Detection of celiac disease-specific autoantibodies

Abstract

The invention relates to cell-assembled matrices, which resemble the endomysium and which are not taken from animal or human tissue, and which comprise antigen for celiac disease-specific antibodies; as well as solid phase matrix preparations comprising said cell-assembled matrices. Furthermore, methods for producing the cell-assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix are also disclosed. In addition, uses of the cell-assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix in a method for detecting celiac disease-specific antibodies in a sample or in a method for the in vitro diagnosis or evaluation of celiac disease in a subject are also disclosed.

Description

[0001] Detection of celiac disease-specific autoantibodies

[0002] FIELD OF THE INVENTION

[0003] The invention relates to cell-assembled matrices which resemble the endomysium (i.e. cell-assembled matrices comprising in vitro produced reticulin fiber network organized into honeycomb pattern close to the morphology of endomysium) and which are not taken from animal or human tissue, and which comprise in said honeycomb pattern antigen for celiac disease-specific antibodies; as well as solid phase matrix preparations comprising said cell-assembled matrices. Furthermore, methods for producing the cell-assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix are also disclosed. In addition, uses of the cell-assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix in a method for detecting celiac disease-specific antibodies in a sample or in a method for the in vitro diagnosis or evaluation of celiac disease in a subject are also disclosed.

[0004] TECHNICAL BACKGROUND

[0005] Celiac disease (CD) is a systemic autoimmune disorder induced in 1.5-2% of the Caucasian populations by the consumption of gluten-containing cereals, wheat, rye and barley. Gluten-derived gliadin peptides activate gliadin-specific T lymphocytes which then provide help to plasma cells to produce antibodies against a number of gliadin fractions and an autoantigen, tissue (type-2) transglutaminase (TG2). TG2 is a protein-cross linking enzyme which can use also gliadins as substrate, and this intimate binding is regarded as the basis for the generation of TG2-specific autoantibodies by a hapten-carrier mechanism when TG2-specific B lymphocytes act as antigen presenting cells. Since the antigen presentation occurs at mucosal surfaces, CD antibodies are dominantly produced in IgA class and they also appear in the circulation as diagnostic biomarkers, whereas IgM class antibodies are also produced locally, but they rarely circulate. IgG class antibodies are the relevant diagnostic antibodies in subjects with IgA deficiency. Gliadin (gluten-directed) antibodies have little diagnostic value, since they are not found in all celiac patients, but are often positive in non-diseased persons as well [Dios et al. 2022],

[0006] In CD, the generated immune reaction and inflammation leads to villous atrophy in the small bowel and to organ dysfunctions at various extraintestinal sites. Clinical symptoms thus include malabsorption, abdominal distension, bloating, abdominal pain, iron and vitamin deficiencies, and osteopenia or osteoporosis. Elevated liver enzymes, hepatic failure, myopathy, cardiomyopathy, glomerular and tubular renal dysfunction, infertility, neuropathy and cerebellar ataxia are also known extraintestinal manifestations. An itchy skin disease is associated with CD, called dermatitis herpetiformis, where IgA antibodies deposit in the papillary dermis of the skin. Type- 1 diabetes mellitus and autoimmune thyroid diseases are also associated with CD. IgA autoantibody deposition can be found in the affected organs along endomysial and reticulin structures or in close relation to vessel walls [Korponay-Szabo et al. 2004, Myrsky et al. 2009], corresponding to TG2 exposure in tissues. Most clinical symptoms are reversible (except diabetes, hypothyreosis and ataxia with permanent cell loss) after a gluten elimination diet, which also leads to the cessation of autoantibody production and restoration of the small bowel villous structure. However, new exposure to gluten activates the immune reaction, albeit the spectrum of clinical symptoms may vary. In general, measurement of IgA class celiac autoantibodies from the blood is the best indicator of disease activity and this tool has prominent role both in the initial diagnosis [Husby et al. 2020] and in the medical surveillance of patient condition during treatment [Mearin et al. 2023], Moreover, detection of celiac antibodies also reveals subclinical cases among relatives of CD patients and in the general population. Celiac antibodies can be detected with a number of immunoassays, of which the most specific is the so- called endomysium antibody (EMA) assay [Giersiepen et al., 2012], which offers the celiac autoantigen in its natural context with the exposure of the disease -relevant conformational TG2 epitopes, mainly epitope-2, targeted by approximately 90% of circulating TG2 -directed autoantibodies [Dios et al., 2022], Traditionally, frozen-cut sections from normal tissues containing smooth muscle structures (such as monkey esophagus or human umbilical cord) are used for this test, which, after incubation with appropriately diluted serum or plasma samples bind the celiac antibodies in a distinct pattern along endomysium, the outer sheet of muscular structures (Figure 1A-C). The resulting honeycomb binding pattern is very specific for CD and can be visualized by microscopy by adding anti-IgA secondary antibodies labelled with fluorophores, most commonly fluorescein isothiocyanate (FITC) or alternatives (rhodamine, Texas-red or various Alexa Fluor dyes). Earlier studies have shown that the celiac antigen in the endomysium is extracellularly localized TG2 bound to fibronectin and collagens [Ladinser et al., 1994; Korponay-Szabo et al., 2004; Cardoso et al., 2015],

[0007] The EMA test has been introduced into CD diagnosis in 1984 and since then, it remained a highly appreciated etalon test in clinical practice due to its very easy procedure and excellent specificity and diagnostic performance in experienced laboratories. Currently it is mainly applied as a second, confirmatory test, and forms part of the ESPGHAN 2020 criteria for the non-biopsy CD diagnosis in children and adolescents and of equivalent guidelines used also for adults in some countries, e.g. Finland. However, in the last years, a shortage of tissue section sources emerged. The classical substrates for the EMA test are monkey esophagus and human umbilical cord, since only primate tissues contain the major celiac epitope (epitope 2) of TG2, the active antigenic component of endomysium, and thus the EMA substrate cannot be replaced without loss of sensitivity by mouse or rodent antigens which only contain epitope 1 and other minor epitopes but not the human-specific epitope 2 [Sblattero et al., 2002], Celiac serum antibodies have in approximately 90% epitope 2 specificity and only a minor fraction is directed to epitope 1 and other less important TG2 epitopes [Dios et al. 2022], Moreover, the conformational structure of epitope 2 [Simon-Vecsei et al. 2012] necessitates the use of unfixed frozen sections. Common fixatives and embedding procedures destroy the antigenicity of epitope 2, therefore attempts to convert the EMA assay to peroxidase or other enzyme-based immunohistochemistry methods remained unsuccessful. Since monkeys are endangered species, companies providing monkey tissue sections replaced esophagus with other monkey tissues (uterus, urinary bladder or other muscle structures) from which sections can be made more abundantly, or, mostly, simply stopped production. Human tissues have never been sold commercially for ethical reasons and uncertain supply, moreover, centers cutting home-made human umbilical cord sections often have difficulties in the evaluation due to frequent non-specific reactions caused by anti-smooth muscle and other common non-celiac autoantibodies. For this reason, lack of available tissue substrates may compromise the adherence to guidelines in the everyday diagnostic workup.

[0008] The EMA test is a very simple and robust method, but requires for the reasons mentioned above frozen sections, fluorescent microscopy equipment and expertise in reading the results. Further, the throughput is not very high and the EMA testing provides only semi -quantitative results by titration. Therefore, immunoassays applying purified or recombinant human TG2 are now also in use for the initial testing and screening for CD. Although initial detection and quantitation of anti-TG2 antibodies can be performed on a larger scale by higher throughput or automated methods, such as ELISA or ELIA assays [Sulkanen et al., 1998a], the obtained numerical results differ between tests from different manufacturers, i.e. quantitative values are not comparable between commercial kits due to the lack of standardisation, and positive predictive value for the final CD diagnosis based on histology reached >99% values in prospective studies only at much higher concentration results than the manufacturers’ declared cut-off [Werkstetter et al. 2017], In line with this, the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) still requires in its most recent (2020) diagnostic guidelines for CD IgA TG2 antibody serum concentration above 10 times of the upper limit of normal (>10xULN) plus a positive EMA test result from a separately drawn blood specimen for the final CD diagnosis, if it is intended to be made with the omission of endoscopy and small bowel histology [Husby et al. 2020], In fact, endoscopy in children requires general anaesthesia and hospital stay, and it is much more often refused by the parents (and nowadays often by adult patients as well) than in previous times. Therefore, the non-invasive diagnosis is currently preferred both by medical professionals and patients, if celiac antibodies in serum are high and their presence is confirmed by a method different from ELISA or ELIA, i.e. by an independent tissue-based EMA test, which enables the visual identification of patient antibodies bound to the specific TG2 tissue target sites in a distinct pattern. This approach tries to minimize two common risks for false life-long diagnosis of CD: potential mix-up of laboratory samples and non-specific sticking of antibodies to recombinant TG2s. In fact, many autoimmune patients have sticky IgA antibodies that also bind to plastic surfaces without antigen [Sulkanen et al. 1998a], may recognize the non-coding tag of the recombinant protein [Csipo et al. 2012] or bind to TG2 at other epitopes distinct from the celiac ones [Simon-Vecsei et al. 2012], TG2 has several conformations important for celiac autoantibody recognition, and recombinant TG2 proteins are not always true imitators of the natural autoantigen in this respect due to poor folding and even mismatching amino acid sequences. It should be emphasized that the celiac epitopes of TG2 are highly dependent on the correct folding, stability and good quality of the protein needed for specific results. Currently, no quality control is in force for recombinant TGs, although it is known that proteins produced in E. coli have less stable conformation than those made in insect cells. In addition, early cloning errors are still causing that many commercial TG2 proteins have at position 224 glycine, whereas the real TG2 proteins in humans contain valine according to recent sequence databases. The224Val TG2 protein has a superior antigenicity [Kanchan et al. 2013] and such changes may lead to different results. Further, non-primate animal TG2s may be defective in their epitope composition. Although the EMA test requires fluorescent microscopy and some expertise in the reading, it provides natural TG2 and high specificity, and therefore it is particularly suited for the confirmation of correct celiac antibody positivity.

[0009] A shortage of EMA tissue section sources has recently emerged, since monkeys are endangered species and most manufacturers stopped selling monkey tissue sections. Possible alternatives, such as human umbilical cord [Ladinser et al., 1994] or normal appendix have long been applied by many clinical centers capable of cutting frozen sections (Figures 1C and ID), but their evaluation is more difficult due to frequent nonspecific reactions, and increasing ethical concerns for using human tissues may restrict also their use in the future.

[0010] To improve the applicability of the EMA test, we explored the exposure and distribution of the natural TG2 in tissues. TG2 is present in all cells but in different amounts and different cell types externalize the protein in variable amounts. It is known that the intracellular TG2, retained in closed, enzymatically inactive conformation by high levels of GDP and low concentrations of intracellular calcium, is not a good antigen for the celiac antibodies which bind to extracellular matrix components containing TG2 in calcium-activated conformation. Staining studies have shown that celiac antibodies bind to TG2 on the extracellular surface of basement membrane laminin [Korponay-Szabo et al. 2004] in endomysial structures. TG2 is present extracellularly on the surface of fibronectin [Korponay-Szabo et al. 2000], having a specific binding site for fibronectin, and both are components of reticulin fibers identifiable by silver-impregnation staining, the connective tissue network of parenchymal organs, to which celiac antibodies bind well. Before the identification of TG2 as the main autoantigen for celiac disease, this binding was also known as „reticulin” antibodies (ARA), but studies already early on confirmed the identity of EMA and ARA by competition studies and by the use of TG2 -depleted and TG2- / - tissues, which have no celiac antigenicity despite being organized into endomysial or reticulin structures morphologically [Korponay- Szabo et al. 2000 and 2003] . In consequence, the endomysium can be regarded as a TG2-rich fine reticulin network around muscle fibers and around cells in general and has a honeycomb appearance in sections when visualized for example by silver impregnation [Gdmori, 1937; Krutsay, 1988; Hallstrom, 1989], Further, the classical EMA binding pattern of celiac antibodies includes the staining of reticulin fibers in the subpapillary region (Figure IB) and in mucosal lymphatic follicles of the monkey esophagus as obligatory non-muscular components of the positive EMA reaction. Of note is that celiac antibodies, however, do not bind to the submucosal connective tissue of the esophagus or of other tissues, which represent common extracellular tissue in the body made up mainly from collagen I, nor to fibronectin in certain tissues (e.g. cartilage [Korponay-Szabo et al. 2000]) which does not comprise TG2.

[0011] Celiac antibodies were shown to deposit in the tissues of patients along small blood vessels, with a prominent staining of myofibroblasts’ cell bodies also in cell culture. It was attempted to use human umbilical cord (myo)fibroblasts from the Wharton jelly cultured in ELISA wells to measure celiac antibody binding to their cell bodies for diagnostic purposes by densitometry in ELISA [Sulkanen et al. 1998b]. However, the results had lower specificity compared to the EMA testing, because patient serum samples often contained also other, nonceliac antibodies against cellular components. In addition, such cells are highly differentiated and do not proliferate in culture, thus it was difficult to reproduce the results in larger scale. In summary, this way of celiac antibody detection did not prove useful for clinical diagnostics, because this system did not offer endomysial structures for the testing.

[0012] Whelan et al. (1996) detected IgA and IgG celiac antibodies by indirect immunofluorescence employing trypsinized cell suspension from human umbilical vein endothelial cells (HUVECs) dried on poly-L-lysine-coated glass slides. They measured intracellular celiac antigens, for which the HUVEC cells were fixed in paraformaldehyde. They found a cytoplasmic reactivity on these cells, while no surface positivity was obtained by flow cytometry analyses. The same results were claimed with similarly treated human umbilical cord artery endothelial cells, human umbilical cord artery smooth muscle cells and human umbilical cord vein muscle cells. Thus, while they demonstrated that HUVEC cells possess antigens that react with IgA celiac antibodies, they suggested that these antigens are found intracellularly, i.e. inside the cells. Furthermore, their test system did not contain adherent cells cultured on a solid surface and they did not evaluate extracellular matrix components or structures. Further, they maintained endothelial cells in RPMI medium and not in endothelial cell media. The same method was used by Feighery et al. (1998) who compared it with endomysial IgA antibody detection on monkey esophagus and on human umbilical cord tissue sections in a larger clinical patient material.

[0013] Assuncao et al. (2020) provide a brief overview over the methodologies used to facilitate the deposition and manufacturing of cell-derived extracellular matrices. They state that extracellular matrix (ECM) has a composition specific for each tissue. It is comprised of a complex and highly organized three-dimensional macromolecular network of biomolecules. These include fibrous proteins (such as collagens) and glycosaminoglycan-based components. They also state that it can be decellularized, but this publication does not teach how a celiac antigenic matrix, being distinct from common extracellular matrices and mimicking endomysial substrate tissue can selectively be produced or its production induced. Carvalho et al. (2019) teach a strategy to produce decellularized cultured cell-derived ECM (dECM) obtained from mesenchymal stem / stromal cells (MSCs) and human umbilical vein endothelial cells (HUVECs), as well as the co-culture of MSC:HUVEC and investigate the effects of its various compositions on cell metabolic activity, osteogenic differentiation, and angiogenic properties of human bone marrow (BM)-derived MSCs which are vital features for adult bone tissue regeneration and repair. The authors have found that ECM from co-cultures of MSCs and HUVECs have promising effect suggesting the potential use of MSC:HUVEC ECM as a therapeutic product to improve clinical outcomes in bone regeneration. In contrast to the MSC:HUVEC co-culture, HUVECs alone maintained according to their protocol (i.e. throughout in EGM2 medium) neither produced an organized fibrillar network, nor exposed substantial amounts of fibronectin on their ECM as the authors show in their published figure (figure 2 in that paper) and state in the text, and MSC alone produced an ECM rich in collagen I. Carvalho et al. are absolutely silent about using HUVEC ECM in celiac disease diagnosis or exposing TG2 as an antigen.

[0014] WO98 / 15836A1 discloses a sandwich-type method for diagnosing celiac disease, based on the measurement of anti-endomysial human serai IgA, which may be present in the serum of an individual. This method comprises the successive stages of: a) immobilizing an antigen which reacts with anti-endomysial antibodies on a solid phase; b) bringing the solid phase obtained in a) into contact with serum which may contain anti-endomysial human serai IgA; c) bringing the solid phase obtained in b) into contact with an aqueous phase containing an anti-human IgA antibody bound to a marker (preferably an enzyme); d) detecting the presence of this marker in the solid phase obtained in c). The antigen is prepared from homogenized human umbilical cord tissue, and examples are shown for ELISA (i.e. wherein the marker is an enzyme).

[0015] W02003 / 008976A2 suggests a method for diagnosing celiac disease, by an ELISA assay, in a patient comprising (a) providing an antibody sample from the patient to be tested, (b) contacting said antibody sample with cross-linked extracellular matrix (ECM) material immobilized on a surface and (c) detecting binding of antibodies present in the sample to the ECM-coated surface, wherein the binding of antibodies present in the sample to the ECM-coated surface being indicative of a positive diagnosis.

[0016] However, their ELISA assay is not a reticulin fiber network based assay and the assay is not tissuetransglutaminase 2 (TG2) dependent.

[0017] While examples are shown wherein the ECM is produced by 3T3 Swiss mouse embryo fibroblasts transfected with a nucleic acid molecule encoding TG2, or human foreskin dermal fibroblasts, the authors emphasized that their assay focuses on antigens other than TG2.

[0018] The authors have found a very low difference, if any, in the reactivity of TG2 expressing and non-expressing fibroblast derived ECM, and state that transglutaminase is not a significant antigen in their assay in the diagnosis of celiac disease (see their Example 1). For this reason, this cross-linked extracellular matrix material would not be a suitable substrate for the EMA assay, because in the EMA reaction, the only antigen is the TG2 [Korponay- Szabo et al., 2003], thus the cross-linked extracellular matrix is not specific for celiac disease. Furthermore, the authors do not mention either that endothelial cells (e.g. HUVECs) could be suitable in the preparation of a cell assembled matrix in the diagnosis of celiac disease, in particular in an EMA-like assay, which is based on a visually recognizable diagnostic pattern.

[0019] W02010 / 113025A2 discloses a diagnostic method for gluten-induced autoimmune disease (celiac disease), in which antibodies of a subject are contacted with a reference transglutaminase protein with intact main celiac epitope and with at least one test transglutaminase protein with an altered main celiac epitope, and if the antibodies bind to the reference protein but not the test protein, this indicates the gluten-induced autoimmune disease in the subject. They state that celiac-specific antibodies recognize TG2 in the form as it is bound to the surface of fibronectin in muscular tissue section substrates used for the endomysial antibody detection by immunofluorescence. However, the authors did not disclose or suggest any cell-assembled matrices.

[0020] EP0912898B1 discloses a method for the diagnosis or therapy control of sprue or celiac disease, characterized in that antibodies against tissue transglutaminase (which is also called type 2 transglutaminase or TG2) from body fluids are detected by means of an immune reaction with tissue transglutaminase, the immunoreactive sequences or analogues thereof, wherein the immune reaction is not carried out with a tissue section of animal or human tissue. The identification studies of the TG2 antigen were carried out on various fixed cell lines. However, none of the cell lines were endothelial cells, and, in addition, the cytoplasmic signals were analyzed.

[0021] None of the above-mentioned publications disclose a matrix that is not derived from animal or human tissue, and that can be used as a substrate for detecting celiac disease-specific antibodies in indirect immunofluorescence methods, i.e. that can replace the traditionally used monkey esophagus tissue or human umbilical cord tissue sections.

[0022] The inventors have created a cell -culture assembled (bio)matrix which, surprisingly, is suitable for EMA- test of celiac disease instead of traditionally used monkey esophagus tissue or human umbilical cord tissue sections. The (bio)matrix was also prepared on a support and provided high specificity and substantially improved sensitivity for EMA-IgA detection. The inventors’ results demonstrate the high diagnostic performance of the new EMA test using “artificial endomysium”. Thus, the EMA testing on the novel cell-culture assembled (bio)matrix substrate had high diagnostic accuracy for CD and was even more sensitive than EMA testing on tissues.

[0023] BRIEF DESCRIPTION OF THE INVENTION

[0024] The invention relates to a cell-assembled matrix (“CAM”, may also be called a cell-culture assembled (bio)matrix or cell-assembled extracellular matrix or artificial endomysium) produced in vitro which has a reticulin fiber network comprising fibronectin, and which comprises, bound to the fibronectin surface, antigen for celiac disease-specific antibodies (celiac antigen).

[0025] The invention also relates to a use of an in vitro cell culture derived cell-assembled matrix in a method for detecting celiac disease-specific antibodies in a sample, said cell-assembled matrix comprising a reticulin fiber network comprising fibronectin, and comprising, on the surface of fibronectin, antigen for celiac disease-specific antibodies (celiac antigen).

[0026] Preferably, the cells of the in vitro cell culture are cells which are capable of producing endomysium-like matrix and / or a reticulin fiber network, preferably under specific culturing conditions or with a long culturing time.

[0027] In a highly preferred embodiment, the cell-assembled matrix is an endothelial cell culture derived extracellular matrix, and particularly preferably the cell -assembled matrix is derived from a cell culture of human umbilical cord vein-derived endothelial cells (HUVECs).

[0028] Preferably the celiac antigen is type 2 transglutaminase (TG2). More preferably, the celiac antigen is a celiac epitope of TG2. Preferably, the cell-assembled matrix comprises intact celiac epitopes of TG2, i.e. the integrity of celiac epitopes of TG2 is maintained.

[0029] Preferably the CAM is used in the in vitro diagnosis or evaluation of celiac disease in a subject. Preferably the CAM is used in a method for detecting celiac disease-specific antibodies according to the invention.

[0030] The celiac antigen is, in particular, type 2 transglutaminase (TG2). Preferably, the celiac antigen is a celiac epitope of TG2. Preferably, the cell-assembled matrix comprises intact celiac epitopes of TG2.

[0031] The cell-assembled matrix of the invention has endomysial characteristics, i.e. it is an endomysium-type (or endomysium-like) cell-assembled matrix mimicking and / or modelling the structure of natural endomysium in tissue sections.

[0032] In particular, the cell-culture assembled matrix is an artificial endomysium, modelling the structural and macromolecular composition features of tissue endomysium.

[0033] Thus, the cell-culture assembled matrix is not derived from a tissue, for example it is not derived from animal or human tissue.

[0034] Preferably, the reticulin fiber network is a morphologically recognizable reticulin fiber network.

[0035] In a preferred embodiment, the endomysial characteristics or features of the endomysium-type CAM, preferably the endomysium-like character of the matrix can be tested by any method useful for testing endomysial features, in particular by the classical silver impregnation method.

[0036] In fact, endomysial features are not limited to the endomysium (traditionally muscle endomysium); however, endomysial features may be present in any tissue and in turn tissue preparation or tissue slice which comprises a honeycomb-pattern reticulin fiber network.

