WO2007025977A2

WO2007025977A2 - Anti-tnf alpha producing lactic acid bacteria for the treatment of chronic enterocolitis

Info

Publication number: WO2007025977A2
Application number: PCT/EP2006/065803
Authority: WO
Inventors: Pieter Rottiers; Klaas Vandenbroucke
Original assignee: Actogenix Nv; Ablynx N.V.
Priority date: 2005-08-30
Filing date: 2006-08-30
Publication date: 2007-03-08
Also published as: US9017662B2; EP1948206A2; US20080274084A1; WO2007025977A8; EP3181141A1; ES2626610T3; JP5049973B2; EP1948206B1; CA2925307C; WO2007025977A3; JP2009506095A; CA2619748C; AU2006286563A1; CA2925307A1; CA2619748A1; AU2006286563B2

Abstract

The present invention relates to a novel treatment of chronic enterocolitis. More specifically, the invention relates to the production of a medicament comprising anti-TNFalpha antibody producing lactic acid bacteria, and the use of this medicament in the treatment of chronic enterocolitis.

Description

NOVEL TREATMENT OF CHRONIC ENTEROCOLITIS

The present invention relates to a novel treatment of chronic enterocolitis. More specifically, the invention relates to the production of a medicament comprising anti-TNFα producing lactic acid bacteria, and the use of this medicament in the treatment of chronic enterocolitis. Inflammatory bowel disease (IBD) refers to a group of gastrointestinal disorders characterized by a chronic non-specific inflammation of portions of the gastrointestinal tract. The most prominent examples of IBD in humans are ulcerative colitis (UC) and Crohn's disease (CD). The etiology or etiologies of IBD are unclear. IBD diseases appear to result from the unrestrained activation of an inflammatory response in the intestine. This inflammatory cascade is thought to be perpetuated through actions of proinflammatory cytokines and selective activation of lymphocyte subsets. UC and CD are associated with many symptoms and complications, including growth retardation in children, rectal prolaps, blood in stools, wasting, iron deficiency and anemia. UC refers to a chronic, non-specific, inflammatory and ulcerative disease having manifestations primarily in the colonic mucosa. It is frequently characterized by bloody diarrhea, abdominal cramps, blood and mucus in the stoll, malaise, fever, anemia, anorexia, weight loss, leukocytosis, hypoalbuminemia and an elevated erythrocyte sedimentation rate.

Crohn's disease shares many features in common with ulcerative colitis. Crohn's disease is distinguishable in that lesions tend to be sharply demarcated from adjacent normal bowel, in contrast to the lesions of ulcerative colitis which are fairly diffuse. Additionally, Crohn's disease predominantly afflicts the ileum (ileitis) and the ileum and colon (ileocolitis). In some cases, the colon alone is diseased (granulomatous colitis) and sometimes the entire small bowel is involved. Colon cancer is a known complication of chronic IBD. It is increasingly common in those patients who have IBD for many years. The risk for cancer begins to rise significantly after eight to ten years of IBD, making a fast and efficient treatment of IBD even more important. The most commonly used medication to treat IBD includes anti-inflammatory drugs such as corticosteroids and sulicilates such as sulphasalazine and its derivatives. For people that not respond to these drugs, immunosuppressive drugs such as cyclosporine A, mercaptopurin and azathropine are used. However, these medicaments all have serious side effects. A recent, successful development in the treatment of IBD consists in the use of compounds, blocking the working of TNF or its receptor. In that respect is the use of TNF antibodies one of the most promising new therapies. Tumor necrosis factor α (TNFα) is a cytokine produced by numerous cell types, including monocytes and macrophages, that was originally identified based on its capacity to induce the necrosis of certain mouse tumors (see e.g., Old, L. (1985) Science 230:630-632). TNFα has been implicated in the pathophysiology of a variety of other human diseases and disorders, including sepsis, infections, autoimmune diseases, transplant rejection and graft-versus-host disease (see e.g. Moeller, A. et al. (1990) Cytokine2: 162-169; U.S. Patent No. 5,231 ,024 to Moeller et al ; European Patent Publication No. 260 610 (B1 ) by Moeller, A. et al.; Vasilli. P.(1992) Annu. Rev.lmmunol. 10:41 1-452; Tracey, K.J. and Cerami, A. (1994) Annu. Rev. Med. 45:491-503). Because of the harmful role of human TNFα (hTNFα) in a variety of human disorders, therapeutic strategies have been designed to inhibit or counteract hTNFα activity. In particular, antibodies that bind to, and neutralize, hTNFα have been sought as a means to inhibit hTNFα activity.

