ZA200901517B

ZA200901517B - Method for detecting antibodies or antigens rapidly in a two-particle system that requires mixing

Info

Publication number: ZA200901517B
Application number: ZA2009/01517A
Authority: ZA
Inventors: Frankie Chi Hang Tam; Pak Leong Lim
Original assignee: Frankie Chi Hang Tam; Pak Leong Lim
Priority date: 2008-03-03
Filing date: 2009-03-03
Publication date: 2009-12-30
Also published as: HK1134213A2

Description

ros

LL

UL

Background to the invention | : 1. Field of the Invention

The present invention relates to a method of detecting antibodies or antigens in liquid samples, which method involving the rapid mixing of two types of reagent particles and which samples may comprise body fluids or secretions or excretions, cells or : : 0 tissue extracts, or culture media. In particular, but not exclusively, the invention : relates to a method for detecting antibodies and antigens in non-clear sera such as hemolytic, icteric or lipemic using a rapid immunodiagnostic assay test and to a method of improving the specificity of a rapid immunodiagnostic assay test for antibodies or antigens. 2. Background Information : :

Antibodies are proteins produced in the blood or other bodily fluids of a person or an animal in response to the presence of a foreign object, commonly called an antigen, in the body. The antibody-causing antigen can be any object, such as a bacterium or : 20 virus, and any size, such as a whole bacterium or small component parts of a bacterium such as a small chemical group. The resultant antibodies are a highly specific mirror image of various component parts of the particular antigen causing them and are produced by the body primarily to combat infection by binding to the infectious agent. However, antibodies also serve as useful diagnostic markers of an infection and are often more easily detected than antigens because they are amplification products of the immune system and are found abundantly in a person during an infection. Detection of both antibodies and antigens is widely used in healthcare and public health as it aids in the diagnosis of a variety of diseases including the many different types of infections and cancers, allergies and autoimmune diseases.

1 3 ) : Rapid immunodiagnostic tests employ immunological reagents such as antibodies or antigens or both of these together in an assay system to detect an infection in a person or an animal in a rapid manner. Typical rapid immunodiagnostic tests give a result in less than 15 minutes. Many different assay methods of antibody or antigen detection have been developed over the years. These test vary in assay format, the time required, sensitivity and specificity. Recently, the trend has moved towards rapid tests that require minimal time and effort and no instrumentation and so can be used as point-of-care aids in the field or by the bedside. The first rapid test invented many decades ago was the slide agglutination test. Initially, whole bacteria were used in this test as an antigen for detecting relevant antibodies in the serum of a patient suffering from a suspected disease. A drop of serum is simply mixed with a drop of a suspension of the bacteria on a glass slide by gently rocking the slide for about § minutes. A positive reaction is shown by clumping (agglutination) of the bacteria, : which is visible to the naked eye. A later modification of the slide agglutination test : used latex microspheres about the size of a bacterium coated with selected components of the bacterium as the indicator reagent. Instead of bacterial components, antibodies and other proteins or macromolecules, such as human chorionic gonadotropin (hCG), can be coated onto latex microspheres and used as the indicator reagent. Thus, latex particles coated with antibodies specific for hCG were widely used in *80s to detect hCG in urine to diagnose pregnancy. A problem with slide agglutination tests, however, is the general lack of sensitivity and specificity; the results are often difficult to read and the test is unfavourably influenced by various non-specific factors in the specimen.

Improvement in the reading system for agglutination tests were recently advanced by using special clear plastic devices instead of glass slides to perform the test. These devices contain channels through which the reactants travel that are narrowed in certain areas to promote the reaction. Such tests have been used for detecting antibodies to HIV in suspected AIDS individuals.

The next development in rapid diagnostics was the immuno-concentration or flow- through test. This test employs a small device that contains a porous membrane

Poa coated with the antigen (e.g. HIV) or antibody (e.g. to hCG) of interest, and an absorbent pad underneath. A liquid specimen (serum or urine) applied to the device flows through the reagent membrane, where reaction takes place, to the absorbent pad.

A wash solution is next applied, followed by a signal reagent to develop the assay.

The indicator reagent initially used was enzyme, a reagent used commonly in enzyme- linked immunosorbent assay (ELISA), but coloured latex particles or colloidal gold particles became more widely used later on. Commercial tests based on this system that take 5-15 minutes to perform have been developed for pregnancy and HIV testing.

At about the same time, “dipstick” rapid tests were developed that utilized strips of non-porous membrane coated with the antigen or antibody of interest. To perform the test, the strip (dipstick) is incubated with the patient’s serum or urine, washed, incubated with a solution of the developing reagent, and washed again. As - with flow-through tests, the assay is developed using enzyme labels as in ELISA, or using indicator coloured latex particles or colloidal gold particles. The set-back of all these tests is that multiple steps involving both incubation and washing are often required.

More recently, immunochromatographic or lateral-flow tests have become the mainstay of rapid diagnostics today. This test was originally developed for the detection of hCG to diagnosis pregnancy. The test uses a strip of nitrocellulose impregnated with the antibody (or antigen) of interest and a signal reagent. A liquid specimen is applied to the wick end of the strip, from which the fluid moves along the strip by capillary action to the absorbent (drainage) end, carrying with it the antigen or antibody from the specimen. The antigen or antibody together with the signal reagent becomes trapped by a specific ligand (antibody or antigen) coated on the nitrocellulose strip in the viewing-window area. The whole process, which comprises a single step, takes about 10 minutes. Lateral-flow tests are sensitive for antigen detection (e.g. hCG) but appeared less so for antibody. This is because in the various assay formats used, washing that is normally required for a similar format in traditional ELISAs, is sacrificed for the sake of convenience.