[0037] In the present invention, in particular, the cell-assembled matrix, usefiil as an artificial endomysium comprising the reticulin fiber network comprises fibronectin which in the present invention has, bound to the fibronectin surface, celiac antigen, preferably type 2 transglutaminase.

[0038] The cell-assembled matrix of the invention is a cell -culture derived (or cell-culture produced) matrix. Thus, the matrix is prepared from cell-culture, preferably endothelial cell culture and is not prepared from animal or human tissue.

[0039] In a particular embodiment, the reticulin fiber network comprises fibronectin and TG2.

[0040] In a particular embodiment, the cell-assembled matrix is an endothelial cell culture derived extracellular matrix, preferably an endomysium-type or endomysium-like matrix, organized into a reticulin fiber network.

[0041] Preferably the CAM, preferably the CAM organized into a reticulin fiber network comprises type III collagen, in particular COL3A1, fibronectin, laminin, and preferably type IV collagen. Preferably it comprises at most a low level of type I collagen, preferably it is essentially free of type I collagen.

[0042] Preferably, the cell-assembled matrix resembles an endomysial structure or the endomysium and is not derived from animal or human tissue, and comprises antigen for celiac disease -specific antibodies.

[0043] Preferably said CAM has a mesh-like appearance, preferably a honeycomb-like appearance, when visualized by silver impregnation or by any other appropriate method for visualization of such fiber network.

[0044] The cell-assembled matrix (CAM) or artificial endomysium may also be called a cell-assembled extracellular matrix, or a biomatrix. It may be an antigenic biomatrix, a specialized reticulin-type extracellular matrix (ECM), an endomysium-type (or endomysium-like) ECM, an ECM with the characteristics of endomysial structures, an ECM resembling an endomysial network, an ECM with celiac antigenicity and / or a cell culture.

[0045] Preferably, the cell-assembled matrix has endomysial characteristics or resembles an endomysial structure or the endomysium and is not derived (i.e. taken or prepared) from animal or human tissue, and which comprises antigen for celiac disease-specific antibodies. Preferably, the cell-culture assembled matrix is modelling the structural and macromolecular composition features of tissue endomysium and can be called artificial endomysium.

[0046] The cell-assembled matrix is different from an animal or human tissue section.

[0047] In a particular embodiment, the endomysium-type matrices comprise collagen type III, in particular COL3A1; preferably fibers composed of type III collagen.

[0048] In alternative wording, the endomysium-like matrices comprise granulation tissue type collagen.

[0049] Alternatively, the endomysium-like matrices comprise reticular fibers or a fiber network of reticulin type.

[0050] In a particular embodiment, the fiber network comprising fibronectin is argyrophilic and thus the endomysium-like characteristic can be shown by silver staining of the matrix.

[0051] In a particular embodiment, the endomysium-type characteristic of the matrix can be shown by i) removing the type 2 transglutaminase from the fibronectin and optionally ii) administering type 2 transglutaminase or a tissue or culture comprising type 2 transglutaminase. Each of steps i) and ii) can be monitored by antibodies, e.g. fibronectin binding and type 2 transglutaminase binding antibodies, respectively. Type 2 transglutaminase can be removed e.g. by potassium thiocyanate (KSCN).

[0052] Preferably, the cell-assembled matrix is produced by endothelial cells, preferably by primary endothelial cells or immortalized cells derived therefrom, more preferably by human umbilical cord vein-derived endothelial cells (HUVECs).

[0053] Preferably, in any of the cell-assembled matrices above, the antigen for celiac disease-specific antibodies is a transglutaminase, preferably type 2 transglutaminase (TG2) or any (celiac) epitope thereof.

[0054] Preferably, the cell-assembled matrix is useful for detecting celiac disease-specific antibodies in a sample.

[0055] Preferably, the celiac disease-specific antibodies are endomysial antibodies or anti-endomysium antibodies (EMAs) or anti-transglutaminase antibodies, preferably TG2-specific antibodies or autoantibodies. Preferably, the antibodies are IgA and / or IgG antibodies.

[0056] Preferably, the cell-assembled matrix is used in an endomysium antibody (EMA) assay as an EMA substrate, preferably as an artificial endomysium or instead of a tissue section.

[0057] Preferably, the cell-assembled matrix is an organized fiber network which comprises reticulin fibers containing collagen III, collagen IV, fibronectin and laminin, and wherein type 2 transglutaminase (TG2) is present on the surface of fibronectin, and it has low or absent collagen I content that characterize other types of connective tissues and extracellular matrices. More preferably, the cell-assembled matrix is not derived from a tissue, for example it is not derived from animal or human tissue.

[0058] In particular, the cell-assembled matrix shows high similarity to the tissue endomysium morphologically and in its macromolecular composition. More specifically, the cell-assembled matrix shows high degree of similarity to tissue endomysium composed of reticulin fiber patterns and comprising macromolecules collagen III, collagen IV, fibronectin, laminin and TG2.

[0059] Preferably, any of the cell-assembled matrices above show a honeycomb binding pattern with celiac disease-specific antibodies (preferably anti-TG2 IgA and / or anti-TG2 IgG antibodies), and / or the cell-assembled matrix has a similar morphology than that of the endomysium and / or the cell-assembled matrix is an organized, preferably highly organized matrix.

[0060] Preferably, in any of the cell-assembled matrices above, the matrix comprises the cells that produced it, or the cells have been removed from the matrix. More preferably, the cell-assembled matrix either comprises only one type of cell, i.e. the endothelial cells (e.g. HUVECs) that produced it, or it does not comprise any cells. In one embodiment, the cell-assembled matrix comprises the cells that produced it, wherein preferably the cells are endothelial cells, preferably primary endothelial cells or their immortalized descendent cell lines, more preferably human umbilical cord vein-derived endothelial cells (HUVECs). More preferably, the cells are HUVECs which retain their endothelial characteristics even in the culture stage with mature endomysium-type matrices expressing vascular endothelial (VE)-cadherin and do not show transition to other mesenchymal cell types and do not express muscle cell markers, like desmin. In an embodiment, the cells that produce the cell- assembled matrix are of a primary endothelial cell derived immortalized cell line, which retains the morphological properties of primary endothelial cells and express VE-cadherin and optionally von Willebrand factor and CD31.

[0061] In another embodiment, the cell-assembled matrix is a decellularized matrix, i.e. it does not comprise the cells that produced said matrix.

[0062] Preferably, any of the cell-assembled matrices above, which comprises the cells that produced it, is an organized, TG2-rich extracellular matrix around primary endothelial cells, more preferably around HUVEC cells.

[0063] Preferably, any of the cell-assembled matrices above, irrespective of whether they comprise the cells that produced it or not, is capable of binding celiac disease-specific antibodies - preferably anti-TG2 IgA and / or anti- TG2 IgG antibodies - in a mesh-like binding pattern, preferably in a honeycomb binding pattern. More preferably, the honeycomb binding pattern corresponds to the argyrophilic network detectable by silver staining.

[0064] Preferably, any of the cell-assembled matrices above form a 2-dimensional structure.

[0065] Any of the cell-assembled matrices comprising the cells that produced it may be a monolayer culture.

[0066] Preferably, any of the cell-assembled matrices above is suitable for use in an immunofluorescence method, preferably in an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies.

[0067] In another embodiment, any of the decellularized cell -assembled matrices defined above is suitable for use in an immunoassay, for example in an enzyme-linked immunosorbent assay (ELISA) or in an immunofluorescence method, preferably in an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease -specific antibodies.

[0068] The invention also relates to a solid phase matrix preparation which comprises any of the cell- assembled matrices defined above, or elsewhere in the description, on a support, preferably prepared on a support.

[0069] Thus, in particular, the cell -culture assembled (bio)matrix or cell-assembled extracellular matrix produced in vitro has a reticulin fiber network comprising fibronectin, and which comprises, bound to the fibronectin surface, antigen for celiac disease-specific antibodies (celiac antigen), in particular, type 2 transglutaminase (TG2) or any (celiac) epitope thereof.

[0070] The invention also relates to a use of the solid phase matrix preparation in a method for detecting celiac disease-specific antibodies in a sample, said cell-assembled matrix comprising a reticulin fiber network comprising fibronectin, and comprising, on the surface of fibronectin, antigen for celiac disease -specific antibodies (celiac antigen). Preferably the celiac antigen is type 2 transglutaminase (TG2) or any (celiac) epitope thereof.

[0071] Preferably the solid phase matrix preparation is used in the in vitro diagnosis or evaluation of celiac disease in a subject. Preferably the solid phase matrix preparation is used in a method for detecting celiac disease -specific antibodies according to the invention.

[0072] Preferably, the solid phase matrix preparation is used in an EMA assay. The cell-assembled matrix of the invention is an endomysium-type (or endomysium-like) cell-assembled matrix mimicking and / or modelling the structure of natural endomysium in tissue sections.

[0073] Thus, in particular, the cell -culture assembled matrix is an artificial endomysium, modelling the structural and macromolecular composition features of tissue endomysium which is not derived from a tissue, for example it is not derived from animal or human tissue.

[0074] In a preferred embodiment, the endomysial characteristics or features of the endomysium-type CAM, can be tested by any method useful for testing endomysial features, in particular by the classical silver impregnation method. In fact, endomysial features may be present in any tissue and in turn tissue preparation or tissue slice which comprises a reticulin fiber network. In particular, the cell-culture assembled matrix comprises a reticulin fiber network comprising fibronectin which has, bound to the fibronectin surface, celiac antigen, preferably type 2 transglutaminase.

[0075] The cell-assembled matrix in the solid phase matrix preparation is a cell-culture derived (or cell-culture produced) matrix. Thus, the matrix is prepared from cell -culture, preferably endothelial cell culture and not from animal or human tissue and therefore is different from an animal or human tissue section.

[0076] In a particular embodiment, in the solid phase matrix preparation, the endomysium-type (or endomysium- like) matrices comprise collagen type III, in particular COL3A1; preferably fibers composed of type III collagen, in other words, granulation tissue type collagen. In a preferred embodiment, the matrix comprises at most a low level of type I collagen, preferably it is essentially free of type I collagen.

[0077] Alternatively, the endomysium-like matrices comprise reticular fibers or a fiber network of reticulin type.

[0078] In a particular embodiment, the fiber network comprising fibronectin is argyrophilic and thus the endomysium-like characteristic can be shown by silver staining of the matrix.

[0079] In a particular embodiment, the endomysium-type characteristic of the matrix can be shown by i) removing the type 2 transglutaminase from the fibronectin and optionally ii) administering type 2 transglutaminase or a tissue or culture comprising type 2 transglutaminase. Each of steps i) and ii) can be monitored by antibodies, e.g. fibronectin binding and type 2 transglutaminase binding antibodies, respectively. Type 2 transglutaminase can be removed e.g. by potassium thiocyanate (KSCN).

[0080] Preferably, the cell-assembled matrix in the solid phase matrix preparation is produced by endothelial cells, preferably by primary endothelial cells or immortalized cells derived therefrom, more preferably HUVECs.

[0081] Preferably, in any of the cell-assembled matrices above, the antigen for celiac disease-specific antibodies is a transglutaminase, preferably type 2 transglutaminase (TG2) or any (celiac) epitope thereof. Preferably, celiac disease-specific antibodies or autoantibodies, more preferably TG2 -specific IgA and / or IgG antibodies can bind to the antigen, preferably to TG2.

[0082] Preferably, the cell-assembled matrix is an organized fiber network which comprises reticulin fibers, containing collagen III, collagen IV, fibronectin and laminin, and wherein type 2 transglutaminase (TG2) is present on the surface of fibronectin, and it has low or absent collagen I content that characterize other types of connective tissues and extracellular matrices. More specifically, the cell-assembled matrix shows high degree of similarity to tissue endomysium composed of reticulin fiber patterns and comprising macromolecules collagen III, collagen IV, fibronectin, laminin and TG2.

[0083] Preferably, any of the cell-assembled matrices above show a mesh-like binding pattern, preferably a honeycomb binding pattern with celiac disease-specific antibodies (preferably anti-TG2 IgA and / or anti-TG2 IgG antibodies), and / or the cell-assembled matrix has a similar morphology than that of the endomysium and / or the cell-assembled matrix is an organized, preferably highly organized matrix.

[0084] Preferably, in any of the cell-assembled matrices above, the matrix comprises the cells that produced it, or the cells have been removed from the matrix. More preferably, the cell-assembled matrix either comprises only one type of cell, i.e. the endothelial cells (e.g. HUVECs) that produced it, or it does not comprise any cells.

[0085] In another embodiment, the cell-assembled matrix is a decellularized matrix, i.e. it does not comprise the cells that produced said matrix.

[0086] Preferably, any of the cell-assembled matrices above, irrespective of whether they comprise the cells that produced it or not, is capable of binding celiac disease -specific antibodies - preferably anti-TG2 IgA and / or anti- TG2 IgG antibodies - in a mesh-like binding pattern, preferably in a honeycomb binding pattern. More preferably, the honeycomb binding pattern corresponds to the argyrophilic network detectable by silver staining.

[0087] Preferably, any of the cell-assembled matrices above form a 2-dimensional structure.

[0088] Any of the cell-assembled matrices comprising the cells that produced it may be a monolayer culture.

[0089] Preferably, the support is a slide, a chamber slide, a cover slip or a plate. Preferably, the solid phase matrix preparation prepared on the support is suitable for use in an immunoassay method, preferably in an enzyme-linked immunosorbent assay (ELISA) or in an immunofluorescence method, more preferably in an immunofluorescence method, more preferably an indirect immunofluorescence method, more preferably an endomysium antibody (EMA) assay, and / or the solid phase matrix preparation is suitable for use with a fluorescence microscope.

[0090] Preferably, the solid phase matrix preparation is prepared on a support to which endothelial cell (preferably primary endothelial cells, more preferably HUVECs) can adhere. More preferably, the support is a support to which endothelial cells (preferably primary endothelial cells, more preferably HUVECs) can adhere under certain seeding and culturing conditions. Examples for such conditions are described below.

[0091] Preferably, the support is treated physically to be adhesive to cells.

[0092] Preferably, the support has a glass surface or a plastic surface. Preferably, the glass surface or plastic surface is uncoated or is not coated with proteins or with polypeptides. More preferably, the support is not coated with gelatin or fibronectin.

[0093] Preferably, the support has an acetone-resistant surface. Preferably, the support is an acetone-resistant slide, an acetone-resistant chamber slide, an acetone-resistant cover slip or an acetone-resistant plate. More preferably, the support has a surface that is hydrophilic, adhesive to cells and / or slightly negatively charged. More preferably, the support is a product with an ibiTreat surface (ibidi®, Grafeling, Germany) manufactured from acetone-resistant plastic.

[0094] Preferably, the support has an uncoated glass or ibiTreat plastic surface and / or its surface is not coated with proteins or with polypeptides, preferably not coated with either fibronectin or gelatin.

[0095] Preferably, the support is a glass slide, a glass chamber slide, a glass cover slip or a glass plate, and which is not coated with proteins or with polypeptides, preferably not coated either with fibronectin or with gelatin. Preferably, the support is an uncoated glass slide, an uncoated glass chamber slide, an uncoated glass cover slip or an uncoated glass plate.

[0096] Preferably, the support is a slide, a chamber slide, a cover slip or a plate with an ibiTreat plastic surface.

[0097] Preferably, the cell-assembled matrix is fixed to the support to form a solid phase matrix preparation. More preferably, the cell-assembled matrix fixed to the support can still bind antibodies of positive celiac control samples (the matrix shows a positive extracellular staining with a positive celiac control sample), i.e. the celiac epitopes of the TG2 present in the cell-assembled matrix have not been destroyed during the fixation. Preferably, the cell- assembled matrix is fixed by a fixative, wherein the celiac epitopes - preferably the integrity of the celiac epitopes - of TG2 are maintained. More preferably, the fixative is acetone or methanol or a solution of methanol and MES buffer.

[0098] The invention also relates to a method for producing a cell-assembled (cell-culture assembled) matrix, wherein said method comprises the following steps: a) seeding cells, preferably endothelial cells on a support, b) culturing the seeded cells of step a) in a medium suitable for culturing said cells, preferably endothelial cells to obtain a culture with the cell -assembled matrix, c) optionally, removing the cells from the culture obtained in step b), thereby producing cell-assembled matrix without cells, d) fixing the culture obtained in step b) or the cell-assembled matrix without cells obtained in step c) to a support with a fixative that does not destroy celiac epitopes of antigens for celiac disease-specific antibodies (in particular, the celiac epitopes of TG2).

[0099] In particular, the cell-assembled matrix is any of the cell-assembled matrices defined above, or elsewhere in the description, or a (bio)matrix in the solid phase matrix preparation provided on a support, e.g. prepared on a support.

[0100] The invention also relates to a method for producing a cell-assembled (cell-culture assembled) matrix, wherein said method comprises the following steps: a) seeding endothelial cells on a support, wherein the endothelial cells are seeded on the support in a medium for culturing endothelial cells, or in a medium comprising VEGF and / or EGF; b) culturing the seeded cells of step a) in a medium selected from:

[0101] - a medium for culturing endothelial cells (e.g. EGM),

[0102] - a medium comprising VEGF and / or EGF, or

[0103] - Medium 199 supplemented with fetal bovine serum (FBS) and with a medium for culturing endothelial cells; c) optionally, removing the cells from the culture obtained in step b), thereby producing an extracellular matrix without cells, d) fixing the culture obtained in step b) or the extracellular matrix obtained in step c) to the support with a fixative that does not destroy celiac epitopes of antigens for celiac disease-specific antibodies (in particular, the celiac epitopes of TG2) present in the extracellular matrix of the culture or in the extracellular matrix, respectively, preferably wherein a fixative does not destroy the celiac epitopes of antigens (in particular, the celiac epitopes of TG2) if, after fixation, the matrix shows a positive extracellular staining with a positive celiac control sample, preferably wherein the fixative is acetone, methanol, a solution of methanol and MES buffer, or PAXgene tissue fix, more preferably wherein the fixative is acetone, methanol or a solution of methanol and MES buffer, e) thereby obtaining a cell-assembled matrix, preferably in the form of a solid phase matrix preparation comprising the cell-assembled matrix, wherein preferably the cell-assembled matrix resembles or modells an endomysial structure or the endomysium and which is not derived (i.e. taken or prepared) from animal or human tissue, and which comprises antigen for celiac disease-specific antibodies, more preferably wherein the cell-assembled matrix is an organized fiber network which comprises reticulin fibers, collagen, fibronectin and laminin, and wherein type 2 transglutaminase (TG2) is present on the surface of the fibronectin, and / or preferably wherein the cell -assembled matrix is capable of binding celiac disease-specific antibodies - preferably anti-TG2 IgA and / or anti-TG2 IgG antibodies - in a honeycomb binding pattern.

[0104] Preferably, in the method for producing a cell-assembled matrix, the endothelial cells are primary endothelial cells, more preferably human umbilical cord vein-derived endothelial cells (HUVECs).

[0105] Thus, although the starting cells were of non-muscle origin, they conserved their endothelial properties throughout the matrix production process as demonstrated by their VE-cadherin positivity in the mature cellcontaining matrix preparates.

[0106] Preferably, the cell-assembled matrix is characterized with low collagen I and high collagen III and IV content, which is unusual under conventional cell culturing conditions directed to produce common ECM and which results in the accumulation of collagen I (the latter defining connective tissue-type matrix).

[0107] In a preferred embodiment, the cells were induced by a specific sequential use of different cell media described in the present invention.

[0108] Preferably, in any of the methods for producing a cell-assembled matrix, the support is a slide, a chamber slide, a cover slip or a plate. Preferably, the produced cell-assembled matrix, preferably in the form of a solid phase matrix preparation prepared on the support, is suitable for use in an immunoassay method, preferably in an enzyme- linked immunosorbent assay (ELISA) or in an immunofluorescence method, more preferably in an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably in an endomysium antibody (EMA) assay, and / or the solid phase matrix preparation is suitable for use with a fluorescence microscope or with a suitable electronic imaging device.

[0109] Preferably, in any of the methods for producing a cell-assembled matrix, the support is a support to which endothelial cell (preferably primary endothelial cells, more preferably HUVECs) can adhere. More preferably, the support is a support to which endothelial cells (preferably primary endothelial cells, more preferably HUVECs) can adhere under certain seeding and culturing conditions. Examples for such conditions are described above [see step a) and step b)] and below.

[0110] Preferably, the support is treated physically to be adhesive to cells.

[0111] Preferably, in any of the methods for producing a cell -assembled matrix, the support has a glass surface or a plastic surface. Preferably, the glass surface or plastic surface is uncoated or is not coated with proteins or with polypeptides before the cells are seeded. More preferably, the support is not coated with gelatin or fibronectin before the cells are seeded. Preferably, the fibronectin comprised in the cell-assembled (cell-culture assembled) matrix is produced by the cells themselves (i.e. fibronectin is not added externally) or, optionally, the cells incorporate fibronectin into the cell-assembled matrix from the serum content of the culture medium, but the support is not coated with it before the cells are seeded. Preferably, in any of the methods for producing a cell-assembled matrix, the support has an acetoneresistant surface. Preferably, the support is an acetone -resistant slide, an acetone-resistant chamber slide, an acetone-resistant cover slip or an acetone-resistant plate. More preferably, the support has a surface that is hydrophilic, adhesive to cells and / or slightly negatively charged. More preferably, the support is a product with an ibiTreat surface (ibidi®, Grafeling, Germany) manufactured from acetone-resistant plastic.

[0112] Preferably, in any of the methods for producing a cell-assembled matrix, the support has an uncoated glass or ibiTreat plastic surface and / or its surface is not coated with proteins or with polypeptides, preferably not coated with either fibronectin or gelatin before the cells are seeded.

[0113] Preferably, the support is a glass slide, a glass chamber slide, a glass cover slip or a glass plate, and which is not coated with proteins or with polypeptides, preferably not coated either with fibronectin or with gelatin. Preferably, in any of the methods for producing a cell -assembled matrix, the support is an uncoated glass slide, an uncoated glass chamber slide, an uncoated glass cover slip or an uncoated glass plate.

[0114] Preferably, in any of the methods for producing a cell-assembled matrix, the support is a slide, a chamber slide, a cover slip or a plate with an ibiTreat plastic surface.

