WO1990005304A1 - Structural and enzymatic assays on glycocompounds, their use in the diagnosis of cancer, kits for use in the assays and a device for taking a sample of secretion via a body opening - Google Patents

Abstract

Altered saccharide structures on glycoproteins, glycolipids, poly- and oligosaccharides are detected and quantified in a body fluid by contacting the body fluid with a first ligand binding to all saccharide structures and with a second ligand which either binds only to a specifically altered terminal portion of the saccharide structures or only to the normal terminal portion, one of said ligands being immobilized on a solid phase, and the other being labelled. A glycosyltransferase enzyme producing such altered saccharide structures is detected and quantified in a body fluid or a cell or tissue homogenate by similar use of an immobilized sugar chain which is an acceptor substrate for the enzyme and a nucleotide-sugar which is a donor substrate for the enzyme. A device for taking a sample of secretion via a body opening and for storing and processing the sample comprises a container (14) and a rod (2) which is provided at one end portion thereof with an absorbing material (8) and at the other end portion slidably extends through a cover means (12) of the container (14). Various kits for use in the above assays are also described.

Description

Structural and enzymatic assays on glycocompounds, their use in the diagnosis of cancer, kits for use in the assays and a device for taking a sample of secretion via a body opening

This invention relates to a method of detecting and quantifying altered saccharide structures on glyco¬ proteins, glycolipids, poly- and oligosaccharides (in the following termed glycocompounds) or a glycosyltransferase responsible for such altered saccharide structures. This method is useful for the diagnosis of disease which leads to altered glycosyiation of glycocompounds in the body, and in particular for the diagnosis of cancer. The in- vention also relates to kits for use in carrying out the method, and to a device for taking a sample of secretion via a body opening and for storing and processing the sample.

An important function of cells and tissues is the syn¬ thesis of oligosaccharide chains. These chains are a) used by the cells, as the chains are incorpotated into glyco- proteins and glycolipids and thereby become part of cell membranes, b) exported out of the cells as a secretion product like intestinal secretion, secretion of the cervix uteri, glycoprotein hormones, serum glycoproteins etc. , c) stored in the cells as oligosaccharide chains.

These oligosaccharide chains are important for a large number of physiological processes as they regulate the turnover of glycoprotein hormones, the binding of bacteria to cells, the specificity of receptors, and often are in¬ volved in the tertiary structure of glycoproteins. In ad¬ dition, they are often strongly immunogenic, exemplified by the blood group ABH related antigens, that are all oligosaccharides. When the behaviour of cells is changed due to disease, this is often associated with a change in the synthesis of oligosaccharide chains, especially in the most terminal portions. This has been demonstrated in diseases like alcoholic liver cirrhosis, pancreatitis, colitis and especially in cancer of various organs. This change in the synthesis of oligosaccharide chains is caused by a change in the relative activity of the gene-encoded glycosyl- transferases that form the oligosaccharide chains. In some diseases, like bladder carcinomas, certain cells loose activity of a transferase, and, hence, the oligosaccharide chain-elongation is stopped, and a precursor structure is accumulated. As this precursor structure is an antigen it¬ self that can be recognized by antibodies and lectins, a disease associated oligosaccharide is produced. In other diseases or in other cells, the disease process may lead to increased activity of glycosyltransferases and the formation of longer or more branched structures than hitherto produced by the cells. This happens in the distal colon where ABH blood group antigens are synthesized in rectal carcinomas but not in normal mucosa.

By far the best known changes in the oligosaccharide chains, associated with disease, occur in the terminal part of the oligosaccharide chain. This is the part of the chain which protrudes from a glycoprotein or glycolipid, whereas the opposite end is bound to the protein or lipid. The terminal part may therefore be regarded as the variable part of the oligosaccharide chain, whereas the non-terminal part of the chain may be regarded as the constant part, which is rarely changed during disease. In addition, the terminal part is very often glycosylated according to the genetic make-up of the individual e.g. blood group antigens. Single cells are capable of producing a wide range of glycoproteins. Some of these may be structural compounds, others receptors or hormones. The difference between these molecules is mostly due to their protein portion whereas the glyco-portion (the oligosaccharide chains) of these molecules can be very similar e.g. lactoseries chains with repeating Gal01-3GlcNAc sequences, or repeating GlcNAc or mannose sugars. When a disease associated loss or gain of glycosyltransferase activity is leading to a change in terminal oligosaccharides structures, it may, thus, affect the glycosyiation of several different molecules in the same way.

The carbohydrate portion of glycoproteins and glycolipids, are changed in various disease states like cancer and in¬ flammatory diseases (ref. 1). These changes can be de¬ tected by the use of carbohydrate specific ligands like lectins or mono/polyclonal antibodies raised against these carbohydrate structures. In case of cancer of the colon and rectum, it has been known for some years that the carbohydrate part of mucins (mucins are high molecular weight glycoproteins consisting of more than 40% carbo¬ hydrate, the rest being a peptide backbone, from which the carbohydrate oligosaccharide chains jut out) secreted by the cancer cells and the surrounding transitional mucosa is changed. Immunofluorescence and immunohistochemistry has shown that both normal and malignant mucosa contains mucin goblets that bind WGA and sialosyl-Le, whereas only malignant mucosa of the rectum contains mucin in glandular lumens or in surrounding transitional mucosa, that bind PNA and UEA1 lectins as well as antibodies against Lewis b, Le , Le , T, Tn, and ABH antigens. On the other hand, it is known that the DBA lectin only binds to normal mucosa and not to malignant mucosa. Colorectal carcinomas are very common in the western world, where they are among the three major cancers, accounting for about 20% of all cancer deaths in the United States of America. As no curative therapy, apart from surgery, is known (the death rate has not changed for 40 years) it is essential for a benign disease course that colorectal carcinomas are diagnosed early, before they in¬ vade and metastasize (2). As these tumors are relatively symptomless, various screening procedures have been used in the past to detect colorectal carcinomas. The most used is screening for occult blood in faeces. However, this screening procedure is ambiguous, and influenced by many false positive tests, mostly due to blood-containing meat in the diet (3). In recent years serological tumor markers have been investigated, based on the principle that tumors shed antigens into the circulation. The first and most well examined example of such a marker is the Carcino- embryonal-antigen CEA. Unfortunately CEA levels in serum are too often false negative for diagnostic use (4).

The carbohydrate determinants on glycoproteins and especially those on mucins have gained interest in recent years as colorectal tumor markers. They have been in¬ vestigated by histochemical methods as mentioned above, and especially as serological tumor markers. However, like CEA, they have the drawback that when the epitope they detect is highly tumor associated (very specific), the number of false negative colorectal carcinoma bearing patients is too high (low sensitivity). In addition the principle has the general drawback that is seems that tumors have to be invasive before the antigens occur in the circulation. It would be much better to detect carcinomas at an earlier stage in their development.

With the purpose of detecting carcinomas at an earlier stage, Shamsuddin et al. (5) developed a non-quantitative galactose oxidase test on rectal mucus, removed from the patient by a gloved finger. This test detected all cancer patients but also reacted with 8.5% of non-carcinoma patients giving false positive results. Patients with in- flammatory bowel diseases were not investigated.