1 3 oo 4-

A rapid test recently introduced that also requires no washing step is the TUBEX test for typhoid fever, marketed by IDL Biotech, Sollentuna, Sweden. This test takes about 5 min to perform and has been found by several investigators in different regions of the world to be quite accurate (House et al., 2001; Olsen et al., 2004;

Kawano et al, 2007; Rahman et al., 2007). The test uses a unique assay system to detect serum antibodies in typhoid patients to the O9 antigen (or epitope as more specifically called) found in the lipopolysaccharide (LPS) of the causative organism,

Salmonella enteritidis serotype Typhi (S. Typhi). These antibodies are detected by their ability to block the binding between a reagent anti-O9 monoclonal antibody oo 10. (mAb) that is coupled to blue-coloured indicator particles, and the reagent O9 antigen prepared in the form of S. Typhi LPS-coupled magnetic particles. When aqueous suspensions of both reagents are mixed together in the absence of inhibiting antibodies, the indicator antibody-particles bind to the magnetic antigen-particles and both indicator and magnetic particles sediment to the bottom of the reaction well 1 15. when the well is placed on a magnet stand 2 (see Fig.1). The resultant appearance of the reaction mixture is actually colourless due to the absence of any suspended particles, but for easy visualization, the background has been artificially coloured red. : If, on the other hand, inhibiting antibodies are present in the liquid sample, these will prevent binding between the two types of particles to an extent dependent on the concentration of these antibodies. A blue colour results from the unsedimented blue indicator particles, ranging from light reddish blue (low antibody concentration) to dark blue at the extreme of inhibition. The results which are thus semi-quantitative, are read against a colour chart provided, with scores ranging from 0 (red, most negative) to 10 (dark blue, most positive). The TUBEX test system can potentially be applied to a variety of diseases and not just typhoid fever, but the cardinal feature of this system is: a reagent indicator particle must be rapidly but thoroughly mixed in : solution with the test specimen and a reagent magnetic particle in specially-contoured wells that aid in the mixing. The colour reading system of TUBEX allows easy visualization of the results, which is ideal for point-of-care use. However, there are problems with this system when it comes to specimens that are heavily hemolytic, lipemic or icteric. This is a serious drawback since hemolytic sera, for example, can account for a significant percentage of a routine laboratory’s specimens that would

1 3 : have to be discarded otherwise. In addition, even for clear serum samples, it is sometimes difficult to interpret results that are borderline positive. A solution to this problem that we herein describe is to add a wash-step to the test after incubating the : liquid sample with the magnetic particles, and re-suspending the washed magnetic particles in a suitable volume for the next step. In this way, any colour or interfering substance present is removed from the liquid sample, and a bigger volume of liquid sample can be used which is useful for ascertaining the status of borderline positive cases. Herein we also describe a simple method of washing involving the use of either a pipet or an absorbent device.

Another problem with the TUBEX test is this. It has been claimed by the manufacturer and reported in all published articles that it detects anti-O9 antibodies ‘ only and consequently, only detects infections caused by S. Typhi and other Serogroup

D Salmonella organisms but not those caused by other types of Salmonella organisms.

A new unexpected finding that we herein reveal, however, shows that this test can in : fact detect infections due to S.Paratyphi A. The reason for this is the presence of a : common epitope - called the O-12 antigen - that is found in both S.Typhi and : S.Paratyphi A organisms, and antibodies made to this antigen in a patient can interfere in the TUBEX test by steric hindrance due to the close physical proximity between the O-9 and 0-12 antigens. In this disclosure, we also describe a solution to this cross-reactivity problem by adding a solution of S.Paratyphi A LPS antigen to the reagent particles in the TUBEX test to block the anti-O12 antibodies, thus making the test more specific for typhoid fever.

Summary of the Invention

According to the first aspect of the invention there is disclosed herein a method for detecting antibodies or antigens in a liquid sample, the method comprising: adding a liquid sample to reagent particles coated with a first reagent, incubating the liquid : sample and reagent particles, separating the reagent particles from the liquid sample, thoroughly mixing the separated reagent particles in an indicator solution comprising

Tf B] } indicator particles coated in a second reagent, separating the reagent particles from the indicator solution, and detecting any change in appearance of the indicator solution.

Preferably, the liquid sample is selected from a group comprising whole blood, or serum or a body fluid that lack a clear appearance.

Preferably, the liquid sample is selected from a group comprising hemolytic, icteric or lipemic sera.

Preferably, the first reagent is an antigen and the second reagent is an antibody. :

Preferably, the first reagent is an antibody and the second reagent is an antigen.

Preferably, the method is performed in a plurality of physically linked together micro- 15. tubes.

Preferably, the liquid sample and reagent particles are incubated for between 1 and 10 minutes.

Preferably, the reagent particles are magnetic and separating the reagent particles from the liquid sample or indicator solution comprises introducing the indicator solution to a magnetic field.

Preferably, separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles by aspiration with a pipet or an absorbent.

Preferably, separating the reagent particles from the liquid sample further comprises washing the reagent particles with a washing solution and removing the washing solution from the reagent particles by aspiration with a pipet or an absorbent.

1 3 .

Preferably, separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles in a plurality of physically : linked together micro-tubes by aspiration with a corresponding plurality of physically linked together absorbents. :

Preferably, thoroughly mixing the reagent particles and the indicator solution comprises mixing for between 1 and 10 minutes.

Preferably, the indicator solution is introduced to the magnetic field for less than 5 minutes before detecting any change in appearance of the indicator solution.

Preferably, any change in appearance of the indicator solution is detected by oo spectrometry, colorimetry, chemiluminescence or fluorimetry. 15s Preferably, prior to adding the liquid sample to the reagent particles a blocking agent is added to the reagent particles.

Preferably, the first reagent is an antigen having a specific epitope and an adjoining non-specific epitope, and the blocking agent is an antigen having the non-specific epitope but not the specific epitope.

Preferably, separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles in a plurality of physically

¥ ) N linked together micro-tubes by aspiration with a corresponding plurality of physically : linked together absorbents. :

Preferably, the method is performed in a plurality of physically linked together micro- tubes.

Preferably, the first reagent is an antibody having non-specific epitopes, and the blocking agent is an antigen having the non-specific epitopes.

Preferably, the step of separating the reagent particles from the liquid sample is : omitted.

According to a second aspect of the invention there is provided a method for detecting antibodies or antigens in a liquid sample, the method comprising: adding a blocking agents to reagent particles coated with a first reagent, adding a liquid sample to the ; reagent particles, incubating the liquid sample and reagent particles to bind any antibodies or antigens in the sample to the first reagent, thoroughly mixing the separated reagent particles in an indicator solution comprising second particles coated in a second reagent, separating the reagent particles from the indicator solution, and : detecting any change in appearance of the indicator solution.

