NEUTRALISING ANTIBODIES TO THE MAJOR EXOTOXINS TCDA AND TCDB OF CLOSTRIDIUM DIFFICILE

Abstract

This present invention describes the derivation and selection of antibodies capable of neutralising the major exotoxins; TcdA and TcdB of Clostridium difficile. The invention also describes novel neutralisation and antigen binding properties of individual Mabs and mixtures thereof.

Claims

1-41. (canceled)

42. A pharmaceutical composition comprising one or more monoclonal antibodies specific to antigen TcdB1234 wherein the antibody has high affinity of 600 pM or less for the target antigen TcdB1234 and said one or more monoclonal antibodies are independently selected from: i) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:124 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 125 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 126 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:121 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:122 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:123 for CDR-L3; ii) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:154 for CDR-H1, a CDR having the sequence given in SEQ ID NO:155 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:156 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:151 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:152 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:153 for CDR-L3; iii) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:144 for CDR-H1, a CDR having the sequence given in SEQ ID NO:145 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:146 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:141 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:142 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:143 for CDR-L3; iv) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:164 for CDR-H1, a CDR having the sequence given in SEQ ID NO:165 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:166 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:161 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:162 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:163 for CDR-L3; v) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:64 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 65 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 66 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:61 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:62 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:63 for CDR-L3; vi) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:74 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 75 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 76 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:71 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:72 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:73 for CDR-L3; vii) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:84 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 85 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 86 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:81 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:82 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:83 for CDR-L3; viii) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:134 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 135 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 136 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:131 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:132 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:133 for CDR-L3.

43. A pharmaceutical composition according to claim 42, wherein said one or more monoclonal antibodies are independently selected from: i) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:124 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 125 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 126 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:121 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:122 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:123 for CDR-L3; ii) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:154 for CDR-H1, a CDR having the sequence given in SEQ ID NO:155 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:156 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:151 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:152 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:153 for CDR-L3; iii) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:144 for CDR-H1, a CDR having the sequence given in SEQ ID NO:145 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:146 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:141 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:142 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:143 for CDR-L3; and iv) a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:164 for CDR-H1, a CDR having the sequence given in SEQ ID NO:165 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:166 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:161 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:162 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:163 for CDR-L3.

44. A pharmaceutical composition according to claim 42, wherein a monoclonal antibody which specifically binds TcdB1234 comprises a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:124 for CDR-H1, a CDR having the sequence given in SEQ ID NO: 125 for CDR-H2 and a CDR having the sequence given in SEQ ID NO: 126 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:121 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:122 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:123 for CDR-L3.

45. A pharmaceutical composition according to claim 44, wherein the monoclonal antibody has a heavy chain comprising the sequence given in SEQ ID NO:129 and a light chain comprising the sequence given in SEQ ID NO:127.

46. A pharmaceutical composition according to claim 42, wherein a monoclonal antibody which specifically binds TcdB1234 comprising a heavy chain wherein the variable domain of the heavy chain comprises a CDR having the sequence given in SEQ ID NO:154 for CDR-H1, a CDR having the sequence given in SEQ ID NO:155 for CDR-H2 and a CDR having the sequence given in SEQ ID NO:156 for CDR-H3, and a light chain wherein the variable domain of the light chain comprises a CDR having the sequence given in SEQ ID NO:151 for CDR-L1, a CDR having the sequence given in in SEQ ID NO:152 for CDR-L2 and a CDR having the sequence given in SEQ ID NO:153 for CDR-L3.

47. A pharmaceutical composition according to claim 46, wherein the monoclonal antibody has a heavy chain comprising the sequence given in SEQ ID NO:159 and a light chain comprising the sequence given in SEQ ID NO:157.

48. A pharmaceutical composition according to claim 42, wherein the monoclonal antibody has a heavy chain comprising the sequence given in SEQ ID NO:149 and a light chain comprising the sequence given in SEQ ID NO:147.

49. A pharmaceutical composition according to claim 42, wherein the monoclonal antibody has a heavy chain comprising the sequence given in SEQ ID NO:169 and a light chain comprising the sequence given in SEQ ID NO:167.

50. A pharmaceutical composition according to claim 42, wherein at least one of said monoclonal antibodies is a neutralizing antibody which is effective against ribotypes 003.

51. A pharmaceutical composition according to claim 42, further comprising two or more antibodies specific to TcdB.

52. A pharmaceutical composition according to claim 42, further comprising one or more antibodies specific to TcdA.

53. A pharmaceutical composition according to claim 52, wherein the antibody which specifically binds TcdA comprises a heavy chain sequence given in SEQ ID NO: 19 and a light chain sequence given in SEQ ID NO: 17.

54. A pharmaceutical composition according to claim 52, wherein the antibody which specifically binds TcdA comprises a heavy chain sequence given in SEQ ID NO: 29 and a light chain sequence given in SEQ ID NO: 27.

55. A pharmaceutical composition according to claim 52, wherein the antibody which specifically binds TcdA comprises a heavy chain sequence given in SEQ ID NO: 49 and a light chain sequence given in SEQ ID NO: 47.

56. A pharmaceutical composition according to claim 52, wherein the antibody which specifically binds TcdA comprises a heavy chain sequence given in SEQ ID NO: 59 and a light chain sequence given in SEQ ID NO: 57.

57. A pharmaceutical composition according to claim 52, wherein the antibody which specifically binds TcdA comprises a heavy chain sequence given in SEQ ID NO: 39 and a light chain sequence given in SEQ ID NO: 37.

58. A pharmaceutical composition according to claim 52, wherein the antibody which specifically binds TcdA comprises a heavy chain sequence given in SEQ ID NO: 9 and a light chain sequence given in SEQ ID NO: 7.

59. A pharmaceutical composition according to claim 52, which further comprises a pharmaceutically acceptable excipient.

60. A method of treatment or prophylaxis of Clostridium difficile infection or complications therefrom comprising administering a pharmaceutical composition as defined in claim 42 to a patient in need thereof.

61. A method according to claim 60, wherein the composition is co-administrated with a compound selected the group comprising metronidazole, vancomycin, clindamycin, fidaxomicin and combinations thereof.

Description

FIGURES

[0429] FIG. 1-10 shows various antibody and fragment sequences

[0430] FIG. 11 shows sera titres for TcdA and TcdB

[0431] FIG. 12 shows anti TcdA (Ribotype 003) in-vitro neutralization data for single Mabs

[0432] FIG. 13 shows anti TcdA (Ribotype 003) in-vitro neutralization data for single and paired Mabs

[0433] FIG. 14-15 shows anti TcdA (Ribotype 003) in-vitro neutralization data for paired Mabs

[0434] FIG. 16-18 shows anti TcdA (Ribotype 003) in-vitro neutralization data for three Mab mixtures

[0435] FIG. 19-20 shows anti TcdA (Ribotype 003) in-vitro neutralization data for four and five Mab mixtures

[0436] FIG. 21-22 shows anti TcdA (Ribotype 003) in-vitro neutralization data for single and paired Mabs at different TcdA concentrations

[0437] FIG. 23-24 shows anti TcdA (Ribotype 003) in-vitro neutralization data for single and to five Mab mixtures at different TcdA concentrations

[0438] FIG. 25-26 shows anti TcdB (Ribotype 003) in-vitro neutralization data for single Mabs

[0439] FIG. 27-30 shows anti TcdB (Ribotype 003) in-vitro neutralization data for paired Mabs

[0440] FIG. 31-33 shows anti TcdB (Ribotype 003) in-vitro neutralization data for three Mab mixtures

[0441] FIG. 34-40 shows anti TcdB (Ribotype 003) in-vitro neutralization data for two Mab mixtures at different toxin concentrations

[0442] FIG. 41-45 shows anti TcdB (Ribotype 003) in-vitro neutralization data for two Mab mixtures at different relative Mab ratios and different toxin concentrations

[0443] FIG. 46-59 shows TcdB neutralisation data for single antibodies and pairs of antibodies

[0444] FIG. 60 shows the amino acid sequence for TcdA

[0445] FIG. 61 shows the amino acid sequence for TcdB

[0446] FIG. 62 shows TEER assay data for TcdA in a histogram format

[0447] FIG. 62A shows TEER assay data for TcdA in line graph format

[0448] FIG. 63 shows a meier-kaplan curve for the combination of antibodies 997, 1125 and 1151, high concentration is 50 mg/Kg and low concentration is 5 mg/Kg 50 mg/kg dose gave 100% protection to day 11, 82% protection to day 28. 5 mg/kg dose resulted in non-durable and incomplete protection.

[0449] FIG. 64 shows bodyweight changes for vancomycin and vehicle treated hamsters

[0450] FIG. 65 shows the bodyweight for low dose antibodies 5 mg/Kg and high dose antibodies 50 mg/Kg

[0451] FIG. 66 shows photographs of a colon where the animal received treatment with antibodies according to the present disclosure vs a control

[0452] FIG. 67-68 show effects of vortexing on antibody stability

[0453] FIG. 69 shows a comparison of aggregation stability for various antibodies

[0454] FIG. 70-73 show neutralisation of TcdA for various ribotypes

EXAMPLES

Antibody Generation

[0455] A range of different immunogens and screening reagents were either purchased or produced by conventional E. coli expression techniques in order to provide a diverse and broad immune response and to facilitate identification and characterisation of monoclonal antibodies (listed in Table 1). In cases where recombinant proteins or peptides were generated, sequences were based on ribotype 027. The sequence for TcdA from ribotype 027 is given in SEQ ID NO: 171 (Uniprot accession number C9YJ37) and the sequence for TcdB from ribotype 027 is given is SEQ ID NO: 172 (Uniprot accession number C9YJ35).

