AU707114B2 - Method for the diagnosis of coeliac disease - Google Patents

Description

WO 97/10508 PCT/IE96/00061 1 Description METHOD FOR THE DIAGNOSIS OF COELIAC DISEASE Technical Field This invention relates to a method for diagnosing coeliac disease or gluten-sensitive enteropathy and related conditions as hereinafter defined.

Background Art Coeliac disease is a permanent intolerance to gliadin (wheat protein) leading to intestinal villous flattening and crypt hyperplasia in susceptible subjects. Immune reactions to gliadin are considered to play a part in the pathogenesis of the disease, although the immunological mechanisms have not been fully elucidated.

Antigliadin antibodies (AGA), reticulin antibodies and endomysium antibodies (EmA) are found in serum samples from patients with coeliac disease. However, AGAs are not specific to coeliac disease and high serum AGA titres are found in individuals affected by other gastrointestinal disorders.

EmAs are closely related to reticulin antibodies and are found in most individuals with active coeliac disease and in the majority of individuals with dermatitis herpetiformis, who are also gluten sensitive.

By related condition herein is meant any condition which is characterised by the presence of one or more of antigliadin, reticulin or endomysium antibodies in a body fluid. These conditions are referred to collectively hereinafter as coeliac disease.

The diagnosis of coeliac disease is dependent on the histological evaluation of the small intestinal mucusal biopsy, more specifically the WO 97/10508 PCT/IE96/00061 2 jejunal biopsy. However, serological assessment is an increasingly important method of screening for the condition, especially as it represents a less invasive test therefor.

The measurement of IgA EmA antibodies is currently considered to be the most important single serological test (Ferreira, M. et al (1992); Gut 33 1633-1637). The current method for IgA EmA determination involves the use of microscope-mounted sections of monkey oesophagus.

Human umbilical cord is also used for measuring EmA antibodies (but requires lengthy processing (Ladinser, B. et al (1994); Gut 35 776-778).

For the detection of reticulin antibodies rat kidney, stomach or liver tissue is used.

Reticulin antibodies and EmA are considered by some to be closely related (Ferreira, M. et al supra). However, the exact nature of either antigen is not known.

Foetal lung fibroblasts have been shown to secrete noncollagenous protein molecules which bind to serum IgA from patients with coeliac disease. The purified molecules removed antibodies against reticulin and endomysium but not against gliadin from serum samples from coeliac disease patients (Marttinen, A. and Miki, M.

(1993); Pediatric Research 34 No. 4 420).

The requirement for monkey or human tissue for the serological screening of individuals with coeliac disease poses ethical and moral problems. Also the use of tissue samples for the detection of antibodies precludes the use of standard techniques such as enzyme linked immunosorbent assay (ELISA) and flow cytometry.

L_ D'Cruz, D.P. et al. (1991) Clin. Exp. Immunol. 85, 254-261 describes the use of human umbilical vein endothelial cells (HUVECs) as the solid phase in an ELISA assay to identify antibodies to endothelial cells in the sera of patients with systemic lupus erythematosus and to investigate whether these antibodies may be a marker for disease pattern and activity with particular reference to nephritis and vasculitis.

AMENDED

SHEET

WO 97/10508 PCT/IE96/00061 3 The present invention overcomes these problems by providing an in vitro method for the diagnosis of coeliac disease based on the use of cell suspensions specific for antibodies indicative of a condition of coeliac disease in a patient.

Disclosure of the Invention Accordingly, the invention provides a method for diagnosing coeliac disease and related conditions in a subject known or suspected of having such a condition, which comprises contacting a body fluid from said subject containing antibody particular to said condition with a suspension of cells from an immortal cell line having an epitopic site specific for said antibody and determining the extent of binding of antibody to said cells.

It will be appreciated that cell suspensions, especially single cell suspensions, are easier to examine than the tissue sections typical of the known methods. The use of a cell line means that substantial amounts of material for the antibody estimation can be prepared by culturing the cells and storing them for future use at a very low cost.

By immortal cell line herein is meant a cell line which can be repeatedly cultured and does not die.

Preferably, the cells are embryonic in nature.

A common feature of the tissues that are involved in the detection of coeliac disease specific antibody is their involvement at the mucosal surfaces as indicated by the examples of oesophagus and foetal lung fibroblasts mentioned above.

A further feature that can be identified as being common to tissues that bind antibodies found in patients with coeliac disease is that they are embyronic in nature. Marttinen, A. and Maki, M.