Preferably, the first reagent is an antigen having a specific epitope and an adjoining non-specific epitope, and the blocking agent is an antigen having the non-specific epitope but not the specific epitope. :

4 A

Preferably, separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles in a plurality of physically * linked together micro-tubes by aspiration with a corresponding plurality of physically linked together absorbents.

Preferably, the method is performed in a plurality of physically linked together micro- : tubes.

Preferably, the indicator solution comprises indicator particles coated in antibody. oo

Preferably, the first reagent is an antibody having specific epitope and an adjoining non-specific epitope, and the blocking agents is an antibody having the non-specific epitope but not the specific epitope.

Preferably, the indicator solution comprises indicator particles coated in antigen.

Preferably, the liquid sample and reagent particles are incubated for between 1 and 10 : minutes.

Preferably, the reagent particles are magnetic and separating the reagent particles from the indicator solution comprises introducing the indicator solution to a magnetic field.

y 4 } : Preferably, any change in appearance of the indicator solution is detected by spectrometry, colorimetry, chemiluminescence or fluorimetry.

Further aspects of the invention will become apparent from the following description.

Brief Description of the Drawings

An exemplary form of the present invention will now be described by way of example only and with reference to the accompanying drawings, in which:

Figure 1 is a schematic illustration of the use of a plurality of physically linked cotton- wool bud absorbent 3 with a plurality of physically linked micro-tubes 1 placed on a magnet stand 2. : :

Figure 2 is a pictoral results for 2 hemolytic human serum samples and a buffer control all spiked with an anti-O9 mAb (50 pg/ml) in the first well or without such an antibody in the second well, obtained in TUBEX method and the method of the invention. - 20 .

Figure 3 is a bar chart showing the results of mice immunized against S. Typhi LPS antigen or not immunized examined in the method of the invention using whole blood or in TUBEX method using serum and the results for serum of different amounts from the same mice in the method of the invention and TUBEX method. 25 .

Figure 4 is a schematic illustration of chemical structure of 09 or 02 antigen linked directly to the O12 antigen as a repeating unit of the LPS structure in the cell wall of

Salmonella bacteria

Figure S is a graphical illustration of the test results for detecting anti-O12 antibodies in typhoid and paratyphoid A patients using a prototypic TUBEX test for such : antibodies. :

. i

Description of the Exemplary Embodiments

BE An exemplary embodiment of the invention will now be described as practiced in a method to detect antibodies in a liquid sample from a patient. This is not however intended to limit the scope or functionality of the invention, which may equally be practiced in a method to detect antigens in a liquid sample from a patient or antibodies from an animal. The method detects antibodies and antigens in non-clear sera such as hemolytic, icteric or lipemic sera. However, the method may be used to detect antibodies or antigens in clear body fluids or secretions or excretions, cells or tissue extracts, or culture media. 3

In the exemplary embodiment magnetic reagent particles having a diameter of between 50 and 1500 nm are coated with an antigen for the antibody one wants to detect. Typically, the magnetic reagent particles are polystyrene microspheres of about 1000 nm in diameter with a core of high magnetite content (greater than 60%) : and having superparamagnetic properties. The magnetic reagent particles are combined with a liquid sample under test in a transparent reaction well 1 or micro- tube at ambient temperature for sufficient time to allow any antibodies of interest in the liquid sample to bind with the specific antigen coated onto the magnetic reagent particles. This typically takes between 1 and 10 minutes, but 2 minutes is typically sufficient for an accurate test. During incubation any antibodies of interest in the sample will bind with the specific antigen coated onto the magnetic reagent particles.

The reaction well 1 is then placed on a magnetic stand 2 in order to separate the magnetic reagent particles from the liquid sample by sedimentation of the magnetic particles to the bottom of the reaction well 1. The magnetic reagent particles are then carefully washed to remove the supernatant liquid sample. Washing of the magnetic particles is achieved by aspiration with a pipet or an absorbent 3. Where the liquid sample is whole blood, it is necessary that the magnetic particles be rinsed after aspiration with a rinse solution comprising glycine-buffered saline and 0.9% NaCl and the supernatant rinse solution removed by a second aspiration with a pipet or an absorbent 3.

‘ »

After washing the reaction well | is removed from the magnetic stand 2 and an indicator solution is added to the magnetic reagent particles in the reaction well 1.

The indicator solution comprises a suspension of indicator particles having a diameter of between 50 and 1500 nm coated with the antibody of interest. The indicator particles are preferably monodisperse coloured latex microspheres such as styrene- butadiene copolymer, polystyrene, styrene-divinylbenzene copolymer or acrylic copolymer. Any colour can be used, but the preferred colour is red, blue, green or yellow. The concentration of the indicator suspension may vary from 0.001% for an instrumental read assay to 0.2% for visual examination. For most assays in which the results are read visually, a preferred concentration is from 0.01 to 0.08%, using particles preferably about 800nm in diameter. : The reaction well 1 containing the indicator solution and magnetic reaction particles :

Is are vigorously and thoroughly agitated for at least 1 minute and possibly upto 10 minutes. The agitation should be vigorous and thorough enough to rapidly cause antibodies coated onto the indicator particles to bind onto any unbound antigens coated onto the magnetic reaction particles. Typically this will take 2 minutes. The agitation may be done manually or by means of a suitable mixer or agitation device. 200 When manual agitation is employed, a set of physically linked together micro-tubes is : used and mixing may conveniently be effected by holding the set of reaction well 1 in one hand and tapping it repeatedly at a speed of about 150 times per minute against the palm of the other hand or some other object. Alternatively the set of reaction well may be held flat on its side in one hand and shaken back-and-forth in the horizontal plane through an arc of about 40 degrees pivoting substantially about the elbow of the arm. The speed of motion should be not less than 50 back-and-forth movements per minute and preferably 160 back-and-forth movements per minute.

As will be appreciated by those skilled in the art, the number of antibody-bound antigens will depend on the presence and load of target antibodies in the liquid sample previously incubated with magnetic reaction particles. After agitation the reaction well | is returned to the magnetic stand 2 and allowed to sit for 1-5 minutes to allow

. . sedimentation of the magnetic particles to the bottom of the reaction well 1. The results may be read by the naked eye or by an electronic device, viewing through a face of the transparent reaction well to determine a change in colour of the : supernatant indicator solution. If the liquid sample contained no target antibodies all of the coloured indicator particles will sediment with the magnetic reaction particles due to binding of the coated antibodies and antigens and the supernatant will be colourless due to the absence of any suspended particles. If, on the other hand, inhibiting antibodies were present in the liquid sample, these will bind with the antigen coated reaction particles during incubation and prevent binding between the two types of particles to an extent dependent on the concentration of the antibodies I in the liquid sample. A coloured supernatant results from unsedimented coloured indicator particles, ranging from lightly coloured for low antibody concentration in the liquid sample to darkly coloured at high antibody concentration in the liquid sample. The result is thus semi-quantitative, and can be read against an empirically calibrated colour chart.