[0115] Preferably, in any of the methods for producing a cell-assembled matrix, the medium for culturing endothelial cells of step a) is a medium for culturing primary endothelial cells, more preferably a medium for culturing HUVEC cells. Preferably, the medium of step a) is selected from the following:

[0116] - a medium comprising a growth factor selected from the group consisting of insulin-like growth factor (IGF), epithelial growth factor (EGF), fibroblast growth factor (FGF) and vascular endothelial growth factor (VEGF) ; preferably a low-serum (at most 2% V / V) medium that contains a growth factor selected from the group defined above; preferably a medium selected from:

[0117] - a medium comprising vascular endothelial growth factor (VEGF), human epithelial growth factor (hEGF), human insulin-like growth factor-I (R3-IGF-1), human fibroblast growth factor (hFGF-B), hydrocortisone, ascorbic acid, heparin, fetal bovine serum (FBS), gentamicin and amphotericin-B, and basal EBM (Lonza CC-3156), and optionally which does not comprise bovine brain extract;

[0118] - an EGM-2 medium (Lonza CC-3156 plus CC-4176),

[0119] - a medium comprising VEGF and / or EGF.

[0120] Preferably, in any of the methods for producing a cell-assembled matrix, the medium for culturing endothelial cells of step b) is a medium for culturing primary endothelial cells, more preferably a medium for culturing HUVEC cells. Preferably, the medium of step b) is selected from the following:

[0121] - a medium comprising a growth factor selected from the group consisting of insulin-like growth factor (IGF), epithelial growth factor (EGF), fibroblast growth factor (FGF) and vascular endothelial growth factor (VEGF); preferably a low-serum (at most 2% V / V) medium that contains a growth factor selected from the group defined above; preferably a medium selected from:

[0122] - a medium comprising vascular endothelial growth factor (VEGF), human epithelial growth factor (hEGF), human insulin-like growth factor-I (R3-IGF-1), human fibroblast growth factor (hFGF-B), hydrocortisone, ascorbic acid, heparin, fetal bovine serum (FBS), gentamicin and amphotericin-B, and basal EBM (Lonza CC-3156), and optionally which does not comprise bovine brain extract;

[0123] - an EGM-2 medium (Lonza CC-3156 plus CC-4176); - a medium comprising VEGF and / or EGF;

[0124] - a medium comprising potassium chloride, sodium chloride, calcium chloride, magnesium sulfate, amino acids, glucose, cholesterol, pyrimidines, vitamins - including thiamine, riboflavin, and biotin -, fetal bovine serum (FBS), and vascular endothelial growth factor (VEGF), human epithelial growth factor (hEGF), human insulin-like growth factor-I (R3-IGF-1), human fibroblast growth factor (hFGF-B), hydrocortisone, ascorbic acid, heparin, gentamicin and amphotericin-B;

[0125] - Medium 199 supplemented with fetal bovine serum (FBS) and with a medium for culturing endothelial cells, preferably primary endothelial cells, more preferably HUVEC cells; more preferably Medium 199 supplemented with 5-15 V / V% FBS and 5-15 V / V% EGM-2 medium, more preferably Medium 199 supplemented with 8-12 V / V% FBS and 8-12 V / V% EGM-2, most preferably Medium 199 supplemented with 10 V / V% FBS and 10 V / V% EGM-2.

[0126] More preferably, in any of the methods for producing a cell-assembled matrix, after seeding the cells in step a), the cells are cultured in a medium for culturing endothelial cells (preferably primary endothelial cells, more preferably HUVEC cells) until the cells adhere to the support, preferably for 2 hours to 72 hours (preferably for 2 hours to 2 days); and after that, the cells are cultured in Medium 199 supplemented with FBS and with a medium for culturing endothelial cells (preferably primary endothelial cells, more preferably HUVEC cells). In an embodiment, the cells are maintained before preparing the matrix (i.e. before seeding the cells to the support) in Medium 199 supplemented with 5-15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) FBS and with 5- 15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) EGM-2.

[0127] Preferably in step b), the culturing of the seeded cells is carried out in EGM, preferably EGM2 medium for 2 hours to 72 hours, which is then replaced by a Medium 199 supplemented with fetal bovine serum (FBS) and EGM2, preferably supplemented with 5-15 V / V% FBS and 5-15 V / V% EGM, preferably EGM2.

[0128] In an embodiment, the cells are seeded to the solid support in undiluted EGM-2 medium (day 0), cultured for 1 day in undiluted EGM-2 medium, and then from the second day the cells are cultured in Medium 199 supplemented with 5-15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) FBS and with 5-15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) EGM-2.

[0129] Preferably, in any of the methods for producing a cell -assembled matrix, the EGM-2 medium is an undiluted, i.e. 100 V / V% EGM-2 medium. More preferably, the undiluted EGM-2 medium comprises only 2% FBS.

[0130] Preferably, in any of the methods for producing a cell -assembled matrix, the medium comprising VEGF and / or EGF is a medium for culturing endothelial cells. More preferably, the medium comprising VEGF and / or EGF is a medium for culturing macrovascular endothelial cells. More preferably, the medium comprises both VEGF and EGF.

[0131] More preferably, in any of the methods for producing a cell-assembled matrix, the endothelial cells are HUVECs and the medium containing VEGF and / or EGF is a medium for culturing macrovascular endothelial cells, more preferably wherein the medium comprises both VEGF and EGF.

[0132] Preferably, in any of the methods for producing a cell-assembled matrix, in step b) the culturing is conducted for 1 day to 28 days, preferably for 1 to 14 days, more preferably for 1 to 6 days, preferably for at least 72 hours, preferably for 4, 5 or 6 days.

[0133] In an embodiment, the method comprises the following steps: - prior to seeding, maintaining endothelial cells in Medium 199 supplemented with 5-15 V / V% (preferably 8- 12 V / V%, more preferably 10 V / V%) FBS and with 5-15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) EGM-2;

[0134] - seeding the endothelial cells on a support, wherein the cells are seeded on the support in EGM2 medium;

[0135] - culturing the cells in undiluted (100% V / V) EGM2 medium for 2 hours to 72 hours (preferably for 12 hours to 60 hours, more preferably for 12 hours to 48 hours, more preferably for 24 hours to 48 hours, preferably for about 24 hours or more preferably for 36 hours to 48 hours);

[0136] - then changing the cell-culture medium to Medium 199 supplemented with 5-15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) FBS and with 5-15 V / V% (preferably 8-12 V / V%, more preferably 10 V / V%) EGM-2, and culturing the cells for a further 1 to 28 days, preferably for a further 1 to 14 days, more preferably for 1 to 6 days, preferably for at least 48 hours, preferably for 2, 3, 4, 5 or 6 days, thereby obtaining a cell culture;

[0137] - optionally, removing the cells from the obtained cell culture, thereby producing an extracellular matrix without cells (i.e. a decellularized matrix),

[0138] - fixing the obtained culture or the decellularized matrix obtained to the support with a fixative that does not destroy celiac epitopes of antigens for celiac disease-specific antibodies (in particular, the celiac epitopes of TG2) present in the extracellular matrix of the culture or in the extracellular matrix, respectively, preferably wherein a fixative does not destroy the celiac epitopes of antigens (in particular, the celiac epitopes of TG2) if, after fixation, the matrix shows a positive extracellular staining with a positive celiac control sample, preferably wherein the fixative is acetone, methanol, a solution of methanol and MES buffer, or PAXgene tissue fix, more preferably wherein the fixative is acetone, methanol or a solution of methanol and MES buffer,

[0139] - thereby obtaining a cell-assembled (cell-culture assembled) matrix.

[0140] In an embodiment, the method for producing a cell-assembled matrix does not comprise step c). In another embodiment, the method for producing a cell-assembled matrix comprises step c).

[0141] In an embodiment, the method comprises step c) and in step c) the cells are removed from the matrix before fixing by an agent that does not destroy the celiac epitopes of the TG2 antigen, such as with sodium deoxycholate or TritonX 100. In a preferred embodiment, the cells are removed using 0.1% to 1% TritonXIOO or 0.1% to 1% sodium deoxycholate, and then the matrix is washed 3 times with a suitable buffer, such as PBS, Tris buffer or HEPES buffer.

[0142] Preferably, in any of the methods for producing a cell -assembled matrix, the antigen for celiac diseasespecific antibodies present in the extracellular matrix of the culture or in the extracellular matrix is a transglutaminase, preferably type 2 transglutaminase (TG2) or any (celiac) epitope thereof.

[0143] Preferably, in any of the methods for producing a cell-assembled matrix, in step d) the fixative does not destroy the celiac epitopes of the TG2 antigen. More preferably, a fixative does not destroy the celiac epitopes of TG2 wherein, after the fixation, the cell-assembled matrix fixed to the support shows a positive extracellular staining with a known celiac serum sample (positive celiac control sample). More preferably, the fixative is acetone or methanol or a solution of methanol and MES buffer.

[0144] More preferably, in any of the methods for producing a cell-assembled matrix, the antigen for celiac disease-specific antibodies present in the extracellular matrix of the culture or in the extracellular matrix is type 2 transglutaminase (TG2) or a celiac epitope thereof, and in step d) the culture or extracellular matrix is fixed with a fixative that does not destroy the celiac epitopes of the TG2 antigen present in the extracellular matrix of the culture or in the extracellular matrix, preferably wherein the fixative is acetone or methanol or a solution of methanol and MES buffer.

[0145] Preferably, in any of the methods for producing a cell-assembled matrix, the produced cell-assembled matrix has endomysial characteristics, such as a honeycomb binding pattern with celiac disease-specific antibodies, and / or the produced cell-assembled matrix has a similar morphology than that of the endomysium, and / or the produced cell-assembled matrix comprises an organized, preferably highly organized network of argyrophilic reticulin fibers which show a positive reaction with silver impregnation.

[0146] Preferably, in any of the methods for producing a cell-assembled matrix, the produced cell-assembled matrix has a fiber network comprising fibronectin and comprises, bound to the surface of fibronectin, antigen for celiac disease-specific antibodies (celiac antigen). In particular, the celiac antigen is a transglutaminase, more preferably TG2.

[0147] In a preferred embodiment, the method for producing a cell -assembled matrix does not comprise step c), and thus the produced cell-assembled matrix comprises the endothelial cells (preferably primary endothelial, more preferably HUVEC cells) that produced said matrix.

[0148] In a preferred embodiment, in the method for producing a cell-assembled matrix, step c) is absent and the produced cell-assembled matrix is an organized, TG2-rich extracellular matrix around endothelial cells, preferably around primary endothelial cells, more preferably around HUVEC cells.

[0149] In another preferred embodiment, the method for producing a cell -assembled matrix comprises step c), and thus the produced cell-assembled matrix is a decellularized matrix, i.e. a matrix that does not comprise cells.

[0150] Preferably, in any of the methods for producing a cell-assembled matrix, the produced cell-assembled matrix is a 2-dimensional structure. If step c) is absent from the method, the produced cell-assembled matrix may be a monolayer culture.

[0151] Preferably, in any of the methods for producing a cell-assembled matrix, the produced solid phase matrix preparation comprising the cell-assembled matrix is suitable for use in an immunofluorescence method, preferably in an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies. Preferably, the celiac disease-specific antibodies are anti-endomysium antibodies or anti-transglutaminase antibodies, preferably TG2 -specific antibodies or autoantibodies. Preferably, the antibodies are IgA and / or IgG antibodies, preferably IgA antibodies, IgG antibodies, or both IgA and IgG antibodies.

[0152] In another embodiment, when the method comprises step c), i.e. the removal of the cells, the produced solid phase matrix comprising the decellularized cell-assembled matrix is suitable for use in an immunoassay, for example in an enzyme-linked immunosorbent assay (ELISA) or in an immunofluorescence method, preferably in an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies.

[0153] Preferably, in any of the methods for producing a cell-assembled matrix, in step e) the cell-assembled matrix is produced as a solid phase matrix preparation comprising the (optionally decellularized) cell-assembled matrix.

[0154] The invention also relates to the cell-assembled matrix produced by any of the above-mentioned methods (or methods disclosed elsewhere in the description) for producing a cell-assembled matrix, or the solid phase matrix preparation comprising the cell-assembled matrix produced by any of the above-mentioned methods (or methods disclosed elsewhere in the description) for producing a cell-assembled matrix.

[0155] The invention also relates to the use of any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix in a method for detecting celiac diseasespecific antibodies in a sample.

[0156] The invention also relates to the use of any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations, preferably prepared on supports and comprising the cell -assembled matrix for detecting celiac disease-specific antibodies in a sample.

[0157] Preferably, the method for detecting celiac disease-specific antibodies in a sample is an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an endomysium antibody (EMA) assay. Preferably, wherein the cell-assembled matrix is a decellularized matrix, the method for detecting celiac disease -specific antibodies in a sample is an immunoassay method, for example an enzyme-linked immunosorbent assay (ELISA) or an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies.

[0158] Preferably, in the method for detecting celiac disease -specific antibodies in a sample, the celiac diseasespecific antibodies are anti-endomysium antibodies (endomysial antibodies) or anti-transglutaminase antibodies, preferably TG2-specific antibodies or autoantibodies. Preferably, the antibodies are IgA and / or IgG antibodies, preferably IgA antibodies, IgG antibodies, or both IgA and IgG antibodies.

[0159] Preferably, in any of the methods for detecting celiac disease-specific antibodies mentioned above, the sample is a sample that can potentially comprise celiac disease-specific antibodies, preferably any of the antibodies mentioned above, more preferably anti-TG2 antibodies. More preferably, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease-specific cell culture, such as a TG2- producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0160] Preferably, in any of the methods for detecting celiac disease -specific antibodies mentioned above, the sample is obtained from a human individual. More preferably, the sample is a human serum sample or a human plasma sample.

[0161] The invention also relates to the use of any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix in a method for in vitro diagnosing or evaluating celiac disease in a subject.

[0162] The invention also relates to the use of any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations, preferably prepared on supports and comprising the cell-assembled matrix for in vitro diagnosing celiac disease.

[0163] Preferably, the method for in vitro diagnosing or evaluating celiac disease comprises an immunofluorescence method - preferably an indirect immunofluorescence method or an endomysium antibody (EMA) assay - for detecting celiac disease -specific antibodies. Preferably, when the cell-assembled matrix is a decellularized matrix, the method for in vitro diagnosing or evaluating celiac disease comprises an immunoassay method, for example an enzyme-linked immunosorbent assay (ELISA) or an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies.

[0164] Preferably, the method for in vitro diagnosing or evaluating celiac disease in a subject comprises the following steps:

[0165] - providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix,

[0166] - providing a sample obtained from the subject,

[0167] - incubating the sample with the substrate,

[0168] - contacting the incubated sample with detection antibodies (such as fluorophore-conjugated secondary antibodies),

[0169] - detecting the detection antibodies and interpreting the results (e.g. with a fluorescence microscope) to diagnose or evaluate celiac disease.

[0170] Preferably, the method for in vitro diagnosing or evaluating celiac disease in a subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating detecting antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound detecting antibodies, g) visualizing or quantifying the detecting antibodies bound to the washed substrate of step 1), and h) based on the image or quantitative result obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment.

[0171] Preferably, the method for in vitro diagnosing or evaluating celiac disease in a subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating fluorophore-conjugated secondary antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound fluorophore-conjugated secondary antibodies, g) viewing under or visualizing by a fluorescence microscope of the washed substrate of step f), and h) based on the pattern observed or on the image obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment.

[0172] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the sample obtained from the subject is a sample that can potentially comprise celiac disease -specific antibodies, preferably anti-endomysium antibodies (endomysial antibodies) or anti-transglutaminase antibodies, preferably TG2-specific antibodies or autoantibodies, preferably IgA or IgG antibodies, more preferably anti-TG2 antibodies. More preferably, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease-specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0173] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the subject is a vertebrate, preferably a mammal, preferably a primate, more preferably a human. The subject may also be a transgenic animal, such as a human transgene animal, for example a zebrafish.

[0174] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the washing of step d) and / or step f) is carried out using phosphate buffered saline (PBS), a Tris buffer, a HEPES buffer or any other suitable buffer. The washing of step d) and / or step 1) may be carried out using the same or different buffers.

[0175] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the antibodies of step a) and / or step d) are human antibodies.

[0176] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the detecting antibodies are selected from fluorophore -conjugated antibodies, enzyme-conjugated antibodies or antibodies with chemiluminescent label. More preferably, the detecting antibodies are secondary antibodies that can bind to celiac disease-specific (human) antibodies.

[0177] Preferably, the detecting antibodies also include using two secondary antibodies, i.e. a first secondary antibody capable of binding to the celiac disease-specific (human) antibodies and a second secondary antibody labelled with a fluorophore, an enzyme or a chemiluminescent label that is capable of binding to the first secondary antibody. In this case, the method comprises another pair of incubating and washing step, i.e. after incubating the substrate of step a) with the sample of step b) and washing said incubated substrate, the substrate would be first incubated with the first secondary antibody and would be washed, and then the substrate would be incubated with the second secondary antibody and then washed before the visualizing or quantifying of step g) is carried out. This is a known technique for amplifying the detecting signal.

[0178] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the fluorophore-conjugated secondary antibodies of step e) are fluorescein-labelled antibodies or secondary antibodies labelled with fluorophores. More preferably, the antibodies of step a) and step d) are IgA or IgG antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step f) are anti-IgA or anti-IgG secondary antibodies labelled with fluorophores, respectively. More preferably, the antibodies of step a) and step d) are human antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step 1) are fluorophore-labelled anti-human antibodies. Most preferably, the antibodies of step a) and step d) are human IgA antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step f) are anti-human anti-IgA secondary antibodies labelled with fluorophores.

[0179] Preferably, the fluorophore is selected from xanthene derivatives, such as fluorescein (FITC), rhodamine (TRITC), Oregon green, eosin, Texas red or Alexa Fluor dyes; cyanine derivatives, such as cyanine, indocarbocyanine, oxacarbocyanine, thiacarbocyanine or merocyanine; squaraine derivatives; squaraine rotaxane derivatives; naphthalene derivatives; coumarin derivatives; oxadiazole derivatives, such as pyridyloxazole, nitrobenzoxadiazole or benzoxadiazole; anthracene derivatives, such as anthraquinones, including DRAQ5, DRAQ7, and CyTRAK Orange; pyrene derivatives, such as cascade blue; oxazine derivatives, such as Nile red, Nile blue, cresyl violet or oxazine 170; acridine derivatives, such as proflavin, acridine orange or acridine yellow; arylmethine derivatives, such as auramine, crystal violet or malachite green; tetrapyrrole derivatives, such as porphin, phthalocyanine or bilirubin; and dipyrromethene derivatives, such as BODIPY or aza-BODIPY.

[0180] More preferably, the fluorophore is selected from fluorescein isothiocyanate (FITC), rhodamine, Texas red or AlexaFluor dyes.

[0181] In an embodiment, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, step h) is carried out using an automated method, such as an automated image (or data) processing, or an artificial intelligence-based image (or data) processing.

[0182] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the celiac disease treatment of step h) is (recommending) a gluten-free diet.

[0183] In another preferred embodiment, the method for in vitro diagnosing or evaluating celiac disease in a human subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the human subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating anti-human IgG antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound anti-human IgG antibodies, g) incubating the washed substrate of step 1) with fluorophore-conjugated secondary antibodies against the anti-human IgG antibodies, h) washing the incubated substrate of step g) to remove unbound fluorophore-conjugated secondary antibodies against the anti-human IgG antibodies, i) incubating the washed substrate of step h) with fluorophore-conjugated anti-human IgA antibodies, j) washing the incubated substrate of step i) to remove unbound fluorophore-conjugated anti-human IgA antibodies, k) viewing under or visualizing by a fluorescence microscope of the washed substrate of step j), and l) based on the pattern observed or on the image obtained in step k), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment. Preferably, in the method using both IgG and IgA secondary antibodies described above, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease -specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0184] Preferably, in any of the methods using both IgG and IgA secondary antibodies described above, the washing of step d) and / or step f) and / or step h) and / or step j) is carried out using phosphate buffered saline (PBS), a Tris buffer, a HEPES buffer or any other suitable buffer. The washing of step d) and / or step f) and / or step h) and / or step j) may be carried out using the same or different buffers.

[0185] Preferably, in any of the methods using both IgG and IgA secondary antibodies described above, the fluorophore-conjugated antibodies of step g) and step i) are secondary antibodies labelled with a fluorophore selected from xanthene derivatives, such as fluorescein (FITC), rhodamine (TRITC), Oregon green, eosin, Texas red or Alexa Fluor dyes; cyanine derivatives, such as cyanine, indocarbocyanine, oxacarbocyanine, thiacarbocyanine or merocyanine; squaraine derivatives; squaraine rotaxane derivatives; naphthalene derivatives; coumarin derivatives; oxadiazole derivatives, such as pyridyloxazole, nitrobenzoxadiazole or benzoxadiazole; anthracene derivatives, such as anthraquinones, including DRAQ5, DRAQ7, and CyTRAK Orange; pyrene derivatives, such as cascade blue; oxazine derivatives, such as Nile red, Nile blue, cresyl violet or oxazine 170; acridine derivatives, such as proflavin, acridine orange or acridine yellow; arylmethine derivatives, such as auramine, crystal violet or malachite green; tetrapyrrole derivatives, such as porphin, phthalocyanine or bilirubin; and dipyrromethene derivatives, such as BODIPY or aza-BODIPY. More preferably, the fluorophore-conjugated antibodies of step g) and step i) are secondary antibodies labelled with a fluorophore selected from fluorescein isothiocyanate (FITC), rhodamine, Texas red or AlexaFluor dyes. More preferably, the fluorophore-conjugated antibodies of step g) and step i) are labelled with fluorophores of different colors and applied in a way excluding cross-reactions. For example, the fluorophore-conjugated antibodies of step g) are labelled with Alexa568, and the fluorophore-conjugated antibodies of step i) are labelled with fluorescein isothiocyanate (FITC).

[0186] Optionally, any of the methods using both IgG and IgA secondary antibodies described above, further comprise step m) of determining whether the subject diagnosed with celiac disease is a subject with IgA deficiency.

[0187] A kit comprising

[0188] - any of the solid phase matrix preparations comprising a cell-assembled matrix disclosed herein,

[0189] - detecting antibodies, preferably fluorophore-conjugated secondary antibodies,

[0190] - positive control,

[0191] - negative control, and

[0192] - buffer.

[0193] The invention also relates to a method for detecting celiac disease-specific antibodies in a sample, which comprises the use of any of the cell-assembled matrices disclosed herein or any of the solid phase matrix preparations comprising the cell-assembled matrix disclosed herein as the substrate comprising the antigens for the celiac disease-specific antibodies (in particular the TG2 antigen).

[0194] Preferably, the method for detecting celiac disease -specific antibodies in a sample is an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an endomysium antibody (EMA) assay. Preferably, wherein the cell-assembled matrix is a decellularized matrix, the method for detecting celiac disease -specific antibodies in a sample is an immunoassay method, for example an enzyme-linked immunosorbent assay (ELISA) or an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies.

[0195] Preferably, in the method for detecting celiac disease -specific antibodies in a sample, the celiac diseasespecific antibodies are endomysial antibodies (anti-endomysium antibodies) or anti-transglutaminase antibodies, preferably TG2-specific antibodies or autoantibodies. Preferably, the antibodies are IgA and / or IgG antibodies, preferably IgA antibodies, IgG antibodies, or both IgA and IgG antibodies, each giving / providing a separate signal.