(supra) have shown that sera from patients with coeliac disease react WO 97/10508 PCTJIE96/00061 4 with foetal fibroblasts. However, foetal fibroblasts undergo a limited number of passages and, accordingly, do not satisfy the criteria for an immortal cell line required as a reagent for use in a routine diagnostic assay of the type described herein.

One type of preferred cell for use in accordance with the invention is a human umbilical cell.

Especially suitable cells for use in accordance with the invention are human umbilical vein endothelial cells (HUVECs).

Our studies have shown that HUVECs possess antigens that react with antibodies which closely correlate with EmA.

The HUVECs can be, for example, cells from a commercially available cell line such as ECV 304 cell line, No. 92091712.

Further suitable human umbilical cells are human umbilical artery smooth muscle cells and human umbilical vein smooth muscle cells. An example of the former type of cells are those available from Technoclone, Austria under Catalogue No. 62019/L/1 and an example of the latter type of cells are those available from the same organisation under the Catalogue No. 6209/L/1.

The antibodies determined in accordance with the invention will typically be of the IgA type. The reason why IgA EmA is the predominant antibody response in coeliac disease is not fully understood. However, it is postulated to be related to antigen presentation at the gastrointestinal level.

However, in the case of endomysium, IgG antibodies thereto have been detected in the sera of some patients.

Suitably, the cells are ruptured prior to or during contact with said body fluid.

I The body fluid can be any body fluid but is preferably serum.

Because the antigen for use in immunoassays in accordance with the invention is a cell line suspension, more especially a single cell suspension, various techniques can be used to detect bound antibody.

Especially suitable techniques are flow cytometry and enzyme immunoassay. However, other techniques that can be used include a fluorescent microscopic procedure.

The invention also provides a kit when used for carrying out the method hereinbefore defined containing a suspension of cells from S: an immortal cell line.

The invention will be further illustrated by the following Examples.

Modes for Carrying Out the Invention Example 1 S 15 Measurement of HUVEC antibody response by fluorescent microscopy The sera used in this Example were obtained from patients attending the Gastroenterology Clinic at St. James's Hospital, Dublin.

Twenty five patients had untreated coeliac disease, 16 were on a gluten free diet and 16 who had non-specific symptoms and a normal small intestinal biopsy acted as controls.

HUVEC ECV 304 No. 92091712 (European Collection of Animal Cell Culture) were grown in Rosemount Park Memorial Institute (RPMI) 1640 medium containing 10% foetal calf serum and gentamycin at 37 0 C in a humidified incubator with 5% CO 2 until confluent. The cells were harvested using trypsin EDTA and washed 2X in RPMI 1640.

WO 97/10508 PCT/IE96/00061 6 Immunofluorescence studies: 30g1 aliquots of 2X106 cells/ml were allowed to air dry on poly-L-lysine (Sigma) coated glass slides at room temperature The slides were then washed for 30 mins. in M phosphate buffered saline (PBS) pH 7.2. 30gl volumes of dilutions of a patient or control serum in PBS were layered over the cells and allowed to react for 30 mins. at RT. The slides were again washed in PBS for a further 30 mins. The cells were then stained with gl volumes of fluorescein isothiocyanate (FITC) labelled rabbit anti human IgA or IgG (Dako) diluted 1/20 for 30 mins. and washed for mins. in PBS. The slides were covered with a coverslip in glycerol/saline mounting media and visualised by immunofluorescence using a Leitz Ortholux (Ortholux is a trade mark) microscope with a watt mercury vapour lamp. Positivity was denoted by staining of the cell cytoplasm.

EmA and reticulin antibodies were measured using standard techniques. Briefly, EmA antibodies were measured using commercially available tissue sections of monkey oesophagus (Medica California). Patients' sera were diluted 1/5 in PBS and treated as for the HUVECs. Similarly, a composite block of rat kidney, liver and stomach was used to detect reticulin antibodies. Sections 4 micron thick were cut and mounted onto glass slides, sera were diluted and immunofluorescence carried out as described in the HUVECs and EmA studies. The sera were randomised with the normal controls and the observer was blinded as to the origin of each slide.

The results are shown in Table 1.

Table 1 Control Control Contd/...

WO 97/1 0508 PCTJIE96/00061 7 Table 1 contd/...