As discussed earlier an antigen may be any size including a whole bacterium or a small component part of a bacterium or a small chemical group. Generally most antigens are constructed of linked epitopes (sometimes also referred to as antigens).

Each antigen can comprise a specific epitope and an adjoining non-specific epitope, the latter being common to many types of antigens. The specific epitopes and non- ) specific epitopes may be packed very tightly together within the antigen. For ‘ practicality purposes the antigen coated on to the reagent particles includes both the specific epitope and the adjoining non-specific epitope. This can lead to false positive results when particular antibodies bind to the non-specific epitope in the reagent particles. Because antibodies to these epitopes or antigens may be a large structure they block binding of the target antibody to the specific epitope. In order to avoid this condition, a blocker is added to the liquid sample prior to combining with the reagent particles. The blocker is another antigen which has the non-specific epitope but not the specific epitope of the target reagent antigen. The blocker antigen will bind with any antibodies having specificity for the non-specific epitope and thus leave the t v reagent particles uninhibited to receive antibodies having specificity for the specific epitope. : : The reaction well 1 of the present invention preferably comprises a plurality of physically linked together micro-tubes, to allow multiple tests to be performed simultaneously. Preferably there are between 3 and 8 micro-tubes, typically 6. The : set of micro-tubes may be formed integrally, e.g. by casting from a single mould as a single unit or by machining tubes in a block of suitable material. Alternatively, the set : of micro-tubes may be comprised of individual micro-tubes made separately but thereafter held more or less immovably in a holder therefore that allows the reaction contained within the micro-tubes to be observed. The plurality of two or more micro- tubes (of the same or different kinds) are thus physically linked or held together. The material used for the set of micro-tubes is not of importance but should be at least translucent to allow the result of any reaction to be observed. Preferably the set of 15° micro-tubes is cast from a plastics material. Although the specific dimensions of the plurality of micro-tubes and of the individual tubes is not critical, it is important that the tubes be relatively small. If the tubes are small then only a relatively small volume of reagent and sample is required. Preferably each micro-tube may hold from 0.05 to 0.5 ml of reaction mixture. Preferably, the mouth of each tube is wide for easy delivery of the liquid sample and particles, and the tube tapers downwardly to a narrow width so that the small volume of reaction mixture can attain reasonable . height in the tube for easy visual examination. In this way, each tube has a flat V- shaped face large enough in area for the reaction mixture to be mixed adequately.

Moreover, each tube preferably has sufficient depth to allow for weak concentrations of the indicator solution to be used and seen. The face of the set of micro-tubes is preferably flat for easy viewing of a reaction.

Removal of the supernatant (50 pl) from the sediment of the reaction mixture or following washing could be performed using a pipet or more conveniently using a cotton-wool bud absorbent. The advantage of the latter is that the cotton-wool bud absorbent can be inserted as far into the V-shaped well as possible without actually Co

‘ \ : touching the sediment while the well is placed on the magnet stand, and yet is able to drain all of the supernatant.

Where a plurality of physically linked micro-tubes 1 are used, a plurality of physically ~ linked cotton-wool bud absorbent 3 as substantially shown in Fig. 1 is preferably used. The latter comprises a comb-like structure composed of individual cotton-wool buds that form the teeth of the structure which are held together by a support that acts as the handle of the device. It can be conveniently constructed in various ways such as by assembling commercially-available cotton-wool buds in a specially designed support or holder made of light plastics or wood.

Worked Examples of the Invention ~The following test procedure was used to verify the method of the invention against . the TUBEX test, marketed by IDL Biotech, Sollentuna, Sweden. A drop (25 pl) of a test sample was mixed with a two drops (50 pl) of antigen-coated magnetic particles - and stood for 2 minutes. After 2 minutes the reaction well was stood on the magnet stand to sediment the magnetic particles and the supernatant of the reaction mixture was carefully removed using a pipet. Only in situations where the liquid sample is : whole blood, the reaction particles were then rinsed by adding 50 pi of glycine- ~ buffered saline (GBS; comprising 0.1 M glycine [Sigma Chemical Co., St. Louis,

MO] and 0.9% NaCl, final pH 8.2) to the reaction well, without disturbing the sediment, while the reaction well was still on the magnet stand. The wash is carefully : removed again using a pipet. In all situations, a buffer comprising 50 ul of GBS containing 1% bovine serum albumin (Sigma) was then added to the sediment, the reaction well was removed from the magnet stand, and the sediment was lightly and gently mixed in the buffer. Two drops (50 pl) of antibody-coated blue indicator particles were added to the reaction well the reactants thorough mixing for 2 minutes, followed by magnetic sedimentation of the magnetic particles. Various trials were : undertaken using this procedure and TUBEX test for comparison. The results of these

Cl16- tests were scored using TUBEX scores (‘0° being most negative and ‘10° being most positive).

CT TEST : :

The test sample was a GBS buffer spiked with various amounts of an anti-09 antibody. This was a mouse IgM antibody purified from mouse ascites by cryoprecipitation and quantitated using the BCA™ Protein Assay kit (Pierce,

Rockford, IL). The results are given in the following table 1. } : .

Vol test . Concentration of spiked mAb (ug/ml) sample | 1 1 125 31.3 7.8

TE eee

Invention : : : : 100 | 9 S| 8 6 | 4) 2 0

Table 1. A comparison of the efficiency of detection of spiked anti-09 mAb in various amounts of buffer between TUBEX method and the method of the invention. : (Results shown as TUBEX scores.)