[0196] Preferably, in any of the methods for detecting celiac disease -specific antibodies, the sample is a sample that can potentially comprise celiac disease-specific antibodies, preferably any of the antibodies mentioned above, more preferably anti-TG2 antibodies. More preferably, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease -specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0197] Preferably, in any of the methods for detecting celiac disease -specific antibodies, the sample is obtained from a human individual. More preferably, the sample is a human serum sample or a human plasma sample.

[0198] The invention also relates to a method for in vitro diagnosing or evaluating celiac disease in a subject, which comprises the use of any of the cell-assembled matrices disclosed herein or any of the solid phase matrix preparations comprising the cell-assembled matrix disclosed herein.

[0199] Preferably, the method for in vitro diagnosing or evaluating celiac disease comprises an immunofluorescence method - preferably an indirect immunofluorescence method or an endomysium antibody (EMA) assay - for detecting celiac disease -specific antibodies. Preferably, when the cell-assembled matrix is a decellularized matrix, the method for in vitro diagnosing or evaluating celiac disease comprises an immunoassay method, for example an enzyme-linked immunosorbent assay (ELISA) or an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease-specific antibodies.

[0200] Preferably, the method for in vitro diagnosing or evaluating celiac disease in a subject comprises the following steps:

[0201] - providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix,

[0202] - providing a sample obtained from the subject,

[0203] - incubating the sample with the substrate,

[0204] - contacting the incubated sample with detection antibodies (such as fluorophore-conjugated secondary antibodies),

[0205] - detecting the detection antibodies and interpreting the results (e.g. with a fluorescence microscope) to diagnose or evaluate celiac disease.

[0206] Preferably, the method for in vitro diagnosing or evaluating celiac disease in a subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating detecting antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound detecting antibodies, g) visualizing or quantifying the detecting antibodies bound to the washed substrate of step 1), and h) based on the image or quantitative result obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment.

[0207] Preferably, the method for in vitro diagnosing or evaluating celiac disease in a subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating fluorophore-conjugated secondary antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound fluorophore-conjugated secondary antibodies, g) viewing under or visualizing by a fluorescence microscope of the washed substrate of step f), and h) based on the pattern observed or on the image obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment.

[0208] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the sample obtained from the subject is a sample that can potentially comprise celiac disease -specific antibodies, preferably anti-endomysium antibodies (endomysial antibodies) or anti-transglutaminase antibodies, preferably TG2-specific antibodies or autoantibodies, preferably IgA or IgG antibodies, more preferably anti-TG2 antibodies. More preferably, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease-specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0209] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the subject is a vertebrate, preferably a mammal, preferably a primate, more preferably a human. The subject may also be a transgenic animal, such as a human transgene animal, for example a zebrafish. Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the washing of step d) and / or step f) is carried out using phosphate buffered saline (PBS), a Tris buffer, a HEPES buffer or any other suitable buffer. The washing of step d) and / or step 1) may be carried out using the same or different buffers.

[0210] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the antibodies of step a) and / or step d) are human antibodies.

[0211] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the detecting antibodies are selected from fluorophore -conjugated antibodies, enzyme-conjugated antibodies or antibodies with chemiluminescent label. More preferably, the detecting antibodies are secondary antibodies that can bind to celiac disease-specific (human) antibodies.

[0212] Preferably, the detecting antibodies also include using two secondary antibodies, i.e. a first secondary antibody capable of binding to the celiac disease -specific (human) antibodies and a second secondary antibody labelled with a fluorophore, an enzyme or a chemiluminescent label that is capable of binding to the first secondary antibody. In this case, the method comprises another pair of incubating and washing step, i.e. after incubating the substrate of step a) with the sample of step b) and washing said incubated substrate, the substrate would be first incubated with the first secondary antibody and would be washed, and then the substrate would be incubated with the second secondary antibody and then washed before the visualizing or quantifying of step g) is carried out. This is a known technique for amplifying the detecting signal.

[0213] Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the fluorophore-conjugated secondary antibodies of step e) are fluorescein-labelled antibodies or secondary antibodies labelled with fluorophores. More preferably, the antibodies of step a) and step d) are IgA or IgG antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step f) are anti-IgA or anti-IgG secondary antibodies labelled with fluorophores, respectively. More preferably, the antibodies of step a) and step d) are human antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step f) are fluorophore-labelled anti-human antibodies. Most preferably, the antibodies of step a) and step d) are human IgA antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step 1) are anti-human anti-IgA secondary antibodies labelled with fluorophores.

[0214] Preferably, the fluorophore is selected from xanthene derivatives, such as fluorescein (FITC), rhodamine (TRITC), Oregon green, eosin, Texas red or Alexa Fluor dyes; cyanine derivatives, such as cyanine, indocarbocyanine, oxacarbocyanine, thiacarbocyanine or merocyanine; squaraine derivatives; squaraine rotaxane derivatives; naphthalene derivatives; coumarin derivatives; oxadiazole derivatives, such as pyridyloxazole, nitrobenzoxadiazole or benzoxadiazole; anthracene derivatives, such as anthraquinones, including DRAQ5, DRAQ7, and CyTRAK Orange; pyrene derivatives, such as cascade blue; oxazine derivatives, such as Nile red, Nile blue, cresyl violet or oxazine 170; acridine derivatives, such as proflavin, acridine orange or acridine yellow; arylmethine derivatives, such as auramine, crystal violet or malachite green; tetrapyrrole derivatives, such as porphin, phthalocyanine or bilirubin; and dipyrromethene derivatives, such as BODIPY or aza-BODIPY.

[0215] More preferably, the fluorophore is selected from fluorescein isothiocyanate (FITC), rhodamine, Texas red or AlexaFluor dyes.

[0216] In an embodiment, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, step h) is carried out using an automated method, such as an automated image (or data) processing, or an artificial intelligence-based image (or data) processing. Preferably, in any of the methods for in vitro diagnosing or evaluating celiac disease mentioned above, the celiac disease treatment of step h) is (recommending) a gluten-free diet.

[0217] In another preferred embodiment, the method for in vitro diagnosing or evaluating celiac disease in a human subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell -assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the human subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating anti-human IgG antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound anti-human IgG antibodies, g) incubating the washed substrate of step 1) with fluorophore-conjugated secondary antibodies against the anti-human IgG antibodies, h) washing the incubated substrate of step g) to remove unbound fluorophore-conjugated secondary antibodies against the anti-human IgG antibodies, i) incubating the washed substrate of step h) with fluorophore-conjugated anti-human IgA antibodies, j) washing the incubated substrate of step i) to remove unbound fluorophore-conjugated anti-human IgA antibodies, k) viewing under or visualizing by a fluorescence microscope of the washed substrate of step j), and l) based on the pattern observed or on the image obtained in step k), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment.

[0218] Preferably, in the method using both IgG and IgA secondary antibodies described above, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease -specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0219] Preferably, in any of the methods using both IgG and IgA secondary antibodies described above, the washing of step d) and / or step f) and / or step h) and / or step j) is carried out using phosphate buffered saline (PBS), a Tris buffer, a HEPES buffer or any other suitable buffer. The washing of step d) and / or step f) and / or step h) and / or step j) may be carried out using the same or different buffers.

[0220] Preferably, in any of the methods using both IgG and IgA secondary antibodies described above, the fluorophore-conjugated antibodies of step g) and step i) are secondary antibodies labelled with a fluorophore selected from xanthene derivatives, such as fluorescein (FITC), rhodamine (TRITC), Oregon green, eosin, Texas red or Alexa Fluor dyes; cyanine derivatives, such as cyanine, indocarbocyanine, oxacarbocyanine, thiacarbocyanine or merocyanine; squaraine derivatives; squaraine rotaxane derivatives; naphthalene derivatives; coumarin derivatives; oxadiazole derivatives, such as pyridyloxazole, nitrobenzoxadiazole or benzoxadiazole; anthracene derivatives, such as anthraquinones, including DRAQ5, DRAQ7, and CyTRAK Orange; pyrene derivatives, such as cascade blue; oxazine derivatives, such as Nile red, Nile blue, cresyl violet or oxazine 170; acridine derivatives, such as proflavin, acridine orange or acridine yellow; arylmethine derivatives, such as auramine, crystal violet or malachite green; tetrapyrrole derivatives, such as porphin, phthalocyanine or bilirubin; and dipyrromethene derivatives, such as BODIPY or aza-BODIPY. More preferably, the fluorophore-conjugated antibodies of step g) and step i) are secondary antibodies labelled with a fluorophore selected from fluorescein isothiocyanate (FITC), rhodamine, Texas red or AlexaFluor dyes. More preferably, the fluorophore-conjugated antibodies of step g) and step i) are labelled with fluorophores of different colors and applied in a way excluding cross-reactions. For example, the fluorophore-conjugated antibodies of step g) are labelled with Alexa568, and the fluorophore-conjugated antibodies of step i) are labelled with fluorescein isothiocyanate (FITC).

[0221] Optionally, any of the methods using both IgG and IgA secondary antibodies described above, further comprise step m) of determining whether the subject diagnosed with celiac disease is a subject with IgA deficiency.

[0222] The invention also relates to a method of treatment of celiac disease in a subject, which comprises the use of any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix.

[0223] Preferably, the method of treatment comprises in vitro diagnosis or evaluation of celiac disease using an immunofluorescence method - preferably an indirect immunofluorescence method or an endomysium antibody (EMA) assay - for detecting celiac disease-specific antibodies. Preferably, wherein the cell-assembled matrix is a decellularized matrix, the method of treatment comprises in vitro diagnosis or evaluation of celiac disease using an immunoassay method, for example an enzyme-linked immunosorbent assay (ELISA) or an immunofluorescence method, preferably an indirect immunofluorescence method, more preferably an indirect immunofluorescence method for detecting celiac disease -specific antibodies.

[0224] Preferably, the method of treatment comprises the following steps:

[0225] - providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix,

[0226] - providing a sample obtained from the subject,

[0227] - incubating the sample with the substrate,

[0228] - contacting the incubated sample with detection antibodies (such as fluorophore-conjugated secondary antibodies),

[0229] - detecting the detection antibodies and interpreting the results (e.g. with a fluorescence microscope) to diagnose or evaluate celiac disease,

[0230] - treating the subject diagnosed with celiac disease.

[0231] Preferably, the method of treatment of celiac disease in a subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating detecting antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound detecting antibodies, g) visualizing or quantifying the detecting antibodies bound to the washed substrate of step 1), h) based on the image or quantitative result obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment; and i) treating the subject diagnosed with celiac disease or evaluated as not responding to celiac disease treatment in step h), wherein the treatment is a celiac disease-specific treatment, preferably wherein the celiac disease-specific treatment is (recommending) a gluten-free diet.

[0232] Preferably, the method of treatment of celiac disease in a subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell-assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating fluorophore-conjugated secondary antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound fluorophore-conjugated secondary antibodies, g) viewing under or visualizing by a fluorescence microscope of the washed substrate of step f), and h) based on the pattern observed or on the image obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment, i) treating the subject diagnosed with celiac disease or evaluated as not responding to celiac disease treatment in step h), wherein the treatment is a celiac disease-specific treatment, preferably wherein the celiac disease-specific treatment is (recommending) a gluten-free diet.

[0233] Preferably, in any of the methods of treatment mentioned above, the sample obtained from the subject is a sample that can potentially comprise celiac disease -specific antibodies, preferably anti-endomysium antibodies (endomysial antibodies) or anti-transglutaminase antibodies, preferably TG2-specific antibodies or autoantibodies, preferably IgA or IgG antibodies, more preferably anti-TG2 antibodies. More preferably, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease-specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0234] Preferably, in any of the methods of treatment mentioned above, the subject is a vertebrate, preferably a mammal, preferably a primate, more preferably a human. The subject may also be a transgenic animal, such as a human transgene animal, for example a zebrafish.

[0235] Preferably, in any of the methods of treatment mentioned above, the washing of step d) and / or step 1) is carried out using phosphate buffered saline (PBS), a Tris buffer, a HEPES buffer or any other suitable buffer. The washing of step d) and / or step f) may be carried out using the same or different buffers. Preferably, in any of the methods of treatment mentioned above, the antibodies of step a) and / or step d) are human antibodies.

[0236] Preferably, in any of the methods of treatment mentioned above, the detecting antibodies are selected from fluorophore-conjugated antibodies, enzyme-conjugated antibodies or antibodies with chemiluminescent label. More preferably, the detecting antibodies are secondary antibodies that can bind to celiac disease -specific (human) antibodies.

[0237] Preferably, the detecting antibodies also include using two secondary antibodies, i.e. a first secondary antibody capable of binding to the celiac disease -specific (human) antibodies and a second secondary antibody labelled with a fluorophore, an enzyme or a chemiluminescent label that is capable of binding to the first secondary antibody. In this case, the method comprises another pair of incubating and washing step, i.e. after incubating the substrate of step a) with the sample of step b) and washing said incubated substrate, the substrate would be first incubated with the first secondary antibody and would be washed, and then the substrate would be incubated with the second secondary antibody and then washed before the visualizing or quantifying of step g) is carried out. This is a known technique for amplifying the detecting signal.

[0238] Preferably, in any of the methods of treatment mentioned above, the fluorophore-conjugated secondary antibodies of step e) are fluorescein-labelled antibodies or secondary antibodies labelled with fluorophores. More preferably, the antibodies of step a) and step d) are IgA or IgG antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step f) are anti-IgA or anti-IgG secondary antibodies labelled with fluorophores, respectively. More preferably, the antibodies of step a) and step d) are human antibodies, and the fluorophore- conjugated secondary antibodies of step e) and step f) are fluorophore-labelled anti-human antibodies. Most preferably, the antibodies of step a) and step d) are human IgA antibodies, and the fluorophore-conjugated secondary antibodies of step e) and step 1) are anti-human anti-IgA secondary antibodies labelled with fluorophores.

[0239] Preferably, the fluorophore is selected from xanthene derivatives, such as fluorescein (FITC), rhodamine (TRITC), Oregon green, eosin, Texas red or Alexa Fluor dyes; cyanine derivatives, such as cyanine, indocarbocyanine, oxacarbocyanine, thiacarbocyanine or merocyanine; squaraine derivatives; squaraine rotaxane derivatives; naphthalene derivatives; coumarin derivatives; oxadiazole derivatives, such as pyridyloxazole, nitrobenzoxadiazole or benzoxadiazole; anthracene derivatives, such as anthraquinones, including DRAQ5, DRAQ7, and CyTRAK Orange; pyrene derivatives, such as cascade blue; oxazine derivatives, such as Nile red, Nile blue, cresyl violet or oxazine 170; acridine derivatives, such as proflavin, acridine orange or acridine yellow; arylmethine derivatives, such as auramine, crystal violet or malachite green; tetrapyrrole derivatives, such as porphin, phthalocyanine or bilirubin; and dipyrromethene derivatives, such as BODIPY or aza-BODIPY.

[0240] More preferably, the fluorophore is selected from fluorescein isothiocyanate (FITC), rhodamine, Texas red or AlexaFluor dyes.

[0241] In an embodiment, in any of the methods of treatment mentioned above, step h) is carried out using an automated method, such as an automated image (or data) processing, or an artificial intelligence -based image (or data) processing.

[0242] Preferably, in any of the methods of treatment mentioned above, the celiac disease treatment of step h) is (recommending) a gluten-free diet.

[0243] Preferably, in any of the methods of treatment mentioned above, in step i) the treatment of the subject comprises restricting the subject to a gluten-free diet or recommending to the subject a gluten-free diet. In another preferred embodiment, the method of treatment of celiac disease in a human subject comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is any of the disclosed cell -assembled matrices or any of the disclosed solid phase matrix preparations comprising the cell-assembled matrix, b) providing a sample obtained from the human subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating anti-human IgG antibodies with the washed substrate of step d), f) washing the incubated substrate of step e) to remove unbound anti-human IgG antibodies, g) incubating the washed substrate of step 1) with fluorophore-conjugated secondary antibodies against the anti-human IgG antibodies, h) washing the incubated substrate of step g) to remove unbound fluorophore-conjugated secondary antibodies against the anti-human IgG antibodies, i) incubating the washed substrate of step h) with fluorophore-conjugated anti-human IgA antibodies, j) washing the incubated substrate of step i) to remove unbound fluorophore-conjugated anti-human IgA antibodies, k) viewing under or visualizing by a fluorescence microscope of the washed substrate of step j), and l) based on the pattern observed or the image obtained in step k), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment. m) treating the subject diagnosed with celiac disease or evaluated as not responding to celiac disease treatment in step 1), wherein the treatment is a celiac disease -specific treatment, preferably wherein the celiac disease-specific treatment is (recommending) a gluten-free diet.

[0244] Preferably, in the method of treatment using both IgG and IgA secondary antibodies described above, the sample is selected from serum; plasma; whole blood; saliva; breast milk; the supernatant of a celiac disease - specific cell culture, such as a TG2 -producing cell culture; the supernatant of an organ culture obtained from a biopsy sample incubated with gliadin; or the supernatant from a homogenized biopsy sample. More preferably, the sample is a serum sample or plasma sample.

[0245] Preferably, in any of the methods of treatment using both IgG and IgA secondary antibodies described above, the washing of step d) and / or step f) and / or step h) and / or step j) is carried out using phosphate buffered saline (PBS), a Tris buffer, a HEPES buffer or any other suitable buffer. The washing of step d) and / or step f) and / or step h) and / or step j) may be carried out using the same or different buffers.

[0246] Preferably, in any of the methods of treatment using both IgG and IgA secondary antibodies described above, the fluorophore-conjugated antibodies of step g) and step i) are secondary antibodies labelled with a fluorophore selected from xanthene derivatives, such as fluorescein (FITC), rhodamine (TRITC), Oregon green, eosin, Texas red or Alexa Fluor dyes; cyanine derivatives, such as cyanine, indocarbocyanine, oxacarbocyanine, thiacarbocyanine or merocyanine; squaraine derivatives; squaraine rotaxane derivatives; naphthalene derivatives; coumarin derivatives; oxadiazole derivatives, such as pyridyloxazole, nitrobenzoxadiazole or benzoxadiazole; anthracene derivatives, such as anthraquinones, including DRAQ5, DRAQ7, and CyTRAK Orange; pyrene derivatives, such as cascade blue; oxazine derivatives, such as Nile red, Nile blue, cresyl violet or oxazine 170; acridine derivatives, such as proflavin, acridine orange or acridine yellow; arylmethine derivatives, such as auramine, crystal violet or malachite green; tetrapyrrole derivatives, such as porphin, phthalocyanine or bilirubin; and dipyrromethene derivatives, such as BODIPY or aza-BODIPY. More preferably, the fluorophore-conjugated antibodies of step g) and step i) are secondary antibodies labelled with a fluorophore selected from fluorescein isothiocyanate (FITC), rhodamine, Texas red or AlexaFluor dyes. More preferably, the fluorophore-conjugated antibodies of step g) and step i) are labelled with fluorophores of different colors and applied in a way excluding cross-reactions. For example, the fluorophore-conjugated antibodies of step g) are labelled with Alexa568, and the fluorophore-conjugated antibodies of step i) are labelled with fluorescein isothiocyanate (FITC).

[0247] Optionally, any of the methods of treatment using both IgG and IgA secondary antibodies described above, step 1) further comprises determining whether the subject diagnosed with celiac disease is a subject with IgA deficiency.

[0248] DEFINITIONS

[0249] “Celiac disease” or “coeliac disease” (also called coeliac sprue or non-tropical sprue) is an illness caused by an immune reaction to eating gluten-containing foods, such as cereals, wheat, rye and barley. It is a systemic autoimmune disorder. In celiac disease, the generated immune reaction and inflammation leads to villous atrophy in the small bowel and to organ dysfunctions at various extraintestinal sites. Clinical symptoms may include malabsorption, abdominal distension, bloating, abdominal pain, iron and vitamin deficiencies, osteopenia, osteoporosis, elevated liver enzymes, hepatic failure, myopathy, cardiomyopathy, glomerular and tubular renal dysfunction, infertility, neuropathy, or cerebellar ataxia. Celiac disease is characterized by the presence of transglutaminase 2 (TG2) serum autoantibodies.

[0250] The term “cell-assembled matrix” (CAM), which may also be called herein cell-assembled extracellular matrix or biomatrix, refers to a matrix produced by cells in vitro, which resembles the extracellular matrix surrounding the cells in vivo. The cell-assembled matrix may comprise the cells which produces it, or the cells may be removed from it, leaving only the extracellular matrix -like structure behind. Preferably, the cells producing the cell-assembled matrix are endothelial cells, preferably primary endothelial cells, more preferably HUVECs.

[0251] The term “extracellular matrix” or “ECM” refers to a network of proteins and other molecules that surround, support and give structure to cells. In other words, the extracellular matrix is an intricate network composed of an array of multidomain macromolecules organized in a cell-specific or tissue-specific manner. As used herein, the extracellular matrix also includes in vitro produced extracellular matrices, which comprise the cells that produce them or which does not comprise the cells that produce them (decellularized matrix).

[0252] The term “endomysium” refers to the connective tissue surrounding each individual muscle fiber or muscle cell. It contains collagen (mainly type III), reticulin fibers, fibronectin and laminin. The endomysium can be regarded as a TG2-rich fine reticulin network around muscle fibers and around cells in general. The endomysium is further defined by its positive staining in the silver impregnation method.

[0253] An “endomysium-like” or “endomysium-type” matrix (also called endomysium-like or endomysium-type ECM, matrix which has endomysial characteristics, a matrix resembling an endomysial network / structure or the endomysium defined by the positivity with the silver impregnation method) is a cell-assembled matrix which has a similar morphology than that of the endomysium. It is a highly organized matrix, which offers celiac autoantigen in its natural context (but which is produced not from a tissue section, but by cells in vitro). For example, it exposes celiac disease-relevant transglutaminase epitope (e.g. TG2 epitopes) and gives a positive staining by the silver impregnation method. For the silver impregnation method, see Example 9, or Gomori, 1937; Slidders et al., 1958 orKrutsay, 1988.