[0456] Sprague Dawley rats and half-lop rabbits were immunised with either synthetic peptides mapping to regions common to both TcdA and TdcdB full-length toxin, formaldehyde-inactivated toxoid A, binding domain fragments of Toxin A (CROPs1,2,3 or CROPs4,5,6) or binding domain fragment of Toxin B (CROPs1,2,3,4), or in some cases, a combination of the above. Following 2 to 6 immunisations, animals were sacrificed and PBMC, spleen and bone marrow harvested. Sera were monitored for binding to Toxin A domains, toxin B domains, toxin or toxoid by ELISA. Sera titres of 2 such immunisations are shown in FIG. 11. UCB SLAM was used as a means to generate monoclonal antibodies. B cells were cultured directly from immunised animals (Zubler et al., 1985). This step enabled sampling of a large percentage of the B cell repertoire. By incorporating the selected lymphocyte antibody method (SLAM) (Babcook et al., 1996) it was possible to deconvolute positive culture wells and identify antigen-specific antibody-secreting cells. Here we used a modified version of SLAM (UCB SLAM (Tickle et al. 2009)) that utilises a fluorescence-based method to identify antigen-specific B cells from culture wells. B cell cultures were set up and supernatants were first screened for their ability to bind a relevant purified toxin domain (binding, translocation or catalytic) in a bead-based assay using an Applied Biosystem 8200 cellular detection system. This was a homogeneous assay using B cell culture supernatant containing IgG, biotinylated toxin domains coated onto streptavidin beads and a goat anti-rat/rabbit Fc-Cy5 conjugate. Cell cultures positive for binding to TcdA or TcdB components from this assay were selected for use in cell-based functional assays to identify neutralisers of toxin-induced cytotoxicity. Approximately 12,000 toxin-specific positives were identified in the primary binding screen from a total of 4050-plate experiments. This equated to the screening of approximately 0.5 billion B cells. Heavy and light variable region gene pairs were isolated from single cells harvested by micromanipulation from approximately 100 toxin-neutralising wells following reverse transcription (RT)-PCR. These V-region genes were then cloned as mouse IgG1/kappa full-length antibodies for rat variable regions and rabbit IgG/kappa full-length antibodies for rabbit variable regions. Antibodies were re-expressed in a HEK-293 transient expression system. These recombinant antibodies were then retested for their ability to neutralise toxin in cell based assays. Recombinant antibodies were also screened by BIAcore to determine affinity for a given toxin domain and to also determine the specificity and approximate the number of binding events of antibody to toxin. Based on in vitro activity in cell based assays and affinity measurements, lead candidates were selected for humanisation. Unless otherwise stated, all the data herein was generated using the humanised antibodies.

[0457] A panel of recombinant, E. coli-produced toxin fragments (TcdA), C. difficile-derived toxin or toxoid (A) and synthetic peptides (B) were generated or purchased from commercial sources.

TABLE-US-00053 TABLE 1 Toxin A (TcdA) sequence related reagents for screening and immunizations. Fragment Residue number Source TcdA catalytic M1-E659 UCB E. coli expression TcdA translocation K577-D1350 UCB E. coli expression TcdA CROPS.sub.123 (TcdA123) S1827-D2249 UCB E. coli expression TcdA CROPS.sub.456 (TcdA456) G2205-R2608 UCB E. coli expression TcdA CROP.sub.1 S1827-N1978 UCB E. coli expression TcdA CROP.sub.2 G1966-N2133 UCB E. coli expression TcdA CROP.sub.3 G2100-D2249 UCB E. coli expression TcdA CROP.sub.4 G2213-N2381 UCB E. coli expression TcdA CROP.sub.5 G2328-N2494 UCB E. coli expression TcdA CROP.sub.6 G2462-N2609 UCB E. coli expression TcdA CROP.sub.7 R2554-D2701 UCB E. coli expression TcdB catalytic M1-A593 UCB E. coli expression TcdB translocation R576-D1349 UCB E. coli expression TcdB binding (TcdB1234) S1833-E2366 UCB E. coli expression TcdB CROP.sub.1 S1833-S1981 UCB E. coli expression TcdB CROP.sub.2 G1968-D2113 UCB E. coli expression TcdB CROP.sub.3 G2100-E2247 UCB E. coli expression TcdB CROP.sub.4 G2234-E2366 UCB E. coli expression Toxin A Full length purchased Toxin B Full length purchased Toxoid A Full length purchased

TABLE-US-00054 TABLE2 ToxinB(TcdB)sequencerelatedreagentsfor screeningandimmunizations. ToxinDomain AminoacidSequence Catalytic SPVEKNLHFVWIGGEVSDSEQIDNO:173 Catalytic NLAAASDIVRLSEQIDNO:174 Catalytic CGGVYLDVDMLPGIHSEQIDNO:175 Catalytic CGGVYLDVDMLPGIHSDLFKSEQIDNO: 176 Catalytic CWEMIKLEAIMKYKSEQIDNO:177 Catalytic CTNLVIEQVKNRSEQIDNO:178 Catalytic PEARSTISLSGPSEQIDNO:179 Catalytic CSNLIVKQIENRSEQIDNO:180 Catalytic TEQEINSLWSFDQASEQIDNO:181 Catalytic TEQEINSLWSFDPEARSTISLSGPCSEQID NO:182 Translocation NVEETYPGKLLLCSEQIDNO:183 Translocation Acetyl-CANQYEVRINSEGRSEQIDNO: 184 Translocation VNTLNAAFFIQSLICSEQIDNO:185 Translocation YAQLFSTGLNTICSEQIDNO:186 Translocation CAGISAGIPSLVNNELSEQIDNO:187 Translocation DDLVISEIDFNNNSICSEQIDNO:188 Translocation MEGGSGHTVTSEQIDNO:189 Translocation AVNDTINVLPTITEGIPIVSTILDGINLGAAIK ELSEQIDNO:190 Binding CGFEYFAPANTDANNIEGQASEQIDNO: 191 Binding CGYKYFAPANTVNDNIYGQASEQIDNO: 192 Binding CKYYFNTNTAEASEQIDNO:193 Binding CKYYFDEDTAEASEQIDNO:194
Expression and Purification of C. difficile Anti-Toxin Mabs

[0458] Separate light chain and heavy chain mammalian expression plasmids were combined in equimolar ratios and used to transfect HEK-293 or CHO-S cells. For small scale expression studies lipofectamine and HEK-293 cells were used whereas for production of larger batches of IgG electroporation into CHO-S was preferred.

[0459] Culture supernatants were loaded onto a Mab Select SuRe column (in PBS pH 7.4). Antibody was eluted with 100% 0.1M Sodium Citrate pH 3.4 buffer. Samples were neutralized to pH7.4 with Tris.Cl pH8.0. Aggregate was removed by Superdex 200 Gel Filtration column in PBS pH 7.4.

TABLE-US-00055 TABLE 3 Cell Volume of Expression Amount Antibody type SN (L) type purified (mg) CA164_00997.g1_P3 CHO 10 Transient 755.93 CA164_00922.g1_P3 CHO 0.5 Transient 129.36 CA164_01125.g2_P3 CHO 10 Transient 498.96 CA164_01151.g4_P3 CHO 5 Transient 262.43

Example 1 In-Vitro Neutralization of TcdA Activity by Purified Mabs

[0460] All neutralisation screening assays were run in 96 well polystyrene plates. The assay uses CACO-2 cells grown, and screened in MEM+20% FCS, 2 mM Q, and NEAA. Any antibody combinations are at equal molar ratios unless stated otherwise. Day 1: Cells were plated @ 3000 per well in 50 l media, and incubated for 24 hrs; Day 2: Purified samples of humanised Mab were added to 96 well round bottomed polypropylene sterile plates; Spike PP plates with toxin A at a concentration sufficient to generated the appropriate lethal dose i.e. LD.sub.80 or above and incubate for 1 hr, at 37 C.; Add 50 l of this mixture to cell plates and incubate for 96 hrs; Day 5: Add Methylene blue (0.5% Methylene Blue 50% ethanol); Incubate for 1 hr at room temperature; Lyse the cells with 1% N-Lauryl Sarcosine, and Read on the BIOTEK Synergy2 plate reader at 405 nm. The results are shown in FIGS. 12 to 24. EC.sub.50 and % maximum neutralization of TcdA activity shown confirm that the selected antibodies have very high potencies as single agents. Combinations of 2 to 5 of these did not improve upon the best EC.sub.50 or % maximum neutralization. Lack of any synergy when combining Mabs CA922, 923, 995, 997 and 1000 is an important observation and may be due to the fact the each antibody alone has exceptionally high levels of affinity and potency. Supporting data in Example 5 also show that some of the Mabs (e.g. CA997) are capable of binding to TcdA subdomains many times. Hence it seems probable that these 5 Mabs represent that the maximum affinity, potency and valency that is achievable when targeting the C-terminal cell binding domain of TcdA. The antibodies were also effective at neutralising very high toxin concentrations ranging from LD80 to greater than LD.sub.95 (LD.sub.max) but some modest increases in EC.sub.50 (i.e. decreases in potency) were observed with very high levels of [TcdA]. These data are also surprising since others have shown substantial reductions in potency when testing elevated TcdA concentrations (20).