The test results show that for the same volume of test material (25 pl), the sensitivity of detection of the spiked antibody was very similar (31.3 pg/ml; using score < 2 as negative) between the method of the invention and the TUBEX method. Since bigger volumes of test material could be used with the method of the invention, but not with the TUBEX method, it was investigated whether increasing the volume would increase the sensitivity of the assay. Table 1 shows that this was indeed the case, with the sensitivity increasing to 15.6, 7.8 and 3.9 pg/ml for 50, 100 and 200 pl volume of test sample used, respectively.

! .

TEST 2

The procedure was repeated using clear human serum to verify that this medium had no adverse effect on the method of the invention. The procedure was undertaken using 25 pl, 50 pl and 100 pl of the test sample and the serum was spiking with the | . : 3h1 mAb obtained from a pool from 4 healthy people.. The results are given in the following table 2.

Volume of test sample mAb added? | | I yes yes no | yes | no (25 pg/ml) :

IC FC FC HL 0

Table 2 NA = not applicable (fixed volume used in TUBEX) : 10 A comparison of the efficiency of detection of spiked anti-O9 mAb in various : amounts of normal human serum between TUBEX method and the method of the invention. (Results shown as TUBEX scores.)

This was not only proven to be true but importantly, again, increasing the volume of test material from 25 pl to 50 pl or 100 pl, increased the positive test score correspondingly (from score 4 to score 6 and score 8, respectively). :

TEST 3

The procedure was repeated using hemolytic human sera. Ten such samples were randomly selected from specimens sent for routine autoantibody analysis to the Prince of Wales Hospital in Hong Kong. All samples were not testable in the TUBEX method but were clearly negative when used in the method of the invention, and were clearly positive in the latter when these samples were spiked with 50 pg/ml] of the 3hl

- mAb (data not shown). The pictoral results for two of these sera obtained in the two methods are shown in Figure 2.

Verification that whole blood could also be used in the method of the invention was further verified using heparinized blood obtained from a healthy person, which was spiked with 3h1 mAb (25 pg/ml). The results are shown in the following table 3,

Volume of test sample 100 ul* | 250 uss mAb added? : yes yes no (25 pg/ml) * 150 pl 2x GBS buffer added for buffering ** 39 ul 10x GBS buffer added for buffering : Table 3

Detection of spiked anti-O9 mAb in different amounts of whole human blood by method of the invention. (Results shown as TUBEX scores.)

The results verify that whole blood can be used in the method of the invention, but a second wash with GBS buffer (50 pl) was necessary to remove all traces of blood. As in buffer or serum, the TUBEX® score was found to increase with increasing amounts of spiked mAb (from score 6 [100 pl blood] to score 8 [250 pl}. In a further study, no reactivity was observed when the whole blood of 14 other healthy individuals were similarly examined, but when these samples were spiked with 3h1 mAb (25 ug/ml), all became positive, yielding TUBEX® scores of 4-6 (data not shown).

TEST 4 _ To verify that the method of the invention worked with real specimens blood samples were taken from mice immunized against S. Typhi LPS antigen. The main portion

(100-150 pl) of the blood samples was placed in a tube containing 3 pl heparin (1,000

IU/ml; Mayne Pharma Pty Ltd, Melbourne, Australia), while the remainder was allowed to clot for use as serum. The whole blood (100 pl) was tested by the method of the invention and the serum (25 pl) tested using the TUBEX test. The results obtained between the two methods are shown in Figure 3a. Both methods detected antibodies (TUBEX score 3-7) in § of 8 immunized mice with comparable a reactivities, with an additional mouse being marginally positive in the method of the invention, and all 5 non-immunized (control) mice were not reactive in both methods.

It thus seems that 25 ul of serum contain roughly the same amount of relevant : 10 antibodies in 100 pl of blood. The alternative explanation that the bigger volume of fluid used in the blood sample contains more antibodies but a longer incubation is : required to complete the rezation is not likely in view of the studies shown in Table 1- 3. Itis possible, indeed, that if more blood (e.g. 200 pl) is used in the method of the invention, detection might in fact be improved. This is not possible to verify in the .I5 present study because of the small blood volume in the mouse, but using clinical specimens in future will be able to ascertain this likely prospect.

An important implication of these studies is that the method of the invention might actually be superior to the original method even for clear serum samples. This is because (a) a greater sample volume (e.g. 100 pl) can be used that can enhance the sensitivity of the test, and (b) the wash step, which only requires minimal effort and time, reduces any background color of the specimen and thus standardizes the specimens used. To verify this supposition, further blood samples were taken from the test mice used in the above study one month later. The serum obtained from each animal was used in parallel in the TUBEX method, using 25 pl serum, and in the method of the invention using twice as much serum (50 pl). The results shown in

Figure 3b suggests that the method of the invention is more sensitive than the TUBEX method in almost all cases (5/6 responder mice), detecting 6 of the immunized mice with TUBEX scores 5-9, compared to scores of 4-7 with the TUBEX for the same animals. The remaining two mice, as well as all of the 5 un-immunized mice, were negative in both methods.

20-0

In the study described hereunder, the specificity of the TUBEX test was examined.

The question asked was whether it detected typhoid fever only or would it also detect a very similar and related disease called paratyphoid A fever that is clinically indistinguishable from typhoid fever, which is caused by another Salmonella organism called Salmonella Paratyphi A (see Table 5). In theory, the TUBEX test which detects antibodies to the O9 antigen that S.Paratyphi A lacks, should be specific for typhoid fever only. Conversely, detecting antibodies to the O2 antigen found in S.

Paratyphi A (the counterpart of 09 in S.Typhi) but Jacking in S. Typhi organisms, should be specific for paratyphoid A fever only. To investigate these possibilities, a prototypic TUBEX test for paratyphoid A (TUBEX-PA) that specifically detects : anti-O2 antibodies was developed. The components of this fest are similar to those : used in the TUBEX test for typhoid fever (herein called TUBEX-TF for the sake of clarity) except that S.Paratyphi A LPS is used instead of S.Typhi LPS, and an IgM

I5 anti-O2 monoclonal antibody is used instead of the anti-O9 antibody. The test procedure remains the same as in the TUBEX test.

In the first study aimed at finding out whether TUBEX-TF was truly specific for typhoid fever, 15 serum samples from 14 culture-confirmed paratyphoid A patients from Kumming, China, were examined by both TUBEX-PA and TUBEX-TF tests.