[0254] The term “endomysium-like” or “having endomysial characteristics” or “resembling an endomysial structure” means that the (cell-assembled) matrix or substrate shows high similarity to the tissue endomysium morphologically and in its macromolecular composition. More specifically, the cell -assembled matrix shows high degree of similarity to tissue endomysium comprising or composed of a network of argyrophilic reticulin fibers and comprising macromolecules collagen III, as well as preferably collagen IV, fibronectin, laminin and TG2, while it has a low content of collagen I or essentially free of collagen I. Preferably, the network of fibers shows a mesh-like pattern, preferably a honeycomb pattern. In other words, the term “endomysium-type” (“endomysium- like”) or “having endomysial characteristics” or “resembling an endomysial structure” means that the (cell- assembled) matrix or substrate comprises a reticulin fiber network which gives a positive staining with the silver impregnation method, wherein said reticulin network comprises fibronectin and comprises antigen for celiac disease-specific antibodies, i.e. celiac antigen (in particular, TG2) bound to the fibronectin surface. Alternatively, these expressions refer to the fact that such a substrate or (cell-assembled) matrix behaves in an endomysium antibody (EMA) assay as the endomysium in tissue sections, i.e. it can bind celiac disease-specific antibodies in a distinct net-like pattern (“honeycomb binding pattern” or “mesh-like pattern”).

[0255] A “reticulin fiber network” as used herein is an extracellular network of fibers, preferably in connective tissue (specialized connective tissue network), wherein the network of fibers is formed at least partly by type III collagen, preferably also by type IV collagen and fibronectin, and the fiber network, once visualized, has a reticular (mesh-type) structure, preferably a honeycomb or net-like pattern; wherein visualization can be carried out by silver-impregnation staining, HE (hematoxylin-eosin) method, celiac disease-specific antibodies, and / or specific staining of collagen III and preferably collagen IV and / or fibronectin.

[0256] Preferably the “reticulin fiber network” is argyrophilic reticulin fiber network which shows honeycomb or mesh-like fiber structures and which stain positively (black or purple-black) with the silver staining while, preferably, there are other connective tissue components which are non-antigenic for celiac disease-specific antibodies and are negative in the silver staining (e.g. remain brown).

[0257] Preferably the “reticulin fiber network” is a TG2-rich fiber network which can be visualized by silver impregnation (see for example Gomori, 1937; Slidders etal., 1958; Krutsay, 1988; Hallstrom, 1988 or Example 9).

[0258] An “endothelial cell” is a mesenchymal cell type that normally lines the interior surface of blood vessels and lymphatic vessels. Endothelial cells form the barrier between vessels and tissue and control the flow of substances and fluid into and out of tissue.

[0259] A “primary endothelial cell” is a terminally differentiated endothelial cell that can be isolated from a tissue or organ of interest. It is an endothelial cell taken directly from a living tissue (e.g. biopsy material) and established for growth in vitro (i.e. it is directly cultured from its source organ tissue). Human primary endothelial cells can be isolated from human umbilical vein (HUVEC), aorta, pulmonary arteries, coronary arteries, iliac arteries, cardiac microvascular tissue, lung microvascular tissue and dermal microvascular tissue (HMVEC). In a broader sense a primary endothelial cell involves an immortalized cell line, which is derived from primary endothelial cells - preferably HUVEC cells - and which retain the morphological properties of primary endothelial cells and express von Willebrand factor, CD31, VE-cadherin and TG2.

[0260] The term “HUVEC” or “human umbilical vein endothelial cell” refers to a primary endothelial cell isolated from human umbilical cord vein.

[0261] “Type 2 transglutaminase” or “transglutaminase 2” or “TG2”, which is also called tissue transglutaminase (tTG), cell transglutaminase or erythrocyte transglutaminase, is a calcium-dependent enzyme (EC 2.3.2.13) of the protein-glutamine y-glutamyltransferases family (also called transglutaminase family), which has a molecular weight of 78 kDa. The TG2 is the autoantigen in celiac disease.

[0262] The “celiac disease-specific antibodies” are autoantibodies which are most sensitive and specific for the confirmation of celiac disease. The celiac disease -specific antibodies include anti-transglutaminase 2 (anti-TG2, or also called anti-tissue transglutaminase or anti-tTG) antibodies, endomysial antibodies (EMAs), and Rl-type reticulin antibodies (ARA). These antibodies are usually of IgA isotype, except if a subject has an IgA-deficiency, in which case the afore-mentioned antibodies are of IgG isotype.

[0263] An “immunoassay” is a method for detecting or measuring specific proteins or other substances through their properties as antigens or antibodies. Immunoassays include enzyme-linked immunosorbent assays (ELISAs) and immunofluorescence (IF) methods.

[0264] A “support” (which may also be called solid support material or solid phase) is a solid material to which the cell-assembled matrix can be affixed. Preferably, it is a slide, a chamber slide, a cover slip, a plate, a dish, a well, a bottle, a test tube, a flask or any other traditionally used laboratory equipment. More preferably, it is a slide, a chamber slide, a cover slip or a plate.

[0265] A “medium for culturing endothelial cells” is a “medium useful for culturing endothelial cells”, preferably a “medium designed for culturing endothelial cells”, i.e. a medium comprising factor(s) which are specifically beneficial to or support the culturing of endothelial cells. The medium may be optimized for culturing endothelial cells.

[0266] The “endothelial cell growth medium 2” or “EGM-2 medium” is a low-serum (2% V / V) medium that contains insulin-like growth factor (IGF), epithelial growth factor (EGF), fibroblast growth factor (FGF) and vascular endothelial growth factor (VEGF). EGM-2 medium is made of a basal medium with product code CC-3156 from Lonza (Basel, Switzerland) and a supplement with product code CC-4176 from Lonza (Basel, Switzerland).

[0267] “Medium 199” or “M199” is a cell culture medium developed by J.F. Morgan in 1950. The composition of Medium 199 can vary slightly depending on the specific requirements of the cell line being cultured, however, the basic formulation typically includes salts such as potassium chloride, sodium chloride, calcium chloride, and magnesium sulfate. Medium 199 also contains various essential and non-essential amino acids, glucose, cholesterol, pyrimidines, and vitamins, including thiamine, riboflavin, and biotin. It does not contain proteins, lipids, or growth factors; instead, M199 utilizes a sodium bicarbonate buffer system (2.2 g / L) and requires a 5- 10% CO2 environment to maintain a physiologically suitable pH. Medium 199 is available in several different formulations, which can include Earle's salts or Hanks' salts, L-Glutamine, HEPES, stable glutamine, and sodium bicarbonate, and is available in both liquid and powder forms. In the uses and methods described herein, any Medium 199 may be used.

[0268] “ibiTreat” (ibidi®, Grafeling, Germany) is a physical surface modification, which makes the surface hydrophilic and adhesive to cells, and the ibiTreat surface is slightly negatively charged. An “immortalized” cell line refers to a population of cells from a multicellular organism which would normally not proliferate indefinitely but, due to mutation (natural or induced), have evaded cellular senescence and instead can keep undergoing division. Such cells can therefore be grown for prolonged periods in vitro.

[0269] The term “subject” as used herein shall refer to a vertebrate, preferably a homeothermic (mammalian or avian, preferably mammalian) subject, particularly a human being.

[0270] “Diagnostic accuracy” of a test is its ability to differentiate the patient and healthy cases correctly. To estimate the accuracy of a test, the proportion of true positive plus true negative cases in all evaluated cases is calculated.

[0271] The term “comprise(s)” or “comprising” or “including” are to be construed herein as having a non- exhaustive meaning and to allow the addition or involvement of further features or method steps or components to anything which comprises the listed features or method steps or components. Such terms can be limited to “consisting essentially of’ or “comprising substantially” which is to be understood as consisting of mandatory features or method steps or components listed in a list, e.g. in a claim, whereas allowing to contain additionally other features or method steps or components which do not materially affect the essential characteristics of the use, method, composition or other subject matter. Moreover, these terms or solutions can also be limited to those consisting of a component or components which is or which are essential so that a feature may be achieved.

[0272] As used in this specification and the appended claims, the singular forms “a”, “an” and “the” include plural references, and should be construed as including the meaning “one or more”, unless the content clearly dictates otherwise. In general, it is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0273] ABBREVIATIONS

[0274] ARA anti-reticulin antibody

[0275] CAM cell-assembled matrix

[0276] CD celiac disease

[0277] ECM extracellular matrix

[0278] EGF epithelial growth factor

[0279] EGM-2 endothelial cell growth medium 2

[0280] ELISA enzyme-linked immunosorbent assay

[0281] EMA anti-endomysium antibody (also called endomysial antibody)

[0282] FBS fetal bovine serum

[0283] FGF fibroblast growth factor

[0284] FITC fluorescein isothiocyanate

[0285] HUVEC human umbilical vein endothelial cell

[0286] IGF insulin-like growth factor

[0287] PBS phosphate buffered saline

[0288] TG2 type 2 transglutaminase (also called tissue transglutaminase or tTG)

[0289] TGA anti-transglutaminase antibody IgA

[0290] VE vascular endothelial

[0291] VEGF vascular endothelial growth factor BRIEF DESCRIPTION OF THE FIGURES

[0292] Figure 1. Classical EMA (endomysial antibody) reaction patterns of celiac IgA antibody binding on frozen sections of monkey esophagus outer muscular layer (A), mucosa and tunica muscularis mucosae (B), human umbilical cord vein (C), human appendix (D), and staining of celiac IgA antibodies on umbilical cord-derived myofibroblasts in culture after acetone fixation showing minimal extracellular matrix production at arrows (E). Panel F shows endomysial antibody binding pattern of celiac IgA antibodies to the artificial endomysium of the invention (HUVEC-ECM).

[0293] Figure 2. Morphological organization and antigenicity of extracellular matrix produced by human umbilical cord vein endothelial cells seeded and cultured in EGM-2 medium. Reaction with celiac IgA antibodies (A) and with monoclonal mouse anti-TG2 antibodies CUB7402 (B). High magnification image (600x).

[0294] Figure 3. Morphological organization and antigenicity for celiac IgA antibodies of mature HUVEC-ECM produced at 72 hours in culture using a medium change after 24 hours of the seeding from EGM-2 to Medium 199 with 10 V / V%) FBS and 10 V / V% EGM-2. Magnification lOOx.

[0295] Figure 4. Development of HUVEC-ECM in different cell culture media. The images were taken with the same exposure time.

[0296] Figure 5. Celiac antibody binding in relation to transglutaminase 2 and other extracellular or adhesion proteins in methanol-MES fixated HUVEC-ECM. HUVEC-ECM stained with celiac IgA antibodies (A) and double stained with celiac IgA and monoclonal antibodies CUB7402 to transglutaminase 2 (B) rabbit polyclonal antibodies to fibronectin (C), laminin (D), monoclonal antibodies to vinculin (E) and to VE-cadherin (F). Human IgA binding was visualized by FITC-conjugated secondary antibodies in green, and rabbit or mouse antibodies to ECM or cellular components were visualized by Alexa568-conjugated anti-rabbit or anti-mouse antibodies, as appropriate, in red.

[0297] Figure 6. Binding of celiac IgG antibodies to HUVEC-ECM.

[0298] Figure 7. Binding of IgA competent (A,B,C) and IgA deficient (D,E,F) celiac serum sample to decellularized HUVEC-ECM. Panels A and D show the staining with FITC-conjugated anti-human IgA secondary antibodies in green, panels B and E with polyclonal rabbit anti-human IgG secondary antibodies and AlexaFluor 568-comjugated goat anti-rabbit antibodies in red. On panel C, IgA and IgG stainings merge to yellow, while for the IgA-deficient patient no IgA staining is visible and thus the merged IgA and IgG results (F) remain red. In this way IgA competent and IgA deficient celiac antibody positive serum samples can be differentiated.

[0299] Figure 8. Macromolecular composition of the decellularized biomatrix in comparison with endomysial structures of monkey esophagus. Collagens I, III and IV were detected by monoclonal mouse antibodies and Alexa Fluor 568 secondary antibodies in red; fibronectin, transglutaminase 2 and celiac IgA antibody binding were detected by Alexa Fluor 488 secondary antibodies or by FITC-conjugated anti-human IgA in green. Merging of red and green labels into yellow indicates co-localisation. FBN means fibronectin, TG2 means transglutaminase 2.

[0300] Figure 9. Silver impregnation staining demonstrated positivity for endomysial reticulin structures (black) in the muscular layers of monkey esophagus (A, asterisk) and for the reticulin lining under the epithelium (arrow) . Also the fiber networks organized into honeycomb pattern in both the cell -containing (D) and decellularized (E) HUVEC-ECM were positive. The silver staining positive structures corresponded to the structures recognized by celiac disease IgA in the EMA test (C, asterisk) and in the HUVEC-ECM (F). The submucosal connective tissue was neither antigenic for celiac patient samples (C, triangle), nor gave a positive staining in the silver impregnation (A, brown, triangle) and corresponded to collagen I expression in immunofluorescence (B, triangle). Symbols: asterisks indicate endomysial layers, triangles indicate collagen I-rich non-endomysial connective tissue and circle indicates the epithelium in the esophagus.

[0301] Figure 10. Flowchart of the diagnostic evaluation of the prospectively tested patient cohort.

[0302] Figure 11. Evaluation of the positive EMA staining (A) in comparison with a reference TG2 staining in invisible wavelength range (B) using a computerized image analysis for co -localization (C). In panel A, fluorescent green (FITC) signal was detected at 488 nm (visible to human eye); in panel B, CUB7402 signal was detected by a camera at 647 nm (invisible to human eye); panel C is a superimposition by a computer.

[0303] DETAILED DESCRIPTION OF THE INVENTION

[0304] The present invention is directed to the improved and more cost-effective detection of celiac diseasespecific autoantibodies by producing a suitable antigenic biomatrix for the testing. Surprisingly, human umbilical cord vein endothelial cells under appropriate conditions and on suitable surfaces are able to produce in 72 hours an organized, endomysium-like or endomysium-type extracellular matrix. This matrix can replace more complex systems, such as monkey tissue sections, for the endomysial antibody (EMA) detection, a currently indispensable verification test for the non-biopsy diagnosis of celiac disease (CD).

[0305] In contrast to umbilical cord fibroblasts used by Sulkanen et al. (1998b), which contain high amounts of antigenic TG2 in their cell bodies, but externalize this antigen only in small amounts to the matrix and in noncharacteristic patterns (Figure IE), another easily cultivable cell type from the umbilical cord, vein endothelial cells (HUVECs) with much lower cellular TG2 content, have, surprisingly, the capacity to build up a well- organized extracellular matrix with the characteristics of endomysial structures (HUVEC-ECM) and high antigenicity for celiac antibodies when cultured under the conditions given in Example 1. Here we show that this specifically prepared HUVEC-ECM can be utilized with high diagnostic accuracy and high selectivity for TG2- directed antibodies for the evaluation of CD, not achieved by the commonly applied cultivation of fibroblasts and by their general connective tissue-type extracellular matrix products in earlier studies. In addition, the here developed endomysial-type HUVEC-ECM provided higher diagnostic sensitivity and diagnostic accuracy for CD than the use of tissue sections with natural endomysial structures (e.g. monkey esophagus).

[0306] Serum anti-transglutaminase antibody IgA (TGA) above ten times of the upper limit of normal (> lOxULN), verified by endomysial antibody (EMA) positivity from a second sample, are accepted as the basis for the non-biopsy diagnosis of celiac disease (CD). EMA detection is a visually evaluated indirect immunofluorescence test. EMA detections on the classical substrates, monkey esophagus or human umbilical cord are less sensitive than TGA (although EMA is an excellent, nearly 100% specific confirmatory test for celiac antibody positivity) and both the preparation of the tissue slides and the reading are time-consuming and need expertise. Moreover, it is getting difficult to get commercial slides for the testing. Monkey esophagus, the classical EMA substrate, provides easy evaluation, but it is from an endangered species and commercial selling of slides has recently been terminated by most companies. Human umbilical cord sections are often used instead, however, this alternative is challenged by the need to cut frozen sections and to have high expertise in the reading, especially, when the positivity is faint and disturbing non-celiac antibodies are also present.

[0307] In this work, we searched for an easier available alternative EMA substrate and evaluated whether the use of a cell-assembled biomatrix containing the human EMA antigen is accurate for celiac disease diagnostic testing in clinical practice. We developed a cell-assembled antigenic biomatrix with EMA characteristics produced in cell culture by human umbilical cord vein-derived endothelial cells (HUVECs). We show here that slides with this biomatrix, even in cell-free format, can be utilized with high sensitivity and specificity for the diagnostic detection of celiac autoantibodies and can replace the traditionally used tissue sections in clinical practice. The cell-assembled biomatrix can be applied with similar positive predictive value for the final CD diagnosis as conventional tissue - based EMA testing, has a potential for unlimited supply and is containing human antigen.

[0308] Human umbilical cord vein-derived endothelial cells (HUVECs) produce in cell culture a regular extracellular matrix (ECM) resembling endomysial network with similar exposure of celiac -relevant transglutaminase epitopes as in tissue sections. One of the aims of this work was to evaluate the diagnostic performance of EMA testing on HUVEC-ECM (see Examples).

[0309] The disclosed HUVEC-ECM has high similarity to endomysium in tissues both in its morphological and macromolecular composition, and gives a positive staining with the classical silver impregnation method originally described by Gomori in 1937, and which is widely used in several modifications in histology to define in tissue sections endomysial structures composed of argyrophilic reticulin fibers. Classical silver staining demonstrated that both monkey endomysium and the biomatrix presented herein are built up of a structurally similar network of reticulin fibers, i.e. the cell-free biomatrix network was positive with the classical silver staining in the same pattern as is defining endomysial structures in histopathology.

[0310] The endomysial-type HUVEC-ECM contains tissue transglutaminase in its natural context with the exposure of the celiac epitopes. The macromolecular composition of the HUVEC-ECM biomatrix is identical to natural tissue endomysium; it contains in the same morphological arrangement collagen III, collagen IV and fibronectin, with TG2 bound to them.

[0311] This TG2 acts as the active antigenic component for CD antibodies, similarly to the EMA antigen in tissues. [Korponay-Szabo et al., 2003], Earlier attempts to replace the EMA tissues with cells cultured on ELISA plates [Sulkanen et al., 1998b, W02003 / 008976A2], cell suspensions dried to slides [Whelan et al., 1996] or by homogenised tissue extracts did not prove useful in clinical practice because they offered mainly intracellular antigens for the testing with frequent nonspecific background reactions to non-celiac cellular antigens, or they offered a fibroblast-derived extracellular matrix with general connective tissue components where TG2 was only a minor antigen. It is known that the intracellular TG2, retained in closed, enzymatically inactive conformation by high levels of GDP and low concentrations of intracellular calcium, is not a good antigen for the celiac antibodies which bind to extracellular matrix components containing TG2 in calcium-activated conformation [Simon-Vecsei et al., 2012], In the final format, the here developed biomatrix is decellularized and contains only extracellular matrix components with active TG2 and is free from disturbing non-specific reactivity to cell bodies ensuring high specificity.

[0312] The easily available HUVEC-ECM substrate provided high sensitivity and specificity for EMA-IgA detection and also may improve the evaluation of celiac autoimmunity when TGA is low positive or borderline, or if the subject is having humoral IgA deficiency.

[0313] Thus, a cell-assembled matrix (also called extracellular matrix) is disclosed, which can be used in indirect immunofluorescence methods, such as endomysial antibody (EMA) detection tests.

[0314] In an embodiment, the cell-assembled matrix does not comprise the cells that produce it. Preferably, the cells have been removed from the matrix before fixing by an agent that does not destroy the celiac epitopes of the TG2 antigen, such as with sodium deoxy cholate or TritonX 100. The cells that produce the cell-assembled matrix do not need to be present in the fixed cell-assembled matrix that is to be used in an immunofluorescence method (e.g. indirect immunofluorescence) for detecting the binding of the celiac disease-specific antibodies of a sample to the antigens present in the cell-assembled matrix, since the antibodies bind to the antigens present in the cell-assembled matrix, i.e. antigens outside the cells. However, the presence of the cells does not affect the performance of the immunofluorescence method, thus in another embodiment, the cells need not be removed before fixation.

[0315] However, if the cells are removed from the cell-assembled matrix (decellularized matrix), the cell- assembled matrix can also be used not just in immunofluorescence methods, but in other immunoassays, as well. If the cells are removed, the cell-assembled matrix will not produce background staining (staining of the cell bodies) which would interfere with the readings. Thus, for example, the decellularized cell-assembled matrix may also be used in ELISA, not just in indirect immunofluorescent methods (such as EMA assays).

[0316] Removing the cells also has the added advantage that it is not necessary to teach or practice which samples are considered positive or negative for celiac antibodies because when the cells are absent, only the matrix can be stained. Thus, if staining is observed (visually or even by automated image processing), the sample may be considered positive for celiac antibodies, and when the staining is absent, the sample may be considered negative for celiac antibodies. This works both with IgA and IgG antibodies.

[0317] In case of detecting antibodies of IgG isotype, removing the cells from the cell -assembled matrix before fixation is preferred. In fact, the removal of the cells from the cell-assembled matrix makes the detection of IgG antibodies possible and much easier (see for comparison Figure 6 and Figure 7).

[0318] The detection of IgG antibodies - in case of IgA-deficient subjects - is unreliable using the classical serological tests (such as EMA assays). The ESPGHAN most recent diagnostic guidelines (2020) recommend that IgA-deficient subjects who are positive for an IgG-based serological test should be biopsied, thus it is not recommended that IgA-deficient patients be diagnosed without biopsy.

[0319] The decellularized cell-assembled matrix disclosed herein can be used not just for detecting IgA antibodies, but also reliably detecting IgG antibodies from plasma or serum samples.

[0320] It is also possible to use the cell-assembled matrix (both the cell-containing and the decellularized) described herein for double staining, i.e. staining for both IgA and IgG antibodies, each producing a separable signal. This has the advantage of not needing to know beforehand whether the subject is IgA deficient or not. In case of the double staining, decellularized matrix is preferred.

[0321] The EMA testing carried out on the cell-assembled matrices disclosed herein (especially the HUVEC- ECM) is more sensitive than the classical EMA testing on tissue sections.

[0322] The advantages of the disclosed cell-assembled matrix are that there is no need for preparing or buying tissue sections, but instead the cell-assembled matrix produced by endothelial cells (preferably primary endothelial cells, more preferably HUVECs) can be used for detecting the binding of celiac antibodies in the indirect immunofluorescence method (EMA reaction). Furthermore, when the cell-assembled matrix is decellularized, it may also be used as a substrate in other immunoassay methods, such as ELISA, as well.

