WO1991003572A1

WO1991003572A1 - Test for the detection of staphylococcal fibronectin-receptor antibodies

Info

Publication number: WO1991003572A1
Application number: PCT/US1990/005065
Authority: WO
Inventors: Richard A. Proctor
Original assignee: Wisconsin Alumni Research Foundation
Priority date: 1989-09-08
Filing date: 1990-09-07
Publication date: 1991-03-21
Also published as: EP0444174A4; EP0444174A1; JPH04503114A

Abstract

A test for Staphylococcal fibronectin-receptor antibodies in a test sample employs purified fibronectin-receptor exopolysaccharide derived from Staphylococcus aureus as the antigen. A kit for performing an ELISA test is also disclosed.

Description

TEST FOR THE DETECTION OF STAPHYLOCOCCAL FIBRONECTIN-RECEPTOR ANTIBODIES

The invention relates to diagnostic tests or assays for the detection of staphylococcal fibronectin-receptor antibodies in body fluids, such as sera.

The mechanisms by which Staphylococcus aureus, an infectious microorganism, adheres to host tissues have been studied because adherence is the first step in the develop¬ ment of an infection. If bacteria are unable to adhere to a surface, they will be swept away by the body fluids that normally bathe the tissues and an infection will not occur. Hence, adherence is the crucial first step in initiating and spreading infections. It is generally accepted that microorganisms adhere to surface components of the host tissues.

It has been found that a specific receptor on S. aureus interacts with a host protein called "fibronectin" to medi¬ ate adhesion to host tissues. Fibronectin is a large glyco- protein which is a major component of the material found on the surfaces of cells, in clots and also in the inter¬ cellular matrix. Fibronectin is essential to the well-being of the host as it serves as the "glue" which links one cell to another cell and plays a major role in wound healing. There are several lines of evidence which suggest that fibronectin plays an important role in the adherence of S. aureus to host tissues. Supporting this concept are the following observations: (i) S. aureus has a specific receptor for fibronectin, (ii) the S. aureus fibronectin- receptor (FN-R) is expressed on the surface of the bacterium where it can interact with host tissues, (iii) the fibro¬ nectin-receptor is expressed in greater numbers on clinical isolates of S>. aureus that have invaded the host as compared to noninvasive isolates, (iv) purified FN-R decreases Si. aureus interactions with fibronectin, (v) antibodies directed against FN-R reduce fibronectin binding to S. aureus, (vi) fibronectin is found at sites frequently infected by S. aureus, (vii) fibronectin enhances S. aureus adherence, (viii) specific removal of fibronectin from a complex mixture of host proteins decreases S. aureus binding while removal of other proteins does not, and (ix) anti¬ bodies against fibronectin inhibit S . aureus binding to host tissues.

There also are data which suggest that the fibronectin- receptor polysaccharide plays a major role in the initiation of S. aureus infections: (i) S. aureus mutants that are isogenic with the parent strain except for the ability to express fibronectin-receptor colonize the heart valves of rats 120-fold less effectively than the parent strain and (ii) a monoclonal antibody which recognizes the most invasive strains of S. aureus interacts preferentially with the fibronectin-receptor.

Recently, an exopolysaccharide has been isolated which is expressed on S . aureus and which specifically binds fibronectin. It can be highly purified and it functions as a fibronectin-receptor.

It is an object of the present invention to disclose a diagnostic test or assay for the detection of Staphylococcal fibronectin-receptor antibodies. It is a further object to disclose an ELISA test which can be used to quantify antibiotic production in response to a purified fibronectin-receptor.

I have discovered that an antigen based diagnostic test or assay for the detection of staphlococcal fibronectin- receptor antibodies can be prepared using as the antigen a purified fibronectin-receptor exopolysaccharide which is expressed on Staphylococcus aureus and which specifically binds to fibronectin.

