AU663481B2 - Defined coating with recombinant fusion proteins composed of constant fusion partner and variable antigen portion in diagnostic test systems - Google Patents

Abstract

The invention relates to solid support materials, e.g. microtitre plates, beads or microbeads, magnetic particles, films, filter papers and the like, which are coated with antibodies against recombinant fusion proteins, and to processes for the reproducible and defined coating of these surfaces with such recombinant fusion proteins. The recombinant fusion proteins contain a constant fusion partner and a variable antigen portion which constitutes the specificity of the test system. The abovementioned antibodies are directed against the constant fusion partner, that is to say the part of each fusion protein which is intrinsically not of analytical interest and pure.

Description

P/00/Ohi 28IS/9l Regulation 3.2(2)

AUSTRALIA

Patents Act 1990

ORIGINAL

COMPLETE SPECIFICATION STANDARD PATENT 0. Ir Application Number: Lodged: Invention Title: DEFINED COATING WITH RECCMBINANT FUSION PROTEINS CCM4POSED OF CONSTANT FUSION PARTNER AND VARIABLE ANTIGEN PORTION IN DIAGNOSTIC TEST SYSTEMS.

~0 The following statement Is a full desriptlcei of this Invention, Including the best method of performing it known to us -7 BEHRINGWERKE AKTIENGESELLSCHAFT 91/B 003 Ma 804 Dr.BO /Wr Defined coating with recombinant fusion proteins composed of constant fusion partner and variable antigen portion in diagnostic test systems The invention relates to solid support materials, for example, microtiter plates, beads or microbeads, magnetic particles, films, filter papers and the like, which are coated with antibodies against recombinant fusion proteins, and to processes for the reproducible and defined coating of these surfaces with such recombinant fusion proteins. The recombinant fusion proteins contain a 0 0o constant fusion partner and a variable antigen part which 0° o..o 15 constitutes the specificity of the assay system. The 1 oo::o 0abovementioned antibodies are directed against the constant fusion partner, that is to say the part of each fusion protein which is intrinsically not of analytical interest and uniform.

20 When coating surfaces with antigen, for example proteins which are employed in diagnostic test systems, for example ELISA, latex agglutination tests etc, the chemical affinity of the antigen is crucial for the homogeneous coating of a surface, especially when the antigens are in purified form. The homogeneity of the coating depends, inter alia, on the chemical properties of the antigen, such as hydrophobic't.y or polarity, which in turn depend on physicochemical parameters such as pH, ionic strength, molecular weight, partition coefficients or solubility behavior.

If, in addition, further components are also involved in a coating system, the complexity of the reaction mechanism increases owing to competition for the available binding sites. The competition in turn depends on the various chemical properties of the coating antigens and L Ji 2 can in part be controlled by the choice of the coating concentrations, the coating conditions such as pH and ionic concentration or composition of the coating buffer, or else by sequential coating. However, there is always expected to be a considerable loss of individual components of the system during this, because on simultaneous coating of weakly and strongly reacting components, the competition can be s ted in favor of the weaker reacting only by increasing the concentration thereof. Since the different partition coefficients of the proteins determine their binding properties, it is also possible to describe their mixtures by Nernst's partition law.

To date, the problem of competition between several 15 recombinant antigens during the coating of surfaces employed in diagnostic detection methods has been partly avoided by a temporally and/or spatially separate coatingO sequence with the individual recombinant proteins. Thus, for example, in the coating of beads with recombinant antigens first hemisphere A has been coated with antigen 1 and, in a subsequent step, hemisphere B has been coated with antigen 2. This principle entails various technical problems and, in the case of a multifactorial component system, can be employed only with great difficulty.

25 In the coating of microtiter plates for ELISA, sequential coating of the cavities with the individual proteins is conceivable, in such a way that first antigen 1 is S. inserted to fill a volume of, for example, 50 pl, and then antigen 2 to fill a volume of 75 pi etc.

This process also has the intrinsic disadvantage of timeconsuming sequential coating of the cavities, and the elaborate washing and coating steps are particularly susceptible to error. In addition, there is no certainty in this method that the components actually are bound in the required concentration to the solid phase.

-I W 1PW MM F

L

3 3 The invention has the object of providing solid support materials which are coated with monoclonal or polyclonal trapping antibodies which have at least one specificity and are directed against one fusion partner in a recombinant fusion protein. This support material makes it possible to immobilize recombinant proteins via their fusion partner in defined alignment on the surface, the extent to which the recombinant protein samples are also contaminated with extraneous components being immaterial.