[0461] The high potency and affinity of the Mabs described here, e.g. for CA997; is not due solely to their high valency of binding. Others (20 and WO06/071877) describe anti-TcdA Mabs capable of binding up to 14 times. These Mabs only had affinities in the range 0.3 to 100 nM and showed incomplete protection against TcdA mediated cell killing, alone (26-63% protection) or in pairs (31-73% protection). Hence it has been demon-strated that high valency of binding to TcdA does not necessarily invoke either high affinity of binding to or neutralisation of TcdA. Neither the affinities nor valency of binding to TcdA were described for Mab CDA-1 (18 and U.S. Pat. No. 7,625,559). Thus Mabs described herein to have surprising affinity, potency and valency.

TABLE-US-00056 TABLE 4 Anti TcdA 1, 2 & 3 Mab combinations at a single TcdA conc. (LD.sub.80) Final (highest) EC.sub.50 Antibody Mab conc.ng/ml (ng/ml) 922 500 1.21 923 500 160.42 995 500 37.64 997 500 6.25 1000 500 19.73 922 + 923 500 3.58 922 + 925 500 3.326 922 + 997 500 2.88 922 + 1000 500 2.64 923 + 995 500 60.23 923 + 997 500 7.54 923 + 1000 500 9.24 995 + 997 500 7.29 995 + 1000 500 19.63 997 + 1000 500 4.46 922 + 923 + 995 500 4.72 922 + 923 + 997 500 3.23 922 + 923 + 1000 500 3.21 922 + 995 + 997 500 2.22 922 + 995 + 1000 500 2.85 922 + 997 + 1000 500 2.22 923 + 995 + 997 500 5.04 923 + 995 + 1000 500 9.49 995 + 997 + 1000 500 5.84 922 + 923 + 995 + 997 500 2.75 922 + 923 + 995 + 1000 500 3.75 922 + 995 + 997 + 1000 500 3.46 923 + 995 + 997 + 1000 500 4.81 922 + 923 + 997 + 1000 500 3.06 922 + 923 + 995 + 997 + 1000 500 4.72

TABLE-US-00057 TABLE 5 Anti TcdA single, paired, and triplet Mab combinations at various TcdA concentrations, where TcdA is at its LD.sub.80, LD.sub.90, LD.sub.95 and LD.sub.max. Final Mab EC.sub.50 Toxin TcdA Sample conc.ng/ml (ng/ml) @ 3000 pg/ml 922 500 4.89 (LD.sub.MAX) 997 500 10.99 1000 500 50.17 922 + 997 500 7.18 922 + 1000 500 6.99 997 + 1000 500 9.437 922 + 997 + 1000 500 10.80 922 + 997 + 1000 + 995 500 15.03 922 + 997 + 1000 + 995 + 923 500 7.16 @ 1000 pg/ml 922 500 1.24 (LD.sub.95) 997 500 3.42 1000 500 9.60 922 + 997 500 1.85 922 + 1000 500 2.51 997 + 1000 500 3.61 922 + 997 + 1000 500 2.40 922 + 997 + 1000 + 995 500 2.74 922 + 997 + 1000 + 995 + 923 500 2.38 @ 700 pg/ml 922 500 0.84 (LD.sub.90) 997 500 2.40 1000 500 6.23 922 + 997 500 1.19 922 + 1000 500 1.33 997 + 1000 500 2.68 922 + 997 + 1000 500 1.84 922 + 997 + 1000 + 995 500 2.17 922 + 997 + 1000 + 995 + 923 500 2.06 @ 350 pg/ml 922 500 0.39 (LD.sub.80) 997 500 1.18 1000 500 2.76 922 + 997 500 0.67 922 + 1000 500 0.85 997 + 1000 500 2.06 922 + 997 + 1000 500 0.83 922 + 997 + 1000 + 995 500 0.97 922 + 997 + 1000 + 995 + 923 500 0.98

Example 2 Anti TcdB In-Vitro Neutralization by Purified Mab

Assay Methods Description:

[0462] All neutralisation screening assays were run in 96 well polystyrene plates.

[0463] The assay uses CACO-2 cells grown, and screened in MEM+20% FCS, 2 mM Q, and NEAA. Unless stated all Ab combinations are in equal ratios. [0464] Day 1: Cells are plated @ 3000 per well in 50 l media, and incubated for 24 hrs [0465] Day 2: Purified samples of humanised Mab were added to 96 well round bottomed polypropylene sterile plates [0466] Spike PP plates with toxin B lot #031 and incubate for 1 hr, at 37 C. [0467] Add 50 l of this mixture to cell plates [0468] Incubate for 96 hrs [0469] Day 5: Add Methylene blue (0.5% Methylene Blue 50% ethanol) [0470] Incubate for 1 hr at room temperature [0471] Lyse the cells with 1% N-Lauryl Sarcosine [0472] Read on the BIOTEK Synergy2 plate reader at 405 nm

[0473] The data in FIGS. 25 to 33 show that single Mabs alone were relatively ineffective at neutralizing TcdB, both in terms of % maximum neutralization and activity (EC.sub.50). However, when the antibodies were combined in two's and three's considerable improvements in both % maximum neutralization and activity (EC.sub.50) were observed. 1125 and 1151 were selected as a best pairing, although other good pairings were observed: 1125+1153, 1125+1134.

[0474] The most effective pairs of Mabs were selected empirically and were found retrospectively to make unexpected and surprising combinations when regarding the individual potencies of each Mab. For example, in Table 6 only CA927 had a TcdB neutralisation potential which could result in a defined EC.sub.50 whilst the TcdB neutralisation potential of both CA1125 and CA1151 were insufficient under these assay conditions to result in a defined EC.sub.50. However, CA927 was not found to be the most effective Mab to use within a combination. The best CA927 containing combination had an EC.sub.50 of 13.5 ng/ml whereas other two Mab combinations had EC.sub.50's as low as 2.59 and 4.71 ng/ml. In another example, in Table 8 CA1099 had the lowest TcdB neutralisation EC.sub.50 under the assay conditions used. However, CA1099 was not found to be the most effective Mab to use within a combination. The best CA1099 containing combination had an EC.sub.50 of Eng/ml whereas other two Mab combinations had EC.sub.50's as low as 2 and 1 ng/ml. We might speculate that the most effective pairings of Mabs are defined by their cooperative binding modalities especially as defined by having non-overlapping epitopes.

TABLE-US-00058 TABLE 6 Anti-TcdB Mab combinations and relative Mab ratios at constant toxin concentration. Final Mab Sample conc.ng/ml EC.sub.50(ng/ml) 1125.g2 1000 >1000 1134.g5 1000 >1000 927.g2 1000 12.89 1153.g8 1000 >1000 1102.g4 1000 >1000 927 + 1099 1000 >1000 927 + 1102 1000 >1000 927 + 1114 1000 >111.111 927 + 1125 1000 13.55 927 + 1134 1000 51.58 1099 + 1114 1000 >1000 1102 + 1114 1000 >333.333 1102 + 1125 1000 15.51 1114 + 1134 1000 19.70 1114 + 1151 1000 25.69 1114 + 1153 1000 27.48 1125 + 1134 1000 2.59 1125 + 1151 1000 4.71 1125 + 1153 1000 21.23 1125 + 1134 + 1114 1000 3.77 1125 + 1134 + 927 1000 2.63 1125 + 1151 + 1114 1000 4.90 1125 + 1151 + 927 1000 5.69 1125.g2 + 1134.g5 + 927.g2 1000 5.83 1125.g2 + 1134.g5 + 1153.g8 1000 9.89 1125.g2 + 1134.g5 + 1102.g4 1000 2.72

Example 3 Neutralisation of TcdB by Combinations of Purified Mab

[0475] All neutralisation screening assays were run in 96 well polystyrene plates.

[0476] The assay uses CACO-2 cells grown, and screened in MEM+20% FCS, 2 mM Q, and NEAA. [0477] Day 1: Cells are plated @ 3000 per well in 50 l media, and incubated for 24 hrs [0478] Day 2: Purified samples of humanised Mab were added to 96 well round bottomed polypropylene sterile plates [0479] Spike PP plates with toxin B (VPI 10463) and incubate for 1 hr, at 37 C. [0480] Add 50 l of this mixture to cell plates [0481] Incubate for 72 hrs [0482] Day 5: Add Methylene blue (0.5% Methylene Blue 50% ETOH) [0483] Incubate for 1 hr at room temperature [0484] Lyse the cells with 1% N-Lauryl Sarcosine [0485] Read on the BIOTEK Synergy2 plate reader at 405 nm

[0486] The results are shown in FIGS. 34 to 45.

[0487] These data show that the best pair of Mabs for neutralizing TcdB at a range of toxin concentrations was CA1125 and CA1151. Moreover, the 1125+1151 combination was largely unaffected by changes in the relative molar ratios which is in contrast to 1125+1153.