[0323] The disclosed cell-assembled matrices are especially advantageous, because they can be used as substrate in immunofluorescence methods. According to the current (2020) ESPHGAN guidelines, immunofluorescence (IF) method is still prescribed as a confirmatory test for the non-invasive (without biopsy) diagnosis of celiac disease, and the commercial availability of tissue sections is becoming scarce. Important aspects of the present invention are that endothelial cells are used, and that the staining of the matrix (and not of the cell body) is evaluated in the CD diagnostic application. Furthermore, it had to be contrived how to culture the endothelial cells and how to fix the produced cell-assembled matrix in such a way as not to destroy the celiac epitopes of antigens (preferably the TG2 antigen) in the process and to produce an organized cell-assembled matrix, which results in an unequivocally positive reaction in the silver staining (Figure 9), thus can be defined by these classical means as endomysial-type structure. The starting cells were of non-muscle origin and conserved their endothelial properties throughout the matrix production process as demonstrated by their VE- cadherin positivity in the mature cell-containing matrix preparates. According to the present invention an extracellular matrix with low collagen I and high collagen III and IV content was produced, which is unusual under conventional cell culturing conditions directed to produce common ECM and which results in the accumulation of collagen I. (the latter defining connective tissue-type matrix [Carvalho et al. 2019]). In a particularly preferred example the cells were induced by a specific sequential use of different cell media described in the present invention to achieve this feature. As shown in Figure 9, connective tissue rich in collagen I of the esophagus (submucosa) is not antigenic for celiac antibodies and does not give a positive staining with the silver impregnation method, thus it is not useful for specific CD antibody detection. Similarly, not all kind of tissue sections with various connective tissues are suitable for the detection of CD-specific TG2 antibodies in clinical practice, only those which contain a reticulin fiber network which has been shown by the silver impregnation in early diagnostic studies [Hallstrom 1989], In this respect, the appendix, spleen and liver, which contain organized argyrophilic reticulin fiber network were found also antigenic for CD antibodies and the specific CD antibody binding pattern (Rl-ARA) was restricted to the reticulin structures. However, this type of positivity was diagnostically more difficult to evaluate than the simple honeycomb EMA pattern on the esophagus, and thus ARA tests gradually disappeared from clinical diagnostics by the 21st century. Of note is, that the CD antigenic component in the HUVEC-ECM described in the present invention, is also restricted to the honeycomb fiber structures which stain positively (black or purple-black) with the silver staining as reticulin fibers, while other ECM components representing common connective tissue are non-antigenic for celiac disease-specific antibodies and remain brown (negative in the silver staining).

[0324] The organized matrix was not produced by culturing the endothelial cells in a traditionally used, fibronectin- coated or collagen-coated (gelatin-coated) plate or other suitable surface and the present inventors did not add external fibronectin or collagen to the cells. Thereby, letting the cells themselves assemble the fibronectin- containing fiber network, the process resulted in an organized, endomysium-type matrix.

[0325] Preferably, the method for producing a cell-assembled matrix comprises the following steps: a) seeding endothelial cells on a support, wherein the endothelial cells are seeded to the support in undiluted (i.e. 100 V / V%) EGM-2 medium or in a medium containing VEGF and / or EGF, b) culturing the seeded cells of step a) in a medium selected from EGM-2 medium; Medium 199 with 5-15 V / V% FBS and 5-15 V / V% EGM-2; or a medium containing VEGF and / or EGF; c) fixing the culture obtained in step b) to the support with a fixative that does not destroy celiac epitopes of the antigens for celiac disease-specific antibodies present in the extracellular matrix of the culture, preferably wherein the fixative is acetone or methanol or a solution of methanol and MES buffer, d) thereby obtaining a cell-assembled matrix, wherein preferably the cell-assembled matrix has endomysial characteristics or which resembles an endomysial structure or the endomysium (i.e. it is close in its structure to tissue endomysium as defined by silver impregnation), and which comprises antigen for celiac disease-specific antibodies.

[0326] Preferably the cells are seeded and cultured for 2 hours to 2 days in undiluted (100% V / V) EGM-2 medium, and then the cells are cultured in Medium 199 supplemented with 5-15 V / V% FBS and 5-15 V / V% EGM-2.

[0327] Preferably the antigen for celiac disease -specific antibodies is TG2.

[0328] The shifting of the culturing medium from EGM2 to the supplemented Medium 199 after about 2 hours - 2 days (preferably after about 6 hours to 2 days, preferably after about 12 hours to 36 hours, preferably after about 1 day) promotes the extemalization of the TG2 produced by the endothelial cells into the cell-assembled matrix. (Endothelial cells cultured solely in EGM-2 produce large intracellular quantities of TG2 but are slow to externalize them.)

[0329] In an embodiment, Medium 199 supplemented with 5-15% V / V (preferably 10 V / V%) FBS and with 5-15% V / V (preferably 10 V / V%) EGM-2 medium is used to maintain the HUVEC cells when they are cultured in a flask before plating them to the solid surface (support) and starting HUVEC-ECM production and this way was found more advantageous. In this procedure, HUVECs are abruptly exposed to 100% (V / V) EGM-2 containing a shot of high level of growth factors for a limited period of time. This has promoted better cell adhesion and also induced high level of TG2 expression, but in order to produce an endomy sium-type biomatrix and let it grow properly, this concentrated EGM-2 had to be withdrawn and be substituted with Medium 199 with only 10 V / V% EGM-2 but having higher, 10 V / V% FBS (see Example 3). This is contrary to earlier observations [Assuncao et al., 2020] derived from experiments producing common connective tissue-type ECM, where low FBS content resulted in higher ECM production and also contrary to the report and pictures published by Carvalho et al., 2019 who used only (undiluted) EGM-2 for growing the endothelial cells. The advantage of using Medium 199 with only 10 V / V% EGM-2 and 10 V / V% FBS for growing the cells after the second day is not only to induce more abundant and bigger honeycomb structures, as expected from the use of higher FBS concentrations, but also to promote the extemalization of TG2 from cell bodies and thus having lower background at the staining with patient serum samples which often contain nonspecific and non-celiac autoantibodies to various cellular antigens.

[0330] The support may be uncoated with either fibronectin or with gelatin before the cells that produce the cell- assembled matrix are seeded onto it. However, it may also be possible to coat the support using very diluted fibronectin or gelatin, in a quantity that later will not be visible on the support. In this case, visibility means that the externally added fibronectin or gelatin will not impede the visualization of the cell-assembled matrix binding celiac antibodies.

[0331] Preferably the endothelial cells (preferably primary endothelial cells, more preferably HUVECs) are cultured according to good practice, meaning that the cells are cultured on the solid surface at the latest until they reach confluency or 2-3 days after they reach confluency . After the cells become confluent, there are less nutrients for the individual cells, and they start dying. In other words, the cells are cultured until they remain viable cells. This maximum culturing period depends on many factors, such as the number of the cells with which the culturing is started, and the size or volume of the support or chamber used. For example, starting with fewer cells and / or larger surface size or volume, the cells reach confluency later, and vice versa. Generally, HUVEC cells cultured according to standard practice may reach confluency in 2-14 days of culturing.

[0332] Preferably, in any of the methods for producing a cell -assembled matrix defined above, in step d) the culture or extracellular matrix is fixed with acetone at -20°C with prior and after washings in phosphate buffered saline (PBS); or with a solution containing 90% methanol and 10% 2-(4-morpholino)ethanesulfonic acid (MES) buffer. Preferably, in step d) the fixative is not paraformaldehyde or formaldehyde, as they destroy the celiac epitopes of the antigen (in particular celiac epitopes of TG2) present in the extracellular matrix of the culture or in the extracellular matrix.

[0333] Preferably, in any of the cell-assembled matrices, solid phase matrix preparations comprising a cell- assembled matrix or in any of the methods disclosed herein, transglutaminase is not added externally (either purified or recombinant transglutaminase).

[0334] In another aspect, an immortalized cell line is disclosed, which is derived from primary endothelial cells - preferably HUVEC cells - and which retain the morphological properties of primary endothelial cells and express von Willebrand factor, CD31, VE-cadherin and TG2.

[0335] A method for producing an immortalized cell line comprises i) providing primary endothelial (preferably HUVEC) cells, ii) growing the primary endothelial (preferably HUVEC) cells in a suitable medium, preferably EGM -2 or Medium 199 supplemented with 5-15 V / V% FBS and 5-15 V / V% EGM-2, iii) immortalizing the cells of step ii) by the viral delivery of telomerase gene, preferably transfecting the cells with a viral vector comprising the telomerase gene, and iv) selecting the transfected cells, thereby obtaining the immortalized cell line.

[0336] Preferably, the immortalized cell line is used in any of the methods mentioned above, i.e. for producing the cell-assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix.

[0337] Below, Examples are provided to illustrate the invention, but they are not intended to limit the scope of protection.

[0338] In Example 1, the development of the novel matrix is described - human umbilical cord vein-derived endothelial cells (HUVECs) were induced to produce in cell culture extracellular matrix with endomysial characteristics.

[0339] In Example 2, ninety patients consecutively referred to the inventors’ centre between May - September 2023 in reason of positive celiac antibody results in outside laboratories or by various rapid tests were prospectively tested. Results were compared with those of transglutaminase antibody ELISA and EMA tests performed on monkey esophagus, human umbilical cord and appendix sections. The results were that EMA-IgA on the biomatrix, EMA-IgA on tissues and transglutaminase IgA antibody ELISA were positive in 67, 62 and 73 of the 90 prospectively tested patients, with 95.9, 91.9% and 97.3% respective diagnostic accuracies for the final celiac disease diagnosis.

[0340] In Examples 3 to 5, different conditions for HUVEC -ECM preparation were tested. In Example 3, the effects of culturing time and growth media changes were assessed; in Example 4, the adhesion and extracellular matrix production of HUVEC cells were evaluated; and in Example 5, further fixatives were tested.

[0341] In Example 6, it was assessed whether the HUVEC-ECM would also be suitable to detect celiac antibodies in IgA deficient patients. According to the results, IgG celiac antibodies produced a similar staining as IgA celiac antibodies.

[0342] In Example 7, the preparation of a decellularized matrix is described. Furthermore, sensitivity was evaluated on the decellularized biomatrix, as well, using 61 stored transglutaminase antibody positive serum samples from untreated CD patients negative for EMA on tissues. Decellularization of the matrix improved sensitivity and enabled simple reading without previous training.

[0343] In Example 8, the macromolecular composition of the HUVEC-ECM was determined by immunofluorescent studies, both in cell-containing matrix and in decellularized matrix, and in Example 9, the classic silver impregnation staining was applied to define the HUVEC-ECM as in vitro produced endomysial-type structure.

[0344] In Example 10, decellularized matrix preparations were double-stained. The first stain showed bound celiac disease-specific antibodies, while the second stain showed TG2.

[0345] Thus, the conclusion is that the cell-assembled biomatrix substrate provided high specificity and substantially improved sensitivity for EMA-IgA detection.

[0346] Our results demonstrate the high diagnostic performance of the new EMA test in high-pretest probability setting in a tertiary centre, which was chosen as appropriate for a test meant mainly for confirmatory use. Our reference test was the final diagnosis of CD either by histology or by the ESPGHAN 2020 non-biopsy criteria, which have repeatedly been shown reliable [Sheppard et al., 2022] to indicate the presence of CD. Our prospectively enrolled study population had a high usage of self -initiated home testing or of antibody tests performed before the evaluation by a specialist, which is typical for the current situation in clinical practice.

[0347] Our work also demonstrates that confirmatory testing is indeed needed. Patients enrolled in the prospective part of our study (Example 2) were referred to our centre with an earlier positive celiac antibody result, however, these results were heterogeneous (rapid tests, isolated IgG antibody positivity besides low and high positive TGA- IgA) and almost 20% of these patients (17 out of 90) turned out to have negative results at our testing. Borderline and low positive TGA values are nowadays very frequent in the clinical practice and histology results are often negative in such cases. EMA testing has important role also for making a decision for performing a biopsy, because there is higher rate of positive predictability for a future final CD diagnosis in EMA positive than in EMA negative cases [Kurppa et al. 2009],

[0348] EMA positivity also indicates that the celiac -relevant epitopes of TG2 are targeted by the investigated serum antibodies. In contrast, TGA positivity only means a reaction with TG2 in general, which may or may not be CD related [Simon-Vecsei et al. , 2012], The HUVEC-ECM exposes in the same way epitope-2 as EMA tissues, because the TG2 antigen is bound to fibronectin in both. TG2 has a specific binding site for fibronectin and this complexing interferes with non-specific sticking of antibodies to the antigen or to non-celiac epitopes of it, making the test result highly specific for CD. Recombinant TG2 proteins utilised in ELISA and other immunoassays as antigens have different quality and are not always true imitators of the natural autoantigen due to poor folding and even mismatching amino acid sequences [Kanchan et al., 2013],

[0349] The EMA test has been criticised as being laborious and having observer-dependent and less sensitive results than ELISA or automated tests. The here presented biomatrix is produced by monolayer cell culture, thus there is no need to cut sections. Removal of the cells before the testing improves sensitivity and makes possible the reading even without previous training. Since the biomatrix itself is antigenic and no background is generated by cell bodies, any positivity seen indicates EMA reaction. This type of reaction will be suitable also for automated reading and high-throughput assays. The HUVEC-ECM preparations both in cell-containing and cell-free formats were more sensitive to detect CD antibodies than traditional EMA tissues due to the much lower background, which improves the applicability of this novel EMA test. IgA deficiency is a further challenge for CD antibody detection. IgA deficiency occurs with a prevalence of 1 in 400 in the normal population and these subjects have higher risk for CD [Korponay-Szabo et al., 2004], but might remain undetected if only IgA class antibodies are measured. ESPGHAN guidelines suggest the determination of total serum IgA concentration and IgA class TGA tests as the initial step in diagnostics [Husby et al., 2020], but many laboratories offer instead IgA and IgG or combined IgA+IgG TGA tests. The IgG-TGA tests are less well optimised than the IgA-TGA tests and more often yield elevated results [Absah et al., 2017], Nowadays patients with isolated positive IgG class TGA result constitute an increasing bulk for confirmatory testing in tertiary centres, also for the EMA testing. IgG class EMA positivity is highly predictive for CD in IgA deficient subjects [Korponay-Szabo et al., 2004; Werkstetter et al., 2017] and is an indication for small intestinal biopsy, whereas IgG class TGA positivity alone has poor predictive value for villous atrophy in IgA competent cases [Absah et al., 2017] and thus makes the invasive procedure unnecessary if IgG class EMA is negative. Monkey esophagus is not a good substrate for IgG class EMA determination due to high background reactions, thus the IgG testing is more often done on the lower background umbilical cord of premature babies or on human appendix sections, where additional reticulin components ease the evaluation. The decellularized HUVEC-ECM biomatrix with its low background enabled us to detect both IgA and IgG class EMA antibodies at the same time on the same matrix specimen, utilising double fluorescent staining (green for IgA and red for IgG) (see Example 6). This method allows to detect IgA deficient celiac patients also in the absence of previous total serum IgA determination.

[0350] Endoscopy in children requires general anaesthesia and hospital stay, and it nowadays often refused by the parents and by adult patients as well. The use of the here presented biomatrix substrate improves the availability and applicability of the EMA test to perform the required confirmatory testing for the non-invasive diagnosis according to the ESPGHAN 2020 criteria, which are currently gradually extended for adults in many countries. The endothelial cells producing the matrix are commercially available and since they are not any more present in the biomatrix made ready for use, the novel system poses less ethical concerns than the recent proposal to use monkey liver for the confirmatory testing [Wolf et al., 2016],

[0351] In conclusions, the EMA test is still diagnostically important, especially in children, for the safe diagnosis of CD without invasive endoscopy and for the evaluation of unclear cases with low antibody positivity. Cell- assembled endomysial-type biomatrix (in particular after decellularization) offers a reliable, easy-to-read and ethically acceptable alternative for monkey esophagus in EMA testing.

[0352] EXAMPLES

[0353] EXAMPLE 1: Development of novel EMA substrate

[0354] Patient samples

[0355] Serum samples from 3 IgA competent CD patients with positive transglutaminase antibodies (TGA), EMA antibodies, and Marsh III (IIIB or IIIC) histology in the small bowel, and from 3 non-celiac patients with normal small bowel were used during the development, for the initial evaluation and optimization.

[0356] Materials and Methods

[0357] Selection of cell lines

[0358] Umbilical cord-derived fibroblasts, myofibroblasts, and endothelial cells (HUVECs) are known to contain high amounts of TG2, which, after conventional culturing and fixation with paraformaldehyde, can be recognized by staining with monoclonal TG2 antibodies CUB7402 targeting a linear epitope, but not by CD autoantibodies. It was, thus, concluded that the conformational epitopes of TG2, which are important for celiac antibody binding, did not survive the fixation procedure. When other fixatives, such as acetone or methanol (MES-buffered) (see also Example 5), and acetone-resistant plasticware were applied, the fibroblasts and myofibroblasts showed prominent CD antibody binding, but mainly in their cytoplasm, with only a minimal extemalization of the antigenic TG2 to the extracellular matrix (Figure IE). Although the high fetal bovine serum (FBS) content of the medium promoted the export of TG2 to the matrix, the protein appeared in bunches on the cell surface without a definite pattern. HUVECs, however, started to produce under the cells a very fine spider net-like extracellular structure with CD antigenicity. Therefore, HUVECs were selected for fiirther study, although we note that other cells (such as fibroblasts and myofibroblasts) may also be able to produce endomysium-like matrix under specific culturing conditions or with a longer culturing time.

[0359] Cell culture

[0360] HUVEC cells were obtained from Lonza (Cambrex Bio Science, Walkersville, MD, USA) or prepared from anonymously donated human umbilical cords by collagenase digestion as described by Palatka et al. (2006). Cells were maintained in culture in CO2 incubator at 37°C with 5% CO2 in complete endothelial medium EGM-1 (Clonetics, San Diego, CA, USA) or in Medium 199 (HyClone, Logan, Utah, USA) supplemented with 10 V / V% foetal bovine serum (FBS) (Merck-Millipore), 10 V / V% EGM-2 Endothelial Growth Medium with added SinglcQuots™ (CC-3156 and CC-4176, Lonza, Basel, Switzerland), 20mM HEPES (Biosera, Nuaille, France), 100 U / mL Penicillin, 100 mg / mL Streptomycin and 2.5 mg / mL Amphotericin B (all from Biosera, Nuaille, France) using 25 cm2plastic flasks. Cells were immortalized by the viral delivery of telomerase gene using pBABE-neo-hTERT (Addgen) as earlier described [Shaw et al. 2021] to enable long-term culturing and more reproducibility. Transfected cells were selected using 300 pg / mL G418 (Merck-Millipore). Immortalized cells completely retain the morphological properties of primary endothelial cells and express von Willebrand factor, CD31, VE-cadherin and TG2. In some experiments, human umbilical cord-derived fibroblasts (prepared as described by Sulkanen et al. 1998b) grown in Medium 199 without EGM2 supplements were used as controls.

[0361] Celiac antibody staining (celiac antibody detection on biomatrix)

[0362] For the evaluation of celiac antibody binding, HUVEC cells were seeded on coverslips or to various chamber slides with glass or plastic surfaces from Sarstedt, Becton-Dickinson or to 18-well p-Slide chamber slides (Ibidi, Grafelfing, Germany) manufactured from acetone -resistant plastic for optimization with or without coating with 0.2 mg / mL gelatin or 1.25 pg / mL fibronectin (Merck-Millipore) for 1 hour at 37°C. Cells were cultured for 1-6 days (e.g. for 3-6 days) before fixation of the chambers in 4% paraformaldehyde at +4 °C or in precooled acetone at -20°C with prior and after washings in phosphate buffered saline, pH 7.2 (PBS) [for fiirther fixatives, see Example 5], Serum samples diluted in PBS were added for 30 minutes at room temperature. Initially we used 1:10 serum dilutions, because 100 pL solution (four times more than usually used for individual tissue sections) was needed to fill a chamber, and in this way, we applied approximately similar amounts of antibodies per specimen as when using 1:2.5 serum dilutions on tissues in clinical laboratory practice. In later measurement, both 1 :2.5 and 1 : 10 dilutions were tested. After washings with PBS, bound IgA antibodies were visualized with FITC- labelled rabbit anti-human IgA polyclonal antibodies (DAKO, Roskilde, Denmark) and evaluated with an Olympus CKX41 inverted fluorescent microscope as positive (visually graded 1-3) or negative. Samples for the diagnostic evaluation were tested in a blinded fashion for the results of other antibody tests and biopsies. In some experiments, the biomatrix preparates were also stained with mouse monoclonal antibodies to TG2 (CUB7402, Neomarkers, Fremont, CA, USA) diluted 1:200 in PBS, followed by secondary goat anti -mouse antibodies conjugated with Alexa-Fluor 568 or 647 (Molecular Probes), or with primary polyclonal rabbit antibodies to fibronectin, laminin (Merck-Millipore), collagen III (Chemicon / Merck-Millipore) or with mouse monoclonal antibodies to collagen I, collagen III, collagen IV (Chemicon / Merck-Millipore), vinculin, VE-cadherin (Merck- Millipore) or desmin (Abeam, Cambridge, UK) followed by Alexa Fluor 568 or 488 conjugated goat antibodies to rabbit or mouse immunoglobulins, as appropriate, or with cloned celiac patient-derived antibodies selectively recognizing celiac epitope 1 or 2 of the TG2 protein. These antibodies were produced with human IgG framework and were recognized by rabbit anti-human IgG (DAKO) and AlexaFluor 568-labelled goat anti rabbit secondary antibodies (Molecular Probes), or were recognized by anti-V5 tag monoclonal mouse antibodies followed by Alexa Fluor conjugated secondary antibodies. Alternatively, also biotin-conjugated horse anti-mouse antibodies (Vector Laboratories, Newark, CA, USA) and Texas-red conjugated streptavidin were used for the recognition of CUB7402 with enhanced signal.

[0363] Results

[0364] Initially, HUVEC cells were grown conventionally in EGM-1 and seeded, as customary for primary cells which dislike to adhere to glass or uncoated plastic surfaces, to chamber slides coated with gelatin or fibronectin. After 1-6 days in culture, chambers were treated with paraformaldehyde for fixation. Under this conditions, CUB monoclonal TG2 antibodies recognized TG2 within HUVEC cell bodies, but celiac serum samples did not show any positive staining. It was concluded that conformational epitopes of TG2 important for celiac antibody binding did not survive the fixation procedure. Thus, although human endothelial cells are known to express TG2 intracellularly and related to adhesion complexes after conventional cell culturing and fixation in paraformaldehyde, incubation with celiac disease serum samples did not produce a positive binding.

[0365] In next experiments, only acetone was applied as a fixative. Acetone did not influence the recognition and typical binding pattern of celiac antibodies in classical EMA substrate tissues. Since most plastic chamber slides are not compatible with acetone, the chamber had to be removed from the glass slide in case of Becton-Dickinson chamber slides. The developed fixation procedure was as follows: chambers were washed 3 times with PBS, then the glass slides devoid of the chambers were immersed at -20°C for 10 minutes into acetone precooled to -20°C, then the acetone solution was changed to a new one and the 10 minutes incubation was repeated in the same way. After fixation, the slides were air-dried and stored at -20°C until staining. In later experiments, ibidi p-slides 18 (ibidi®, Grafeling, Germany) were used, which are acetone-resistant and thus could undergo acetone fixation as a whole.