Several antibody preparations have been tested for the treatment of IBD. Although polyclonal antibodies have been tested in phase Il clinical tests, monoclonal antibodies are clearly preferred. Infliximab is a chimeric human-mouse monoclonal antibody of the IgGI K subclass, which specifically targets and irreversibly binds to TNFα on cell membranes and in blood. Single intravenous doses, ranging from 5 to 20 mg/kg of the antibody infliximab resulted in a drastic clinical improvement in active Crohn's disease; it has been launched on the market to treat Crohn's disease in 1998. To solve possible problems linked to chimeric antibodies, the human monoclonal TNFα adalimumab was developed, which is currently tested in phase III clinical trials for the treatment of Crohn's. To improve the half-life time of the antibody in patients, Celltech developed Certolizumab pegol, which is a humanized monoclonal pegylated anti- TNFα antibody, which is currently also tested in phase III clinical trials for the treatment of Crohn's.

However, in all those cases, the antibodies are applied in a systemic way, mainly by subcutaneous injection. Systemic administration of anti TNF-α antibody may result in rather serious unwanted effects, including headache, abscess, upper respiratory tract infection and fatigue. The unwanted effects associated with systemic delivery could be solved by local delivery on the place of the inflammation. A promising system for delivery of biological active compounds in the intestine has been disclosed in WO97/14806 whereby non-invasive gram positive bacteria such as lactic acid bacteria are used to deliver biological active compounds in the gut. WO00/23471 discloses that this system can be used to deliver IL-10 to the ileum, whereby this strain can be used to treat IBD. WO01/98461 discloses an alternative method for intestinal delivery using yeast. However, although the delivery of biologically active compounds is described, these documents do not teach the delivery of antibodies in the intestine. The in situ production of active antibodies in the intestine is far from straightforward, as both folding and secretion of the antibody are critical. Especially, the stabilization of the structure by sulfur bridges may cause problems for the production of antibodies in bacteria or yeasts. Moreover, whereas cytokines like IL-10 fulfill a catalytic function, TNF antibodies need to be produced in a sufficient amount to inactivate the endogenous produced TNF. Surprisingly, we found that the local delivery of anti TNF-α antibody by a genetically engineered micro-organism can be used in an efficient way to treat IBD.

A first aspect of the invention is the use of a genetically modified micro-organism, producing an anti TNF-α antibody, for the preparation of a medicament to treat IBD. The term antibody, as used here, includes, but is not limited to conventional antibodies, chimeric antibodies, single chain antibodies, bifunctional antibodies, bivalent antibodies, diabodies, and camelid antibodies, including antibody fragments such as VHHs; Preferably said antibody is a camelid antibody fragment (VHH; further called nanobody), even more preferably said antibody is a bivalent anti-TNFα nanobody. Bivalent antibodies have the advantage to inhibit TNF binding to its receptor, in a significantly more efficient fashion than does monovalent (EC50 = 16 pM and 6.7 nM, respectively). Surprisingly we found that the production of bivalent antibodies in Lactococcus is as high or even higher than that of monovalent antibodies. Preferably, said genetically modified micro-organisms is a lactic acid bacterium or a yeast. Delivery of biologically active polypeptides into the animal body by lactic acid bacteria has been disclosed in WO9714806; intestinal delivery of peptides by yeast has been described in WO0198461. However, none of these documents mention the delivery of antibodies or nanobodies in the intestine. Production, secretion and delivery in vivo of biological active antibodies or nanobodies is far from evident, as a correct folding and secretion of the antibody is required, and sufficient antibody is required to obtain a neutralizing activity. In one preferred embodiment said genetically modified micro-organism is a Lactococcus lactis strain, preferably said genetically a Lactococcus lactis ThyA mutant. A specially preferred embodiment is the use of a Lactococcus lactis ThyA mutant, whereby the gene encoding the TNF-α antibody has been used to disrupt the THYA gene. In another preferred embodiment, a yeast is be used to deliver the anti TNF-α antibody. Preferably said yeast is Saccharomyces cerevisiae, even more preferably said yeast is Saccharomyces cerevisiae subsp. Boulardii.