The exopolysaccharide contains no lipids, less than 2% protein and the sugars making up the exopolysaccharide are aminohexoses. The purified exopolysaccharide cross-reacts with monoclonal antibody directed against the type 8 capsular polysaccharide of S. aureus, but the fibronectin- receptor polysaccharide of the present invention contains, galacturonic acids and thus is distinct from type 8 capsular material. The fibronectin receptor exopolysaccharide of the present invention has an average molecular weight of -1000 kdal and it competes with intact organisms for fibronectin binding.

The fibronectin-receptor exopolysaccharide is harvested from intact Staphylococcus aureus cells by gently releasing the surface material, including the expressed fibronectin- receptor polysaccharide, on the cells without rupturing or killing the cells by non-destructive procedures, such as gentle sonication. The exopolysaccharide is then purified by DEAE ion exchange and fibronectin affinity chroma- tography. The preferred S . aureus is designated strain 6850.

In an indirect ELISA test to screen patients for anti- fibronectin-receptor antibodies, the exopolysaccharide, which serves as a ligand, is bound to a solid phase immunosorbent support. The test sera is reacted with the bound exopolysaccharide. Then fibronectin is added, and the mixture incubated and washed. Finally, an alkaline phosphatase-conjugated, anti-fibronectin-antibody is added. The test concludes with a detection step, compatible with the label used, designed to measure the amount of antibodies bound to the exopolysaccharide in the test sera. Monoclonal antibodies against the fibronectin- receptor exopolysaccharide are used as standards in the ELISA test employing the fibronectin-receptor polysaccharide as an antigen. The monoclonal antibodies employed can be the previously described monoclonal antibodies directed against type 8 capsular polysaccharide or antibodies obtained by injecting the fibronectin-receptor exopolysaccharide into mice, removing the spleens of the mice showing antibody that blocked fibronectin binding to FN-R, cloning the antibodies, preparing hybridomas by the method of Kohler and Milstein and selecting the monoclonal antibodies that demonstrate blocking activity and inhibit fibronectin binding to FN-R. Other objects and advantages of the invention will be apparent to those skilled in the art.

In the preferred practice of the invention, the diagnostic test is an ELISA test in which fibronectin- receptor exopolysaccharide (FN-R) is absorbed on the polystyrene walls of microtiter plates, washed and dried.

The plates are incubated at room temperature. The test sera is added and the mixture is incubated for about 60 minutes. After washing, polyclonal rabbit anti-human fibronectin antibody conjugated to alkaline phosphatase (1:2000 dilution) is added and incubated for 2 hours and washed. An enzyme substrate is added and the mixture is incubated for 2 hours. It is then read with an automated fluoresence spectrophotometer to determine the amount of Staphylococcal fibronectin-receptor antibodies present in the test sera.

The invention is further illustrated by the examples which follow:

Example 1

Method of Preparation of Purified Fibronectin-Receptor Exopolysaccharide

S. aureus (strain 6850) is grown in chemically defined or dialyzed media for 16 hours. Bacteria are harvested by centrifugation, washed twice in Hank's balanced salt solution, and then once in sterile distilled water. The organisms are then resuspended in distilled water and gently sonicated (two, 2.5 minute bursts in a cup horn Heat Systems sonicator at 18-20% output). The sonication removes cell surface material without killing the S . aureus, as determined by no change in the viability counts and by lack of DNA release. The sonicated mixture is then centrifuged

5 . to remove bacteria. The supernatant is saved and lyophilized. The lyophilized material is then placed on a QAE Sephadex column and eluted with a linear salt (NaCl) gradient. Fractions are collected and the presence of fibronectin receptor polysaccharide (FN-R) determined by

10 competitive inhibition of radiolabeled fibronectin with S. aureus ATCC 25923. Fractions containing the FN-R are then passed through a G-200 Sephadex column. Fractions are collected as they come off this second column and the

,_ presence of FN-R determined. Those fractions containing the

15 FN-R are then pooled and lyophilized. This material is purified approximately 1000-fold as compared to the starting material which was shed off the bacteria. The chemical characterization of this material shows that it is a polysaccharide which is less than 2% protein, contains no

20 lipids, and is primarily carbohydrate. The sugars making up the carbohydrates are aminohexoses, and contain galacturonic acids. The FN-R has an average molecular weight of -1000 kdal. This polysaccharide cross-reacts with monoclonal antibody directed at the type 8 capsular polysaccharide of

25 S. aureus, but the chemical characterization shows that it is distinct from the type 8 capsular material which contains only mannosa ino uronic acids and fulosamine.