CE EE

«0 EEE B

EEE oo . 6* Cs] A 7+ EE 8 8* | 1 6 8 ; TTT 100 EY 6 8 SE

EE ER

IE EC

IE EE I

Co 11/15x 100 | 14/15 x 100

Sensitivity =73.3% =03.3% * —, negative. TUBEX positive scores: higher, more positive ! duration of fever at sampling; ?, unknown period * from same outbreak; Sa and 5b same individual

Table 4 Table showing sensitivity of detection of paratyphoid A fever in febrile

S.Paratyphi A-confirmed patients by TUBEX-TF and TUBEX-PA.

Eight of these individuals were from the same outbreak of paratyphoid A fever.

Fourteen (93.3%) of the samples were positive in TUBEX-PA, attesting to the high sensitivity of the test. The odd negative sample (#13) was an early serum obtained after 4 days of fever. Except for a patient (#4) who had a low TUBEX score (score 4), all other cases were strongly positive (mostly score 8). The same paratyphoid A patients were then examined to see if they could be cross-detected by TUBEX-TF.

Indeed, 11 of the 15 paratyphoid A patients (73.3%) were found to be positive in

TUBEX-TF.

TEST 6

In another study, the converse was examined to see whether TUBEX-PA could detect typhoid fever. Indeed, out of 11 culture-confirmed typhoid patients, 7 of these individuals (50%) were found to be positive (data not provided). This suggests that the test cross-detected a significant proportion, but not all, of typhoid patients, whereas TUBEX-TF, although not used on these typhoid patients, is known from previous published studies to be able to detect many more of typhoid patients (75- 95%). :

The mutual cross-detection observed between both tests (TUBEX-TF and TUBEX-

PA) may be explained by the presence of a common, minor antigen in S. Typhi and

S.Paratyphi A. A likely candidate is the O12 antigen as shown in Table 5 below and

Figure 4.

Group Salmonella 0 Antigens ~ Hantigens

S. Stanley 1, 4, (5), 12, 27 d 1,2 : S. Typhimurium _ 1, 4, (5), 12 i 1.2. : S. Choleraesuis 6,7 c 1,5 : C2 S. Manhattan 6, 8 d 1.5 : 1 D S. Sendai 1, 9,12 a 1,5

S. Typhi 9,12, Vi d -

Table 5 shows a small representation of the family of Salmonella organisms grouped : according to their lipopolysaccharide (‘0’) and flagellar (‘H’) antigens, a classification originally devised by Kauffmann and White. Note in particular that S.

Typhi belongs to a subgroup that has the unique O9 antigen, while S.ParatyphiA belongs to another subgroup with another unique antigen (02), but both organisms share the O12 antigen.

Antibodies to 012, particularly of the IgM class that are structurally bulky and produced only during current infections, can perceivably inhibit the binding of the indicator anti-O9 or anti-O2 monoclonal antibody to the target LPS antigen by steric hindrance, due to the close provimity between the 09 or O2 and O12 antigens (see

Fig. 4). oo oo

Test samples from typhoid and paratyphoid A patients from Guangxi, China, as well as several typhoid patients from Hong Kong and the Philippines, were examined with : : both TUBEX-PA and TUBEX-TF. For cases that were positive in TUBEX-PA, the : test was repeated in the presence of soluble (free-form) S.Typhi LPS to see if this could block the activity of any anti-O12 antibodies present. Anti-O12 antibodies, but not anti-O2 antibodies would be expected to bind to S. Typhi LPS and would thus be neutralized. The LPS solution was simply added to the initial reaction mixture (comprising the test serum and the LPS-coated magnetic particles), while the rest of the test procedure remained unchanged. Conversely, cases positive in TUBEX-TF were repeated using S.Paratyphi A LPS as the blocking agent. Thus, in theory, the reactivity would be significantly inhibited by the heterologous LPS if the case was not typhoid in the TUBEX-TF test, or not paratyphoid A in TUBEX-PA. On the other hand, no significant change in results would be expected for true typhoid cases in

TUBEX-TF or for true paratyphoid A in TUBEX-PA. The results of the study are displayed in Table 6-8 below. | oo ’

Table 6 Comparative performance of TUBEX-TF and TUBEX-PA in typhoid patients

LPS LPS | LPS LPS

LPS LPS

— | : mp TTR ee Tw ™ 7 | + [7] 160/320] 40 ND Hm | oH H | H HH

Cl al ree

BA EA REAR

EAE RAE EARAE

Ghee eee wt ee]

I i A LA RA EA LAS "SENSITIVITY No case of TUBEX PA-positive when TUBEX.TF is negative

Table 7 Comparative performance of TUBEX-TF and TUBEX-PA in paratyphoid A patients

S.Paratyphi A isolated (all true paratyphoid A patients)

TUBEX-PAJ WIDAL (liter) | "ELISA IgM | ELISA IgG + IgM

CC LAA

EEE 8 [40160 0 - | M HM H | 0 |= oe ee fe

PAG EEE 40 160]1280 8 H | H | H| H HH

PAT | = | = | 7 | 4 No ND [ND RET CM ERER H aE RAR lL ALIEN REAR]

Te p ERERES “Tw — = — rT FTE “Palo 6 = 8 E -| =o] M |u| H | H hall | 4 BE 160 — | — ND] TL H | H

RINE 8 BEE 160 M | L H | H

EERE we [ee ak

Ea CN RCT

SENSITIVITY No case of TUBEX TF-positive when TUBEX-PA is negative oT Table 8 Results of patients bacteriologically negative for typhoid and paratyphoid A examined by TUBEX-TF and TUBEX-PA (column headings as in Table 6-7)

S.Typhi and S.Paratyphi A not isolated, febrile subjects :

ERDEDREEES

Ce ERE FREE ET - 400 et eked _ [meee

EEE opel ol eee ‘Table 6, 7, 8. —, negative result; ND, test not done; TUBEX ‘screen’ = normal test procedure, TUBEX ‘check’ = inhibitor (heterologous LPS) used in test procedure, TUBEX positive scores 4 to 10, increasing intensity; ELISA results: H (high), M (medium), L (low) antibody levels based on OD values > mean of oo -26- healthy subjects + 4SD, 2SD (but < 4SD) and 1SD (but < 2 SD) respectively; S para A, S.Paratyphi A;

S.tyrm, S.Typhimurium : : tall subjects from Guangxi except for some typhoid patients from Hong Kong (T7-T9) and the

Philippines Philippines (T10-T11) ® serum obtained 3-7 days after onset of fever © meningococcus isolated ¢ Salmonella group C isolated ° streptococcus isolated

The test results confirm, firstly, that TUBEX-TF (10/11 or 90.9%) was slightly more sensitive than TUBEX-PA (9/11 or 81.8%) in detecting typhoid patients. Conversely,

TUBEX-PA (13/15 or 85.7%) detected many more paratyphoid A patients than

TUBEX-TF (7/15 or 46.7%). Notably, there was not a case of typhoid fever that was negative for TUBEX-TF but positive for TUBEX-PA, nor a case of paratyphoid A fever with the converse results. Results of the Widal test also performed in this study for the paratyphoid A patients indicated that many of these patients (7/12 or 58.3%) developed significant antibody titers (1:40 - 1:1280) fo AO, and most of these : 75° individuals also had antibodies to TO.