In an attempt to develop an assay which

1. can be made quantitatively,

2. is more specific and sensitive than previous approaches,

3. can be performed at the bedside or in the laboratory,

4. is based on material that may be collected by the patient himself,

5. besides testing the tumor marker also tests the quality of the sample, thus minimizing false negative results,

I invented a new method of detecting and quantifying altered saccharide structures on glycoproteins, glyco¬ lipids, poly- and oligosaccharides in a body fluid, which is characterized by contacting the body fluid with a first ligand binding to the non-terminal (constant) portion of the saccharide structures and with a second ligand binding to a specifically altered terminal (variable) portion of the saccharide structures, but not to the normal terminal portion, or binding to the normal terminal portion of the saccharide structures, but not to the specifically altered terminal portion, one of said ligands being immobilized on a solid phase, and the other being labelled, and there¬ after detecting labelled immobilized glycocompounds and/or determining the amount of labelled ligand bound to im¬ mobilized glycocompounds or the amount of labelled ligand not bound to immobilized glycocompounds.

Another aspect of the invention is a method of detecting and quantifying a glycosyltransferase enzyme producing altered saccharide structures on glycoproteins, glyco¬ lipids, poly- and oligosaccharides in a body fluid or a cell or tissue homogenate, which is characterized by contacting the body fluid or homogenate with an im¬ mobilized sugar chain which is an acceptor substrate for the enzyme and with a nucleotide-sugar which is a donor substrate for the enzyme, adding a labelled ligand binding to the saccharide structure of the coupling product, and thereafter detecting labelled immobilized coupling product and/or determining the amount of labelled ligand bound to immobilized coupling product or the amount of labelled ligand not bound to immobilized coupling product.

These methods are suited for the diagnosis of disease which leads to altered glycosyiation of glycocompounds in the body and especially for the diagnosis of cancer.

The body fluid to be assayed is conveniently selected from the group consisting of rectal, vaginal, uterine cervical and pulmonary secretions. Most conveniently the body fluid is rectal secretion. In these circumstances the pre¬ dominant glycocompounds in the secretion are mucins.

For taking samples of secretion via a body opening it is expedient to use a special device according to the inven- tion, which is characterized in that it comprises a con¬ tainer and a rod which is provided at one end portion thereof with an absorbing material fixed to the rod within the end portion in question, and which slidably extends through a cover means at the other end portion, said cover means being adapted to be fixed in an opening of the con¬ tainer. Thus, a sample of secretion absorbed in the ab- sorbing material is squeezed out into a liquid in the con¬ tainer by displacement of the rod with respect to the cover means so that the absorbing material is compressed against the inner side of the cover means of the con¬ tainer, which may be filled with a buffer liquid for this purpose, e.g. pH neutral isotonic phosphate-buffered brine to which a bacteriostatic (e.g. antibiotics) or bacteri¬ cidal (e.g. antibiotics or sodium azide) agent is optio¬ nally added.

Particularly expediently the absorbing material opposite the end portion may be fixed to the rod in a rupturable manner so that when the rod is displaced outwardly with respect to the cover means, the absorbing material is com- pressed against the inner side of the cover means for squeezing the sample of secretion out of the absorbing ma¬ terial.

Advantageously the absorbing material may consist of a suitable foamed plastic which is preferably surrounded by a net-shaped sleeve, which is adapted to protect the ab¬ sorbing material against lateral compression, but allows said axial compression of the absorbing material with a view to squeezing the sample of secretion out into the collection liquid.

The device of the invention will be explained more fully below with reference to the drawing, in which

fig. 1 is a perspective view of an embodiment of a sample- taking device according to the invention,

fig. 2 shows a container for the sample-taking device according to the invention, while fig. 3 shows how the sample material is squeezed out of the absorbing material in the sample-taking device according to the invention.

The sample-taking device shown in fig. 1 comprises a rod 2 of a suitable foamed plastic and with a handle 4. Inside an outer rounded end portion 6 of the rod 2, it is sur¬ rounded by an absorbing material 8, e.g. foamed rubber or foamed plastic, which is in turn surrounded by a helical net-shaped sleeve 10 of plastics connected with the end portion 6. The net sleeve 10 serves to prevent lateral compression of the absorbing material 8, so that the sample material is prematurely squeezed out of the mate¬ rial 8, which is temporarily fixed to the rod 2 opposite the end portion 6. However, the net sleeve 10 may be com¬ pressed axially together with the foamed rubber material 8, which thus acts like a sponge.

Somewhat spaced from the sponge 8 the rod 2 is provided with a cover means 12 which fits tightly around the rod 2, which, however, may be displaced with respect to the cover means 12. The sample-taking device of the invention more¬ over comprises a container 14 which is provided at an opening 16 with internal threads 18 for fixing the cover means 12 in a liquid-tight manner to the container 14 by means of external threads 20 on the cover means 12. Also a separate closing plug 22 is associated with the container 14, said plug being used for liquid-tight sealing of the container 14 - prior to and optionally also after sample- taking - the container 14 preferably containing a suitable liquid, such as pH neutral isotonic phosphate buffered brine, destilled water or the like in which the sample ma- •terial may be steeped.

The sample-taking device of the invention is used in the following manner, there being referred to the taking of a rectal mucus sample in this example:

The rod 2 with the foam rubber sponge 8 is preferably supplied as a sterile unit in a sealed plastics bag - and preferably some cream or vaseline is applied to the end portion 6 to facilitate insertion of the rod 2 through the sphincter of the anus. The container 14 contains a suit¬ able collection liquid, optionally admixed with a bacteri¬ cidal agent, the container 14 being closed and sealed by means of the closing plug 22.

The end portion 6 of the rod 2 with the foam rubber sponge 8 is inserted into the rectum, and the rod 2 is pulled out very carefully after some time. The closing plug 22 is removed from the container 14, and the rod 2 with the foam rubber sponge 8, which now contains a mucous sample, is placed in the container 14, and the cover means 12 is screwed tightly into the container opening 16 by means of the threads 18, 20, following which the container 14 is shaken and turned thoroughly to wash out the mucus sample from the sponge 8.

To squeeze as much of the mucus sample as possible into the collection liquid in the container 14, the rod 2 is now pulled upwardly with respect to the cover means 12 until the foam rubber sponge is compressed completely against the inner side of the cover means 12. This proce¬ dure is optionally repeated some times with simultaneous turning of the container 14. Then the cover means 12 is again screwed off the container 14, and the rod 2 with the compressed sponge 8 is removed from the container 14, which is again closed by means of the separate closing plug 22.

The closed container 14 containing the mucus sample sus¬ pended in the liquid can now be passed on for more de- tailed laboratory examination. If the sample cannot be analysed immediately, it may be stored in the container 14 - optionally frozen.

If sample- aking is performed at a hospital or a clinic with associated laboratory, the mucus sample can advan¬ tageously be prepared additionally right away, in that the container 14 contains a suitable detergent solution, e.g. containing "Triton® X-100", it being possible to use the actual container 14 during the subsequent determination of the mucus sample 14 for activity of transferases and/or quantitative analysis of carbohydrate structures.

The example below refers to cancer in the colon, but also other organs may be examined by the described method, e.g. the cervix uteri, the urinary system and their glands as well as neck and lungs; of course, the actual sample- taking device is to be modified depending upon the con¬ crete organ to be examined.

Most of the malignant tumors of the intestine and their prestages can normally be located to the sigmoideum and to the rectum. Another approximately 20% is present far away from the coecum. It is known, however, that these tumors in the coecum are very frequently accompanied by changes of a less severe nature in the sigmoideum, and it is be¬ lieved that the entire intestine is somehow subjected to impacts which change the cells to cancer cells.