The prior published techniques for FN-R purification have relied on harvesting the bacteria from complex media

30 and using broken open bacteria. This causes problems in the interpretation of validity of the material that has been harvested for two reasons. First, bacteria can adsorb complex media components onto their surface. The media used in the previous reports are made from crude digests of

35 mammalian tissues and could quite possibly contain components which are known to interact with fibronectin (such as collagen and fibrin). The nonchemically defined edia contain many macromolecules which could stick to the surface of the £3. aureus and be purified as if they were produced by the bacteria. Second, by using broken open S . aureus, many intracellular proteins are being released, some of which may interact with fibronectin, but they may not be expressed on the surface of the microorganism. It has been observed that broken open S . aureus contain many more fibronectin receptors than do intact bacteria. The fact that antibodies produced against the protein fibronectin receptor do not block fibronectin binding to S . aureus suggests that these protein materials are not the critical ligands for promoting fibronectin-S. aureus interactions. This is in contrast to antibodies which react with the carbohydrate FN-R of the present invention, wherein the binding of fibronectin is almost completely blocked. Hence, breaking open the bacteria may release fibronectin receptor materials which are not expressed normally. These "released" materials are different from the FN-R expressed on the outside of the bacteria. Consequently, the novel technique of harvesting the surface material without breaking open the bacteria prevents the harvesting of potentially nonexpressed, nonsurface fibronectin receptor material. Thus, the harvesting techniques are unique and have led to the isolation of a carbohydrate FN-R polysaccharide which is important in the pathogenesis of S_. aureus infections.

Example 2

Production of Monoclonal Antibodies Against The Fibronectin-Receptor Exopolysaccharide The FN-R polysaccharide obtained by the above method was used as an antigen to prepare monoclonal antibodies. One hundred micrograms of the antigen was mixed with complete Freund's adjuvant and injected subcutaneously into multiple sites on BALB/c mice. The mice were then injected with 100 yg of antigen intraperitoneally after 2 and 4 weeks. At 6 weeks, the mice were injected with 100 μg intravenously. Two and three days later, the mice were given 100 ι_g of antigen intraperitoneally. The mice were bled and those showing blocking antibody (antibody which blocked fibronec¬ tin binding to the FN-R in the ELISA test) had their spleens removed. After screening approximately 800 clones, several clones were made into hybridomas by fusing the clones with cancerous tumor cells (NS-1 cells) derived from myeloma tumors of bone marrow after the method of Kohler and Milstein. The hybridomas were screened to select ones producing the desired antibody and grown in culture to produce monoclonal antibodies. Seven monoclonal antibodies were produced which demonstrated blocking activity. Two monoclonal antibodies gave 70-92% inhibition of fibronectin binding to the FN-R in an ELISA assay when these antibodies were diluted 1:100.

Example 3

Diagnostic Assay (ELISA) for Determining the Presence of Anti-FN-R Antibodies in Patients' Sera