Importantly, the reactivities (TUBEX “screen” scores 4-8; see Table 6, of all typhoid patients in TUBEX-PA were found to be completely abolished (negative “check” Co scores) when S.Typhi LPS was added to the reaction mixture. Likewise, the : reactivities (scores 4-6; see Table 7) for all paratyphoid A patients in TUBEX-TF were abolished by S.Paratyphi A LPS. This demonstrated the high specificity of the reaction in both cases i.e. the reactivity was real and not an assay artifact, which was

Co mediated by anti-O12 antibodies.

As predicted, the TUBEX-TF score of many typhoid patients (8/10 or 80.0%; see Table 6) remained essentially unchanged (positive) in the presence of S.Paratyphi

A LPS, suggesting the reactivity was due largely to anti-O9 antibodies. However, for : some typhoid patients (#23 and B2) who had weakly positive TUBEX-TF ‘screen’ scores, the ‘check’ results based on LPS inhibition became negative, suggesting anti-

O12 antibodies may be predominant in these cases. Abundant amounts of IgM and

IgG anti-O12 antibodies were in fact found in all typhoid patients examined, evidenced by the extremely high anti-LPS ELISA reactivities of the sera, especially

-27- : Co when both IgG and IgM antibodies were measured, that were indistinguishable between the homologous antigen (S.Typhi LPS, 09" and 012") and the heterologous antigens (S.Paratyphi A LPS and S. Typhimurium A LPS, both 09" and 012%.

However, more IgM anti-O9 antibodies appeared to be produced by the patients than | IgM anti-O12 antibodies judging from (a) the IgM anti-LPS reactivities of 2 of the patients, and (b) the greater sensitivity of TUBEX-TF over TUBEX-PA in detecting these individuals. It is possible that different amounts of anti-012 and anti-09 antibodies are made in different patients and at different stages of the disease in a person. : oo

Similar observations were made of the paratyphoid A patients (see Table 7) following re-testing of the sera in the presence of S. Typhi LPS in TUBEX-PA, in which the results of 10/13 or 76.9% of the cases remained positive, demonstrating the important role anti-O2 antibodies (presumably IgM) played in the reactivity. These antibodies appeared to be more dominant and more common than the IgM anti-O12 antibodies.

This is shown not only by the greater sensitivity of TUBEX-PA than TUBEX-TF in detecting these patients, but more clearly, by the greater ELISA reactivity of the IgM antibodies in most of these sera for the homologous antigen (S.Paratyphi A LPS, 02° and O12") than for the heterologous antigen (S.Typhi LPS or S. Typhimurium LPS, both 027, and 012%). The abundant amounts of IgG anti-O12 antibodies in the same sera resulted in very high ELISA activities for all 3 types of LPS. Presumably, anti-

O12 antibodies in these paratyphoid A patients were responsible for the crossreactivity seen in the typhoid (TO) Widal test for many of the cases.

Thus, when used together, TUBEX-TF and TUBEX-PA can provide a robust screening against typhoid and paratyphoid A fever. By repeating the test using : heterologous LPS inhibition in positive cases, it is possible to ascertain whether the case is typhoid or paratyphoid A in the majority of cases. The usefulness of this : approach is illustrated in the study shown also in Table 8, where 9 febrile patients from whom neither S.Typhi nor S.Paratyphi A was isolated were examined. First, 7 of the subjects clearly did not have typhoid or paratyphoid A fever since they were negative for the TUBEX tests and negative (or weakly positive in 2 cases) for the anti-

_ LPS ELISAs. Four of these individuals were in fact infected with meningococcus,

Salmonella Group C spp or Streptococcus spp. This attests to the high specificity of the TUBEX tests. More interestingly, individual N3 appeared to have true typhoid oo because TUBEX-TF was positive which remained so after S.Paratyphi A LPS was added, while TUBEX-PA was also positive which became negative after the addition of S.Typhi LPS. Supportive evidence comes from the strong TO, TH and ELISA activities found in this person. Individual N1 was positive in both TUBEX-TF and : TUBEX-PA, and in both cases, the reactivity was abolished by LPS. Together with the positive TO and ELISA results, it is likely that this person had typhoid (serologically similar to patients #T7 and T11) although paratyphoid A (or B) fever cannot be ruled out completely. Cases such as N1 may pose a serological dilemna but these are real Salmonella infections and not infections caused by other organisms because the 09, 02 and O12 stigens are only found in Salmonella. is A corollary of these findings demonstrating the usefulness of adding an inhibitory or blocking substance to make a test more specific is the possibility that in a TUBEX test | : aimed at detecting antigen rather than antibody, an antibody similar or identical to the first reagent may be similarly used as a blocker to verify the specificity or authenticity of the reaction. :

Test 8

Support for the diagnostic importance of anti-O12 antibodies in typhoid and paratyphoid A patients comes from another study in which another prototypic

TUBEX test was developed to detect these antibodies directly. An IgM anti-O12 monoclonal coupled to blue-colored indicator particles and S.Typhi LPS coated- magnetic particles was used in the same inhibition format and assay procedure as

TUBEX-TF and TUBEX-PA. This test detected 66.7% (22/33) of culture-confirmed paratyphoid A patients, and 75.0% (9/12) of culture-confirmed typhoid patients from the Guangxi population (Fig.5). None (0%) of 18 non-febrile control subjects were positive in the test. These detection rates are consistent with the cross-detection rates observed in TUBEX-TF and TUBEX-PA. Thus, detecting anti-O12 antibodies is not

: as sensitive as detecting the anti-O9 antibodies in typhoid patients or the anti-O2 antibodies in paratyphoid A patients. :

Claims

\ I B -30- : . Jp—— eT . Claims

1. A method for detecting antibodies or antigens in a liquid sample, the method comprising: adding a liquid sample to reagent particles coated with a first reagent, incubating the liquid sample and reagent particles, separating the reagent particles from the liquid sample, thoroughly mixing the separated reagent particles in an indicator solution comprising indicator particles coated in a second reagent, separating the reagent particles from the indicator solution, and detecting any change in appearance of the indicator solution. 3

: 2. The method of claim 1 wherein the liquid sample is selected from a group comprising whole blood, or serum or a body fluid that lack a clear appearance. oo

3. The method of claim 1 wherein the liquid sample is selected from a group comprising hemolytic, icteric or lipemic sera.