[0366] When we used a fixation with acetone that better preserves conformational epitopes, we observed that CD serum samples bind to HUVEC cell bodies and also recognise some fine fibre structures in their extracellular matrix, which are also positive for TG2 by monoclonal antibodies.

[0367] On acetone-treated slides, HUVEC cells showed intracellular and extracellular staining with celiac IgA antibodies, but the extracellular matrix around the cells was irregular, often blurred, so the pattern was variable and not reproducible. We suspected that proteins (gelatin / fibronectin) used for the coating of the dishes, which is normally used in cell culture to help adhere primary cells to slides, are responsible for the high background and irregularity, and they were omitted. In fact, gelatin contains a complex of extracellular matrix proteins, including small amounts of TG2. Fibronectin as a main TG2 capturing protein amorphously distributed on the surface by the coating might have interfered with the production of the matrix by cells.

[0368] In next experiments, HUVEC cells were added to uncoated glass or IbiTreat plastic surfaces, but they adhered poorly, did not proliferate and produced no or little extracellular matrix when the same growing media (EGM-1) were applied as when they were cultured in the flasks. Same results were obtained with poly-L-lysine coated surfaces. In fact, percentage of successfully adhered cells determines how fast the cells reach confluency, which in turn determines their matrix production by cell -cell contacts. Moreover, material from non-adhered cells that shortly die creates unhealthy conditions to living cells.

[0369] Disturbances in the cell attachment could satisfactorily be overcome when the HUVEC cells were seeded to the chamber in undiluted (100 V / V%) EGM-2 medium after they had been grown in Medium 199 supplemented with 10 V / V% foetal bovine serum (FBS) and with 10 V / V% EGM-2. Surprisingly, under these conditions, after culturing longer than 1 week or more in EGM-2 a highly organized, TG2-rich extracellular matrix was observed below and around the cells where they adhered, in the form of fine spider’s web, that was well recognized both by celiac disease antibodies and TG2-specific monoclonal mouse antibodies (Figure 2). However, the extracellular matrix produced in these conditions did not form a continuous fiber network and a lot of the TG2 remained in the cell bodies.

[0370] Further experiments were directed to explore whether the production of this type of matrix - initially confined to the slide’s surface occupied by individual cells - is reproducible or could be expanded into a bigger fiber network. When the cells were left for days in undiluted EGM-2 which contains only 2 V / V% FBS, both intracellular and matrix binding was observed with the celiac antibodies and the matrix structures grew only very slowly. However, changing the medium to Medium 199 with 10 V / V% FBS and 10 V / V% EGM-2 on the second day after plating resulted in the loss of intracellular reactivity with rapid growth of antigenic matrix structures attaining an aspect of mature endomysium by 72 hours of culture (Figure 3). The binding pattern of celiac antibodies is very similar (Figure IF) to the honeycomb binding pattern of celiac disease antibodies in monkey esophagus. This HUVEC-ECM (biomatrix) was then used after acetone fixation for the evaluation of diagnostic performance as EMA substrate on an extended number of patient samples (see Example 2), as well as for determining the presence of molecular EMA characteristics (see Example 8). The HUVEC-ECM matrix was recognized by TG2’s epitope-2 specific cloned antibodies as well, while epitope-1 directed antibodies showed inconsistent binding.

[0371] In general, unless otherwise specified, this HUVEC-ECM matrix was used in further Examples (as a matrix prepared according to Example 1).

[0372] EXAMPLE 2: Diagnostic testing

[0373] Patient samples

[0374] To assess the sensitivity, specificity and diagnostic accuracy of EMA tested on the prepared HUVEC-ECM (HUVEC biomatrix), serum samples were prospectively collected between May - September 2023 at the Coeliac Disease Centre, Heim Pal National Paediatric Institute, Budapest, and tested from 90 patients (median age: 6.3 years, range 2.4-40) consecutively referred for CD evaluation because of previous positive transglutaminase or gliadin antibody results in outside laboratories or detected by various rapid home tests. Details on demographic data, the initial antibody tests and the clinical symptoms leading to the testing are presented in Table 1. All these patients were evaluated with standard hospital anti-transglutaminase antibody ELISA (TGA-IgA) and EMA tests on monkey esophagus, human umbilical cord and human appendix substrates for IgA class celiac antibodies as part of their diagnostic workup from the same sample which was used number-coded also for the evaluation on the HUVEC biomatrix upon informed consent. None of these patients was IgA deficient, but if the test before referral was positive only for IgG class celiac antibodies, also TGA-IgG and EMA-IgG tests were performed. CD diagnosis was established according to ESPGHAN 2020 guidelines [Husby et al. 2020] by histology or by the non-biopsy route as appropriate.

[0375] Table 1. Initial characteristics of the included patients. TGA means antibodies to transglutaminase 2, DGP means antibodies to deamidated gliadin peptides.

[0376] Thus, serum samples from 90 patients were tested in a prospective fashion using the standard indirect immunofluorescent method. Results were compared with IgA antibody concentrations measured in clinical TGA IgA and with EMA IgA test results on a composite block of monkey esophagus, human umbilical cord and human appendix tissue sections. None of the included patients was IgA deficient. The EMA testing with tissues was performed with a screening serum dilution of 1:2.5 in PBS and if any of the substrates showed a celiac-type positivity, the result was regarded as positive. Since the ibidi chambers required a filling volume of 100 pL for the incubation steps, which is approximately 4 times more than the drops (25 pl) used for incubating tissue sections with serum dilutions, 1:10 dilution of serum samples in PBS was used for the EMA testing on HUVEC-ECM substrate. (Thus, this 1: 10 serum dilutions on the HUVEC-ECM corresponds to 1 :2.5 dilutions on tissue sections.) HUVEC-ECM

[0377] The diagnostic testings were performed with HUVEC-ECM prepared according to Example 1 : HUVEC- ECM was grown for 72 hours in 18-well p-slides with ibi-Treat surface (ibidi®, Grafeling, Germany) manufactured from acetone-resistant plastic with seeding the HUVEC cells in 100 V / V% EGM-2 and changing the medium on the second day to Medium-199 with 10 V / V% FBS and 10 V / V% EGM-2 (the other materials and methods are described in Example 1).

[0378] Clinical antibody testing

[0379] TGA levels were measured by a sensitive red blood cell transglutaminase-based sandwich ELISA described earlier [W02002 / 086509] and validated in the ProCeDe international study [Werkstetter et al. , 2017] . In brief, the TG2 antigen was captured to the ELISA plate coated with polyclonal antibodies to TG2, followed by the incubation with patient serum samples diluted 1: 100 and peroxidase -conjugated anti-human IgA (DAKO). Results were calculated from a 4-parameter fit curve utilizing 6 calibrators (0, 3, 7, 16, 40, 100 U / ml) adjusted to calibrator values of the Phadia Celikey Varelisa test. Cut-off for positivity (IxULN) was 3 U / mL.

[0380] EMA testing on tissues was performed on a composite substrate of unfixed frozen sections of monkey esophagus, human umbilical cord and human appendix tissues using indirect immunofluorescent method with serum diluted 1:2.5. Results were regarded as positive when any of these substrates showed IgA binding around endomysial structures in conjunction with classical additional components of the EMA binding (staining of subepithelial reticulin fibers and of those in lymphoid follicles in the esophagus and appendix, and positivity of Wharton jelly fibroblasts in the umbilical cord).

[0381] Results

[0382] TGA-ELISA results for IgA celiac antibodies were positive (>3 U / mL, IxULN) in 73 of the 90 samples, EMA-IgA on tissues in 62 and EMA-IgA on HUVEC-ECM in 67 samples. In 17 samples all three antibodies were negative. In these seronegative patients the earlier result at the referral site was low TGA -IgA ELISA laboratory positivity in 4 cases, isolated TGA-IgG positivity in 7, DGP-based home test positivity in 4 and doubtful other test positivity in 2 (thus, most of these patients had only TGA-IgG or DGP, thus non-specific antibody positivity earlier). At the time of the present study, none of these patients was on a gluten-free diet, nor did any of the patients with earlier IgG class antibody positivity display TGA-IgG or EMA-IgG positivity in our clinical testing, and they were not IgA deficient.

[0383] EMA results in relation to TGA-IgA concentrations are shown in details in Table 2. Notably, all 51 highly TGA-ELISA positive samples were also positive for EMA detected on HUVEC-ECM, and all these patients reached a final CD diagnosis according to ESPGHAN 2020 criteria (Figure 10). None of the patients negative for TGA-IgA ELISA showed EMA positivity. From the 22 samples found low positive in TGA-ELISA, 16 showed EMA positivity on HUVEC-ECM, whereas only 11 on conventional tissues, thus EMA testing on HUVEC-ECM was more sensitive than classical EMA testing. Even in the very low TGA-IgA positivity range of l-1.9xULN, two samples displayed positivity on HUVEC-ECM. It is well known that in the borderline positivity and low positivity range, TGA-ELISA is usually more sensitive than the EMA test based on microscopic evaluation. However, the novel EMA testing method on HUVEC-ECM in 1 : 10 serum dilutions proved to yield more positive results than evaluation on tissues in 1 :2.5 serum dilutions. Explanation for this can be the very low background in the cell-assembled ECM that made the reticular network of the EMA reaction more easily recognizable, but also the observation that in the HUVEC-ECM cells generated at places very high TG2 density, which trapped antibodies and could made them visible even if they were present in the serum in low quantities.

[0384] Figure 10 shows the detailed flowchart of further diagnostic evaluations performed according to the ESPGHAN 2020 guidelines. In 50 of the 51 the patients with >10xULN TGA-IgA positivity, all ESPGHAN criteria for the non-biopsy diagnosis were present. Patients with positive but <10xULN TG2 -ELISA values underwent endoscopy and biopsy according to the ESPGHAN 2020 guidelines, regardless of EMA positivity. The diagnosis of celiac disease was established in 50 patients by the non-invasive route. 24 patients underwent small intestinal biopsy, of whom 18 had Marsh II -III histology results, confirming celiac disease in altogether 68 (76%) of the 90 enrolled patients. Taking into account the final diagnosis of CD based on villous atrophy Marsh II or III plus ESPGHAN non-invasive criteria, diagnostic accuracy values were 97.3% for TG2-ELISA, 92.9% for conventional EMA on tissues and 95.9% for the here described novel EMA test with HUVEC-ECM. Table 3 shows the sensitivity, specificity, positive predictive and negative predictive values and diagnostic accuracy of the three antibody tests in relation to the final diagnosis. The novel EMA test with HUVEC-ECM was equally specific as the traditional EMA test, but it had higher sensitivity (95.6% versus 91.2%) and diagnostic accuracy (95.9% versus 91.9%) than EMA testing on tissue sections, whereas TGA ELISA was more sensitive but less specific. Given the prospective study was performed in current clinical environment, most patients with negative TGA and EMA did not have justification for endoscopy which decreased specificity values for TGA.

[0385] Table 2. Comparison of EMA results obtained with HUVEC-ECM substrate with results of conventional diagnostic tests for celiac disease in the prospectively tested cohort. HUVEC-ECM means extracellular biomatrix produced by human umbilical cord vein endothelial cells, TGA means antibodies to transglutaminase 2 measured by ELISA, EMA means endomysial antibodies, ULN means upper limit of normal (3 U / mL) in TGA-ELISA.

[0386] Table 3. Diagnostic performance of the EMA testing on the HUVEC-ECM biomatrix, on conventional tissue sections and of the ELISA detection of celiac antibodies in the prospectively investigated cohort in relation to the final diagnostic outcome. EMA means endomysial antibodies, TGA means transglutaminase antibodies, HUVEC-ECM means extracellular biomatrix produced by human umbilical cord vein endothelial cells, 95% confidence intervals are shown in brackets. f Only results of patients with biopsy results or qualifying for the non-biopsy celiac disease diagnosis were taken into account for calculating diagnostic performance parameters.

[0387] {Most patients with negative results did not undergo biopsy and thus were not in the calculation

[0388] EXAMPLE 3: Effect of culturing time and growth media

[0389] Our initial experiments (see Example 1) indicated that human umbilical cord vein-derived endothelial cells (HUVECs) express the celiac TG2 autoantigen which also appears in their extracellular matrix (ECM), but is only visible for celiac serum samples if either acetone or methanol is used as a fixative, that preserve its conformational celiac epitopes [further fixatives were tried later, see Example 5], We observed that two factors are determining the amount and morphological organisation of produced matrix into endomysial structures: adhesion efficiency of the cells and their protein synthesis activity.

[0390] In these experiments, undiluted EGM-2 medium was used for the plating and initial growing of the HUVECs. However, after the second day, we either left the medium unchanged (i.e. HUVEC cells were further cultured in undiluted EGM-2 medium), or the medium was changed to Medium 199 supplemented with 10% V / V FBS and 10% V / V EGM-2 or was changed to Medium 199 supplemented with 10% V / V FBS only.

[0391] Thus, it was explored whether sensitivity could be even more enhanced by culturing HUVEC cells for longer time than 72 hours in different culture media. Six serum samples (3 positives, 2 borderlines and 1 negative) were tested at 4, 5 and 6 days of culture. Results are presented in Table 4 and Figure 4. It was observed that extracellular matrix network structures were more abundant and more brightly stained at 5 days in all cases, but there was no substantial increase in sensitivity for borderline samples when the culturing time extended over 4 days. Again, Medium 199 with 10 V / V% FBS and 10 V / V% EGM-2 proved to be optimal - the organized antigenic biomatrix attaining an aspect of mature endomysium appeared in 3 -4 days and further grew until day 6 with gradual loss of most intracellular positivity. Although keeping the cells in undiluted EGM-2 resulted at day 6 in bigger size extracellular matrix structures, their production was delayed compared to Medium 199 with 10 V / V% FBS and 10 V / V% EGM-2, and cell bodies stained for much longer time (at day 6 it still displayed an intensive intracellular positivity and less matrix positivity), indicating that in EGM-2 cells externalize TG2 slower and less well. Cell body staining is disadvantageous for the diagnostic EMA evaluation, since also other, non-TG2 antibodies can attach to the cytoplasm decreasing specificity. In contrast, at 4 and 5 days only extracellular EMA staining was observed on matrices grown in Medium 199 with 10 V / V% FBS and 10 V / V% EGM-2 which was specific for CD. When Medium 199 with 10 V / V% FBS only was applied, sensitivity has dropped and mostly only cell body staining was observed.

[0392] Table 4. Effect of culturing time and growth media for the formation of HUVEC-ECM matrix antigenicity for celiac disease antibodies

[0393] (a)with bright positivity of cell bodies (only intracellular signal)

[0394] EXAMPLE 4: Adhesion and extracellular matrix production of HUVEC cells

[0395] In order to explore the requirements of HUVEC cells to adhere to chamber surfaces and for HUVEC -ECM production, we performed an experiment where full EGM2 or its components as listed in Table 5 were used for plating the cells and during the initial two days of culture. EGM-2 is a complex medium containing seven growth factor additives. In order to determine which of its components contribute to the production of the organised extracellular matrix, we cultured the cells omitting one additive at a time or adding only one additive to the basal medium. Adhesion was evaluated at 24 hours after plating by phase -contrast microscopy images. Then cells were grown for the following days either in the same medium or in Mediuml99 with 10 V / V% FBS and 10 V / V% EGM-2, and HUVEC-ECM production was evaluated on the 6th day after acetone fixation and stainings by celiac antibodies for IgA binding.

[0396] Results

[0397] Adhesion of HUVEC cells was poor after 24 hours if only the basal medium was used without growth factors and they did not produce CD antigenic ECM by the 6th day. In general, ECM production was inferior if the medium was not shifted to the higher FBS content medium after 2 days. Omission of either heparin, ascorbic acid (a widely used additive to promote ECM production [Assuncao et al., 2019]) or hydrocortison, as well as omission of R3-IGF-I or hFGF-B did not influence significantly either the adhesion or ECM production. HUVEC- ECM production was enhanced by VEGF even if it was used as a single growth factor in the system, but it did not promote adhesion. EGF promoted adhesion and in some extent, also ECM formation. In conclusion, VEGF and EGF seem to play a role in the production of HUVEC-ECM for celiac diagnostic purposes. However, the optimized composition of the complete EGM-2 medium seemed to be more advantageous for cell fitness and consequently, for enhanced and better ECM production. Table 5. Adhesion and extracellular matrix production of HUVEC cells with and without growth factors. Basal EBM means Endothelial Cell Growth Basal Medium-2 (Lonza CC-3156) + 2%FBS, Penicillin-Gentamycin. ECM means extracellular matrix with celiac antigenicity, VEGF means vascular endothelial growth factor, EGF means human epithelial growth factor, R3-IGF-I means human insulin-like growth factor-I, and hFGF-B means human fibroblast growth factor.

[0398] EXAMPLE 5: Fixatives

[0399] Acetone fixation is not compatible with most plastic surfaces used in cell culture and treated slides require storage at -20°C, therefore we explored in further experiments whether alternative fixatives would be suitable to preserve the celiac antigenicity of the TG2 antigen in the produced extracellular matrix. Simple drying was not sufficient to retain the desired structure since cells or ECM often detached during the incubation with serum and secondary antibodies. As shown earlier, paraformaldehyde destroyed the sensitive TG2 antigen and this was similar also with formaldehyde. Next, methanol fixation was tried. For the satisfactory preservation of the structures this required the use of specific buffers of 100 mM 2-(4-morpholino)ethanesulfonic acid (MES) with 1 mM EGTA and 1 mM MgC'F. H 6.9. The final fixation solution (methanol -MES) was applied after 3 washings in PBS for 10 minutes and contained 90 V / V% methanol and 10 V / V% MES buffer precooled to -20°C. In HUVEC-ECM prepared with methanol-MES celiac antibodies were able to bind to the TG2 antigen in similar patterns as after acetone fixation.

[0400] Results

[0401] Double stainings of methanol-MES treated HUVEC-ECM with celiac IgA and with antibodies to various other extracellular matrix proteins revealed a similar structural organization as known from similar experimental stainings on tissue endomysium [Korponay-Szabo et al. 2000, 2003], The celiac antibody binding co-localized with extracellular TG2 recognized by CUB7402 monoclonal anti-TG2 antibodies (NeoMarkers) and was seen on the surface of fibronectin, in close relation with laminin, vinculin and VE-cadherin (Figure 5). In contrast, the celiac positivity pattern was clearly distinct from other cellular antigens, like gamma-tubulin (not shown). Figure 5 also demonstrates that while TG2 is only found in the matrix co-localized with fibronectin, the fibronectin expression is more abundant and consequently not all fibronectin has attached TG2.

[0402] EXAMPLE 6: IgA-deficiency

[0403] In order to explore whether the HUVEC-ECM would also be suitable to detect celiac antibodies in IgA deficient patients, acetone-treated slides with HUVEC-ECM were incubated with IgA deficient celiac serum samples containing IgG class endomysial antibodies (n=5) and with a non-celiac negative IgA deficient serum sample as control. After washings in PBS, rabbit antibodies against human IgG (Dako) followed by Alexa568- conjugated anti-rabbit goat antibodies (Molecular Probes) were used for the visualization of the binding in red. IgG celiac antibodies produced a similar staining as IgA celiac antibodies, but the background staining of cell bodies was slightly higher (Figure 6). Nonetheless, the negative control sample did not produce a positive binding pattern.

[0404] EXAMPLE 7: Decellularized HUVEC matrix

[0405] Further experiments were conducted in order to determine whether decellularized HUVEC-ECM would better perform in some applications, e.g. detection of IgG class celiac antibodies to TG2 or for the combined detection of both IgA and IgG class TG2 -specific celiac antibodies when total serum IgA status is unknown in clinical samples.

[0406] A second, independent, retrospective cohort of 61 untreated CD patients (median age: 9.3 years, range 3-21) with Marsh II-III lesions and positive TGA-IgA in ELISA, but negative for conventional EMA-IgA results on tissues was tested on decellularized HUVEC biomatrix to further assess sensitivity. Serum samples from five IgA- deficient CD patients with negative IgA but positive IgG TGA and EMA antibodies having Marsh III histology and three non-celiac IgA deficient seronegative controls were also tested.

[0407] Clinical samples were used in both 1:2.5 and 1:10 serum dilutions, because removal of cells from the HUVEC-ECM resulted in a very low background at the staining and allowed us using more concentrated serum samples.

[0408] As a first step, acetone-fixated slides with HUVEC-ECM produced according to Example 1 were exposed to various detergents to disrupt cell bodies and washing out cytoplasmic material before incubation with the patient sample. It was observed that celiac antigenicity of ECM remained intact after treatment with 0.1-1% TritonXIOO or 0.1-1% sodium deoxycholate for 20 minutes. This method, however, did not entirely remove the cells, although it resulted in decreased background staining and a well evaluable positivity pattern when IgG class celiac antibodies were tested, whereas IgA-deficient, non-CD controls were negative.

[0409] When the HUVEC-ECM preparates were exposed for 20 minutes to 0.1%, 0.25%, 0.5% or 1% sodium deoxy cholate prior to acetone-fixation and followed by gentle washing with PBS 3 times, cells were completely removed and only CD-antigenic extracellular matrix with previously described endomysial morphological characteristics remained on the slides. When testing the celiac patient cohort negative for EMA on monkey esophagus and on all other clinically used EMA substrate tissues, a clear positive EMA reaction was seen in 1 : 10 dilutions in 16 samples and in 1:2.5 dilutions in 12 more samples on the decellularized HUVEC-ECM, thus altogether in 28 (45%) of the 61 samples, which demonstrates a markedly improved sensitivity over conventional EMA testing.

[0410] The decellularized ECM used as substrate for the immunofluorescent testing provided superior and easily evaluable results for the detection of IgG class celiac antibodies as well (Figure 7) when incubations with the serum samples and secondary antibodies steps were performed as in Example 6, and much less disturbing background was obtained than in Example 6 with not decellularized biomatrix. Moreover, adding a green label for IgA binding as the last step, this method was able to reveal that the sample is IgA deficient, i.e. no IgA binding occurred in green, whereas IgG celiac antibodies bound in the endomysial pattern in red (Figure 7D-F). In contrast, IgA competent celiac patient samples produced bindings both in green (IgA) and red (IgG), which merged to yellow indicating overlap (Figure 7A-C). Thus, this double staining recognizes IgA-deficient CD patients also in the absence of total serum IgA measurement and when EMA IgA is negative. Using the double set of secondary antibodies in the order a) rabbit anti-human IgG antibodies (Dako), b) Alexa568-conjugated goat anti-rabbit antibodies (Molecular Probes), and c) FITC-conjugated rabbit anti-human IgA antibodies (Dako), with PBS washings between these steps eliminated any possible cross-reactions and enabled to detect both IgA and IgG class celiac antibodies on one and the same ECM preparate.

