbiota. Science. 2017;

[0144] 29. Christmann B S, Abrahamsson T R, Bernstein C N, Duck L W, Mannon P J, Berg G, et al. Human seroreactivity to gut microbiota antigens. J Allergy Clin Immunol. 2015;136:1378-1386-5.

[0145] 30. Bunker J J, Flynn T M, Koval J C, Shaw D G, Meisel M, McDonald B D, et al.

[0146] Innate and Adaptive Humoral Responses Coat Distinct Commensal Bacteria with Immunoglobulin A. Immunity. 2015;43:541-53.

[0147] 31. Okai S, Usui F, Yokota S, Hori-i Y, Hasegawa M, Nakamura T, et al. High-affinity monoclonal IgA regulates gut microbiota and prevents colitis in mice. Nat Microbiol.

[0148] 32. Rollenske T, Szijarto V, Lukasiewicz J, Guachalla LM, Stojkovic K, Hartl K, et al. Cross-specificity of protective human antibodies against Klebsiella pneumoniae LPS O-antigen. Nat Immunol. 2018;19:617-24.

[0149] 33. Jørgensen S F, Troseid M, Kummen M, Anmarkrud J A, Michelsen A E, Osnes L T, et al. Altered gut microbiota profile in common variable immunodeficiency associates with levels of lipopolysaccharide and markers of systemic immune activation. Mucosal Immunol. 2016;9:1455-65.

[0150] 34. Favre O, Leimgruber A, Nicole A, Spertini F. Intravenous immunoglobulin replacement prevents severe and lower respiratory tract infections, but not upper respiratory tract and non-respiratory infections in common variable immune deficiency. Allergy. 2005;60:385-90.

[0151] 35. Gomez de Agiiero M, Ganal-Vonarburg S C, Fuhrer T, Rupp S, Uchimura Y, Li H, et al. The maternal microbiota drives early postnatal innate immune development. Science. 2016;351:1296-302.

[0152] 36. Sandolo C, Péchiné S, Le Monnier A, Hoys S, Janoir C, Coviello T, et al. Encapsulation of Cwp84 into pectin beads for oral vaccination against Clostridium difficile. Eur J Pharm Biopharm Off J Arbeitsgemeinschaft Pharm Verfahrenstechnik EV. 2011;79:566-73.

[0153] 37. Juste C, Kreil D P, Beauvallet C, Guillot A, Vaca S, Carapito C, et al. Bacterial protein signals are associated with Crohn's disease. Gut. 2014;63:1566-77.

[0154] 38. Caporaso J G, Kuczynski J, Stombaugh J, Bittinger K, Bushman F D, Costello EK, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7:335-6.

[0155] 39. Cole J R, Wang Q, Cardenas E, Fish J, Chai B, Farris R J, et al. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 2009;37:D141-145.