TABLE-US-00059 TABLE 7 Anti-TcdB Mab combinations and relative Mab ratios at 3 different toxin cones. EC50 values (ng/ml) Antibody combination TcdB LD60 TcdB LD77 TcdB LD85 1125.g2 + 927.g2 (50:50) 2.8 6 11.3 1125.g2 + 1102.g4 (50:50) 4 13 44 1125.g2 + 1114.g8 (50:50) 3.5 7.1 25.4 1125.g2 + 1134.g5 (50:50) 0.48 1.4 4 1125.g2 + 1151.g4 (50:50) 0.85 0.85 1.5 1125.g2 + 1153.g8 (50:50) 2.7 5.2 25.2 1125.g2 + 1134.g5 (25:75) <0.15 0.84 7.2 1125.g2 + 1151.g4 (25:75) 0.73 1 2.1 1125.g2 + 1153.g8 (25:75) 7 10 27 1125.g2 + 1134.g5 (75:25) 0.66 1.2 2.5 1125.g2 + 1151.g4 (75:25) 1.4 1.2 8.3 1125.g2 + 1153.g8 (75:25) 2.9 7.5 30

[0488] The data show that even the most active specific paired combinations have surprisingly and non-predictably different properties relative to each other. The EC.sub.50 of the preferred combination of CA1125 and CA1151 in equimolar ratios is largely unaffected by an increasing [TcdB]. The three relative molar ratios of Mabs tested (i.e. 25:75 vs 50:50 vs 75:25) have very similar EC.sub.50's to each other, suggesting that CA1125 and CA1151 have especially complementary modes of action. This is in contrast to the combination of CA1125 with CA1134 where the increase in EC.sub.50 (i.e. reduction of potency) with higher [TcdB] is more substantial and where the three Mab molar ratios are not equally effective: The CA1125:CA1134 ratio of 25:75 is notably less potent than 50:50 and 75:25. This suggests that the combined potency of CA1125+CA1134 is more dependent upon the CA1125 component. The EC.sub.50 of all three molar combinations of CA1125 and CA1153 is substantially affected by increasing [TcdB] suggesting that CA1153 is a less suitable partner for combination with CA1125. In toto, these data show that CA1125 and CA1151 are a particularly favourable combination since the highest potency is maintained across a range of Mab and TcdB molar ratios.

TABLE-US-00060 TABLE 8 TcdB neutralisation - 1 or 2 anti-TcdB Mabs at constant toxin dose (LD.sub.80). Antibody IC50 (ng/ml) 1099 2 1102 N/A 1114 103 1125 N/A 1134 8 1151 182 1153 260 926 N/A 927 N/A 1099 + 1125 6 1114 + 1125 7 1151 + 1125 2 1134 + 1125 1 1102 + 1125 6 1125 + 1153 12 926 + 1125 42 927 + 1125 4

TABLE-US-00061 TABLE 9 TcdB neutralisation - 1 or 2 anti-TcdB Mabs at various TcdB doses. EC50 values (ng/ml) Maximum neutralisation Antibody combination TcdB LD75 TcdB LD86 TcdB LD90 TcdB LD75 TcdB LD86 TcdB LD90 1125.g2 n/a n/a n/a 40% 25% 15% 1114.g8 n/a n/a n/a 45% 25% 15% 1134.g5 n/a n/a n/a 45% 25% 15% 1151.g4 n/a n/a n/a 45% 25% 20% 1153.g8 28.3 n/a n/a 65% 35% 28% 1125.g2 + 1114.g8 (50:50) 10.1 243.8 n/a 85% 65% 40% 1125.g2 + 1134.g5 (50:50) 1.7 22.6 n/a 87% 60% 40% 1125.g2 + 1153.g8 (50:50) 6.1 32.2 n/a 95% 75% 48% 1125.g2 + 1151.g4 (50:50) 0.8 2.8 19.1 85% 80% 55% 1125.g2 + 1151.g4 (25:75) 1.2 2.8 47.2 85% 75% 60% 1125.g2 + 1151.g4 (75:25) 2.9 3.8 2.6 75% 70% 60%

[0489] These data show that combination of Mabs, especially CA1125 and CA1151 improve both the potency as measured by EC.sub.50 but also as measured by % maximum protection. The % maximum protection is of particular relevance in this assay method since the Mab:TcdB mixture is incubated with cells for a long time (72 h). Since TcdB is toxic to Caco-2 cells in the range of pg/ml in 2-4 h this measure may be considered to be a very difficult test of Mab neutralisation ability and may reflect the ability of Mab mixture with regard to their binding kinetics or modalities. In turn this may reflect the ability of Mab mixtures to protect against the effects of TcdB during an established infection when there may be substantial quantities of TcdB within tissues for many hours. Selected data from Tables 6-9 are further illustrated in FIGS. 46-59.

Example 4 Valency of Binding of Mabs to TcdB Sub-Domains

[0490] The number of moles of binding events of anti-C. difficile TcdB antibodies to TcdB.sub.1234 was determined by Surface Plasmon Resonance (SPR) on a Biacore 3000 (GE Healthcare). Streptavidin was immobilized on a CM5 sensor chip (GE Healthcare) to a level of 4000RU via amine coupling and biotinylated TcdB.sub.1234 was bound at 500-600RU. Two 20 l injections of the same anti-TcdB antibody mixtures (final concentration of each antibody was 500 nM) were injected over this surface at 10 l/min and the saturating binding response recorded. The surface was regenerated after every cycle using HCl. All the data was corrected for background binding using the response to the streptavidin only reference flowcell.

TABLE-US-00062 TABLE 10 Surface plasmon resonance analysis of the number of IgG binding sites on TcdB.sub.1234 No. of Binding Binding relative Antibody binding Response to CA927 combination cycle repeats (RU) average response CA1125.g2 10 750 0.9 CA1151.g4 10 1232 1.6 CA1125_CA1151 4 1941 2.5 CA1125_CA927 3 1570 2.0 CA1151_CA927 3 1959 2.5 CA927 8 791 1.0

[0491] All responses have been expressed relative to a multiple of CA927 average response (final column table 10) since CA927 appears to be representative of a Mab which binds to TcdB.sub.1234 once only.

[0492] Immobilized CA1125, when bound to TcdB.sub.1234, does not allow CA1125 to bind further supporting the idea that CA1125 has one binding site on TcdB.sub.1234 and that after this has been saturated that no other binding site for CA1125 can be found. However, when TcdB.sub.1234 has been saturated by CA1125, CA1151 can still bind. This demonstrates that CA1151 binds at alternative sites to that occupied by CA1125. Together these data show that CA1125 is a single binder of TcdB.sub.1234 whereas 1151 IgG binds to TcdB.sub.1234 more than once, most likely twice. Hence a mixture of CA1125 and CA1151 can bind to TcdB.sub.1234 approximately 3 times.

[0493] All antibodies combinations have an additive binding response showing that there are 2 or more non-competitive sites on TcdB.sub.1234 bound by these combinations.

Example 5 Valency of Binding of Mabs to TcdA Sub-Domains

[0494] The number of moles of binding events of anti-C. difficile TcdA antibodies to TcdA.sub.123 and A.sub.456 were determined by Surface Plasmon Resonance (SPR) on a Biacore 3000 (GE Healthcare). Streptavidin was immobilized on a CM5 sensor chip (GE Healthcare) via amine coupling to a level of 4000RU and biotinylated TcdA.sub.123 was bound to one flowcell and TcdA.sub.456 was bound to a different flowcell to a response of 500RU. Two 30 l injections of the same anti-TcdA antibody at 1 M were injected over both flowcells at 10 l/min and the saturating binding response recorded. The surface was regenerated after every cycle using HCl. All the data was corrected for background binding using the response to the streptavidin only reference flowcell.

TABLE-US-00063 TABLE 11 SPR analysis of the binding responses of IgGs to immobilised TcdA.sub.123 and TcdA.sub.456 CA997 CA1000 CA997/CA1000 ratio TcdA123 1069 166 6 TcdA456 1285 407 3

[0495] Antibodies CA997 and CA1000 bind to TcdA.sub.123 in a ratio of six CA997's to one CA1000 whereas they bind to TcdA.sub.456 in a ratio of three CA997's to one CA1000 (Table 2).

[0496] The maximum antibody response for CA997, corrected for molecular weight and immobilized toxin level is similar for TcdA.sub.123 and TcdA.sub.456. This suggests that CA997 binds TcdA.sub.456 six times and CA1000 binds twice to TcdA.sub.456. Hence antibody CA997 likely binds to TcdA whole toxin (TcdA) approximately 12 times.

[0497] Overall CA997 binds six times or more to A.sub.123 and six times or more to A.sub.456, whereas CA1000 binds at least once to A.sub.123 and twice to A.sub.456.

[0498] Increased valency of binding to TcdA and TcdB may have two important effects in vivo. The first is that any Mab or Mab mixture which is capable of binding TcdB more than once will have increased potential to form inter-toxin binding events and hence immunoprecipitation. Immunoprecipitation can contribute to potency by reducing the solubility of toxin and forming very large macromolecular complexes which hence reduce the effective working concentration of toxin. Such large protein complexes may be taken up by macrophages and monocytes resident in the tissue and may contribute to an augmented host immune response. Antigen:antibody complexes bearing Fc fragments have been specifically shown to be capable of priming a host immune response against a gut pathogen (21). Also, soluble antigen:antibody complexes have been successfully used as a vaccine directed against the antigen in human clinical trials (22). In addition, immune decoration of toxin with Fc bearing IgG may contribute to immune clearance using normal mechanisms through the liver and spleen. In general, higher levels of Fc decoration of antigen lead to faster and more complete levels of clearance (23) Critically, it may be that presence of 2 or more Mab Fc domains per toxin, especially 3 Fc domains per toxin may represent a critical number of Fcs required for very rapid and substantial clearance of toxin (24). The anti-TcdA Mab CA997 is likely capable of binding to TcdA up to 12 times and the combination of CA1125 and CA1151 is likely capable of binding to TcdB 3 times. Hence the 3 Mab mixture is very likely to be capable of providing for these kinds of additional potency mechanisms in vivo.