[0411] EXAMPLE 8: Macromolecular composition of the matrix

[0412] HUVEC-ECM slides prepared according to Example 1 were studied by multicolor fluorescent labelling for a number of extracellular, intracellular or basement membrane proteins and their relation to the observed endomysial type binding pattern of celiac disease antibodies in monkey esophagus. For this experiment, biomatrix preparates were fixated in buffered methanol (see Example 5), to well preserve the structure and intracellular proteins and the matrix was not decellularized. We observed that antigenicity for CD IgA binding was preserved after methanol fixation and it showed complete co -localisation with TG2, partially co-localised with fibronectin and laminin, but did not overlap with cellular antigens, such as vinculin and VE-cadherin, the latter used as endothelial marker (Figure 5). This finding is consistent with the knowledge that the antigenic component for celiac antibodies is extracellularly localised TG2, bound to the surface of fibronectin. Desmin staining was negative, as expected for endothelial cells. Moreover, cloned antibodies selectively targeting coeliac epitope-2 of TG2 also showed the same binding pattern (data not shown).

[0413] To further prove that the celiac antigen exposed in the HUVEC-ECM is extracellular, we removed the cells from the slides by 0.1% sodium deoxycholate prior to fixation. This procedure did not affect either the intensity of CD antibody binding or the structure of the produced biomatrix.

[0414] The decellularized HUVEC-ECM was used for analyzing the macromolecular composition of the biomatrix in comparison with monkey esophagus tissue sections (Figure 8). Monkey esophagus sections and HUVEC-ECM yielded identical results for fibronectin, collagen III and IV positivity and distribution patterns, known major components of endomysium, and both had similarly low expression of collagen I. Taken together, these results indicate that the produced HUVEC-ECM has endomysial character which makes it suitable for EMA testing in diagnostics.

[0415] EXAMPLE 9: Structural identification of the matrix by silver staining

[0416] The endomysium in tissues is defined by histological means because of its very complex and not fully known molecular composition, which extends beyond collagen III, collagen IV, fibronectin, laminin and TG2 to other biomolecules and carbohydrates which give argyrophilic properties to endomysial structures. Hence the endomysium is identified in histology by the positive staining to black (or purple-black) in the silver impregnation method introduced by Gdmori in 1937 for visualizing reticulin fibers which he claimed to be the most sensitive to show fine fibers and to differentiate reticulin fibers from other types of connective tissues and their collagen. The method was improved and simplified by many pathologists (for example by Slidders et al. in 1958, and by Krutsay in 1988).

[0417] For establishing whether the biomatrix produced according to Example 1 represents endomysial type extracellular matrix, we followed the description of Krutsay and stained frozen-cut monkey esophagus tissue sections as well as both cell-containing and decellularized HUVEC-ECM matrices prepared on glass slides after fixation with 4% PBS-buffered formaldehyde (as suggested by the original protocol). In brief, sections were oxidized with 5% potassium permanganate for 5 minutes, and after washings in water between steps they were successively treated with 5% oxalate for 2 minutes, followed by 2% freshly prepared iron-ammonium-sulphate for 2 minutes and by 2.5% freshly prepared silver nitrate in 1% ammonia (also called as ammoniacal silver) for 1 minutes. After this step, the slides were immersed without washing into freshly filtered 5% formaldehyde to develop the color and after extensive washings the slides were fixated with 5% sodium thiosulphate and examined under a light microscope. The expected results of the staining are black color for the endomysium and reticulin fibers and middle brown color for collagen and other ECM components.

[0418] Monkey esophagus sections showed black staining for endomysial structures of the muscular layers (Figure 9 A, asterisks) and for the reticulin lining underneath the epithelium (Figure 9 A, arrow); connective tissue structures in the submucosa and lamina propria stained to middle brown (Figure 9A, triangle). The brown staining corresponded to the connective tissue containing collagen I in monkey esophagus detected by monoclonal collagen I-directed antibodies in serial frozen sections of the same specimen by applying immunofluorescence (Figure 9B). The black silver reaction corresponded to the EMA positive reaction with celiac IgA antibodies on monkey esophagus seen in immunofluorescence (Figure 9C, asterisk).

[0419] In the cell-containing HUVEC-ECM treated in the same way, the fibers antigenic for the celiac disease patient samples and containing TG2 in the previously described examples stained to black, while other ECM components and cell bodies remained brown (Figure 9D). The clear positive silver staining reaction was also demonstrated in the decellularized matrix (prepared according to Examples 1 and 7) in honeycomb pattern, consisting of a well-developed argyrophilic fine fiber network stained black (Figure 9E). This honeycomb pattern corresponds to the structural organisation or the celiac antigen TG2 and to the celiac disease serum binding pattern in immunofluorescence (Figure 9F).

[0420] Taken together, these results confirm that the HUVEC-ECM prepared according to Example 1 has endomysial characteristics and is not a common, connective tissue-type ECM, but it is a highly specialized, endomysial type matrix.

[0421] EXAMPLE 10: Evaluation of positivity for celiac antibodies by computer

[0422] Similarly to that described in Example 5, decellularized matrix preparations were double -stained with 10 celiac or 10 non-celiac patient serum samples followed by FITC-conjugated anti-human IgA antibodies, and as a reference for the expected positive binding pattern, with CUB7402 monoclonal anti-TG2 antibodies labelled with Alexa-Fluor 647 anti-mouse secondary antibodies emitting a fluorescent signal outside the visible wavelength range. During the examination, the fluorescent green (FITC) signal could be well observed, but the CUB7402 signal was invisible for the human eye. However, the CUB7402 signal was detected by the camera attached to the microscope both at 588 nm (red) and 647 nm (near red filters). When both images were captured and analyzed, >99% pixel superimposition was observed for all celiac samples using the Image J software, regardless of the intensity of the FITC stainings, whereas it was lower than 20% for all non-celiac samples. A visual co-localization could only be obtained for the 7 patient samples with a moderately or strongly positive EMA pattern with the patient images in green merged with the respective CUB7402 images in magenta, resulting in grey or white combination pattern and disappearance of the original colors (Fig. 11). Demonstrating the co-localization was not successful when the EMA positivity was faint. Therefore, the computer analysis proved to be superior and it was a useful aid to the evaluation, without disturbing the visual judgment by the human observer. INDUSTRIAL APPLICABILTY

[0423] The cell-assembled matrices disclosed herein can be used as a substrate in methods detecting celiac diseasespecific antibodies. For example, they can be used in EMA assays (indirect immunofluorescence method), replacing the less and less available classical substrates, i.e. tissue sections. Thus, the cell-assembled matrix may be used in a method for in vitro diagnosis of celiac disease. Furthermore, the decellularized cell-assembled matrices disclosed herein may be used not just for detecting IgA celiac antibodies, but also for reliably detecting IgG celiac antibodies, or even both IgA and IgG antibodies with double staining. The double staining has the advantage that IgA-deficiency need not be taken into account, i.e. there is no need to know beforehand whether the subject to be diagnosed is an IgA deficient subject or not. Further, a double or multi-color staining for the investigated sample and TG2 or other ECM antigens may be utilized visually or in an image-processing system that detects degree of co-localization to improve the reliability of the evaluation, or make the evaluation less or not observer dependent.

[0424] REFERENCES

[0425] Absah I, Rishi AR, Gebrail R, Snyder MR, Murray JA (2017). Lack of Utility of Anti-tTG IgG to Diagnose Celiac Disease When Anti-tTG IgA Is Negative. J Pediatr Gastroenterol Nutr.; 64:726-729

[0426] Assuncao, Marisa et al. (2020). Cell-Derived Extracellular Matrix for Tissue Engineering and Regenerative Medicine. Front. Bioeng. Biotechnol. 8:602009

[0427] Cardoso I, Stamnaes J, Andersen JT, et al. (2015). Transglutaminase 2 interactions with extracellular matrix proteins as probed with celiac disease autoantibodies. FEBS J.; 282:2063-75.

[0428] Csipo A, Kiraly R, Toth B, et al. (2011). Reaction with the recombinant tag as cause of false positive transglutaminase antibody results. United European Gastroenterology Week 2011, Stockholm (abstract)

[0429] Dios A, Srinivasan B, Gyimesi J, et al. (2022). Changes in Non-Deamidated versus Deamidated Epitope Targeting and Disease Prediction during the Antibody Response to Gliadin and Transglutaminase of Infants at Risk for Celiac Disease. Int J Mol Sci.; 23:2498.

[0430] Feighery C, Abuzakouk M, Liddy C, et al. (1998). Endomysial antibody detection using human umbilical cord tissue as substrate: reactivity of cells in Wharton's jelly. British Journal of Biomedical Science, 55: 107-110.

[0431] Giersiepen K, Lelgemann M, Stuhldreher N, et al. (2012). ESPGHAN Working Group on Coeliac Disease Diagnosis. Accuracy of diagnostic antibody tests for coeliac disease in children: summary of an evidence report. J Pediatr Gastroenterol Nutr.; 54:229-41.

[0432] Gomori G. (1937). Silver Impregnation of reticulum in paraffin sections. Am J Pathol. 13:993 -1002

[0433] Hallstrom O. (1989) Comparison of IgA-class reticulin and endomysium antibodies in coeliac disease and dermatitis herpetiformis. Gut. 30: 1225-32.

[0434] Husby S, Koletzko S, Korponay-Szabo I, etal. (2020). European Society Paediatric Gastroenterology, Hepatology and Nutrition Guidelines for Diagnosing Coeliac Disease 2020. J Pediatr Gastroenterol Nutr.; 70: 141-156.

[0435] Kanchan K, Ergiilen E, Kiraly R, et al. (2013). Identification of a specific one amino acid change in recombinant human transglutaminase 2 that regulates its activity and calcium sensitivity. Biochem J.; 455:261-72.

[0436] Korponay-Szabo IR, Sulkanen S, Halttunen T, et al. (2000). Tissue transglutaminase is the target in both rodent and primate tissues for celiac disease-specific autoantibodies. J Pediatr Gastroenterol Nutr.; 31:520-7.

[0437] Korponay-Szabo IR, Laurila K, Szondy Z, et al. (2003). Missing endomysial and reticulin binding of coeliac antibodies in transglutaminase 2 knockout tissues. Gut.; 52: 199-204. Korponay-Szabo IR, Halttunen T, Szalai Z, et al. (2004). In vivo targeting of intestinal and extraintestinal transglutaminase 2 by coeliac autoantibodies. Gut.; 53:641-8.

[0438] Krutsay M. (1988) Uber die Silverimpregnation des Bindegewebes. Zbl allg Pathologie 134:518-520.

[0439] KurppaK, Collin P, ViljamaaM, HaimilaK, Saavalainen P, Partanen J, etal. (2009). Diagnosing mild enteropathy celiac disease: a randomized, controlled clinical study. Gastroenterology. ;136:816-23.

[0440] Ladinser B, Rossipal E, Pittschieler K. (1994). Endomysium antibodies in coeliac disease: an improved method. Gut; 35:776-8.

[0441] Mearin ML, Agardh D, Antunes H, et al. (2022). ESPGHAN Special Interest Group on Celiac Disease. ESPGHAN Position Paper on management and follow-up of children and adolescents with celiac Disease. J Pediatr Gastroenterol Nutr. 2022; 75:369-386.

[0442] Myrsky E, Syrjanen M, Korponay-Szabo IR, et al. (2009). Altered small-bowel mucosal vascular network in untreated coeliac disease. Scand J Gastroenterol.; 44: 162-167.

[0443] Palatka K, Serfozo Z, Vereb Z, et al. (2006). Effect of IBD sera on expression of inducible and endothelial nitric oxide synthase in human umbilical vein endothelial cells. World J. Gastroenterol.; 12,1730-1738.

[0444] Sblattero D, Florian F, Azzoni E, et al. (2002). The analysis of the fine specificity of celiac disease antibodies using tissue transglutaminase fragments. Eur J Biochem.; 269:5175-81.

[0445] Shaw A, Toth BB, Kiraly R, et al. (2021). Irisin Stimulates the Release of CXCL1 From Differentiating Human Subcutaneous and Deep-Neck Derived Adipocytes via Upregulation of NFKB Pathway. Front Cell Dev Biol. 2021 Oct 11;9:737872.

[0446] Sheppard AL, Elwenspoek MMC, Scott LJ, et al. (2022). Systematic review with meta-analysis: the accuracy of serological tests to support the diagnosis of coeliac disease. Aliment Pharmacol Ther.; 55(5):514-527.

[0447] Shiha MG, Nandi N, Raju SA, et al. (2024). Accuracy of the No-Biopsy Approach for the Diagnosis of Celiac Disease in Adults: A Systematic Review and Meta-Analysis. Gastroenterology.; 166:620-630.

[0448] Simon-Vecsei Z, Kiraly R, Bagossi P, et al. (2012). A single conformational transglutaminase 2 epitope contributed by three domains is critical for celiac antibody binding and effects. Proc Natl Acad Sci U S A; 109:431-6.

[0449] Slidders W, Fraser DS, Lendrum AC (1958). Silver impregnation of reticulin. J Pathol Bacteriol. 75:478-81.

[0450] Sulkanen S, Halttunen T, Laurila K, et al. (1998a). Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology.; 115(6): 1322-8.

[0451] Sulkanen S, Halttunen T, Marttinen A, et al. (1998b). Autoantibodies in celiac disease: importance of fibroblasts. J Pediatr Gastroenterol Nutr.; 27:206-213.

[0452] Werkstetter KJ, Korponay-Szabo IR, Popp A, et al. (2017). Accuracy in Diagnosis of Celiac Disease Without Biopsies in Clinical Practice. Gastroenterology.; 153:924-935.

[0453] Whelan A, Willoughby R. & WeirD. (1996). Human umbilical vein endothelial cells: a new easily available source of endomysial antigens. European Journal of Gastroenterology & Hepatology, 8:961-966.

[0454] Wolf J, Jahnke A, Fechner K, et al. (2016). Primate liver tissue as an alternative substrate for endomysium antibody immunofluorescence testing in diagnostics of paediatric coeliac disease. Clin Chim Acta.; 460:72-77

Claims

CLAIMS1. Use of an in vitro cell culture derived cell-assembled matrix in a method for detecting celiac diseasespecific antibodies in a sample, said cell-assembled matrix comprising a reticulin fiber network comprising fibronectin, and comprising, on the surface of fibronectin, type 2 transglutaminase (TG2).

2. The use according to claim 1, wherein said cell-assembled matrix is used in an endomysial antibody (EMA) assay as an EMA substrate, preferably as an artificial endomysium or instead of a tissue section.

3. The use according to any one of claims 1 to 2, wherein the cell-assembled matrix is an endothelial cell culture derived extracellular matrix.

4. The use according to any one of claims 1 to 3, wherein the cell -assembled matrix is derived from a cell culture of human umbilical cord vein-derived endothelial cells (HUVECs).

5. The use according to any one of claims 1 to 4, wherein the cell-assembled matrix is an endomysium-type or endomysium-like matrix, organized into a reticulin fiber network which comprises type III collagen, in particular COL3A1, fibronectin, laminin, and preferably type IV collagen, and which is essentially free of type I collagen.

6. The use according to any one of claims 1 to 5, wherein said cell-assembled matrix has a mesh-like appearance, preferably a honeycomb -like appearance, when visualized by silver impregnation.

7. The use according to any one of claims 1 to 6, wherein the cell-assembled matrix comprises only one type of cell.

8. The use according to any one of claims 1 to 6, wherein the cell-assembled matrix does not comprise cells.

9. The use according to any one of claims 1 to 8, wherein- the cell-assembled matrix can bind celiac disease-specific anti-endomysium antibodies or anti-TG2 antibodies in mesh-like binding pattern, preferably in a honeycomb binding pattern, wherein preferably the anti-endomysium antibodies or anti-TG2 antibodies are IgA or IgG antibodies.

10. The use according to any one of claims 1 to 9, wherein the cell-assembled matrix is derived from a culture of HUVEC cells, the cell-assembled matrix comprises intact celiac epitopes of TG2, and the cell-assembled matrix is used in an endomysial antibody (EMA) assay as an EMA substrate.

11. Use of a solid phase matrix preparation, which comprises the cell-assembled matrix as defined in any one of claims 1 to 10 prepared on a support, in a method for detecting celiac disease-specific antibodies in a sample, preferably in an endomysial antibody (EMA) assay.

12. A method for producing a cell-assembled matrix, wherein said method comprises the following steps: a) seeding endothelial cells on a support, preferably a solid phase support, wherein the endothelial cells are seeded on the support in a medium for culturing endothelial cells, preferably in a medium containing VEGF and / or EGF, b) culturing the seeded cells of step a) in said medium; c) optionally, removing the cells from the culture obtained in step b), thereby producing an extracellular matrix without cells, d) fixing the culture obtained in step b) or the extracellular matrix obtained in step c) to the support with a fixative that maintains the integrity of celiac epitopes of type 2 transglutaminase (TG2) present in the extracellular matrix of the culture or in the extracellular matrix, respectively, preferably wherein a fixative maintains the integrity of the celiac epitopes of TG2 if, after fixation, the matrix shows a positive extracellular staining with a positive celiac control sample, e) thereby obtaining the cell-assembled matrix.

13. The method for producing a cell-assembled matrix according to claim 12, wherein said medium is selected from:- a medium for culturing endothelial cells, e.g. EGM,- a medium containing VEGF and / or EGF, or- Medium 199 supplemented with 5-15 V / V% fetal bovine serum (FBS) and with medium for culturing endothelial cells.

14. The method for producing a cell-assembled matrix according to any one of claims 12 to 13, wherein in step b), the culturing of the seeded cells is carried out in EGM, preferably EGM2 medium for 2 hours to 72 hours, which is then replaced by Medium 199 supplemented with fetal bovine serum (FBS) and EGM2, preferably supplemented with 5-15 V / V% FBS and 5-15 V / V% EGM, preferably EGM2.

15. The method according to any one of claims 12 to 14, wherein in step e) the cell-assembled matrix is obtained as a solid phase matrix preparation, wherein the solid phase matrix preparation comprises the cell- assembled matrix as defined in any of claims 1 to 10 prepared on a support, wherein preferably the cell-assembled matrix is endomysium-like matrix, which comprises TG2 antigen for celiac disease-specific antibodies.

16. The method according to any one of claims 12 to 15, wherein the endothelial cells are human umbilical cord vein-derived endothelial cells (HUVECs), the fixative used in step d) is acetone or methanol or a solution of methanol and MES buffer, the support is a slide, a chamber slide, a cover slip or a plate, wherein if acetone is used as a fixative in step d), the support has an acetone-resistant surface, the culturing of step b) is performed for 1 to 28 days, preferably for 1 to 6 days, preferably for at least 72 hours, preferably for 4, 5 or 6 days, and the medium for culturing endothelial cells is a medium for culturing HUVECs, such as EGM-2.

17. The method according to any one of claims 12 to 16, wherein the cells are seeded and cultured for 2 hours to 3 days in undiluted EGM-2 medium, and after that, the medium is changed to Medium 199 supplemented with 5-15 V / V% FBS and with 5-15 V / V% EGM-2 medium.

18. A cell-assembled matrix or a solid phase matrix preparation comprising the cell-assembled matrix produced by the method defined in any one of claims 12 to 17.

19. An in vitro cell-assembled matrix comprising a reticulin fiber network comprising fibronectin, and comprising, on the surface of fibronectin, type 2 transglutaminase (TG2), wherein the cell -assembled matrix is suitable for detecting celiac disease-specific antibodies in a sample, preferably wherein the celiac disease -specific antibodies are anti-endomysium antibodies (EMAs), or antitransglutaminase antibodies, preferably TG2-specific antibodies, wherein more preferably the antibodies are IgA or IgG antibodies, preferably wherein the cell-assembled matrix is as defined in any one of claims 1 to 10.

20. A solid phase matrix preparation which comprises the cell-assembled matrix according to claim 19 prepared on a support.

21. A method for detecting celiac disease -specific antibodies in a sample, wherein said method comprises the use of the cell-assembled matrix as defined in any one of claims 1 to 10, or the use of the solid phase matrix preparation comprising the cell -assembled matrix defined in claim 11, or the use of the cell -assembled matrix according to claim 18 or 19 or the solid phase matrix preparation according to claim 18 or 20, or the use of the cell-assembled matrix or the solid phase matrix preparation prepared according to any one of claims 12 to 17, as a substrate comprising the antigens for the celiac disease -specific antibodies, preferably wherein the celiac disease -specific antibodies are anti-endomysium antibodies (EMAs), or antitransglutaminase antibodies, preferably TG2-specific antibodies, wherein more preferably the antibodies are IgA or IgG antibodies.

22. A method for in vitro diagnosing or evaluating celiac disease in a subject, wherein said method comprises the following steps: a) providing a substrate comprising antigens for celiac disease -specific antibodies, wherein the substrate is the cell-assembled matrix as defined in any one of claims 1 to 10, or the solid phase matrix preparation comprising the cell -assembled matrix defined in claim 11, or the cell -assembled matrix according to claim 18 or 19 or the solid phase matrix preparation according to claim 18 or 20, or the cell- assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix prepared according to any one of claims 12 to 17, b) providing a sample obtained from the subject, c) incubating the substrate of step a) with the sample of step b), d) washing the incubated substrate to remove unbound antibodies, e) incubating detecting antibodies with the washed substrate of step d),f) washing the incubated substrate of step e) to remove unbound detecting antibodies, g) visualizing or quantifying the detecting antibodies bound to the washed substrate of step f), and h) based on the image or quantitative result obtained in step g), diagnosing the subject as having celiac disease or as not having celiac disease, or evaluating the subject as responding to celiac disease treatment or as not responding to celiac disease treatment.

23. The method according to claim 22, wherein the sample is a serum sample or plasma sample, the celiac disease-specific antibodies are anti-endomysium antibodies (EMAs) or anti-transglutaminase antibodies, preferably TG2-specific antibodies, wherein more preferably the antibodies are IgA or IgG antibodies, the subject is a human subject, and the antibodies of step a) and step d) are human antibodies, the detecting antibodies are fluorophore-conjugated anti-human antibodies and the visualization of step g) is carried out using a fluorescence microscope.

24. The cell-assembled matrix as defined in any one of claims 1 to 6 and 8 to 10, the solid phase matrix preparation comprising the cell-assembled matrix defined in claim 11, the cell-assembled matrix according to claim 18 or 19 or the solid phase matrix preparation according to claim 18 or 20, the cell -assembled matrix or the solid phase matrix preparation comprising the cell-assembled matrix produced according to any one of claims 12 to 17, or the method according to any one of claim 21 to 23, wherein the cell-assembled matrix is a decellularized matrix, i.e. a matrix that does not comprise cells.

25. A kit comprising- the solid phase matrix preparation comprising the cell -assembled matrix defined in claim 11, or the cell- assembled matrix or the solid phase matrix preparation according to any one of claims 18 to 20,- detecting antibodies, preferably fluorophore-conjugated secondary antibodies,- positive control,- negative control, and- buffer.

Images