Patient No. 1gA IgA IgA Diagnosis EmA HUVEC ReTIC 3 neg neg neg Control 4 neg neg neg Control neg neg neg Control 6 neg neg neg Control 7 neg neg neg Cont rol 8 neg neg neg Control 9 neg neg neg Control neg neg neg Control I1I neg neg neg Control 12 neg neg neg Control 13 neg neg neg Control 14 neg neg neg Control neg neg neg Control 16 neg neg neg Control 17 neg neg nteg

TCD

18 neg ne g neg

TCD

19 neg neg neg

TCD

neg neg neg

TCD

21 neg neg neg

TCD

22 neg neg neg TC D 23 neg neg neg

TCD

24 neg neg neg

TCD

neg neg neg

TCD

26 neg neg neg

TCD

Contd/...

WO 97/1 0508 PCT/1E96/00061 8 Table 1 contdl...

Patient No.

IgA IEmAJ IgA

HUVEC

IgA Diagnosis 27 w pos pos w pos

TCD

28 w pos pos neg

TCD

29 w pos w pos neg

TCD

w pos w pos neg

TCD

31 pos w pos w pos

TCD

32 pos pos pos

TCD

33 w pos pos neg

UTCD

34 w pos pos w p0s

UTCD

w pos w pos w pos

UTCD

36 pos pos pos

UTCD

37 POS po5 w p0s

UTCD

38 o0 w pos neg

UTCD

39 po5 po5 po5

UTCD

po5 po5 neg

UTCD

41 pos pos w pos

UTCD

42 po5 po5 w p

UTCD

43 po5 po5 pos

UTCD

44 po5 po5 pos

UTCD

pos po5 pos

UTCD

46 po5 pos w pos

UTCD

47 w pos w pos w pos

UTCD

48 p0s p0s w pos

UTCD

49 po5 po5 w pos

UTCD

pos pos

UTCD

Con WO 97/10508 PCT/1E96/00061 9 Table 1 contd/...

Patient No. IgA IgA IgA Diagnosis EmA HUVEC ReTIC 51 pos pos pos

UTCD

52 pos pos pos

UTCD

53 pos pos pos

UTCD

54 pos pos pos

UTCD

pos pos pos

UTCD

56 pos pos pos

UTCD

57 pos pos w pos

UTCD

Key: w pos weakly positive CD Coeliac disease TCD Treated coeliac disease UTCD Untreated coeliac disease In Table 1, the comparisons between the EmA, reticulin and HUVEC IgA antibody responses in patients and controls are shown.

Fifty seven sera were tested, 16 were from normal controls, 16 were from treated coeliac patients (TCD) and 25 were from untreated coeliac disease (UTCD) patients the results show that there is a strong correlation between the presence of all three antibodies, although the association of EmA and HUVEC antibodies is the strongest. Thirty one sera were positive for IgA EmA antibodies and of these the same 31 were also positive for HUVEC IgA antibodies. Of the 31 sera positive for EmA and HUVEC IgA antibodies, 24 of these were also positive for IgA reticulin antibodies and 7 sera were negative, 26 sera were negative for EmA, HUVEC and reticulin IgA antibodies.

A summary of the results is given in Table 2 which shows the correlation between IgA EmA, HUVEC and reticulin antibody responses from fifty seven sera from coeliac patients and control subjects.

I I WO 97/10508 PCT/IE96/00061 Table 2 No. of EmA HUVEC Reticulin sera tested Pos Neg Pos Neg Pos Neg 57 31 26 31 26 24 33 Of the 31 sera positive for EmA antibodies all of these were positive for HUVEC IgA antibodies and 24 of these were positive for reticulin antibodies.

Example 2 Measurement of HUVEC surface and intracellular antibody responses using flow cvtometry To determine whether the HUVEC antigen reacting with the coeliac disease sera was of surface (extracellular) or intracellular origin, studies were carried out on non-permeabilised cells (surface) and permeabilised cells (intracellular) using flow cytometry. Ten EmA positive and 10 EmA negative sera were used.

Surface staining: Fresh HUVECs were harvested from culture as described in Example 1 and adjusted to 2X10 6 /ml. 50 il aliquots of cells were washed once in PBS and reacted with 50tl aliquots of coeliac patients' sera diluted 1/5 in PBS. The cells were incubated on ice for 30 mins. and washed IX with PBS. 20gl aliquots of FITC conjugated rabbit anti human IgA diluted 1/20 were then added to the tubes and allowed to react for 30 mins. The cells were washed once in PBS and read on a FacScan flow cytometer (FacScan is a trade mark) (Becton Dickinson). U937 and HL60 cells which are nonfoetal cell lines, were used as control cells and treated in a similar manner.