The intestinal secretion is collected from the sigmoideum of the patient either using a gloved finger or a similar oblong instrument or by ano/rectoscopy (telescope exami¬ nation of the rectum with a tubular instrument having a length of 5-50 cm). In ano/rectoscopy the sample material is collected e.g. directly on a piece of wadding or another suitable aid which is inserted into the intestine, which should preferably have been cleaned of faeces some hours before by means of enema. In case of finger-collec¬ tion, the finger is simply wiped in a piece of wadding or the like. The mucus sample may then be stored frozen on the wadding, or the mucus sample may be washed from the wadding by means of a buffer, e.g. pH neutral isotonic phosphate buffered brine, and then the mucous sample is spun down by centrifugation (e.g. 800 x g). The super¬ natant is then removed, and the precipitated and washed mucus sample may be stored by freezing to between -20°C and -80°C if the sample is not to be analysed immediately. Frozen samples are to be thawed before analysis.

Prior to analysis of transferase activity the mucous sample is prepared by adding to the washed mucus sample a detergent solution, e.g. containing "Triton® X-100" and subjecting it to mechanical or physical impact, e.g. ro¬ tating knifes and/or ultrasound - followed by centrifu¬ gation of the supernatant which contains dissolved en- zymes. Optionally, washing of the mucus sample may be omitted, and the detergent solution may be added as the first step of preparation.

After the transferases have been brought into solution, the amount of protein in solution is quantified, and the activity of the transferases is determined by quantifying the amount of the product which the transferases can pro¬ duce either in that the transferases incorporate labelled sugars in an acceptor, or in that the product which they produce is specifically identified with antibodies or lec¬ tins. Several known methods of performing this quantifi¬ cation are known.

Various principles may be used for immobilizing the mucus and then applying the labelled ligand (antibody or lec- tin) to it. The mucus may be bonded directly to a solid surface and then be analysed; or the mucus may be bonded via a ligand which is attached to a solid surface. This ligand will typically be an antibody or a lectine which binds to mucus or mucus portions, but not to structures which are irrelevant to the analysis, thereby providing a certain degree of purification.

In addition to the plain unpurified mucus it is possible to use mucus which has been dissolved by various proce- dures by means of detergents or cleaving enzymes, e.g. mercaptoethanol and trypsin. This example uses mucus which has been brought into solution by homogenization in a PBS buffer containing about 1% "Triton®X-100" as well as mercaptoethanol (cleavage of disulfide bonds) and protease inhibitors (phenylmethylsulfonylfluoride/PMSF). After this homogenization the mucus solution is ultracentrifuged 30,000 x g), and the supernatant is repeatedly precipi¬ tated with 75% ethanol. The precipitate includes gluco- proteins, and these are resuspended in H~0, dialysed over- night and freeze-dried. Then the protein content of the sample is measured by the BioRads method, and the sugar content is measured by periodic acid Schiffs reaction. The sample is then applied to either a nitrocellulose or a nylon membrane with a known amount of protein or PAS posi- tive sugar. The membrane is then blocked e.g. with albumin and incubated with lectin or antibody in a suitable dilu¬ tion. The optional binding of lectin or antibody is then evaluated using a radioactive isotope, an enzyme, a metal or the like which gives a signal, e.g. a colour, that can be identified. In the present example the enzyme peroxi- dase is bonded directly to the lectin, while the antibody binding is detected by binding an anti-globulin label with the same enzyme directly to the antibody. The membrane is then washed, and a colour reaction is produced by 4- chloro-1-naphthol which is converted to a blue substance by the enzyme. This colouring matter is particularly suitable because the colour reaction proceeds to an end point within a reasonable period of time (less than 30 minutes) without background colour. This provides a colour product which linearly reflects the peroxidase enzyme concentration. The blue substance is then quantified in a chromatoscanner, which is adjusted for absorption in a maximum wave range for the blue colour (evaluated by spectral analysis) . This provides a quantitative ex¬ pression, since the more ligand binding there is to the mucin, the more enzyme will be bound, and the more colour is obtained and the more light will be absorbed in the scanner. As previously mentioned, a special standard is used for having a basis for comparison from analysis to analysis.

A method as described has shown the presence of blood group antigen H (Fuc l-2 Gal 01-3/4 Glc NAc-R) in rectal mucus from all humans with intestinal cancer, but not in mucus samples from healthy humans.

Examples of antibodies and lectins used include:

Antibodies: Anti-T, Tn, Le , Le^x, Le^y, A (all subtypes), B, H (all subtypes) as well as antibodies against derivatives of these substituted with other sugars, e.g. sialic acid.

Lectins: PNA (Peanut Agglutinin), Jacalin, UEA (Ulex Europaeicus Agglutinin), WA (Viscia Villosa Agglutinin), DBA (Dolichos Biflorus Aggluti¬ nin) . Determination of g-2-L-fucosyl transferase activity in rectal mucus sample

After the transferases have been brought into solution - as described above - the supernatant, which contains the transferases, is incubated in a suitable buffer containing an acceptor suitable for determination of precisely this transferase as well as a radioactively labelled sugar - here fucose. After incubation the product is separated from the other substances by chromatography, e.g. on thin layer plates or on paper. The product is now quantified in a scintillation counter and is compared with the starting amounts, it being established how much of the added sugar has been incorporated in the acceptor by means of the transferase, and thus how great an activity the transfe¬ rase has.

Tests have demonstrated an activity of α-2-L-fucosyl transferases in mucus from the rectum in cancer patients which is more than 10 times higher than the activity in normal healthy patients. Also a permanently increased ac¬ tivity of this transferase has been demonstrated in pa¬ tients who have had prestages of cancer in the rectum, such as polypi.

To compensate for shifts in activity level caused by vari¬ ation in the above-mentioned procedures it may be an ad- vantate to determine fractions between various transfe¬ rases, e.g. between o-2-L-fucosyl transferases and o-4-L- fucosyl transferases or between α-2-L- and α-3-L-fucosyl transferases. Such determinations show a difference be¬ tween healthy humans and cancerous humans of 2-10 times. To obtain optimum results the examined humans may also optionally be divided according to their erythrocyte phaenotype. And to avoid possible repercussions on the results caused by genetic variation in the population, transferases which are known to be relatively stably ex¬ pressed in the population may advantageously be analysed, e.g. 01-3N-acetylglucosaminyl transferase which is known to increase its activity in case of cancer in the colon.

The following is a detailed discussion of the assay ac¬ cording to the invention providing a variety of embodi¬ ments and exemplifications.

Materials and methods

Reagents

The following reagents were used: 96 micro-well titertrays (also called immunoplates), Maxisorp, Nunc, Roskilde, Den¬ mark; WGA (Wheat Germ Agglutinin), PNA (Peanut Agglu¬ tinin), DBA (Dolichos Biflorus Agglutinin), conA (Concanavalin A), MPA (Maσlura Pomifera Agglutinin), UEA 1 (Ulex Europaeus 1 Agglutinin), and BSA 1 (Bandeiraea Simplicifolia 1 Agglutinin) all lectins with and without conjugation with horseradish peroxidase (HRP), bovine submaxillary mucin, asialofetuin, asialomuσin, asialo- glycophorin, D(+)-galactose, D(+)-glucose, o-L-fucose, N- acetyl-D-glucosamine, polystyrene beads, diameter 11.9 μm, styrene divinylbenzene beads, diameter 25.7 urn, ATP, 4- chloro-1-naphtol, and MnCl« were from Sigma Chem. Co., St. Louis, Mo., USA. Bovine serum albumin and "Tween®20" were from Merck, Darmstadt, FRG. Cellulosenitrate, 0.45 am, was from Schleicher and Schuell, Dassel, FRG. Monoclonal anti- bodies against Lewis a, Lewis b, A, and B antigens were from Biotest Serum-Institut, Frankfurt, FRG. Monoclonal antibody against H antigen, rabbit anti-mouse IgG con¬ jugated with HRP, and 1,2-phenylenediamine-dihydrochloride (OPD) tablets 2 mg, were from Dakopatts, Glostrup, Den- mark. Gelatine G was from Grindsted Products, Brabrand, Denmark. Galactose 01-0-PAP-HSA, Lacto-N-tetraose-O- phenyl-HSA, and GDP-fucose were from Biocarb AB, Lund, Sweden.