IBD, as used here, includes but is not limited to chronic colitis, ulcerative colitis and Crohn's disease. Preferably, IBD is chronic colitis. Another aspect of the invention is a pharmaceutical composition for oral administration, comprising at least one genetically modified anti-TNFα VHH producing micro-organism. Preferably, said anti-TNFα VHH is a bivalent antibody. The pharmaceutical composition may be liquid, comprising biological active micro-organisms, or it may be solid, comprising dried micro-organisms that can be reactivated when put in a suitable environment. Micro-organisms may be dried by any system, including freeze drying and spray drying. "Anti-TNFα VHH producing" as used here doesn't imply that the micro-organism is producing the VHH in the pharmaceutical composition, but it means that the micro-organism is viable and can produce the VHH when placed in a suitable environment. Micro-organisms may be coated to facilitate the delivery into the gastro-intestinal tract. Such coating are known to the person skilled in that art and was, amongst others, described by Huyghebaert et al. (2005). The pharmaceutical composition may further comprise agents to improve the viability of the micro-organisms, such as, but not limited to trehalose. Preferably, the micro-organisms are selected from the group consisting of lactic acid bacteria and yeasts. One preferred embodiment is a pharmaceutical composition, whereby the VHH producing micro-organism is a Lactococcus lactis, preferably a ThyA mutant. Another preferred embodiment is a pharmaceutical composition, whereby the VHH producing micro-organism is a Lactobacillus sp. preferably a ThyA mutant. Preferably, said ThyA mutants are obtained by gene disruption, using the VHH encoding construct as insert. Still another preferred embodiment is a pharmaceutical composition whereby the VHH producing micro-organism is Saccharomyces cerevisiae, preferably S. cerevisiae subspecies boulardii.

Another aspect of the invention is a method of preventing, treating and/or alleviating at least one disease or disorder of the gastro-intestinal tract, comprising administering to the gastro- intestinal tract an effective amount of an anti-TNFα VHH producing micro-organism. Preferably, said anti-TNFα VHH is a bivalent antibody. The way of administering can be any way known to the person skilled in the art, and includes, but is not limited to oral and rectal administration. Preferably, the way of administering is oral administration. Preferably, said disease or disorder is a disease or disorder characterized by an imbalance in TNFα production, and can be treated by TNFα inactivating compounds such as TNFα antibodies. Even more preferably, said disease is an irritable bowel disease, including but not limited to chronic colitis, ulcerative colitis and Crohn's disease. Most preferably, said disease or disorder is chronic colitis. Preferably, said genetically modified micro-organisms is a lactic acid bacterium or a yeast. In one preferred embodiment said genetically modified micro-organism is a Lactococcus lactis strain, preferably said genetically a Lactococcus lactis ThyA mutant. A specially preferred embodiment is a Lactococcus lactis ThyA mutant, whereby the gene encoding the TNF-α antibody has been used to disrupt the THYA gene. In another preferred embodiment genetically modified micro-organism is a Lactobacillus sp strain, preferably said genetically a Lactobacillus ThyA mutant. A specially preferred embodiment is a Lactobacillus ThyA mutant, whereby the gene encoding the TNF-α antibody has been used to disrupt the THYA gene.

In another preferred embodiment, a yeast is the anti TNF-α antibody producing microorganism. Preferably said yeast is Saccharomyces cerevisiae, even more preferably said yeast is Saccharomyces cerevisiae subsp. Boulardii. BRIEF DESCRIPTION OF THE FIGURES

Fig . 1 . Time course of heterologous monovalent and bivalent nanobody NANO3F production by GM L. lactis (LL-NANO3F and LL-NANO3F-3F). LL-pTREX1 : vector control; LL-mlL10: L. lactis strain secreting murine interleukin-10. (A) Western blot analysis of proteins secreted from the various strains revealed by anti-Myc Ab. Each lane on the blot represents 250 μl_ of L. lactis culture supernatant obtained after different periods of growth (2 x 10⁷ CFU at time zero).