The FN-R exopolysaccharide employed was harvested from the surface of a strain expressing large numbers of FN-R (S_. aureus strain 6850) (ATCC No. 53657). This strain was discovered by screening several hundred strains of S . aureus. The FN-R polysaccharide was purified as described above. The lyophilized polysaccharide was resuspended in distilled water to a concentration of 1 mg/ml. It was then tested for inhibitory activity in a standard binding assay and adjusted to a concentration that gave 50% inhibition of 3 ug ¹²⁵I-labeled fibronectin binding to 5 x 10° S. aureus ATCC 25923 (this is generally about 1 yg of polysaccha- ride.) Then 10 μg of this FN-R polysaccharide which was to serve as the antigen was placed in each well of a 96 well microtiter polystyrene plate and allowed to dry overnight. Antibody, either from the patient's serum or added as a control, was then added to the well. Sera were pretreated with gelatin-Sepharose to remove plasma fibronectin. One microgram of fibronectin was then added to each well. After a 60 minute incubation at room temperature, the wells are washed and an alkaline phosphatase conjugated to rabbit anti-fibronectin monoclonal antibody is added. After incubation, the wells are washed again and the appropriate substrate is added.

The test is read fluoroscopically. If the patient's serum has no antibody, then the maximal value will be recorded. However, if there is specific antibody present, then this will interfere with fibronectin binding to the FN-R polysaccharide and will give a lower optical fluor¬ escent reading. A direct inverse relationship between antibody concentration and fluorescence obtains which can be used to determine antibody titer. It has been found that antibodies to the FN-R displace the fibronectin binding to the well and that fibronectin concentrations of 1 yg are sufficient to saturate all the receptors in the well.

Example 4

Isolation and Purification of S. aureus Fibronectin-Receptor Crude extraction of exopolymers from intact, live strains was performed by nondisruptive sonication as previously described. Supernatants obtained after sonicated bacteria were pelleted and discarded were filter sterilized, dialyzed versus distilled water, and lypholyzed. Purified S. aureus fibronectin-receptor exopolysaccharide from strain 6850 was prepared using lysostaphin treated bacteria, anion exchange chromatography, and gel filtration.

Example 5

Enzyme-linked Immunoadsorbant Assay (ELISA) Fibronectin-receptor exopolysaccharide from strain 6850 was adsorbed onto 96-well polystyrene microtiter plates, washed, and allowed to air-dry. In a series of control plates, fibronectin and the amino- and carboxy-terminal fragments of fibronectin (50 yg per ml in TBS) were adsorbed onto plates overnite, washed, and air-dried. Before each assay plates were blocked with buffer A (50 mM Tris, 150 mM NaCl, 20 mM MgCl₂, 20 yM ZnCl₂/ 3% bovine serum albumin, and 10 yg per ml polyclonal rabbit IgG [Sigma Chemical Co., St. Louis, MO] at pH 7.4) for 30 minutes. All incubations were at room temperature, reagent dilutions done using buffer A, and washes performed using buffer A minus BSA and rabbit IgG. For fibronectin binding and control assays, 100 yl of fibronectin (0.01, 0.1, 1.0, 10, and 100 yg per ml of buffer A), the 180 kilodalton fibronectin fragment (10 yg per ml), and the 27 kilodalton fibronectin fragment (10 yg per ml) were added to wells and incubated 60 minutes. In inhibition assays, wells were preincubated 60 minutes with 100 yl of serial dilutions of serum or fibronectin fragments prior to adding 40 yl of fibronectin (2.0 yg per ml) with the incubation period continued for 60 minutes. After washing, polyclonal rabbit anti-human fibronectin antibody conjugated to alkaline phosphatase (1:2000 dilution) was added, incubated for two hours, and washed off. The substrate, 4- methylumbelliferyl phosphate (0.2 mg per ml in 1 M 2-amino- 2-methyl-l-propanol, pH 9.9, supplemented with 25 mM ZnCl₂ and 1.0 mM MgCl₂), was added to each well, and the plates incubated two hours prior to reading in an automated fluoresence spectrophotometer. All samples were run in triplicate. Averaged readings from control wells without added fibronectin were subtracted from sample wells, with measurements given in standard fluorescence units (SFU) from 10" to 10⁴. Percent inhibition of fibronectin binding to wells was calculated as 100 X [(ave. SFU from FN wells) - (ave. SFU from wells preincubated with serum or the 27 kd FN fragment) ]/(ave. SFU from FN wells).

