This invention relates to the use of Pellino proteins as diagnostic markers for Crohn's disease. The mammalian Pellino family consists of four members: Pellinol , Pellino2 and splice variants of Pellino3 termed Pellino3 long (Pellino3L; also known as Pellino3a) and Pellino3 short (Pellino3S; also known as Pellino 3b). Each Pellino family member contains an N-terminal forkhead-associated (FHA) domain that recognizes phosphothreonine residues and mediates association with Interleukin- 1 Receptor-Associated Kinases (IRAKs), and a C-terminal RING-like domain that confers E3 ubiquitin ligase activity and an ability to catalyse lysine 63 (Lys63)-linked polyubiquitination of IRAKs.

Pellino3 is a member of the E3 ubiquitin ligase family of Pellino proteins which play a role in TLR (toll-like receptor) signalling. TLRs sense pathogen-associated molecules and respond by inducing cytokines and type I interferon.

Applicant's PCT Application No. PCT/EP2013/077059, filed 18 December 2013 and claiming priority from UK Patent Application No. 12231 14.8, filed 21 December 2012, disclosed a potential protective role for Pellino3 in inflammatory bowel disease (eg. Crohn's disease, Ulcerative Colitis). The latter contention was based on exacerbation of experimental colitis (DSS-induced colitis) in mice that were genetically engineered to lack the gene encoding Pellino3. Furthermore loss of function mutations in the innate immune receptor NOD2 are associated with increased susceptibility to Crohn's disease and this application includes data showing that Pellino3 is a mediator of NOD2 signalling, again consistent with a potential protective role for Pellino3 in Crohn's disease. Given this hypothesis applicant assessed the levels of expression of Pellino3 protein in colonic biopsy samples from control, Crohn's disease and Ulcerative Colitis patients. The data demonstrate that the levels of Pellino3 protein are strongly reduced in colonic tissue from Crohn's disease subjects relative to control or Ulcerative Colitis subjects suggesting that the protein expression levels of any of Pellinol , Pellino2 or Pellino3 may be a strong diagnostic indicator of subjects with Crohn's disease. According to the invention there is provided the use of Pellinol , Pellino2 or Pellino3 protein as an in vivo or in vitro diagnostic marker for Crohn's disease in a subject.

The invention also provides an in vivo or in vitro method for diagnosing Crohn's disease in a subject comprising:

(a) measuring the level of Pellinol , Pellino2 or Pellino3 protein in a sample of a body tissue or fluid from the subject;

(b) comparing the measured level to one or more predetermined threshold values of the protein; and

(c) determining whether the subject has Crohn's disease based on the comparison, whereby a measured level of Pellinol , Pellino2 or Pellino3 protein below the predetermined threshold value(s) indicates that the subject has Crohn's disease. As used herein, the term "Pellino3" is intended to include Pellino3 long (Pellino3L; also known as Pellino3a) and Pellino3 short (Pellino3S; also known as Pellino3b), unless otherwise stated. The body tissue may be selected from colonic and ileal tissue; and the body fluid may be selected from whole blood, blood plasma and blood serum.

The level of Pellinol , Pellino2 or Pellino3 protein may be measured by any suitable method, including enzyme linked immunosorbent assay (ELISA).

Crohn's disease may be ileum or ileo-colon-based. The subject is preferably human.

When the level of expression of Pellinol , Pellino2 or Pellino3 protein is statistically significantly lower than a predetermined threshold of said protein, the subject is diagnosed as having Crohn's disease. For example, a subject having a mean Pellinol , Pellino2 or Pellino3 expression level which is at least 40 times lower than the expression level in a normal subject, i.e., a subject with a normal ileo- colonoscopy and negative histology, would be diagnosed as having Crohn's disease.

Optionally, the diagnostic marker used in the invention is Pellino3 protein and the level of Pellino3 protein is measured.