Example 6 Mab Neutralisation of Loss of TEER Caused by TcdA

[0499] C. difficile monolayer integrity assay is performed using the Becton-Dickinson (BD) Caco-2 BioCoat HTS plate system.

[0500] Day 1Caco-2 cells seeded @ 210.sup.5/ml per well of the plate insert in 500 l Basal seeding medium (provided by BD). 35 ml of Basal seeding medium added to the feeder tray. Cells incubated for 24 hours at 37 C. Day 2Basal seeding medium removed from inserts and feeder tray, and replaced with Entero-STIM differentiation medium (supplied by BD). 500 l added per well insert and 35 ml to the feeder tray. Cells incubate for a further 72 hrs at 37 C. Day 5Antibodies prepared at 2 concentration relative to that to be used in the assay well in a polypropylene plate and toxin A. Toxin A added to antibodies at a concentration of 125 ng/ml and plate incubated for 1 hr at 37 C. 1 ml of Caco-2 growth medium (MEM+20% FCS, 2 mM Q, NEAA) added to each well of a standard 24-well TC plate. BioCoat insert plate transferred to the 24-well TC plate. Entero-STIM medium removed from inserts and replaced with 400 l of toxin:Ab mixture. Loss of tight junctions between gut cells is the key early effect of TcdA on cell monolayers and gut tissue sections and is the primary cause of diarrhoea. Albumin and other serum proteins are lost into the gut lumen along with accompanying serum fluid. The loss of trans-epithelial electrical resistance in differentiated cultured cells which have formed a monolayer is a useful surrogate for the protection against the acute effects of TcdA. Three antibodies shown have good levels of protection against TEER loss, FIG. 62. It is notable and surprising that the abilities of these Mabs in TEER assays do not reflect those seen in toxin neutralization as measured in a cell proliferation assay. CA922 has the best performance in a cell proliferation assay (EC.sub.50=1.21 ng/ml) and yet this is considerably out-performed in the TEER assay by an antibody (CA1000) which has >10 lower potency in a cell proliferation assay (EC.sub.50=19.73 ng/ml). CA997 had the best performance in the TEER assay since it had both high levels of protection and maintained this at the lower Mab concs. CA997 had a substantial potential to neutralize TEER loss with maximal inhibition approaching 80% and an EC.sub.50 of approximately 80 ng/ml at 4 h. These observations are unexpected since the Mabs in question all had high affinities for TcdA domains (CA922 4 pM, CA997 132 pM, CA1000 73 pM). These data suggest that CA997 and CA1000 recognise epitopes important in TEER loss or neutralize TcdA by different mechanism to other Mabs. Furthermore, since CA1000 is estimated to bind to holotoxin twice (once in TcdA.sub.123 and once in TcdA.sub.456) CA1000 may define TEER critical epitopes within the TcdA cell binding regions which might have particular value for defining vaccine immunogens. Results are shown in FIG. 62.

Example 7 Affinity of Anti-C. difficile Toxin Antibodies for Sub-Domains of TcdA and TcdB: TcdA.SUB.123., TcdA.SUB.456 .and TcdB.SUB.1234

[0501] Kinetic constants for the interactions of anti-C. difficile TcdA and TcdB antibodies were determined by surface plasmon resonance conducted on a BIAcore 3000 using CM5 sensor chips. All experiments were performed at 25 C. Affinipure F(ab).sub.2 fragment goat anti-human IgG, Fc fragment specific (Jackson ImmunoResearch) was immobilised on a CM5 Sensor Chip (GE) via amine coupling chemistry to a capture level of 7000 response units (RUs). HBS-EP buffer (10 mM HEPES pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% Surfactant P20, Biacore AB) was used as the running buffer with a flow rate of 10 L/min. A 10 L injection of each antibody at 1 ug/ml or lower was used for capture by the immobilised anti-human IgG, Fc. TcdA123, TcdA456 or TcdB1234 were titrated over captured purified antibodies at doubling dilutions from 12.5 nM at a flow rate of 30 L/min. For antibodies present in culture supernatants, a single concentration of 12.5 nM of TcdA123 or TcdA456 and 50 nM of TcdB1234 was passed over the antibodies at 30 ul/min. Kinetics were calculated on n=2 The surface was regenerated at a flowrate of 10 uL/min by two 10 L injections of 40 mM HCl, and a 5 L injection of 5 mM NaOH. Double referenced background subtracted binding curves were analysed using the BIAevaluation software (version 3.2) following standard procedures. Kinetic parameters were determined from the fitting algorithm.

TABLE-US-00064 TABLE 12 Anti-TcdA Mab affinities and binding kinetics Antibody ID ka (1/Ms) kd (1/s) KD (M) KD (pM) Material/Assay TcdA123 CA164_00922.g1 1.09E+06 4.43E06 4.06E12 4.06 Purified Mab 5 point CA164_00923.g1 5.36E+05 3.47E05 6.47E11 64.7 titration CA164_00995.g1 No binding No binding CA164_00997.g1 7.84E+05 1.03E04 1.32E10 132 CA164_01000.g1 1.33E+05 9.78E06 7.33E11 73.3 CA164_00993.g1 9.00E+05 5.00E06 5.56E12 5.56 Supernatant 2x 1 point titration TcdA456 CA164_00922.g1 1.29E+06 3.33E06 2.59E12 2.59 CA164_00923.g1 6.16E+05 1.92E04 3.12E10 312 Purified Mab 5 point CA164_00995.g1 2.87E+05 3.42E05 1.19E10 119 titration CA164_00997.g1 9.21E+05 6.15E05 6.68E11 66.8 CA164_01000.g1 3.55E+05 2.98E05 8.41E11 84.1 CA164_00993.g1 1.25E+06 5.00E06 4.00E12 4.00 Supernatant 2x 1 point titration

TABLE-US-00065 TABLE 13 Anti-TcdB Mab affinities and binding kinetics Antibody ID ka (1/Ms) kd (1/s) KD (M) KD (pM) Material/Assay TcdB1234 CA164_1125.g2 2.64E+05 3.23E05 1.22E10 122 Purified Mab 3 point titration CA164_1151.g4 7.49E+05 4.13E04 5.51E10 551 Purified Mab 3 point titration CA164_926.g1 1.38E+05 7.12E05 5.16E10 516 Supernatant 2x 1 point titration CA164_927.g2 3.97E+05 3.61E05 9.11E11 91 Purified Mab 3 point titration CA164_1099.g2 5.24E+05 1.63E05 3.10E11 31 Purified Mab 3 point titration CA164_1102.g4 1.17E+05 3.78E04 3.25E09 3250 Supernatant 2x 1 point titration CA164_1114.g2 2.87E+05 1.97E03 6.87E09 6870 Supernatant 2x 1 point titration CA164_1114.g8 2.55E+05 1.85E03 7.25E09 7250 Supernatant 2x 1 point titration CA164_1129.g1 1.89E+05 2.30E04 1.22E09 1220 Supernatant 2x 1 point titration CA164_1134.g5 5.09E+05 2.45E05 4.81E11 48 Purified Mab 3 point titration CA164_1153.g8 1.43E+05 4.48E05 3.14E10 314 Purified Mab 3 point titration

[0502] The anti-TcdA affinities are particularly high compared to the published affinities of other Mabs. We demonstrate that affinities as low as 4 pM are achievable. The preferred CA997 has an affinity of 132 pM, CA1125 122 pM and CA115 551 pM. CA995 clearly shows that it does not bind to CROPs A.sub.123 and hence that demonstrates that the Mab shown here have properties which are different from each other in surprising and unexpected ways. CA922, 923, 997 and 1000 do bind at least once to CROPs A123 and A456. Hence these 4 Mabs confirming that each must bind to holotoxin at least twice. We have been unable to derive affinities for the binding of these Mabs to holotoxin due to technical constraints. However, given the high affinities and valencies demonstrated for the anti-TcdA Mabs it is possible to speculate that the functional affinities against holotoxin may be even stronger than those illustrated for binding to toxin sub-domains. The anti-TcdB Mabs also demonstrated strong affinities reaching as low as 31 pM. In particular CA1125, 1151, 927, 1099, 1134 and 1153 show affinities which surpass those demonstrated by others.