WO 97/10508 PCT/IE96/00061 11 HUVEC intracellular antigens: HUVEC cells were fixed in 2% paraformaldehyde for 30 mins. The cells were adjusted to 2X10 6 /ml in 0.05% Saponin/PBS (SPBS) and given two further washes in SPBS.

Patient and control sera were diluted 1/5 in SPBS and added in 50 il volumes to tubes containing aliquots of 2X10 6 cells. The cells were incubated on ice for 30 mins. and washed IX with SPBS. aliquots of FITC conjugated rabbit anti human IgA diluted 1/20 were then added to the tubes and allowed to react for 30 mins. The cells were washed once in SPBS and read on a FacScan flow cytometer.

The results are shown in Table 3.

Table 3 Surface and intracellular staining of HUVEC antigens using IpA antibodies from EmA positive CD sera and EmA negative normal controls Surface Surface/Intracellular EmA neg* EmA pos* EmA neg* EmA pos* 5 29 106 6 6 22 101 6 8 23 78 6 6 31 7 6 24 192 6 5 19 128 7 46 6 7 31 6 6 22 186 6 6 Each numerical value represents a different patient or control It will be observed that no difference was seen between the EmA positive and EmA negative sera when non-permealised HUVECs were used. The mean fluorescent intensity (MFI) values ranged from 5 to 7 with the EmA negative group and from 5 to 8 WO 97/10508 PCT/IE96/00061 12 with the EmA positive group. When the cells were fixed with 2% paraformaldehyde and permealised with 0.1% saponin there was a marked difference between the two groups. The MFI values for the EmA negative group ranged from 19 to 42, with a mean MFI value of 28.2, whereas the MFI values for the EmA positive group ranged from 40 to 192 with a mean MFI value of 109.5. The instrument settings for the two experiments were the same. However, fixing and permealisation causes the cells to give out a degree of autofluorescence which explains the difference between the MFI values for the two sets of experiments.

Example 3 Measurement of gliadin antibody responses by flow cvtometrv A) Absorption of gliadin onto latex particles: One ml of 1X10 7 /ml 8pm polystyrene beads (Polysciences, Warrington) were washed in distilled H 2 0 and incubated overnight at 4 0 C with gliadin (Sigma) at a concentration of 1 mg/ml in 0.05 M bicarbonate buffer pH 9.6. The beads were then washed 1X in PBS to remove the unbound gliadin and reincubated at 37 0 C for 3 hours in PBS containing bovine serum albumin (BSA) to block unoccupied sites.

Any unbound BSA was removed by washing with PBS containing 0.05% Tween (PBS/Tween).

Measurement of gliadin antibodies: 10gl volumes of beads (1X10 7 /ml) were added to 75 mm tubes containing 50 pl of patients' or control sera diluted 1/5 in PBS/Tween and incubated with gentle mixing for 30 mins. at room temperature. The beads were washed 2X in PBS/Tween to remove unbound sera by centrifuging at 2000 RPM for 5 mins. 20)tl aliquots of a 1/20 dilution of FITC conjugated rabbit anti IgA was then added to each tube and incubated with gentle mixing for 30 mins. at room temperature. The beads were again washed 1X with 2 ml vols of PBS/Tween and resuspended in 1 ml of PBS/Tween for analysis by flow cytometry.

WO 97/10508 PCT/IE96/00061 13 B) Absorption studies Serum samples from coeliac disease patients which tested positive for EmA, reticulin, HUVEC and gliadin antibodies were absorbed with 2X10 7 HUVEC cells previously fixed with 2% paraformaldehyde and permeabilised with SPBS. The absorption studies were carried out by initially incubating 2X107 HUVEC permeabilised cells for 1 hour at room temperature in 10 ml of rabbit serum in SPBS to block non-specific binding. The cells were washed IX in SPBS containing 10% rabbit serum and resuspended at a concentration of 2X10 7 /ml. One ml of a 1/5 dilution of SPBS of the coeliac disease serum was then mixed with 1 ml of 2X10 7 cells and incubated with gentle mixing at 4 0 C overnight. The following day the cells were spun in a refrigerated centrifuge at 2000 RPM for 10 mins and the absorbed serum removed. For comparison, another sample of the patients' serum was treated identically with the exception that no HUVEC cells were added. IgG and IgA measurements were made of the absorbed and non-absorbed sera and the concentration of IgA antibody adjusted so that the levels of antibodies were the same in both specimens. Doubling dilutions from 1/5 to 1/640 of the absorbed and non-absorbed serum were made in SPBS. EmA, reticulin, HUVEC and gliadin antibody levels were measured in the absorbed and non-absorbed sera and differences noted.