Monoclonal antibodies raised against capsular serotypes 5 (MAb 831) and 8 (MAb 828) inhibit FN binding to microtiter wells coated with the purified FN-R. High concentrations of rabbit Ig result in only a slight reduction in FN binding, indicating specificity of the inhibition by the MAb's. Type 8 antiserum was more active than type 5 antiserum. To obtain inhibition, relatively high concentrations of antibody protein were needed, suggesting that the common epitopes were only partially cross-reactive. This is consistent with our data concerning the chemical composition of the FN-R. Although the FN-R is a large exopolysac¬ charide, it is clearly distinct from the chemical composi¬ tion of the types 5 and 8 capasular polysaccharides. Hence, the ELISA assay can recognize antibodies that can inhibit FN binding to the FN-R on S. aureus which may prove useful in 3 clinically related areas: (i) The ELISA may be valuable in screening for antibodies that will protect against staphylococcal adhesion to host tissues. (ii) If these antibodies prove to be protective in animal models, then they can be tested in humans to prevent disease (vaccine or passive antibodies are possibilities). (iii)

The ELISA might also be useful in predicting which patients are at higher risk for the development of invasive staphylococcal infections.

Human plasma normally contains between 300 and 600 yg of FN per microtiter. Although clotting consumes a large amount of the FN, enough FN remains in the serum to compete with anti-FN-R antibodies for FN-R sites in the microtiter well. When the highly purified FN is used, 100 ng of FN can be detected by the assay. The sensitivity of the ELISA can be increased by treating the serum with gelatin-Sepharose which specifically removes FN via its interaction thru its gelatin-binding domain but leaves the antibodies.

A positive correlation was observed between the number of binding sites expressed on different strains of S . aureus and the intensity of reaction obtained in the ELISA from the nonprotein antigen harvested from these strains. However, as this material is not highly purified and because quantification would be extremely difficult to maintain which passing through the multiple steps necessary to produce the highly purified and somewhat labile FN-R, a quantitative conclusion should not be drawn from these data. Perhaps some of the genes encoding for capsular exopolysaccharide expression also control FN-R expression. As noted above, there are cross-reacting epitopes suggesting that a portion of FN-R is similar to type 8 capsular material (although clearly chemically distinct). A more important observation is that capsular serotype and FN-R expression can be separated in some strains. This suggests that the FN-R and encapsulation are independent variables in the pathogenesis of S. aureus. For example, the FN-R may be important for adhesion to host tissues whereas capsules may be important for resisting host defenses, a concept supported by the finding that anti-capsular antibodies enhance phagocytosis of S. aureus by human neutrophils. Because FN-R are expressed on a wide variety of staphylo¬ coccal serotypes, anti-FN-R antibodies might be active against a very wide range of S . aureus, not only for anti- adhesive properties, but also as opsonic antibodies.

TABLE 1. Fibronectin bindin to Sta h lococcus aureus

Capsular FN binding SFU for Serotype strain sites per cell* sonicates**

8 ATCC 25923 7500 ±3200 10 Storish 2330

11 191 3230

*strains grown to log phase in chemically defined medium, washed, and lypholyzed to use in ¹²⁵I-FN binding assay (1 x 10⁹ cells/assay) **strains grown to log phase in chemically defined medium and lightly sonicated to shear off exopolymers; lyopholyzed sonicate-supernatants were adsorbed into microtiter wells (50 yg/ml) for ELISA

The described competitive ELISA test and similar diagnostic tests employing the FN-R exopolysaccharide as an antigen, or the monoclonal antibodies to the FN-R can be useful in predicting which patients are at high risk for developing invasive S. aureus infections.