Example 8 Biophysical Characterisation of C. difficile Anti-Toxin Humanised IgG1 Molecules

Molecules Analysed

Anti-TcdA IgG1:

CA164_00922.g1

CA164_0923.g1

CA164_0995.g1

CA164_0997.g1

CA164_01000.g1

Anti-TcdB IgG1

CA164_01125.g1

CA164_01125.g2

CA164_01134.g4

CA164_01134.g5

CA164_01134.g6

CA164_01102.g1

CA164_01102.g4

CA164_01151.g4

[0503] Antibody combinations need to be made up of Mabs having high levels of stability in order to mitigate potential risks of aggregation during long term storage. Thermal stability (Tm) is used as one measure. Of special value to Mab mixtures is measuring their propensity to aggregate due to physical stress such as agitation or shaking. Aggregates are undesirable components of drug compositions since they may reduce storage life time and may pose a safety risk to patients at certain levels. The Tm data show that all 5 anti-TcdA Mabs have high Tm stability, whilst three (CA922, 923 and 997) have very high Tm's in the range of 79-81 C. Of the anti-TcdB Mabs tested all but two have very high Tm's. Of note is that CA997, CA1125 and CA1151 which were tested in the hamster infection study (Example 9) had very high Tm's (79.2 C., 79.3 C. and 80.8 C. respectively) which makes them suitable for use in a Mab mixture.

[0504] In the shaking aggregation assay, CA997 and 922 had the lowest propensity to aggregate of the 5 anti-TcdA Mabs. Similarly, CA115 and 1151 had the lowest aggregation propensities of the anti-TcdB Mabs. Hence the use of CA997, 1125 and 1151 as a Mab mixture may have special value since they are more likely to survive co-formulation and storage at high protein concentrations.

Estimation of Isoelectric Point (pI) by Capillary IEF

[0505] Samples were prepared by mixing the following: 30 ul Protein sample at 2 mg/ml, 0.35% Methylcellulose, 4% pH3-10 ampholytes (Pharmalyte), synthetic pI markers (4.65 and 9.77), 1 ul of each stock solution, and HPLC grade water to make up the final volume to 200 ul. The mixture was then analysed using iCE280 IEF analyzer (pre-focusing at 1500V for 1 min followed by focusing at 3000V for 6 mins). The calibrated electropherograms were then integrated using Empower software (from Waters)

Thermal Stability (Tm) Measured Via Thermofluor Assay.

[0506] This method uses Sypro orange fluorescent dye to monitor the unfolding process of protein domains. The dye binds to exposed hydrophobic regions that become exposed as a consequence of unfolding which results in a change to the emission spectrum.

[0507] The sample (5 ul at 1 mg/ml) is mixed with a 5 ul of a stock solution of Sypro orange (30) and the volume made up to 50 ul with PBS, pH 7.40.

[0508] 10 ul aliquots of this solution is applied towells in a 384 well plate (n=4).

[0509] The plate is placed in a 7900HT fast real-time PCR system containing a heating device for accurate temperature control. The temperature is increased from 20 C. to 99 C. (Ramp rate of 1.1 C./min). A CCD device simultaneously monitors the fluorescence changes in the wells. An algorithm is used to process intensity data and take into account multiple transitions.

Stressing of Samples by Agitation.

[0510] During manufacture antibody samples are subjected to mechanical stress generated by processes such as pumping and filtration. This may cause denaturation and consequently aggregation due to exposure of the protein to air-liquid interfaces and shear forces resulting in the ultimate loss of bioactivity. Stress by vortexing is a method to screen the robustness of the antibody samples for prediction of aggregation stability.

[0511] Both anti-TcdA and anti-TcdB IgG1 molecules were subjected to stress by agitation, by vortexing using an Eppendorf Thermomixer Comfort at 25 C., 1400 rpm. Sample size was 250 uL, (3 per sample) in a 1.5 mL conical Eppendorf-style capped tube (plastic), in PBS pH 7.4. Each sample was brought to a concentration of 1 mg/ml (using extinction coefficient calculated from sequence) and aggregation was monitored by absorbance at 340 nm and/or 595 nm, by use of a Varian Cary 50-Bio spectrophotometer, measured at intervals for up to 24 hours.

[0512] Results Table 14 provides a summary of the measured pI and Tm data for both anti-TcdA and anti-TcdB IgG1 molecules.

TABLE-US-00066 TABLE 14 Compilation of pI and Tm Data measured Tm(Fab) Anti-TcdA IgG1 pI in PBS Tm(CH2) CA164_00922.g1 8.8 81 69.2 CA164_0923.g1 9.2 79 69.3 CA164_0995.g1 8.5 71 no data* CA164_0997.g1 8.3 79.2 68.4 CA164_01000.g1 7.74 70.5 no data* Anti-TcdB IgG1 CA164_01125.g1 9.2 79.3 69.4 CA164_01125.g2 9.2 79.5 69.3 CA164_01134.g4 9.3 78.4 69.4 CA164_01134.g5 9.2 76.4 69.2 CA164_01134.g6 9.2 76.6 69.6 CA164_01102.g1 9.1 69 no data* CA164_01102.g4 9.1 69.1 no data* CA164_01151.g4 9.2 80.8 69.8 *denotes that it was not possible to discern the Fab and CH2 domains.

Measured pI

[0513] The measured pI of the molecules were high (except for CA164_01000.g1_P3) and away from the pH of formulation buffers such as PBS, pH 7.4 and 50 m sodium acetate/125 mM sodium chloride, pH 5. This may mean that buffers with pH's suitable for co-formulation of two or more Mabs can be selected.

Thermal Stability (Tm) Measured Via Thermofluor Assay

[0514] Since all of the molecules are IgG1, the Tm of the Fc domain (Tm(CH2)) are the same. The difference in thermal stability between the molecules can be determined by the Tm of the Fab domain (Tm(Fab)).

[0515] For the anti-TcdA molecules, the rank order (most stable first) was CA922997>923>995>1000 and for the anti-TcdB molecules (most stable first) was CA1151.g4>1125.g1,g4>1134.g4>1134.g51134.g6>1102.g1=1102.g4.

Stressing of Samples by Agitation It was possible to determine different aggregation stability between the different antibodies, FIG. 67 shows the effect of agitation via vortexing on different anti-TcdA IgG1 molecules in PBS, pH 7.4.

[0516] It was possible to determine a ranking order (most aggregation stable first): CA922997>923995>1000

[0517] FIG. 68 shows the effect of agitation via vortexing on different anti-TcdB molecules.

[0518] It was possible to rank the order of aggregation stability, such that the CA1125 grafts appeared more stable than the CA1134 molecules which were more stable than the CA1102 molecules.

[0519] A further study was performed to compare directly the aggregation stability of the anti-TcdB molecule (CA1151.g4) with the more stable molecule CA1125.g2 (see FIG. 2) and more aggregation stable anti-TcdA molecules (CA922.g1 and CA997.g1). The results can be seen in FIG. 69.

[0520] Further results for these 4 Mabs are also shown in FIGS. 67 and 68.

[0521] For the anti-TcdA molecules, CA922.g1 and CA977.g1, CA922 were preferable based on the analyses above, although apart from CA1000) all molecules could be considered suitable candidates for use as therapeutic IgG1.

[0522] For the anti-TcdB molecules, the biophysical characteristics could be grouped within the family of grafts based on the aggregation stability and Tm, such that the CA1125 grafts potentially proved more stable. The CA1102 grafts showed poorest Tm data and also showed the greatest tendency to aggregate via stress by agitation.

[0523] A study using CA1151.g4 showed that this molecule exhibited slightly increased aggregation stability relative to CA11125.g2 and seemed equivalent to the TcdA molecules (CA922.g1 and CA997.g1. All four molecules showed equivalent Tm values. CA997, CA1125 and CA1151 show very high levels of thermostability and very low levels of aggregate formation after agitation.

Example 9 Anti-C. difficile Toxin Mab Hamster Infection Study

[0524] The hamster infection study was performed by Ricerca Biosciences LLC, Cleveland, Ohio, USA. The study protocol was approved by the Ricerca IACUC committee. Active and control components (composition and dose) were blinded to Ricerca staff until after completion of the planned 28 day study period.

[0525] Golden Syrian male hamsters (weight 82-103 g, 54 days old) were individually housed in HEPA filtered disposable cages and fed Teklad Global Diet 2016 and water ad libitum. After acclimatisation, hamsters were pre-dosed (i.p.) with Mab mixtures or PBS (vehicle control) once a day for each of 4 days: days 3, 2, 1 and 0. Two doses of Mab were investigated: high dose=50 mg/kg each of anti-TcdA and anti-TcdB components and low dose 5 mg/kg each of anti-TcdA and anti-TcdB components.

[0526] The drug combination tested was composed of one anti-TcdA antibody (CA997.g1) which constituted 50% of the injected protein and two anti-TcdB antibodies (CA1125.g2 and CA1151.g4) which together constituted 50% of the injected protein but which alone constituted 25% of the injected protein. Hamsters were sensitised (day 1) with 50 mg/kg of Clindamycin phosphate in PBS (s.c.) before being challenged 1 day later (day 0) with 3.4106 c.f.u. of vegetative cells from strain ATCC43596. Vancomycin was dosed at 5 mg/kg twice a day for 5 days (p.o.) on days 1, 2, 3, 4, 5.

[0527] Viability checks were performed on animals twice a day, animals found to be in extremis were euthanised and counted as dead. Body weights were determined on each day of dosing, then twice weekly and before euthanising survivors. Gross necropsy was performed on all animals. Survival curves were created by the method of Kaplan and Meier. Survival curves were analysed using the P value from the log rank test compared to the Bonferroni corrected threshold of P=0.005. The Vancomycin treated group were not included in the analysis. All statistical tests were done with Prism v5.04. All groups contained 11 animals, except the Vancomycin control group which contained 5 animals.