C) Absorption studies on HUVEC and polystyrene particles using flow cytometry Fifty gl aliquots of absorbed and non-absorbed sera diluted in SPBS were added to 50gl aliquots of paraformaldehyde fixed HUVEC cells at a concentration of 2X10 6 /ml. The cells were incubated at room temperature for 30 mins., washed once in SPBS and spun at 2000 RPM at 4 0 C for 5 mins. Fifty gl of FITC anti IgA diluted 1/20 in SPBS were then added to each tube and allowed to react at room temperature for a further 30 mins. The cells were washed 1X in SPBS, resuspended in SPBS and MFI measured on a FacScan.

WO 97/10508 PCT/IE96/00061 14 D) Studies carried out usin polystrene beads coated with gliadin: Briefly, 50il aliquots of absorbed and non-absorbed sera diluted 1/5 in PBS/Tween were added to 501l aliquots of 1X10 7 /ml polystyrene beads previously coated with gliadin. The beads were incubated at room temperature for 30 mins., washed once in PBS/Tween and spun at 2000 RPM at 4 0 C for 5 mins. Fifty pl of FITC anti IgA diluted 1/20 in PBS/Tween were then added to the each tube and allowed to react at room temperature for a further 30 mins.

The beads were washed IX in PBS/Tween, resuspended in PBS/Tween and the MFI measured on a FacScan.

E) Absorption studies with gliadin Sera positive for IgA antibodies to EmA, reticulin,

HUVEC

and gliadin were absorbed with gliadin coated particles as follows: One ml of sera was diluted 1/5 in PBS/Tween containing 10% BSA and allowed to mix gently with 1 ml of gliadin coated polystyrene particles (1X10 7 /ml) overnight at 4 0 C. Similar volumes of sera from the same patients were treated identically with the exception that they were absorbed with polystyrene coated particles coated with 1% BSA.

Both absorbed and non-absorbed sera were then tested for IgA antibodies to EmA, reticulin, HUVEC and gliadin and differences noted.

The results are shown in Tables 4 and WO 97/10508 PCT/IE96/00061 Table 4 Measurement of IgA antibody responses to gliadin. HUVEC and EmA before and after absortion with HUVEC and liadin

SUBSTRATE

Absorbed with: Exp No Gliadin MFI* EmA Titre HUVEC

MFI*

Not absorbed 112 640 441 Gliadin 1 38 640 443 HUVEC 90 20 Not absorbed 90 640 473 Gliadin 2 35 640 480 HUVEC 70 80 110 *MFI= Median Fluorescence Intensity WO 97/10508 PCT/IE96/00061 16 Table Measurement of IgA antibody responses to gliadin, HUVEC, EmA and reticulin before and after absorption with HUVEC Substrate Exp Gliadin* HUVEC* EMA Titre Reticulin Titre No Pre- Post- Pre- Post- Pre- Post- Pre- Post- Absorb Absorb Absorb Absorb Absorb Absorb Absorb Absorb 1 52 30(42) 583 61(90) 1280 80(94) 320 20(94) 2 59 36(39) 425 101(76) 640 80(88) 160 10(94) 3 21 12(43) 70 12(83) 1280 80(94) 160 40(75) 4 60 40(33) 99 15(85) 640 40(88) 160 20(88) 35 19(46) 218 22(90) 160 10(94) ND

ND

Mean Fluorescent Intensity Changes ND Not Done It will be observed from Table 4 that the antigens in the HUVEC did not cross react with gliadin. The results show that while absorption with gliadin markedly reduced the IgA response to the wheat protein from MFI values of 112 and 90 pre-absorption to MFI values of 38 and 35 post-absorption, respectively, the IgA responses for HUVEC were unaffected with MFI values of 441 and 473 preabsorption and MFI values of 443 and 480 post-absorption.

Similarly, while absorption with HUVEC reduced the HUVEC IgA MFI values from 441, 473 to MFI values of 75 and 110, respectively, they only marginally reduced the IgA gliadin responses from preabsorption MFI values of 112, 90 to post-absorption MFI values of and 70, respectively.