Materials

Secretion from the distal part of the large bowel (rectum and sigmoid) was collected by two different procedures. In the first procedure a piece of cloth was inserted into the gut lumen through the reσtoscope during rectoscopic exa- mination and allowed to suck-up secretion. After this, the piece of cloth was taken out and frozen in an appropriate container until further analysis. In the second procedure, a special device according to this invention, consisting of a rod covered with sucking foam, was used to collect the secretion. In this case no rectoscopic examination was necessary, however digital exploration of the rectum, be¬ fore insertion of the device into the gut, was used to assure that no obstruction of the rectum by tumors, or other abnormalities, was present. The rod, which may be lubricated on the tip by a lubricating gel, was inserted into the rectum by gently pressing the tip of the rod against the anus. When the anus starts to open, the rod covered by the sucking foam was gently moved in direction of the umbilicus (as when inserting a thermometer) until the foam covered part of the rod had passed the anal sphincter muscles. The foam covered part of the rod was then moved in a circular fashion to allow secretion of mucus from the distal part of the gut to be absorbed or adhered to the foam. After a time period (appr. 20 seconds) the rod was pulled out and placed in its con¬ tainer, the container was closed and the rod kept frozen until further analysis was taking place. In the present study there was a time-span from when the secretion or mucin sample was taken and to the time when it was frozen, of appr. 20 minutes to three hours. The individuals examined all had received a laxating medium (KLYX) intrarectal before the sample was taken. The effect of this laxating medium was however individual, as some patients did not contain faeces in the rectum whereas others were rather obstipated with faecal masses.

Preparation of samples

The stored frozen samples were thawed, and 1 ml 0.15 M NaCl containing 1% "Triton®X-100'^! was added. The effect of sonication was tested on some of the samples - but did not seem to be of necessity to solubilize the samples. All samples were centrifuged at 800 x g for 30 min at room temperature, whereafter the supernatant was used for the assay. Protein content of the supernatant was measured in a BioRad assay. A typical protein content of the super¬ natant was in the range of from 1000 ug/ml to 20 000 ,

Preparation of mouse small intestine standard

Twenty mice were sacrificed and the small bowel was extirpated, cut open, and the mucosa was gently scraped off with an object glass. The pooled mucosas were homo- genized with an ultraturrax homogenizer in the presence of ammonium acetate containing 1% "Triton®X-100". The homo¬ genate was centrifuged at 30 000 x g and the supernatent dialyzed for 48 hours against distilled water.

The dialysate was then freeze dried and stored frozen in small aliquots.

Coupling of ligands to Latex particles

500 μl of Latex beads were centrifuged for a few seconds at 800 x g, the supernatant was discarded, and the pellet consisting of the Latex particles was suspended in a 1 mg/ml solution of Ligand (PNA, DBA, anti-Le etc. ) in carbonate buffer, pH 9.6. After incubation for 2 hours at 25 °C, with occasional shaking, the solution was centri- fuged at 800 x g and the supernatant discarded. The pelleted Latex particles coated with ligand were then washed three times in PBS, and finally diluted appro¬ priately (1:5 - 1:20) in PBS.

Enzyme linked lectin solid-phase assay (ELLSA)

The following assay conditions were kept constant in all assays: Coating buffer: 0.1 mol/1 sodiumhydrogencarbonate, pH 9.8. Washing solution, and solution for dilution of samples, ligands and blocking reagents: PBS (0.5 M NaCl, 0.0075 M Na₂HP0₄, 2 H₂0, 0.0025 M NaH₂P0₄, H₂0), pH 7.2 containing 0.1% "Tween®20". Enzyme substrate: 8 mg OPD in 12 ml 0.1 M citric acid/phosphate buffer pH 5.0. 5 μl 30% H₂0₂ were added 5 minutes before use. Scanning: All wells were read in a Tim 200 immunoreader (Teknunc, Roskilde, Denmark), at 492 nm.

In the ELLSA procedure the sequence of steps and their variation were as follows:

1. Coating of wells with 100 μl "catching" lectin over¬ night at 4 °C. Various lectins (PNA,WGA,MLA,DBA) were tested in concentrations ranging from 1 μg/ml to 64 μg/ml.

2. Washing three times in washing buffer.

3. Blocking with blocking buffer. 200 μl of the following solutions were tested: 2% BSA in washing buffer, or 2% Gelatine G in washing buffer, for 10 min at 20 °C 4. Washing once in washing buffer.

5. Incubation with 100 μl sample, standard or control solution in concentrations ranging from 1 μg/ml to 500

6. Washing once in washing buffer.

7. Incubation with 100 μl labelled ligand for one hour at 20 °C. The following ligands were tested WGA-HRP, PNA-HRP,

DBA-HRP, conA-HRP, and UEA 1-HRP, in concentration ranges between 0.1 μg/ml and 1 μg/ml, and monoclonal antibodies against Lea, Leb, A, B, and H antigens diluted 1:100 to

1:5000. The ligands were diluted in washing buffer containing Mn ions, and the pH of the buffer was varied between 6.0 and 10.0 to assure optimal specific binding.

The incubation time was 60 min. In addition 100 μl PNA and

WGA coated latex beads were used as a label which could be read semiquantitatively by the naked eye, or by use of a magnifying glass.

8. In case of lectins as labelled ligand this step in¬ cluded washing in washing buffer. In case of monoclonal antibodies as ligands, this step included incubation with Rabbit anti-Mouse conjugated with HRP, diluted in washing buffer 1:500 to 1:4000 for 60 min, followed by washing once in washing buffer.

9. 100 μl of enzyme substrate for 15 min at room temp.

10. The color development was stopped by addition of 150 μl 1M H₂S0₄ to each well.

11. Scanning of the wells at 492 nm. 12. Calculating the results from the standard curve, using semilogarithmic paper.

Membrane assay

The following assay conditions were kept constant: The nitrocellulose membrane was mounted in a 96 well dot blot applicator (BioRad) and 250 μl PBS were added to each well.

Sample buffer: PBS. Washing buffer: 0.1% "Tween®20" in PBS. Blocking buffer: 5% BSA in washing buffer. Color de¬ veloping reagent: 4-chloro-l-naphtol. Buffer for diluting lectins and antibodies: 0.1% BSA in PBS.

In the membrane assay the following variables were tested:

1. The samples were tested in a concentration range from

2. The washing buffer composition was tested with BSA 0.5% to 5% (w/v), the blocking buffer was tested with BSA from 0.5 to 5% (w/v), and with "Tween®20" from 0.01% to 2%.