[0528] Survival curves can be seen in FIG. 63. Hamsters receiving PBS (control) all died on days +2 and +3, whilst those receiving vancomycin treatment for 5 days all died on days +10 and +11. Hamsters receiving the high dose of UCB Mab mixture all survived until day +11, thereafter only two animals died until the end of the 28 day study. Hamsters receiving the low dose of UCB Mab mixture all survived until day +3, thereafter animals were lost fairly steadily until day +16 when all had died. The data show exceptional levels and duration of protection when compared to published data for use of anti-toxin Mabs in hamsters (18). These in vivo data support the in vitro observations of very high level performance for neutralization and stability.

[0529] There is no apparent link between death and body weight during the acute phase (days 1-5) of the infection, FIGS. 64-65. Hence it may be supposed that hamsters die of overwhelming direct and indirect effects of TcdA and TcdB. Hamsters which survive the acute period due to partial protection (UCB low dose) of neutralizing Mabs lose weight, presumably due to gut damage and altered nutritional status. It was notable that many of the hamsters which went on to survive the 28 period of the study due to the protective effects of the UCB high dose Mabs recovered from weight loss and indeed even gained weight. This may be taken as evidence of the superior protective effects of the UCB Mabs enabling the gut to function as normal.

TABLE-US-00067 TABLE 15 Gross pathology scores Anogenital Black Dark red Red Pink Normal staining Red small Group caecum caecum caecum caecum caecum wet-tail intestine PBS 1 9 1 0 0 1 1 control UCB low 0 4 5 2 0 4 1 UCB high 0 0 1 1 9 3 0

[0530] It is clear that UCB Mabs were able to protect the large and small intestines from the bloody effusions caused by TcdA and TcdB.

[0531] The results are shown in FIGS. 63 to 66

[0532] The photographs in FIG. 66 show typical gross pathologies for the swelling and bloody effusions of caeca caused by TcdA and TcdB (left image, PBS control, animal death on day 2) and a normal stool filled caeca after protection by UCB high dose Mabs (right image, UCB high dose, animal surviving to day 28). These data show that after protection with a high dose of UCB Mabs the large intestine can return to normal morphology and function.

Example 10 Neutralisation of TcdA from Different Ribotyped Strains by Purified Mab

[0533] Clinical infections are caused by a variety of different strains. Strain differences are characterized using a number of different methods of which ribotyping is a key one. Different ribotype strains are observed to have different pathogenicity, infection and sporulation properties. All of the TcdA neutralization shown above used TcdA purified from strain known as VPI10463. However, the predominant aggressively pathogenic strain associated with out-breaks is called ribotype 027. Other key ribotypes include 078, 001, 106. Amino acid sequence difference have been observed between toxins produced by different ribotypes and hence it is important that Mabs are capable of neutralizing toxin from a diverse set of clinical isolates. CA922, 997 and 1000 were tested for their ability to neutralize TcdA from strains 027 and 078 and compared to their abilities against TcdA from VPI10463. Mabs were tested at 4 [TcdA] and found to be capable of neutralizing all toxins without significant difference at LD.sub.80, LD.sub.90 and LD.sub.95

TABLE-US-00068 TABLE 16 EC50 values (ng/ml) - TcdA strain VPI 10463 Antibody LD80 LD90 LD95 LDmax CA164_922 0.27 0.9 1.2 >500 CA164_997 1 2.5 3.5 25.4 CA164_1000 3.6 13.5 19.3 >500

TABLE-US-00069 TABLE 17 EC50 values (ng/ml) - TcdA ribotype 027 Antibody LD80 LD90 LD95 LDmax CA164_922 0.19 0.25 0.41 1.46 CA164_997 0.92 1.27 1.75 7.19 CA164_1000 2.25 2.49 3.52 16.32

TABLE-US-00070 TABLE 18 EC50 values (ng/ml) - TcdA ribotype 078 Antibody LD80 LD90 LD95 LDmax CA164_922 0.11 0.12 0.25 0.68 CA164_997 0.33 0.64 1.11 2.57 CA164_1000 2.04 2.41 5.03 14.16

Example 11 PK Data

[0534] A PK study of a human IgG1 (20 mg/kg) in healthy hamsters. The hamster PK was found a half-life similar to Mabs in mice or rats. (t 6-8 days). i.p. and s.c. dosing were essentially the same. The pharmacokinetics and distribution to the gut of a hIgG1 Mab was studied in normal (non-infected) golden Syrian hamsters. Purified Mab was administered to male hamsters (120-135 g) by CARE Research LLC, Fort Collins, Colo., USA and samples were assayed by UCB Pharma. The study was approved by the CARE IACUC committee. Eight animals each received a single dose of 20 mg/kg of IgG1, four were dosed i.p., four were dosed s.c. Blood was collected at 1, 3, 8, 24, 48, 72, 103 and 168 hours post-dose, serum was separated before storage at 80 C. Blood was also taken from two untreated hamsters in order to provide assay controls. Following euthanasia, a 2 cm length of colon was cut from the caeca junction onwards from each hamster. The colon section was flushed with wash buffer (50% (v/v) PBS containing 50% (v/v) Sigma protease inhibitor cocktail (P2714) before being opened and separation and removal of the mucosa from the underlying muscle. Mucosal samples were placed in 0.5 ml of wash buffer homogenized until visually uniform and stored at 4 C. before immediate shipping on wet ice. For the anti-human IgG1 ELISA Nunc maxisorp 96 well plates were coated overnight in 0.1M NaHCO.sub.3 pH 8.3 with Goat F(ab)2 anti-human IgG-Fc fragment (Jackson 109-006-098), plates were washed with PBS-Tween (PBS/0.1% (v/v) Tween 20) and then blocked with 1.0% (w/v) BSA & 0.1% (v/v) Tween in PBS. Serum samples were diluted in sample-conjugate buffer (1% (w/v) BSA, 0.2% Tween in PBS) and after washing were revealed with goat anti-human kappa-HRP (Cambridge Bioscience 2060-05) in sample-conjugate buffer and TMB with a 2.5M H2504 stop solution.

Gut, Mucosa and Serum Levels:

[0535] Serum samples collected from blood taken at 168 hour time point and colon samples were removed after this.

TABLE-US-00071 20mg/kg IP at 168 hour Sample ng/mL per cm mucosa serum g/mL 1001 23.2 75.0 1002 13.7 90.8 1003 21.8 70.5 1004 53.8 119.4

TABLE-US-00072 20mg/kg SC at 168 hour Sample ng/mL per cm mucosa serum g/mL 2001 41.4 108.7 2002 62.1 76.6 2003 35.6 163.7 2004 37.3 153.3

TABLE-US-00073 Serum Data Hamster i.p. Hamster s.c. SE of SE of Mean mean Mean mean C.sub.max: g/mL 202 12 186 21 T.sub.max: hr 36 7 76 16 AUC .sub.(last): hr .Math. g/mL 22626 1378 22371 2258 AUC .sub.(inf): hr .Math. g/mL 43287 7169 61290 17637 % Extrapolation: 43.7 9.2 54 11.7 CL/F mL/hr/kg 0.50 0.07 0.43 0.13 MRT.sub.inf h 223 53 310 88 t.sub.1/2,z: h 149.2 36.9 188.5 61.9

[0536] The data is also shown in FIGS. 70 and 71

TABLE-US-00074 Hamster ID Mean SE IP serum kinetics C.sub.max: g/mL 202 12 T.sub.max: hr 36 7 AUC.sub.(last): hr .Math. g/mL 22626 1378 AUC.sub.(inf): hr .Math. g/mL 43287 7169 % Extrapolation: 43.7 9.2 CL/F mL/hr/kg 0.50 0.07 MRT.sub.inf h 223 53 t.sub.1/2,z: h 149.2 36.9 SC serum kinetics Hamster ID Mean SE C.sub.max: g/mL 186 21 T.sub.max: hr 76 16 AUC.sub.(last): hr .Math. g/mL 22371 2258 AUC.sub.(inf): hr .Math. g/mL 61290 17637 % Extrapolation: 54 11.7 CL/F mL/hr/kg 0.43 0.13 MRT.sub.inf h 310 88 t.sub.1/2: h 188.5 61.9

[0537] It was also shown that hIgG1 could be found in scrapings of the gut i.e that hIgG1 gets into the vasculature of healthy gutand so could be protective in prophylactic dosing. This effect would be even more profound in humans since they have a cognate hFcRn.

Example 12 Serum Levels in Hamsters with C. difficile Infection

[0538] This study was to determine the serum concentration of CA725.0, CA726.0, CA997.g1 CA1125.g2, and CA01151.g4 following i.p. administration (various doses detailed below) in the Golden Syrian Hamster.

[0539] Humanised Mabs were quantified using liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis following tryptic digestion. Quantitation was achieved by comparison to authentic standard material spiked at known concentrations into blank matrix, with spiked horse myoglobin used as the internal standard.

[0540] A unique (proteotypic) peptide common to all of the humanised Mabs investigated was selected (DTLMISR, a CH2 region peptide) and both samples and calibration samples were tryptically digested as outlined. Tryptic digest of 5 l serum samples was performed overnight using sequencing grade modified Trypsin (Promega, Southampton, UK) following denaturation/reduction with acetonitrile/Tris (2-carboxyethyl) phosphine and carbamido-methylation with iodoacetamide (Sigma-Aldrich, Poole, UK).