E. coli purified monovalent NANO3F (+) and bivalent NANO3F-3F (++) were used as positive control. (B) Concentrations of secreted heterologous Myc-tagged proteins in culture supematants of LL-pTREX1 (O), LL-NANO3F (■), LL-NANO3F-3F (A) and LL-mlL10 (O), as determined by ELISA.

Fig . 2. L. lactis secreted or E. coli purified monovalent and bivalent nanobody NANO3F are able to efficiently neutralize soluble and membrane bound TNF. (A) soluble TNF (20 IU/ml) was neutralized by NANO3F and NANO3F-3F in a 16 hour cytotoxicity assay using the WEHI 164 cl 13 cells in the presence of 1 μg/ml actinomycin D. (B) L. lactis secreted and E. coli purified nanobodies NANO3F and NANO3F-3F were able to inhibit the cytotoxic effects of L929 expressing uncleavable membrane-bound TNF. The gray bars represent wells where purified NANO3F or NANO3F-3F was added (total concentration of 250 ng/mL). The black bars represent wells where 50 μl_ of filtered (0.22 μm) lactococcal supernatant was added. The final concentration of LL-NANO3F and LL-NANO3F-3F was 250 ng/ml in each setting.

Fig. 3. Analysis of morbidity in chronic DSS colitis. (A-E) Representative histology of the middle colon from healthy control mice (A) and mice with chronic DSS colitis either mock- treated (B) or treated with LL-pTREX1 (C), LL-NANO3F-3F (D) or LL-mlL10 (E). (F) Statistical evaluation of the histological score of the middle colon. Bars represent the mean + SEM. White bar represent the healthy control group. Mice with DSS-induced chronic colitis were either mock-treated (hatched bars) or received LL-pTREX1 (black bar), LL-NANO3F-3F (red bar) or LL-mll_10 (gray bar). * * * and * * represent statistical significant differences in comparison with the mock-treated and the vector control groups of P < 0.001 and P < 0.01 , respectively.

Fig . 4 Analysis of morbidity in 20 weeks old 129Sv/Ev IL-10^"'^" mice. Each group received daily for 14 days 2 x 10⁹ CFU LL-pTREX1 (vector control), LL-NANO3F-3F or LL-mlL10, except the mock treated group. (A-D) Representative histology of the middle colon from I L- 10^"'^" mice which were mock-treated (A) or treated with LL-pTREX1 (B), LL-NANO3F-3F (C) or LL-mlL10 (D) (hematoxylin and eosin staining). Statistical evaluation of the MPO levels per mm² colon tissue (A) and histological score of the distal colon (B). Bars represent the mean + SEM. Hatched bars represent the 129Sv/Ev I L- 10^7" mice that were mock treated, the black bars represent the 129Sv/EV IL-I O^''^' mice that received the vector control LL-pTREX1 , the red bars represent the 129Sv/EV IL-10^"'^" mice that received LL-NANO3F-3F and the gray bars represent the 129Sv/EV IL-10⁷ mice that were treated with LL-mlL10. * and * * represent a statistically significant difference in comparison with the vector control group of P < 0.05 and P < 0.01 , respectively.

Fig . 5 Presence of nanobody specific antibodies was assessed by ELISA.

Fig . 6 Pretreatment with NANO3F-3F confers LPS irresponsiveness to Mf4/4 macrophages. WEHI164 cl 13 cells can still be killed with 5 IU TNF

Fig. 7 Statistical evaluation of the histological score of the distal colon in mice with chronic DSS colitis. Bars represent the mean + SEM. White bars represent the healthy control group. Mice with DSS-induced chronic colitis received different L. lactis cultures for 28 days whereafter the mice were killed and analyzed immediately. The black bar represents the vector control LL-pTREX1 treated group, the hatched bar those that received LL-NANO3F (L. lactis secreting monovalent 3F), the gray bar those that received LL-NANO3F-3F (L. lactis secreting bivalent 3F-3F-. * * represents a statistical significant difference in comparison with the vector control LL-pTREX1 and the LL-NANO3F treated groups of P < 0.01.