3. The HRP-labelled lectins tested were: PNA, WGA, WA, Jacalin, UEA 1. Monoclonal antibodies tested were anti- Le , -Le , -A,-B, -Le , and anti-H, as well as anti-CEA (Roche) and Ca 19-9 (Centrocor).

Microwell glycosyltransferase assay

The following example shows an assay for αl-2-L-fucosyl- transferase activity in rectal secretions. The nucleotide sugar used was GDP-fucose, and the detecting ligand was UEA 1 lectin which detects αl-2 linked fucoses. Other assays tested - but not shown - include tests for σl-4-L- fucosyltransferase activity with anti-Lewis antibodies as detecting ligands, and tests for ol-3-L-fucosyltransferase activity using anti-Le antibodies as detectors.

1. The wells were coated with 100 μl Gal01-O-PAP-HSA or Lacto-N-tetraose-phenyl-HSA 3 μg/ml in coating buffer pH 9.8 overnight at 4 °C.

2. Washing in washing buffer.

3. Blocking with 250 μl 5% BSA in Tris/HCl buffer.

4. Washing in Tris/HCL buffer.

5. The following reaction mixture was then added to each well: 7 μl GDP-fucose, (tested at concentrations from 0.16 nmol/μl to 5 nmol/μl), 5 μl ATP 0.1 M, 10 μl MnCl₂ 0.2 M, 50 μl Tris/HCl. The enzyme reaction was initiated by adding 20 μl of the rectal secretion sample to the reac- tion mixture, and incubating the titer-tray for 60 - 120 min at 37 °C.

6. The reaction was terminated by washing the wells in PBS.

7. The wells were tested for reaction product by the ad¬ dition of 100 μl UEA 1-HRP 0.5 - 1.5 μg/ml to each well for 60 min at room temp.

8. Washing once in washing buffer.

9. 100 μl enzyme substrate was added to each well for 15 min at room temp.

10. The reaction was terminated by the addition of 150 μl 1M H₂S0₄. 11. Scanning of the wells at 492 nm.

Results

Enzyme linked lectin solid-phase assay (ELLSA)

Screening of the panel of ligands on nitrocellulose mem¬ branes (see Dot-blot membrane assay) with rectal secretion from normal individuals and individuals with colorectal carcinomas, revealed that WGA lectin was unique in the way that it bound to all samples and standards. It, thus, seemed to be a promising candidate as a semi-discrimi¬ native catching ligand in a solid phase sandwich lectin- lectin or lectin-antibody assay. The effect of the coating concentration of WGA lectin is shown in Figure 4, with PNA-HRP as label.

With WGA lectin as catcher (coated at 5 μg/ml), the rest of the ligand panel was screened as labels using rectal secretion from normal and cancer patients, normal serum, and purified glycoproteins as samples (Table 1). WGA bound to all samples tested, whereas PNA reacted with most of the glycoproteins and secretion from cancer patients, but not with secretion from normal individuals, and not with serum. DBA bound to asialomucin and secretions from normal individuals, but not to secretion from cancer patients and asialofetuin. The lectins BSA and UEA 1 were unreactive. The monoclonal anti-Le^* antibody bound to secretion from cancer patients but not to asialomucin.

Based on this it seemed that PNA and DBA lectins as well as the Le^y antibody were promising candidates as labels for the study of secretions from rectum, immobilized by the WGA lectin, and that a WGA-WGA assay could be used as a measure of the sample content of relevant structures.

Standard curves were made with the purified glycoproteins and with Murine intestinal mucin, and the lectin- and antibody combinations were calibrated against these curves. Typical curves are demonstrated in Figure 5 with asialomucin as sample, and with a rectal secretion sample from a cancer patient. As expected, these curves show that the purified glycoproteins contain more binding sites per μg protein than the crude solutions of rectal secretion. It was decided from these curves to use a sample protein concentration of 250 μg/ml to make single point determi- nations on rectal secretions.

With the purpose of minimizing unspecific binding (appr. 0.4 extinction units) of WGA-HRP to WGA, the pH of the buffer containing WGA-HRP was varied from pH 6-10 as was the NaCl concentration in the buffer from 0.9% (w/v) to

1.6% (w/v), without any effect. The blocking with BSA was changed to blocking with Gelatine G which reduced the un¬ specific binding with approximately 25% (data not shown), but did not abolish it. However, for practical purposes, the relatively low unspecific binding was simply sub¬ tracted from the sample extinctions by blanking on selected wells.

The carbohydrate specificity of the PNA binding was tested by preincubating the PNA-HRP label with D-galactose, D- glucose, L-fucose, and N-acetylglucosamine (Fig. 6). As expected, the PNA binding was inhibited by D-galactose even at very low concentrations of the monosaccharide. D- glucose showed some inhibition whereas the other sugars, whose conformations deviate more from galactose, were unreactive (Fig. 6) .

In order to test the coefficient of variation across the titer tray, all wells were incubated with asialomucin at a concentration of 2 μg/ml. The WGA - PNA-HRP assay showed a mean extinction of 0.854 (Standard deviation (SD) = 0.043, variance = 0.0002), and a coefficient of variation (CV) of 5%. The WGA - WGA-HRP assay showed a mean extinction of 1.497 (SD= 0.05, variance = 0.003) and a CV of 3.3%.

Clinical results

Secretion was sampled from the distal part of the large bowel from 35 individuals (Table 1). Their distribution on diagnosis, and their results in the WGA - PNA-HRP assay is shown in the table. A level of 0.5 μg/ml (calibrated against asialomucin) was chosen as the level when deciding whether a sample was abnormal or not in this assay. This means that the reference values for normal mucin is < 0.5 μg/ml. By the use of this reference value, all 13 normal individuals were negative, and 8 out of 8 tumor patients were positive. One of these patients had a tumor in the sigmoid (PNA-HRP value 9.6 μg/ml), one had a tumor in the coecum (PNA-HRP value 1.5 μg/ml), the other six had tumors in the rectum. In the group of patients with colitis ulcerosa 2 out of 5 were positive. These two individuals were suffering from a highly active disease with multiple daily diarrhoeas and pancolitis. One of them had the en¬ tire colon resected a few weeks after the secretion sample was taken from the rectum. The three negative colitis patients were suffering from relatively mild disease.

These data indicate that the WGA - PNA-HRP assay is highly specific for serious colorectal diseases, and that it de¬ tects carcinomas in the colon with a very high sensitivi- ty. As the patients found positive with colitis ulcerosa were having very severe symptoms they would have been diagnosed without any rectal secretion test. However, as the WGA - PNA-HRP test was only positive in the most severe cases (that in addition has an increased colorectal cancer risk), this assay might be useful for identifying the most severe cases that need their colon removed, and it might give predicitive information.

Compared to the colitis patients, the patients who reacted positive in the WGA - PNA-HRP assay, and who were later found to have colorectal carcinomas by rectoscopi, colo- scopi or X-ray examinations, were relatively symptomless. A third group who reacted positive in this test was patients with carcinomas in other organs (Cancer of the uterine cervix, Primary peritoneal mesothelioma. Cancer of the thyroid, Mai. Lymphoma).

Based on this it seems reasonable to state that this assay has a very high specificity and sensitivity for detecting cancer in the colorectal area - and cancer in other or- gans. If one excludes individuals with severe colitis symptoms, who would be diagnosed anyway, the specificity and sensitivity in this small material are both 100%.