WO 97/10508 PCT/IE96/00061 17 In Table 5, the results of absorption studies using different patients sera are shown. These show that in five separate experiments absorption with HUVEC substantially reduced both the EmA and reticulin titres whereas the IgA gliadin antibody responses were only slightly reduced.

The titres for EmA pre-absorption were 1280, 640, 1280, 640 and 160 whereas, after absorption with HUVEC the titres dropped to 80, 80, 40 and 10, respectively. The reticulin titres pre-absorption were 320, 160, 160 and 160 whereas, the post-absorption titres were 20, 10, 40 and 20, respectively. The gliadin IgA antibodies preabsorption MFI values were 52, 59, 21, 60 and 35 whereas, the postabsorption MFI values were 30, 36, 12, 40, 19, respectively. These data show that the level of reduction in the EmA and reticulin titres were similar in each absorption experiment and varied from 4 to 16 fold, whereas the maximum reduction in the gliadin antibody responses were less than two fold.

Example 4 An experiment was carried out to determine the suitability of various cell types for use in the method according to the invention for the detection of EmA. The results are shown in Table 6.

WO 97/10508 PCT/IE96/00061 18 Table 6 CELL TYPE

RESPONSE

HUVEC Human umbilical artery smooth muscle cells Human umbilical vein smooth muscle cells IMR 90 U937 (ECACC No. negative 85011440) HI 60 (ECACC No. negative 85011431) Hut 78 (ECACC No. negative 88041901) LoVo (ECACC No. 87060101) It will be noted from the above Examples that identical positive and negative correlations were obtained when serum antibodies were measured using monkey oesophagus or HUVEC. Sera that were positive for EmA and HUVEC antibodies did not always show reticulin positivity which may reflect the lack of sensitivity of the tissue used rather than that EmA and reticulin antibodies are different.

The HUVEC antibodies were found by flow cytometry not to react with the surface of the cells but only when the cells were permeabilised with saponin which suggests that the target antigens are intracellular.

Absorption studies using permealised HUVEC gave similar findings to those described by Marttinen, A. and Miki, M. (supra) using foetal lung fibroblasts in that, whereas they showed that IgA antibodies to proteins secreted by these cells could bind to both WO 97/10508 PCT/IE96/00061 19 endomysium and reticulin, we showed that absorption with HUVEC substantially removed reticulin and EmA and only partially reduced gliadin antibody responses. In contrast gliadin coated particles did not remove HUVEC EmA antibody reactivity. These findings suggest that HUVECs contain both reticulin and EmA antigens or that these antigens are one and the same.

The role of EmA or reticulin antigens/antibodies in the pathogenesis of coeliac disease is unknown, as is the mechanism, whereby these antibodies are specifically produced. Since the antibodies are present in active coeliac disease and disappear on gluten withdrawal it may suggest that they are simply the result of the inflammatory process with may expose neo-embryonic antigens to the immune system.

Claims (11)

1. A method for diagnosing coeliac disease and related conditions in a subject known or suspected of having such a condition, which comprises contacting a body fluid from said subject containing antibody particular to said condition with a suspension of cells from an immortal cell line having an epitopic site specific for said antibody and determining the extent of binding of antibody to said cells.

2. A method according to Claim 1, wherein the cells are embryonic in nature.

3. A method according to Claim 1, wherein the cells are human umbilical cells.

4. A method according to Claim 3, wherein the cells are human umbilical vein endothelial cells (HUVECs). A method according to Claim 4, wherein the HUVECs are cells of ECV 304 cell line, No. 92091712.

6. A method according to Claim 3, wherein the cells are human umbilical artery smooth muscle cells or human umbilical vein smooth muscle cells.

7. A method according to any preceding claim, wherein the antibodies are of the IgA type.

8. A method according to any preceding claim, wherein the cells are ruptured prior to or during contact with said body fluid.

9. A method according to any preceding claim, wherein the body fluid is serum. A method according to any preceding claim, wherein bound antibody is estimated by flow cytometry.

11. A method according to any one of Claims 1-9, wherein bound antibody is estimated by an enzyme immunoassay procedure.

12. A kit when used for carrying out a method according to any one of Claims 1-11 containing a suspension of cells from an immortal cell line.

13. A method for diagnosing coeliac disease and related conditions in a subject known or suspected of having such a condition, 10 substantially as hereinbefore described and exemplified.