Fig. 8 L. lactis secreted monovalent and bivalent Nanobody 3E (anti-human TNF Nanobody) are able to efficiently neutralize soluble human TNF. Soluble human TNF (different concentrations ranging from 30 till 3.3 Ill/ml) was neutralized by 3E and 3E-3E in a 16 hour cytotoxicity assay using L929s cells in the presence of 1 μg/ml actinomycin D. Cell survival was calculated relative to healthy cells The white bar represents wells where 50 μL of filtered (0.22 μm) lactococcal supernatant of the vector control LL-pTREX1 was added. Black and gray bars represent wells where 50 μL of filtered (0.22 μm) lactococcal supernatant was added; containing L. lactis secreted 3E or 3E-3E, respectively. The final concentration of L. lactis secreted 3E or 3E-3E was 250 ng/ml in each setting.

EXAMPLES

Material and methods to the examples Bacteria and plasm ids The L. lactis strain MG 1363 was used throughout this study. Bacteria were cultured in GM 17 medium, i.e. M17 (Difco Laboratories, Detroit, Ml) supplemented with 0.5% glucose. Stock suspensions of all strains were stored at -20⁰C in 50% glycerol in GM 17. For intragastric inoculations, stock suspensions were diluted 200-fold in fresh GM 17 and incubated at 30⁰C. They reached a saturation density of 2 x 10⁹ colony-forming units (CFU) per ml_ within 16 hours. Bacteria were harvested by centrifugation and concentrated 10-fold in BM9 medium. (Schotte, Steidler et al. 2000). For treatment, each mouse received 100 μl_ of this suspension daily by intragastric catheter.

Identification and formatting of an anti-muήneTNF Nanobody

The generation of anti-murine TNF nanobody was essentially carried out as described in

WO2004041862. After immunizing llamas with mTNF, subsequent cloning of the VHH repertoire and panning, Nanobody NANO3F (MW-15 kDa) was isolated. The murine TNF specific Nanobody was converted into a bivalent format (coded NANO3F-3F, MW~30 kDa) using the 12 aminoterminal residues of the llama lgG2a upper hinge sequence as a spacer. The cDNA of the NANO3F and NANO3F-3F, extended at their 3' ends with the sequence encoding the HisG and Myc-tag, were fused to the Usp45 secretion signal (van Asseldonk, Rutten et al. 1990) downstream of the lactococcal P1 promotor (Waterfield, Le Page et al. 1995) and expressed in MG 1363 (details of plasmid construction can be obtained from the authors). MG1363 strains transformed with plasmids carrying the NANO3F or NANO3F-3F coding sequence were designated LL-NANO3F and LL-NANO3F-3F respectively. LL-pTREX1 , which is MG1363 containing the empty vector pTREXI , served as control.

Quantification of nanobody in L lactis medium.

Myc-tagged LL-NANO3F and LL-NANO3F-3F were quantified by direct adsorption of crude L. lactis supematants to ELISA plates (Maxisorp F96, Nunc, Rochester, NY) and subsequent detection with a specific mouse mAb against the Myc epitope (Sigma, St. Louis, MO). For quantification of 3F-3F nanobodies secreted in vivo in colon tissue, the entire colon was homogenized in PBS containing 1% BSA and sonicated. The 3F-3F nanobodies were measured in the colon supernatant with the nanobody quantification protocol.

Measurement of anti-Nanobody antibody levels in mouse serum. Mice were injected intraperitoneal^ with 100 μg Nanobody, or intragastically with LL-NANO3F- 3F, daily over a 14 day-period and were subsequently bled. We coated Nanobody at a concentration of 10 μg/ml in microtiterplates (NUNC Maxisorb) overnight at 4°C. The plate was washed 5 times with PBS-Tween and blocked for 2 hours at RT with PBS-1 % casein. The samples were applied at a 1/50 dilution in PBS for 2 hours at RT. The plate was washed 5 times and detection was performed by incubation with rabbit-polyclonal-anti-mouse- immunoglobulin-HRP (DAKO, 3, 000-fold diluted) for one hour at RT, and after washing plates were stained with ABTS/H2O2. The OD405nm was measured. Anti-soluble and membrane-bound TNF bioassay