In most patients the samples were tested in a WGA - WGA- HRP assay. This assay was independent of the diagnosis of the patients as it was always positive, and only cor¬ related with the protein concentration of the sample. In 14 rectal secretion samples applied to wells at 250 μg/ml, the mean extinction was 2.574 (SD = 0.1896) and CV = 7%. The WGA - WGA-HRP assay was very useful in two ways when run simultaneously with the WGA - PNA-HRP assay. First, dilutional errors were detected immediately by an ab- berating value from the mean value, second, the WGA - WGA- HRP assay could be used as a reference value for the con- tent of WGA binding structures in the secretion sample, thus making protein determination unnecessary. This has the advantage compared to protein determination, that only structures able to be caught by the WGA catching ligand, and, thus, only the structures able to react with the PNA- HRP label are measured, irrelevant proteins are neglected by this assay. However, it does not sem that the con- tamination with irrelevant proteins is important, as the WGA - WGA-HRP assay correlated very well with the protein content. The WGA - WGA-HRP values could be used to make a ratio between these values and those obtained from the WGA - PNA-HRP assay, thus, making protein determinations un¬ necessary.

The WGA - DBA-HRP assay showed a linear calibration curve with the asialomucin standard (Figure 5). Several normal and malignant samples were tested with this lectin com¬ bination (Table 3). It was evident that high binding was observed with normal rectal secretion whereas rectal secretion from colorectal cancer patients showed a low binding. One patient having severe colitis ulcerosa showed an intermediate value, and this patient was in addition

PNA positive. The DBA ligand, thus, seems to be complemen¬ tary to the PNA ligand, and could be used alone as a marker of carcinoma in the large bowel and other organs, or as a marker of severe colitis. In this context it seems that a reference value for normal individuals would be a DBA value >0.8 μg/ml (calibrated against asialomucin) - meaning that lower values indicate the presence of car¬ cinoma or severe inflammatory bowel disease. As carcinomas are often occult, the DBA measurement could be used for screening patients or selected populations to get diag¬ nostic information. The disease colitis ulcerosa is usual¬ ly diagnosed based on its symptoms (frequent diarrhoeas, abdominal pain), however the DBA measurement could be of importance as a prognostic marker (colitis patients have an increased risk of colorectal carcinomas), or as a marker of disease progression, as discussed above con¬ cerning the WGA - PNA-HRP assay.

As evidenced by Table 3, the ratio between PNA and DBA binding to a sample could be calculated, and maybe lead to an even higher sensitivity and specificity of these assays for clinical use.

A last possibility for the DBA assay was to measure the ratio between DBA binding and WGA binding to samples caught by the immobilized WGA. As discussed above for the PNA assay, such a procedure could make protein determi¬ nations unnecessary, and be a check for dilutional errors.

A standard was prepared from mouse small intestine. This standard contained crude intestinal material compared to the purified glycoproteins used as standard reagents for calibration. However, it is possible to use this standard instead of the glycoproteins, and to express the lectin binding in arbitrary units.

In the present examples a titer-tray microwell was used as the solid phase for binding of the catching ligand. How¬ ever, any solid material able to bind the ligand could be used, and either the catcher could be bound to the con- tainer (like a tube or well) containing the sample, or the solid support carrying the catcher could be placed in a container housing the sample solution.

Dot blot membrane assay

The results from this assay are shown in table 4. The results are in some ways comparable to those obtained in the titer-tray assay. However, this assay suffers from the drawback compared to the titer-tray assay, that the re- suits cannot be quantitated, and the presence of irrele¬ vant proteins in large amounts in the solution e.g. in the case of intraluminal bleeding (a very common finding in colorectal cancer) can block the membrane, and thus lead to false negative results. One could therefore expect this assay to have a lower specificity and sensitivity com¬ pared to the assay principle where a catcher, in our case the WGA lectin is used to immobilize the relevant struc¬ tures and therefore to partly purify the samples before labelling these with PNA and DBA. However, the figures show a seemingly higher sensitivity, and the problems of a lower specificity that often follows a higher sensitivity. Some normal individuals are positive with PNA as ligand, which was not the case in the titer-tray assay. Whether these normal individuals are completely normal or carry a disease - maybe even a cancer disease - which has not reached a level where it gives symptoms can only be elu¬ cidated by prospective clinical studies. As in the titer- tray assay WGA binds to all samples as does anti-Le after neuraminidase treatment. The latter finding is important as it shows that an antibody against neuraminic acid (or sialic acid) substituted Le structures might be used as a catcher instead of WGA in the titer-tray-assay. The anti¬ bodies against ABH blood group antigens are not very specific in the detection of carcinomas compared to anti- bodies against Leb antigens, and Lev^** antigens. Both anti- bodies react very little with normal secretion, but to some extent with malignant secretion. They could be sup¬ plementary to PNA and DBA as tumor markers. They also both work well in the titer-tray assay together with the WGA catcher (Table 1), and could be supplementary to the PNA- HRP ligand in detecting structures that start to occur in colorectal carcinoma secretion or secretion from indivi¬ duals with colitis or carcinomas of other organs.

Modification of the ELLSA with latex particles

In order to modify the ELLSA so that it could be useful as a bedside test, office test or even at home, the HRP- labelled lectins and antibodies were substituted with identical ligands coupled to large Latex beads. These beads can be seen by the naked eye, if necessary assisted by a magnifying glass. The assay, thus, consisted of a WGA - Lectin or Antibody-Latex sandwich, with the sample structure caught between these ligands. The procedure was exactly like the one used with the HRP-conjugated ligands except that the reaction sequence ended with a wash after incubation with the ligand coated latex particles. When the wells were examined, one could see binding of latex particles to positive samples and even follow the standard curve as an increasing number of latex particles in the microwells. This modification could be used as a semi- quantitative assay for bedside use, with the same ligands, PNA, DBA, WGA, Anti-ABH, anti-Leb, anti-Lev, as used with the enzyme-labelled assay. In addition, other solid sup¬ ports having a size that makes it possible to see them by the naked eye could be used, as well as isotopes, other enzymes etc.

Modification of the assay by use of the sample container as the reaction well

To simplify the procedures, the sample container used to place the sample collection rod inside, could be used as one large reaction "well". Exactly the same reactions that take place in the titer-tray wells can be performed in the container - if this is made of a material, or coated with a material, capable of being coated with the ligands used - preferably WGA. However, other combinations could be used e.g. coating with PNA and labelling with WGA, coating with DBA and labelling with WGA, coating with antibody, and labelling with WGA, or the same procedures but with an antibody against sialosyl-Le as a substitute for WGA, as p_\ anti-sialosyl-Le , like WGA, always binds to rectal secretions - no matter the diagnosis. Altogether the com¬ binations are almost endless. In addition, the container could be coated with two different lectins at the same time, thus permitting simultaneous measurement of binding of tumor associated ligand, and a control ligand (WGA or antibody against sialosyl-Le ) for eliminating false nega¬ tive results.

Clinical results with the glycosyltransferase assay

It now seems well consolidated that the αl-2-L-fucosyl- transferase is increased in rectal secretions from indi¬ viduals with colorectal carcinomas. The newly developed titer-tray assay is unique compared to other previous assays in the way that it is much less time-consuming, much cheaper to perform, does not require any special equipment, and is free from radioactive materials.

In the described micro-well assay we examined rectal secretion from seven PNA positive individuals, six with colorectal carcinomas, and one with primary peritoneal mesothelioma. The secretion in all cases showed αl-2-L- fucosyltransferase activity as it led to binding of UEA 1 lectin that was 2 to 18 times the background binding (which was almost zero). Rectal secretion from normal, PNA negative individuals showed no enzyme activity.