The microorganism, S_. aureus, is a potent infectious pathogen. Interactions between host fibronectin and the S. aureus FN-R play an important role in the initiation of infections. Because antibodies directed against the FN-R polysaccharide block binding of fibronectin to S. aureus, monoclonal or polyclonal antibodies directed against the polysaccharide FN-R can be employed in methods to protect against infection caused by S. aureus. The patients at highest risk for developing a S_. aureus infection can be predicted by use of the previously described diagnostic tests. Therefore, monoclonal antibodies directed against the FN-R may be administered in safe and effective amounts in methods which produce passive immunization in high risk patients. Patients with the following clinical situations would benefit from the passive immunization with anti-FN-R monoclonal antibodies:

1) Patients undergoing prosthetic joint emplacement, especially those receiving artificial hips. 2) Patients receiving heart valves or other implanted devices.

3) Patients on hemodialysis or peritoneal dialysis.

4) Patients with recurrent S. aureus infections, especially patients that have diabetes mellitus, rheumatoid arthritis, and antibody deficiency syn¬ dromes.

5) Patients with major burns.

6) Immunosuppressed patients, especially those patients who are likely to have prolonged periods of neutropenia. 7) Patients who will have indwelling intravascular catheters, e.g., patients receiving hyperalimentation or chemotherapeutic agents.

8) Patients who have major orthopedic procedures, e.g., open reduction of fractures.

The availability of a diagnostic test using FN-R polysaccharide as an antigen, such as described in the ELISA test, to measure the pre-existing levels of antibody direct^ ed against the FN-R polysaccharide can help determine which patient will benefit from this passive immunization. How¬ ever, the monoclonal antibody also can be of value in the treatment of S. aureus infections. Because the FN-R poly¬ saccharide is expressed in the largest numbers on invasive S . aureus and because the FN-R polysaccharide is on the surface of the bacteria, the monoclonal antibody directed against this immunogen also can act as an opsonin as well as a blocking antibody. Hence, it is believed that passive immunization with this antibody decreases the severity of ongoing S . aureus infections, as well as, limits the ability of S. aureus to spread to new sites in the host. Thus, another use for the anti-FN-R monoclonal antibody is in a method of treating patients who already have S . aureus infections, especially those with bacteremia or those that are likely to develop bacteremia. As previously mentioned, I have discovered that, in addition to the monoclonal antibodies to FN-R polysac¬ charide, the monoclonal antibodies to type 8 capsular polysaccharide of S . aureus can be used in the above described methods and diagnostic tests. These monoclonal antibodies are known and described in the publications of

Nelles, et al, Infection and Immunity, Vol. 49, No. 1, July 1985, pp. 14-18 and Hochkeppel et al, Journal of Clinical Microbiology, Vol. 25, No. 3, March 1987, pp. 526-530. However, the uses of the monoclonal antibodies in the above described methods and diagnostic tests are new. The monoclonal antibodies are administered to animals to block or treat staphyloccecal infections by injecting sterile 03572 PCT/US90/05065

-16-

pharmaceutical preparations containing the monoclonal antibodies directly into the blood stream of the animals. The amount to be injected is an amount which is safe and effective as determined by the size and condition of the animal and the severity of its infection, if any.

It will be apparent to those skilled in the art that a number of modifications and changes may be made without departing from the spirit and scope of the invention. Therefore, it is intended that the invention not be limited except by the claims.

Claims

1. A method for the detection of fibronectin-receptor antibodies in a test sample comprising:

(a) coating microtiter plates with purified fibronectin-receptor exopolysaccharide antigen to form antigen coated wells;

(b) mixing the test sample with a labeled reagent and incubating a portion of said mixture with the antigen-coated wells;

(c) separating the bound labeled reagent mixture from the free labeled reagent mixture; and

(d) detecting the amount of bound labeled reagent which measures the amount of fibronectin-receptor antibodies in the test sample.

2. A diagnostic test for testing patients for fibronectin-receptor antibodies which employs as the antigen fibronectin-receptor exopolysaccharide obtained from Staphylococcus aureus.

3. A test kit for the detection of fibronectin-receptor antibodies comprising in combination:

(a) an isoluble surface or support containing wells;

(b) fibronectin-receptor bound to said wells; and

(c) means for detecting the amount of antibodies in a test sample which bind to the fibronectin-receptor.