The inhibitory effect of the NANO3F and NANO3F-3F nanobodies on soluble mTNF (20 IU/mL) was measured in a 16 hour cytotoxicity assay using the mouse fibroblast WEHI 164 cl 13 cells in the presence of 1 μg/ml actinomycin D, as described. (Espevik and Nissen-Meyer 1986) The effect of NANO3F and NANO3F-3F to counteract the cytotoxic effect of membrane-bound TNF was determined on the WEHI 164 cl 13 cells after adding L929 cells, expressing uncleavable, membrane-bound TNF to the cell culture (Decoster et al. 1998).

Stimulation of Macrophages with LPS To measure the effect of NANO3F-3F on the induction of proinflammatory cytokines by LPS, MF4/4 macrophages (Desmedt et al. 1998) were incubated with NANO3F-3F (100 μg/ml). After 1 hour cells were extensively washed (3X) in a sufficient volume of PBS to completely remove all nanobody present in solution. The cells were resuspended and incubated in the presence or absence of LPS for 4 hours. The cells were washed (1X) in PBS and after 4 hours of incubation, the supematans and cells were separated by centrifugation. To measure the soluble TNF release, the WEHI 164 cl 13 cells bioassay was used.

Animals

1 1-week old female BALB/c mice were obtained form Charles River Laboratories (Sulzfeld, Germany). They were housed under SPF conditions. IL-10 knockout mice (129Sv/Ev IL-10^"'^") (Kuhn, Lohler et al. 1993) were housed and bred under SPF conditions. The IL-10 '^' mice were used at 20 weeks of age, at which time chronic colitis had fully developed. All mice were fed standard laboratory feed and tap water ad libitum. The animal studies were approved by the Ethics Committee of the Department for Molecular Biomedical Research, Ghent University (File No. 04/02).

Induction of chronic colitis by DSS

Mice weighing approximately 21 g were induced to chronic colitis by four cycles of administration of 5% (w/v) DSS (40 kDa, Applichem, Darmstadt, Germany) in the drinking water, alternating with 10-day periods of recovery with normal drinking water. (Okayasu, Hatakeyama et al. 1990; Kojouharoff, Hans et al. 1997) Treatment was arbitrarily initiated at day 21 after the fourth cycle of DSS.

Myeloperoxidase (MPO) assay MPO activity in the middle colon tissue was measured as described (Bradley, Priebat et al. 1982). Pure human MPO was used as a standard (Calbiochem, San Diego, CA). Data are expressed as μg MPO/mm² colon tissue. Histological analysis

For histological analysis, the colon was removed, cleaned and opened longitudinally. A segment of 1 cm was taken from the middle part of the colon, embedded in paraffin and sectioned longitudinally. Three sections of 4 μm were cut at 200 μm intervals and stained with hematoxylin/eosin. Colon sections were numbered randomly and interpreted semiquantitative^ in a blinded manner by a pathologist. The histological score is the sum of the epithelial damage and lymphoid infiltration, each ranging from 0 to 4 as described (Kojouharoff, Hans et al. 1997).

Statistical analysis

All data are expressed as mean + SEM Parametric data were analyzed with a 1-way analysis of variance followed by a Dunnett multiple comparisons posttest. Nonparametric data (scoring) were analyzed with a Mann-Whitney test. Example 1 : Anti-TNF-α VHH production by L. lactis in vitro L. lactis was transformed with the plasmids encoding NANO3F and NANO3F-3F. The production of the antibodies was checked by Western blot and ELISA, using a strain transformed with the empty plasmid pTREX and an IL10 producing strain as reference. The results are shown in Figure 1. NANO3F-3F is produced by L. lactis in similar or higher amounts than NANO3F. The amount produced is significantly higher than for IL10.