The micro-well enzyme assay could be performed in other containers than micro-well. Especially the container used for housing the sample rod, would be a relevant site for performing the test. The container could be coated with the sugar acceptor, Galactose-R, and the entire reaction mixture could be present in the container so that just the secretion sample having the enzyme activity, if from car- cinoma patients, would be needed to start the reaction.

The next step would then be washing out the mixture after an incubation time, and labelling the product, if any, with UEA 1 coupled to enzyme (color reaction), isotope (counting) or solid particle (e.g. Latex particles, seen by the naked eye or with a magnifying glass). Assay kits

Various kits can be tailored for different uses of the assay according to the invention.

One kit may be designed for use at home, which implies that it should be simple and must not require extra equipment.

Another kit could be intended for general practitioners, small clinics and out-patient's clinics. Here greater demands may be made as to complexity of the procedures and possibly minor equipment.

A third kit could be intended for laboratories with all the possibilities implied.

A minimum kit for structural assays is characterized by comprising a container which on at least part of its in- terior surface is coated with first or second ligand, a buffer solution containing a detergent, such as "Triton®X- 100", for solubilizing the body fluid, a washing buffer, and a labelled second or first ligand.

Preferably the kit comprises a sample-taking device ac¬ cording to the invention. Then the container in which the absorbing material is placed after the sample-taking may be used for carrying out the assay, the container being coated at the bottom, so that the foamed plastic does not scrape off the catcher ligand.

In a special embodiment the bottom of the container could be coated in a particular symbolic pattern. Thus, for didactic reasons, the interior bottom surface of the con- tainer may be coated with a stripe of WGA and a stripe of PNA the two stripes being essentially at right angles to each other, and the labelled ligand provided with the kit be WGA. Then, if the sample material contains altered saccharide structures, it will bind to both stripes, and when the bound sample material is visualized by labelled WGA a + sign will mark that the assay is positive. If no altered saccharide structures bind to the PNA, a - sign will mark that the assay is negative. If the sample has been taken erroneously and for example only contains water, no symbol will appear at all.

A clinical kit would contain either the device for taking samples of secretion or a number of loose coated tubes in which to carry out the assay, so that it is easier to carry out many at the time. The kit could then be extended by e.g. a magnifying glass for reading the latex reactions through the bottom of the tubes or a sort of mini-scanner for measuring enzyme-developed color.

A titer tray micro-well kit intended for laboratories may be used for precise quantitation of the binding of ligands. It should comprise one or more micro-well plates and a ligand coating liquid or pre-coated micro-well plates; a concentrated solution of a standard glyco- compound or a series of solutions diluted to different known concentrations or a standard series pre-bound to catcher ligand in the tray; and labelled ligands. It may also comprise a high and a low control and, if appro¬ priate, enzyme substrate for a possible color reaction. Usually the kit will also comprise a set of directions for use.

Kits for enzymatic assays are, in principle, designed like the above kits for structural assay, only supplying the reaction mixture as an extra component. A minimum kit for enzymatic assays is characterized by comprising a container which on at least part of its in¬ terior surface is coated with an acceptor substrate, a buffer solution containing a detergent for solubilizing the body fluid, and enzyme reagent buffer containing a donor substrate, a washing buffer and a labelled ligand.

The kit may comprise a sample-taking device according to the invention, the assay being conducted in the container of this device; or it may comprise titer tray micro-wells as explained above.

TABLE 1 ELLSA with WGA as catching lectin, and various other lectins and antibodies as labels

Rectal secretion Purified glycoproteins

Normal Cancer Bovine Asialo- Asialo- Asialo- submaxill. mucin fetuin glycophonin Murine

Label mucin Serum intestine

WGA +

PNA

DBA +

BSA

UEA 1 anti-Le-^*

anti-ABH

TABLE 2

ELLSA of rectal secretion (250 μg protein/well) from 35 individuals. WGA was used as catcher and PNA-HRP as label. A standard curve for calibration of binding was made by the use of asialomucin, and the PNA-HRP binding was calibrated against this curve.

PNA-HRP

Number of

Diagnosis < 0.5 μg/ml > 0.5 μg/ml Mean Patients

Normal 13 0.5 13

Colorectal carcinomas 4.2 8

Colitis ulcerosa 0.28

Small adenomas 0.0

Diverti- culosis 0 0.1

Cancer at other sites 6.0 TABLE 3

WGA - DBA-HRP assay of rectal secretion (250 μg protein/well) from 9 individuals, calibrated against a standard curve obtained with asialomucin.

Diagnosis DBA calibrated PNA Calibrated PNA/DBA

Normal (N=4) 1.8 (0.2-3.3) 0 0

Malignant(N=5) 0.2 (0.1-0.5) 3.2 (0.5-14) >11.4

Colitis(N=l) 0.2 0.6 3

TABLE 4 Membrane dot-blot assay of rectal secretion (200 μl sample containing 25 μg/ml applied in each well). The stained dots were read by the naked eye, strong positive staining was scored as +, weak positive staining as (+). In case of small amounts of material, this was primarily used for PNA determinations. Normal Colorect.cancer Benign- Colrect. Other CA

Ligand Pos. Neg. Pos. Neg- Pos. Neg. Pos. Neg.

PNA 3 + 14 0 0 4(+)

WGA 10 + 0 0 0 K+)

3( + )

Le +Nase 12 0 0 0

Le^J K+) 11 0

A,B,H 0

Benign colorectal diseases included colitis ulcerosa, whereas small adenomas were included in the group of normals.

Other CA: Carinomas in other organs

Le +Nase: Staining with antibody against Le following neuraminidase treatment.

References

1. Hakomori, S. Glycosphingolipids in cellular interaction, differentiation, and oncogenesis. Ann. Rev. Biochem. 5_0: 733, 1981.

2. Isselbacher, K.J., Adams, R.D., Braunwald et al. eds. Harrison's principles of internal medicine 9th ed. McGraw-Hill Int. Book Co., 1979.

3. Simon J.B. Occult blood screening for colorectal carcinoma: a critical review. Gastroenterology 8_8: 820, 1985.

4. Holyoke, E.D., Cooper, CH. CEA and tumour markers. Semin. Oncol. 3_: 377, 1976.

5. Shamsuddin, A.M., Elsayed, A.M. A test for detection of colorectal cancer. Human Pathology _1 : 7, 1988.

Claims

P a t e n t C l a i m s

1. A method of detecting and quantifying altered saccharide structures on glycoproteins, glycolipids, poly- and oligosaccharides (in the following termed glyco¬ compounds) in a body fluid, c h a r a c t e r i z e d by contacting the body fluid with a first ligand binding to the non-terminal (constant) portion of the saccharide structures and with a second ligand binding to a specifi- cally altered terminal (variable) portion of the sac¬ charide structures, but not to the normal terminal portion, or binding to the normal terminal portion of the saccharide structures, but not to the specifically altered terminal portion, one of said ligands being immobilized on a solid phase, and the other being labelled, and there¬ after detecting labelled immobilized glycocompounds and/or determining the amount of labelled ligand bound to im¬ mobilized glycocompounds or the amount of labelled ligand not bound to immobilized glycocompounds.

2. A method according to claim 1, c h a r a c t e r¬ i z e d in that the detection and quantitation of altered saccharide structures is used for the diagnosis of dis¬ ease, which leads to altered glycosyiation of glyco- compounds in the body.