The invention has the further object of providing solid supports which are additionally coated as claimed in claim 5 and 6 with a recombinant antigen or a mixture of recombinant proteins or protein fragments which, for example, are of importance for the diagnosis of infectious diseases, where the recombinant antigen is in each case coupled to an at least substantially constant fusion partner. The standardization by use of an antibody which i is monospecific for the fusion partner results in the following advantages: 1. There is a common solid phase for different specificities.

2. Different recombinant antigens can be applied in any desired mixing ratios so that it is also possible to adjust stoichiometries corresponding to the conditions present in vivo.

3. The coating with the recombinant fusion protein does i not depend on its _'?ysicochemical properties but depends solely on the antibody-antigen reaction.

S4. Mixture of recombinant proteins with different fusion partners is also possible as long as the specific antibodies are applied to the support.

The homogeneity of the coating can be checked by suitable detection methods, for example by antibodies which are directed against the coating antibodies and therefore react with the solid support. This check makes it possible to ensure that the coating is reproducible and of constant quality.

6. The recombinant fusion proteins employed as antigens I I. i4 or containing the antigens do not necessarily have to be presented in highly pure form, so that elaborate purification methods can be dispensed with.

Accordingly, the invention provides a solid support material coated with a monoclonal or polyclonal trapping antibody and at least two different fusion proteins, each of which comprises a constant fusion partner and a variable antigen part, characterized in that the trapping antibody is directed against the constant fusion partner and the variable antigen parts are derived from at least two different antigens of diagnostic interest.

Preferably the solid support materials are the cavities of microtiter plates, beads, microbeads, filters, films or magnetic particles. In the case of the solid support materials being the cavities of microtiter plates, the cavities may be coated with at least 0.05 0.5 gg of a poly- or monoclonal antibody per cavity. In the case that the solid support material is magnetic particles, they may be 15 coated with at least 0.005 gig to 0.5 gig of a monoclonal or polyclonal antibody per cm 2 This invention also relates to the use of the solid support material for S detecting specific antibodies which are directed against a recombinant fusion protein or a mixture of recombinant fusion proteins. Preferably the analyte employed is whole blood, plasma, serum, cerebrospinal fluid, sputum or urine.

4-4C In another aspect of the invention there is provided a process for coating a solid support material with at least two different antigens of diagnostic interest, f wherein a) a monoclonal or polyclona trapping antibody is adsorbed or 25 covalently bound to the solid support material, and b) at least two different fusion proteins are bound to the trapping antibody through an antigen-antibody binding, characterized in that the trapping antibody binds to the constant part of the fusion protein and a substance of diagnostic interest can be bound to the variable antigen part of the fusion protein.

LU

4 I; B- .z 4a Suitable fusion proteins are proteins which are composed of a common segment, the fusion partner and of a protein or protein fragment fused thereto by recombinant DNA techniques, it being immaterial whether the fusion is N- or Cterminal. If the number of proteins or protein fragments to be fused on is large, it may be expedient to select more than a single fusion partner, but the fusion partner is usually constant. When there are several different fusion partners, the solid phase must be coated with antibodies of the appropriate specificities. The said proteins or protein fragments are proteins of diagnostic or therapeutic relevance, which in the former case are used for detecting diseases caused by viruses, bacteria or parasites, and in the second case can be used, for example, for establishing the dose and monitoring humorally administered therapeutics.

In the case of infectious diseases, it is possible for individual proteins to lead to <tcr a reliable diagnosis of the infection (for example hepatitis B-HBsAg; HIV 1 p 24; cytomegalovirus pp 150 protein; Treponema pallidum TMPA protein; Toxoplasma gondii p 61 protein). The origin of the fusion partners of these proteins is immaterial in this connection, so that it is possible to use MS2 polymerase fusions as well as trp fusions. The preferred fusion partner is 3galactosidase from E. coli, in particular a segment of this protein with amino acids 1 to 375. Preferred supports are microtiter plates and magnetic p, iticles, S 20 although it is also possible to use films, beads, microbeads and the like. The advantage of these test systems is the possibility of dealing rapidly with the samples employed. The preferred trapping antibody used is a monoclonal antibody against the N-terminal E R' A~ v 5 segment of P-galactosidase (amino acids 1 to 375), although it is also possible to use polyclonal antibodies for this, albeit with less success, as long as they have a specific reactivity with the fusion partner.