4, The method of claim 1 wherein the first reagent is an antigen and the second reagent is an antibody.

5. The method of claim 1 wherein the first reagent is an antibody and the second reagent is an antigen,

6. The method of claim 1 wherein the method is performed in a plurality of physically linked together micro-tubes.

7. The method of claim 1 wherein the liquid sample and reagent particles are : incubated for between 1 and 10 minutes.

8. The method of claim 1 wherein the reagent particles are magnetic and separating the reagent particles from the liquid sample or indicator solution comprises introducing the indicator solution to a magnetic field. Ss

9. The method of claim 8 wherein separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles by aspiration with a pipet or an absorbent. :

10. The method of claim 9 wherein separating the reagent particles from the liquid sample further comprises washing the reagent particles with a washing solution and removing the washing solution from the reagent particles by aspiration with a pipet or an absorbent. :

11. The method of claim 9 wherein separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles in a plurality of physically linked together micro-tubes by aspiration with a corresponding plurality of physically linked together absorbents.

12. The method of claim 1 wherein thoroughly mixing the reagent particles and the indicator solution comprises mixing for between 1 and 10 minutes.

13. The method of claim 8 wherein the indicator solution is introduced to the magnetic field for less than 5 minutes before detecting any change in appearance of the indicator solution.

14. The method of claim 1 wherein any change in appearance of the indicator solution is detected by spectrometry, colorimetry, chemiluminescence or fluorimetry.

15. The method of claim 1 wherein prior to adding the liquid sample to the reagent particles a blocking agent is added to the reagent particles. :

16. The method of claim 15 wherein the first reagent is an antigen having a specific epitope and an adjoining non-specific epitope, and the blocking agent is an oo antigen having the non-specific epitope but not the specific epitope.

17. The method of claim 16 wherein the reagent particles are magnetic and separating the reagent particles from the liquid sample or indicator solution comprises introducing the indicator solution to a magnetic field.

18. The method of claim 17 wherein separating the reagent particles from the {0 liquid sample comprises removing the liquid sample from the reagent particles by aspiration with a pipet or an absorbent.

19. The method of claim 16 wherein separating the reagent particles from the : liquid sample comprises removing the liquid sample from the reagent particles in a plurality of physically linked together micro-tubes by aspiration with a corresponding plurality of physically linked together absorbents. oo

20. The method of claim 16 wherein the method is performed in a plurality of Co physically linked together micro-tubes. : .

21. The method of claim 15 wherein the first reagent is an antibody having non- . specific epitopes, and the blocking agent: is an antigen having the non-specific epitopes.

22. The method of any one of claims 15 to 21 wherein the step of separating the : reagent particles from the liquid sample is omitted.

23. A method for detecting antibodies or antigens in a liquid sample, the method comprising; . adding a blocking agents to reagent particles coated with a first reagent, adding a liquid sample to the reagent particles,

I Ie y RK ’ incubating the liquid sample and reagent particles to bind any antibodies or antigens in the sample to the first reagent, : thoroughly mixing the separated reagent particles in an indicator solution comprising second particles coated in a second reagent, Co separating the reagent particles from the indicator solution, and detecting any change in appearance of the indicator solution.

24. The method of claim 23 wherein the first reagent is an antigen having a specific epitope and an adjoining non-specific epitope, and the blocking agent is an

10. antigen having the non-specific epitope but not the specific epitope.

25. The method of claim 24 wherein the reagent particles are magnetic and separating the reagent particles from the liquid sample or indicator solution comprises introducing the indicator solution to a magnetic field.

26. The method of claim 25 wherein separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles by aspiration with a pipet or an absorbent.

27. The method of claim 25 wherein separating the reagent particles from the liquid sample comprises removing the liquid sample from the reagent particles in a : plurality of physically linked together micro-tubes by aspiration with a corresponding plurality of physically linked together absorbents.

28. The method of claim 25 wherein the method is performed in a plurality of physically linked together micro-tubes. :

29. The method of claim 23 wherein the indicator solution comprises indicator particles coated in antibody.

RR SE

30. The method of claim 23 wherein the first reagent is an antibody having specific epitope and an adjoining non-specific epitope, and the blocking agent is an antibody having the non-specific epitope but not the specific epitope.

31. The method of claim 23 wherein the indicator solution comprises indicator particles coated in antigen.

32. The method of claim 23 wherein the method is performed in a plurality of physically linked together micro-tubes.

33. The method of claim 23 wherein the liquid sample and reagent particles are Co incubated for between 1 and 10 minutes. : :

34. The method of claim 23 wherein thoroughly mixing the reagent particles and the indicator solution comprises mixing for between 1 and 10 minutes.

35. The method of claim 23 wherein the reagent particles are magnetic and : separating the reagent particles from the indicator solution comprises introducing the indicator solution to a magnetic field. : )

36. The method of claim 35 wherein the indicator solution is introduced to the magnetic field for less than 5 minutes before detecting any change in appearance of the indicator solution.

25. 37. The method of claim 23 wherein any change in appearance of the indicator solution is detected by spectrometry, colorimetry, chemiluminescence or fluorimetry.

38. A method according to claim 1 substantially as herein described as exemplified.

ooo, : : | 35.

39. A method according to claim 23 substantially as herein described or exemplified. DATED THIS 3° DAY OF MARCH 2009 SPOOR & FISHER APPLICANT’S PATENT ATTORNEYS