3. A method according to claim 2, c h a r a c t e r- i z e d in that the disease is cancer.

4. A method according to any of claims 1-3, c h a r a c¬ t e r i z e d in that the body fluid is selected from the group consisting of rectal, vaginal, uterine cervical and pulmonary secretions.

5. A method according to claim 4, c h a r a c t e r¬ i z e d in that the glycocompound is a mucin.

6. A method according to any of claims 1-5, c h a r a c¬ t e r i z e d in that the first and second ligands are selected from the group consisting of lectins and anti¬ bodies.

7. A method according to claim 6, c h a r a c t e r¬ i z e d in that the first ligand is wheat germ agglutinin (WGA) or concanavalin A (conA).

8. A method according to claim 7, c h a r a c t e r¬ i z e d in that the first ligand is immobilized WGA.

9. A method according to any of claims 6-8, c h a r a c¬ t e r i z e d in that the second ligand is a lectin selected from the group consisting of Arachis hypogaea (peanut) agglutinin (PNA), Dolichos biflorus agglutinin (DBA), Maclura pomifera agglutinin (MPA), .Ulex europaeus 1 agglutinin (UEA 1), Bandeirea simplicifolia (Griffonia simplicifolia) 1 agglutinin (BSA 1), Viscia villosa agglutinin (WA), and Limulus polyphemus agglutinin (LPA).

10. A method according to claim 6, c h a r a c t e r¬ i z e d in that the second ligand is an antibody selected from the group consisting of those raised against blood group ABO related terminal saccharide structures such as anti-A,B,H,T,Tn,Lewis a, Lewis b, Le Y, Le X and sialylated, fucosylated and sulfatized derivatives thereof.

11. A method according to any of claims 1-10, c h a r a c t e r i z e d in that the quantitation of altered saccharide structures is made by calibration of the determined amount of labelled ligand bound or not bound to immobilized glycocompounds against a standard curve obtained by corresponding determinations on different known concentrations of asialomucin, asialofetuin or murine small intestine homogenate.

12. A method according to any of claims 1-11 for detecting and quantifying total glycocompounds in the body fluid, c h a r a c t e r i z e d by using the same ligand, either WGA or conA, as both the first and the second ligand.

13. A method of detecting and quantifying a glycosyl- transferase enzyme producing altered saccharide structures on glycoproteins, glycolipids, poly- and oligosaccharides (in the following termed glycocompounds) in a body fluid or a cell or tissue homogenate, c h a r a c t e r i z e d by contacting the body fluid or homogenate with an im- mobilized sugar chain which is an acceptor substrate for the enzyme and with a nucleotide-sugar which is a donor substrate for the enzyme, adding a labelled ligand binding to the saccharide structure of the coupling product, and thereafter detecting labelled immobilized coupling product and/or determining the amount of labelled ligand bound to immobilized coupling product or the amount of labelled ligand not bound to immobilized coupling product.

14. A method according to claim 13, c h a r a c t e r- i z e d in that the detection and quantitation of altered saccharide structures is used for the diagnosis of dis¬ ease, which leads to altered glycosyiation of glyco¬ compounds in the body.

15. A method according to claim 14, c h a r a c t e r¬ i z e d in that the disease is cancer.

16. A method according to any of claims 13-15, c h a r a c t e r i z e d in that the body fluid is selected from the group consisting of rectal, vaginal, uterine cervical and pulmonary secretions.

17. A method according to claim 16, c h a r a c t e r¬ i z e d in that the body fluid is rectal secretion.

18= A method according to any of claims 13-17, c h a r a c t e r i z e d in that the labelled ligand is selected from the group consisting of lectins and mono¬ clonal antibodies against the coupling product.

19. A method according to any of claims 13-18, c h a r a c t e r i z e d in that the enzyme is αl-2-, βl-3- or αl-4-L-fucosyltransferase, α-3-N-acetyl-D- galactosaminyltransferase, 01-3-N-acetylgluσosaminyl- transferase or a 0-galactosyltransferase.

20. A method according to claim 19, c h a r a c t e r¬ i z e d in that the enzyme is ol-2-L-fucosyl ransferase.

21. A method according to claim 20, c h a r a c t e r¬ i z e d in that the labelled ligand^" is UEA-1.

22. A device for taking a sample of secretion via a body opening and for storing and processing the sample, c h a r a c t e r i z e d in that it comprises a con¬ tainer (14) and a rod (2) which is provided at one end portion thereof with an absorbing material (8) fixed to the rod (2) within the end portion in question, and which slidably extends through a cover means (12) at the other end portion, said cover means being adapted to be fixed in an opening of the container (14).

23. A device according to claim 22, c h a r a c t e r¬ i z e d in that the absorbing material (8) opposite the end portion is fixed to the rod (2) in a rupturable manner so that when the rod is displaced outwardly with respect to the cover means (12) the absorbing material (8) is compressed against the inner side of the cover means (12) for squeezing the sample of secretion out of the absorbing material.

24. A device according to claim 22 or 23, c h a r a c- t e r i z e d in that the absorbing material (8) consists of a suitable foamed plastic and is preferably surrounded by a net-shaped sleeve (10) which is adapted to protect the absorbing material (8) against lateral compression, but allows said axial compression of the absorbing material (8) on the rod.

25. A kit for use in carrying out the method according to any of claims 1-12, c h a r a c t e r i z e d by com¬ prising a container which on at least part of its interior surface is coated with first or second ligand, a buffer solution containing a detergent for solubilizing the body fluid, a washing buffer, and a labelled second or first ligand.

26. A kit for use in carrying out the method according to any of claims 13-21, c h a r a c t e r i z e d by com¬ prising a container which on at least part of its interior surface is coated with an acceptor substrate, a buffer solution containing a detergent for solubilizing the body fluid, an enzyme reagent buffer containing a donor sub¬ strate, a washing buffer and a labelled ligand.

27. A kit according to claim 25 or 26, c h a r a c t e r¬ i z e d in that it comprises a device according to any of claims 22-24, the container being the container part of said device.

28. A kit according to any of claims 25-27, c h a r a c¬ t e r i z e d in that the detergent is "Triton® X-100".

29. A kit according to any of claims 25-28, c h a r a c¬ t e r i z e d in that the labelled ligand is a ligand coated onto latex particles.

30. A kit according to any of claims 25-29, c h a r a c¬ t e r i z e d in that it comprises a number of tubes coated with catcher ligand or acceptor substrate on their interior surfaces.

31. A kit according to any of claims 25 and 27-29, c h a r a c t e r i z e d in that it comprises a device according to any of claims 22-24, the container of which is coated on its interior bottom surface with a first ligand or with a second ligand or with both.

32. A kit according to claim 31, c h a r a c t e r¬ i z e d in that the interior bottom surface of the container is coated with a stripe of WGA and a stripe of PNA, the two stripes being essentially at right angles to each other, and that the labelled ligand is WGA.

33. A kit according to claim 25 for quantitation of ligand binding, c h a r a c t e r i z e d in that it comprises one or more micro-well plates, a ligand coating liquid, a concentrated solution of a standard glyco- compound or a series of solutions diluted to different known concentrations, and labelled ligands.

34. A kit according to claim 33, σ h a r a c t e r- i z e d in that the micro-well plates are pre-coated with ligand.

35. A kit according to claim 33 or 34, c h a r a c t e r¬ i z d in that the standard series is pre-bound to catcher ligand in the micro-well plates.