Example 2: LL-NANO3F-3F is bioactive and inhibits both soluble and membrane bound

TNF-α

The inhibitory effect of the NANO3F and NANO3F-3F nanobodies, produced by L. lactis on soluble mTNF was measured in a cytotoxicity assay using the mouse fibroblast WEHI 164 cl 13 cells as described by Espevik and Nissen-Meyer (1986). E. coli produced NANO3F and NANO3F-3F was used as a positive reference. Both the purified nonobodies as well as the nanobodies produced by L. lactis can neutralize the soluble TNF. (Figure 2A) The effect of NANO3F and NANO3F-3F to counteract the cytotoxic effect of membrane-bound TNF was determined on the WEHI 164 cl 13 cells after adding L929 cells, expressing uncleavable, membrane-bound TNF to the cell culture (Decoster et al. 1998). The effect of NANO3F is less pronounced, both with the purified form and the L. lactis produced form, but the effect of the NANO3F-3F nanobody is clear in both cases (Figure 2B)

Example 3: LL-NANO3F-3F effect in vivo on established DSS induced chronic colitis Chronic colitis was induced by DSS as described in materials and methods. Mice were daily treated with 2 10⁹ colony forming units (cfu) of either LL-pTREX1 , LL-NANO3F-3F, or LL- mlL10. A mock treatment, and healthy mice ("watercontrol") were used as additional control. The effect of the nano3F-3F nanobody deliverd by L. lactis is comparable to the protection obtained by the in situ produced IL-10 (Figure 3).

Example 4: LL-NANO3F-3F effect in vivo on established IL-10 ^v enterocolitis To evaluate the protection in IL-10 '^' enterocolitis, morbidity in 20 weeks old 129Sv/Ev IL-10 '^' treated and untreated mice. Each group received daily for 14 days 2 x 10⁹ CFU of either LL- pTREXI (vector control), LL-NANO3F-3F or LL-mlL10, except the mock treated group. The results are summarized in Figure 4. Both the myelperoxidase assay as well as the histological score indicate a significant protection in the LL-NANO3F-3F treated mice.

Example 5: immunogenicity of NANO3F-3F

To evaluate a possible adverse immunogenic effect of LL-NANO3F-3F, mice were treated intragastically over a period of 14 days with LL-NANO3F-3F, using intraperitoneal injection of purified nanobody as control. Anti-Nanobody antibody levels were measured in the mouse seum. The results are shown in Figure 5. While interperitoneal injection of NANO3F-3F is giving a clear immune response, the treatment with LL-NANO3F-3F is not immunogenic and proofs to be safe in that respect.

Example 6: effect of NANO3F-3F on LPS induction of proinflammatory cytokines To measure the effect of NANO3F-3F on the induction of proinflammatory cytokines by LPS, MF4/4 macrophages (Desmedt et al. 1998) were incubated with NANO3F-3F. The cells were washed and then incubated with LPS. Soluble TNF release was measured using the WEHI 164 cl 13 cell toxicity assay. The results are shown in Figure 6. Pretreatment of the macrophages with NANO3F-3F naobody gives a clear protection against LPS induced soluble TNF production.

Example 7: bivalent antibodies perform surprisingly better than monovalent antibodies Although bivalent antibodies are larger than monovalent ones, it doesn't affect the production in Lactococcus. The production of bivalent antibodies is at least as good if not better than for monovalent antibodies (Fig. 1 ). However, even more important is the efficacy of the bivalent antibodies. From the in vivo experiments, it is obvious that a monovalent anti-TNF antibody results only in a marginal, non-significant improvement of the histological score, whereas administration of a L. lactis secreting bivalent antibodies results in a significant improvement (Figure 7). Indeed, the neutralizing effect of bivalent antibodies is, for a comparable concentration of protein, more pronounced than that of monovalent antibodies. As long as no complete neutralization is reached, the improvement is more than a factor 2, indicating the effect is not purely due to the double valence of the nanobody (Figure 8) REFERENCES

Bradley, P. P., D. A. Priebat, et al. (1982). "Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker." J Invest Dermatol 78(3): 206-

9. Espevik, T. and J. Nissen-Meyer (1986). "A highly sensitive cell line, WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes." J Immunol

Methods 95(1 ): 99-105. Huyghebaert, N., A. Vermeire, S. Neirynck, L. Steidler, E. Remaut, and J. P. Remon. (2005)

"Development of an enteric-coated formulation containing freeze-dried, viable recombinant Lactococcus lactis for the ileal mucosal delivery of human interleukin-10."