[0541] The LC-MS/MS system consisted of a CTC HTS-x Autosampler (CTC Analytics, Zwingen, Switzerland), a Agilent 1290 LC system (Agilent Technologies, Stockport, UK) and a Sciex 5500 QTrap MS system (AB Sciex, Warrington, UK), equipped with a Turbo V ion source operated in electrospray mode. Analytes were separated using an Onyx Monolithic C18 column (1004.6 mm, Phenomenex, Macclesfield, UK) with a gradient of 2 to 95% (v/v) water/acetonitrile (0.1% formic acid) delivered at 1.5 mL/min over 6 minutes. The injection volume was 10 L; all of the eluent was introduced into the mass spectrometer source. The source temperature of the mass spectrometer was maintained at 600 C. and other source parameters (e.g. collision energy, declustering potential, curtain gas pressure etc.) were optimized to achieve maximum sensitivity for the peptide of interest. Selective transitions for each proteotypic peptide of interest were monitored. Unique (proteotypic) peptides were selected for all of the analytes of interest; samples were analysed following tryptic digestion.

[0542] Plasma concentrations calculated based on the peptides monitored are outlined below.

[0543] For CA164_00997 and CA164_01151, interfering peaks were observed in the MRM traces. For this reason, these two analytes could not be quantified in the samples.

[0544] Total h-IgG was quantified in all samples using a peptide common to all analytes of interest. This was done using a combined standard curve of all five analytes. The validity of this approach is demonstrated by the fact that the sum of the concentrations observed for CA164_00725 and CA164_00726 are in good agreement (within experimental error) of the concentration observed for total h-IgG.

[0545] Using this approach, the total concentration of h-IgG in the samples of animals dosed with CA164_00997, CA164_01125 and CA164_01151 was determined.

[0546] Overall the data obtained indicate that the exposure of all five analytes of interest was similar for a given dose.

TABLE-US-00075 Study groups Blinded labels Treatment components Grp Treatment Actual Treatments Dose days Anti-toxin A Anti-toxin B 4 Treatment3 Vehicle PBS 5 mL/kg i.p. 3, 2, 1, 0 2 Vancomycin Vancomycin 5 1, 2, 3, 4, 5 mg/kg b.i.d. p.o. 1 Treatment1 UCB LD* 3, 2, 1, 0 CA997.g1_P3 CA1125.g2_P3 CA1151.g4_P3 5 mg/kg A 5 mg/kg i.p. 5 mg/kg 2.5 mg/kg 2.5 mg/kg 5 Treatment4 UCB HD* 3, 2, 1, 0 CA997.g1_P3 CA1125.g2_P3 CA1151.g4_P3 50 mg/kg A 50 mg/kg i.p. 50 mg/kg 25 mg/kg 25 mg/kg 6 Treatment5 CompetitorLD* 3, 2, 1, 0 CA726_P3 CA725_P3 5 mg/kg A 5 mg/kg i.p. 5 mg/kg 5 mg/kg 3 Treatment2 CompetitorHD* 3, 2, 1, 0 CA726_P3 CA725_P3 50 mg/kg A 50 mg/kg i.p. 50 mg/kg 50 mg/kg

TABLE-US-00076 TABLE 19 Group/ Animal Serum conc time Day No Dose g/mLtotal h-IgG 1 1 44 5 mg/kg 997, 280 1 45 2.5 mg/kg 302 1 46 1125,2.5 182 6 45 mg/kg 1151 61 6 47 71 6 49 45 3 1 60 50 mg/kg 3040 1 61 725, 3330 1 62 50 mg/kg 2990 6 62 726 583 6 63 913 6 64 1240 28 64 199 28 65 36 4 1 71 Vehicle nd 1 72 nd 1 73 nd 5 1 82 50 mg/kg 3050 1 83 997, 2790 1 84 25 mg/kg 2370 6 82 1125,25 838 6 83 mg/kg 1151 645 6 84 855 28 82 116 28 83 65 28 84 66 28 85 44 28 86 101 28 87 89 28 88 27 28 89 31 28 90 66 6 1 93 5 mg/kg 725, 335 1 94 5 mg/kg 726 322 1 95 260 6 200 103 6 202 62 6 203 79 28 203 nd nd - not detected (LOQ = 2.5 g/mL for all analytes na - not analysed: interference in the sample was observed for 997 and 1151

TABLE-US-00077 TABLE 20 Antibody CA725 is prior art antibody MDX1388. Antibody CA726 is prior art antibody CDA1 as described (15) A summary of this data is presented in FIG. 72. Small intestine Caecal pathology pathology Dark Dark Group Black Red Red Pink Normal Red Red PBS control 1 9 1 0 0 0 1 MDX high 0 1 4 4 2 1 0 50 mg/Kg x4 UCB high 0 0 1 1 9 0 0 50 mg/Kg x4

REFERENCES

[0547] 1. Kuehne, S et al., The role of toxin A and toxin B in Clostridium difficile Infection Nature (2010) 467: 711-713. [0548] 2. Davies A H et al., Super toxins from a super bug: structure and function of Clostridium difficile toxins Biochem. J (2011) 436: 517-526. [0549] 3. Rothman, S et al., Differential Cytotoxic Effects of Toxins A and B Isolated from Clostridium difficile Infect. Imm. (1984) 46: 324-331. [0550] 4. Du, T and Alfa, M J Translocation of Clostridium difficile toxin B across polarized Caco-2 cell monolayers is enhanced by toxin A Can J Infect Dis. (2004) 15: 83-88. [0551] 5. Kim, Iaconis and Rolfe. Immunization of Adult Hamsters against Clostridium difficile-Associated Ileocecitis and Transfer of Protection to Infant Hamsters Infect. Imm. (1987) 55:2984-2992 [0552] 6. Rupnik J C M (2003) 41:1118-1125 [0553] 7. Chaves-Olarte J B C (1999) 274:11046-11052. [0554] 8. Lylerly, D M et al., Passive Immunization of Hamsters against Disease Caused by Clostridium difficile by Use of Bovine Immunoglobulin G Concentrate Infection and Immunity (1991) 59:2215-2218. [0555] 9. Lylerly, D M et al., Vaccination against Lethal Clostridium difficile Enterocolitis with a Nontoxic Recombinant Peptide of Toxin A Current Microbiology (1990) 21:29-32. [0556] 10. Lylerly, D M et al., Characterization of Toxins A and B of Clostridium difficile with Monoclonal Antibodies Infect. Imm. (1986) 54:70-76. [0557] 11. Corthier et al., Protection against Experimental Pseudomembranous Colitis in Gnotobiotic Mice by Use of Monoclonal Antibodies against Clostridium difficile Toxin A Infect. Imm. (1991) 59: 1192-1195. [0558] 12. Kink J A and Williams J A, Antibodies to Recombinant Clostridium difficile Toxins A and B Are an Effective Treatment and Prevent Relapse of C. difficile-Associated Disease in a Hamster Model of Infection Infect. Imm. (1998) 66:2018-2025. [0559] 13. Ma D, et al., Progenics inc. ASM Poster 27 May 2010 [0560] 14. Hansen, G and Demarest, S J. WO 2006/0718877 A2 [0561] 15. Babcock G J, et al., Human monoclonal antibodies directed against toxins A and B prevent Clostridium difficile-induced mortality in hamster Infect. Imm. (2006) 74:6339-6347. [0562] 16. Lowy I et al., Treatment with Monoclonal Antibodies against Clostridium difficile Toxins NEJM (2010) 362: 197-205. [0563] 17. Zubler, R. H., Erard, F., Lees, R. K., Van, L. M., Mingari, C., Moretta, L. & MacDonald, H. R. (1985). Mutant EL-4 thymoma cells polyclonally activate murine and human B cells via direct cell interaction. J. Immunol. 134, 3662-3668 [0564] 18. Babcook, J. S., Leslie, K. B., Olsen, O. A., Salmon, R. A. & Schrader, J. W. (1996). A novel strategy for generating monoclonal antibodies from single, isolated lymphocytes producing antibodies of defined specificities. Proc. Natl. Acad. Sci. U.S.A 93, 7843-7848 [0565] 19. Tickle, S., Adams, R., Brown, D., Griffiths, M., Lightwood, D. & Lawson, A. (2010). High-Throughput Screening for High Affinity Antibodies., pp. 303-307. [0566] 20. Demarest et al., mAbs (2010) 2:190-198 [0567] 21. Yoshida et al., J. Clin. Invest. (2006) 116: 2142-2151 [0568] 22. Xu et al., Vaccine (2005) 23:2658-2664. [0569] 23. Yousaf et al., Clin. Exp. Immunol. (1986) 66:654-660 [0570] 24. Mannik et al., J. Exp. Med. (1971) 133: 713-739 [0571] 25. Nusrat et al., Infection and Immunity (2001) 69:1329-1336 [0572] 26. Lima et al., Infect Immun (1988) 56:582-588 [0573] 27. Ravichandran et al J of Pharmacology and Experimental Therapeutics. (2006) 318: 1343-1351 [0574] 28. Takahashi et al., (2009) Vaccine 27:2616-2619 [0575] 29. Cohen et al., Infect. Cont. and Hosp. Epidem. (2010) 31: 431-455 [0576] 30. Barbut et al., J. Clin. Microbiol. (2000) 38: 2386-2388 [0577] 31. Wilcox et al., J. Hospital Infection (1998) 38: 93-100.