Human Pellino proteins have the following NCBI accession numbers:

Pellinol (NP_065702.2 Gl :1 1037063) ; Pellino2 (NP_067078.1 G l:10864063) ;

Pellino3long (Pellino3a), Isoform 1 (NP_659502.2 G l:148612798) ; and

Pellino3short (Pellino3b), Isoform 2 (NP_001091980.1 Gl :148612833). In the accompanying drawings, Figure 1 illustrates expression levels of Pellino3 protein in colon biopsy samples from control (Normal), Crohn's disease and ulcerative colitis subjects; and Figure 2 is a histogram showing the relative expression levels of Pellino3 protein in the samples used in Figure 1 . The invention is illustrated in the following Example. Example

Patient Information

Subjects aged 18-80 years undergoing routine colonoscopy as part of their on-going care were prospectively recruited. Following informed consent two additional biopsies were taken from the abnormal (inflamed) colonic or ileal segment in patients with inflammatory bowel disease and from the left colon in controls. Anonymised biopsies were snap frozen and stored at -80°C for batch analysis. Patient demographics, medical history, endoscopy findings and routine histology were recorded. A total of 30 subjects have been enrolled, 19 (63%) with Inflammatory Bowel Disease (IBD) and 1 1 (37%) controls. In all 14 (47%) were male and the mean age was 46.38 +/- 16.4 years. There were more women in the control group 7 of 1 1 (63%) versus 9 of 19 (47%) with IBD and the the mean age of controls was slightly older at 54.1 +/- 16.2 years, (range 18-73) versus 43.9 +/- 16.2 years for IBD (range 25-73), but did not reach statistical significance (p<0.24).

Control patients were defined as subjects with a normal ileo-colonoscopy and negative histology. The controls did not have symptoms suggestive of IBD. Eight were undergoing colonoscopy for polyp surveillance and the remaining 3 had anaemia. Within the IBD group, 14 had established Crohn's disease (10 ileo-colonic and 4 isolated ileal disease (ileal and ileo-colonic diseases are two varieties of Crohn's and represent the site of activity)) and 5 ulcerative colitis. 15 (79%) of the IBD patients were on maintenance therapy for IBD including adalimumab 5 (26%), aminosalicylic acid 7(37%), oral steroids 3 (16%). At endoscopy inflammation was graded as moderate in 6 (32%) and mild in the remaining 13 (68%). Routine histology confirmed active inflammation in 12 cases (63%).

Tissue Processing and Analysis Frozen biopsy samples were thawed and placed in cell lysis buffer (20mM Tris-HCI pH 7.4, containing 150 mM NaCI, 0.2% (v/v) Igepal, 10% (w/v) Glycerol, 50 mM NaF, 1 mM Na ₃ V0 ₄ , 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and complete protease inhibitor cocktail (Roche)) and sonicated. Cell lysates were centrifuged at 14,000g for 10 min to remove cell debris and nuclei. Supernatants were assayed for protein concentration and lysate samples (50 μ9) were resolved by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to nitrocellulose membranes, and analyzed by immunoblotting using an anti-Pellino3 antibody and an anti-B-actin antibody (as a loading control). In some cases immunoblotting was performed in the presence of the immunogenic peptide that was used to generate the anti-Pellino3 antibody in order to define non-specific binding of the antibody. Immunoreactivity was visualized by the Odyssey Imaging System (LI-COR Biosciences) or enhanced chemiluminescence. One of the Crohn's disease patients in Cohort 2 was in an inactive phase at the time of sampling. As illustrated in Figure 1 , all Crohn's disease subjects demonstrated greatly reduced levels of Pellino3 protein relative to samples from control (labelled "Normal" in Figure 1 ) and Ulcerative Colitis subjects. The western blots obtained above of colon biopsy samples from the Control (Con) , Ulcerative Colitis (UC) and Crohn's disease (CD) subjects were analyzed with Image J image analysis software (National Institutes of Health, Bethesda, MD) and intensities of Pellino3-reactive bands were normalized relative to the intensity of the bands corresponding to β-actin. The results, which are illustrated in Figure 2, indicate that the control and ulcerative colitis subjects have mean Pellino3 expression levels which are more than 40 times and more than 20 times, respectively, greater than the mean expression levels in the Crohn's disease subjects.