Eur J Pharm Biopharm 60(3): 349-59 Kojouharoff, G., W. Hans, et al. (1997). "Neutralization of tumour necrosis factor (TNF) but not of IL-1 reduces inflammation in chronic dextran sulphate sodium-induced colitis in mice." Clin Exp Immunol 107(2): 353-8. Kuhn, R., J. Lohler, et al. (1993). "lnterleukin-10-deficient mice develop chronic enterocolitis."

Cell 75(2): 263-74. Okayasu, I., S. Hatakeyama, et al. (1990). "A novel method in the induction of reliable experimental acute and chronic ulcerative colitis in mice." Gastroenterology 98(3): 694-

702. Schotte, L., L. Steidler, et al. (2000). "Secretion of biologically active murine interleukin-10 by

Lactococcus lactis." Enzyme Microb Technol 27(10): 761-765. van Asseldonk, M., G. Rutten, et al. (1990). "Cloning of usp45, a gene encoding a secreted protein from Lactococcus lactis subsp. lactis MG1363." Gene 95(1 ): 155-60. Waterfield, N. R., R. W. Le Page, et al. (1995). "The isolation of lactococcal promoters and their use in investigating bacterial luciferase synthesis in Lactococcus lactis." Gene

165(1 ): 9-15.

Claims

1. The use of an anti-TNFα VHH producing micro-organism for the preparation of a medicament to treat IBD

2. The use according to claim 1 , whereby said anti-TNFα VHH is a bivalent antibody.

3. The use of an anti-TNFα VHH producing micro-organism according to claim 1 or 2, whereby said micro-organism is a lactic acid bacterium.

4. The use according to claim 3, whereby said lactic acid bacterium is Lactococcus lactis.

5. The use according to claim 3, whereby said lactic acid bacterium is a Lactobacillus sp.

6. The use of an anti-TNFα VHH producing micro-organism according to claim 1 or 2, whereby said micro-organism is a yeast.

7. The use of an anti-TNFα VHH producing micro-organism according to claim 6, whereby said yeast is Saccharomyces sp.

8. The use according to any of the preceding claims, whereby said IBD is chronic colitis.

9. The use according to any of the claims 1 to 7, whereby said IBD is Crohn's disease.

10. The use according to any of the claims 1 to 7, whereby said IBD is ulcerative colitis.

1 1. A pharmaceutical composition for oral administration, comprising at least one anti- TNFα VHH producing micro-organism.

12. A pharmaceutical composition for oral administration according to claim 1 1 , whereby said anti-TNFα VHH is a bivalent antibody.

13. A pharmaceutical composition according to claim 1 1 or 12, whereby said microorganism is selected from the group consisting of lactic acid bacteria and yeasts.

14. A pharmaceutical composition according to claim 13, whereby said lactic acid bacterium is Lactococcus lactis.

15. A pharmaceutical composition according to claim 13, whereby said lactic acid bacterium is Lactobacillus sp.

16. A pharmaceutical composition according to claim 13, whereby said yeast is Saccharomyces cerevisiae.

17. Method of preventing, treating and/or alleviating at least one disease or disorder of the gastro-intestinal tract, comprising administering to the gastro-intestinal tract an effective amount of an anti-TNFα VHH producing micro-organism.

18. A method according to claim 17, whereby said anti-TNFα VHH is a bivalent antibody.

19. A method according to claim 17 or 18, whereby said administering is oral administering.

20. A method according to claim any of the claims 17 - 19, whereby said disease or disorder is irritable bowel disease.

21. A method according to claim 20, whereby said irritable bowel disease is selected from the group consisting of chronic colitis, Crohn's disease and ulcerative colitis.

22. A method according to any of the claims 17-21 , whereby said micro-organism is selected from the group consisting of lactic acid bacteria and yeasts.

23. A method according to claim 22, whereby said lactic acid bacterium is Lactococcus lactis.

24. A method according to claim 22, whereby said lactic acid bacterium is Lactobacillus sp.

25. A method according to claim 22, whereby said yeast is Saccaromyces cerevisiae.