Example 1. Preparation of the recombinant fusion proteins The fusion proteins listed herein as examples were expressed in the pSEM plasmid vector system (Fig. 1; S. Knapp et al., BioTechniques 8 280 (1990)) which contains the N-terminal segment of amino acid 1 to 375 of p-galactosidase. A polylinker DNA sequence into which it is possible to insert gene products of diagnostic relevance for the DNA fragments of interest has been Sligated to the DNA sequence coding therefor. Fig. 2 depicts examples of recombinant fusion proteins which contain defined DNA segments from the genome of human '0*4 cytomegalovirus and are expressed in E. coli.

2. Preparation of the monoclonal antibodies against .56-galactosidase from E. coli tj To prepare monoclonal antibodies against intact p-galac- Stosidase from E. coli, Balb/c mice six to eight weeks old were immunized. This entailed each mouse receiving about ug of the enzyme emulsified in complete Freund's t adjuvant injected subcutaneously and in a second case 25 intraperitoneally. A second and third immunization was carried out four to eight weeks later in each case.

Immediately before the actual fusion, the experimental animals additionally received intravenous boosters on four consecutive days. On the day of the fusion, the spleens were removed under sterile conditions and suspended to give single cells. Fusion of 108 spleen cells with 2 x 10 7 cells of the myeloma cell line SP 2/0 or cell line X 63 Ag 8653 (Journal of Immunol 173 (1979), 1548 to 1550) generated hybrid cells which were subsequently ,i 6inoculated in a selection medium (Dulbecco's minimal essential medium, DMEM supplemented with 20 fetal calf serum FCS, 0.1 mmol/l hypoxanthine, 0.4 mmol/l aminopterin and 16 mmol/l thymidine) on culture plates with 24 cavities (supplied by Costar) in a concentration of 6 cells/cavity. Two to three weeks later individual cell colonies were isolated from the cavities and each transferred into one cavity of a new culture plate. After a further two to three days, the culture supernatants were screened using an enzyme immunoassay for the presence of p-galactosidase-specific antibodies. The supernatants were incubated on p-galactosidase-coated microtiter plates and any specific antibodies present were detected by an anti-mouse/peroxidase immunoglobulin conjugate reaction. Positive cell lines were propagated S'.as ascites in Balb/c mice pretreated with IFA (incomplete 'Freund's adjuvant). Purification was carried out by °ammonium sulfate precipitation and protein A chromatography. The purity was checked by HPLC and gel electro- 20 phoresis, and the immunogloblin classes were determined by Ouchterlony immunodiffusion.

3. Characterization of the monoclonal antibodies Cell extracts of the transformants and extracts of the plasmid-free, p-galactosidase-deficient E. coli strain 25 MC4100 Casadaban, J. Mol. Biol. 104, p. 541-555 (1976)) and a protein mix for molecular weight determination were fractionated under denaturing conditions in SDS polyacrylamide gels and subsequently transferred to nitrocellulose filter paper. The individual blots were incubated and developed with the various monoclonal antibodies. The monoclonal antibody 87-55/60, for example, recognized in this system only polypeptides which contain sequences from the region 1 375 of p-galactosidase but no polypeptides from the region 376 to 1021 of p-galactosidase nor any polypeptides specified by the E. coli strain MC4100. It was thus possible, oni the basis of the reaction pattern with defined -7i -1 7 p-galactosidase fragments, to demonstrate that the monoclonal antibody 87-55/60 reacts with an antigenic determinant in the amino-terminal region of p-galactosid-se from serine 1 to aspartic acid 375.

4. Preparation of a mAb-coated microtiter plate A suitable monoclonal antibody which is directed against the N-terminal region of p-galactosidase is adjusted to a concentration of 0.1 pg of protein per ml of solution with physiological phosphate buffer (PBS), and a volume of 100 ul is applied to each cavity of a microtiter plate. This plate is incubated at +4°C for 16 hours for the adsorption, and then the coating solution is removed by aspiration. The surface is subsequently washed with a 50 mM tris/60 mM EDTA buffer (pH 7.2) and subsequently 15 incubated with a solution of protective protein, for j example lactoferrin, human serum albumin, bovine sermu albumin, or else detergents, for example Tween, cholic o o""acid and derivatives thereof, amine oxide and the like for 45 minutes to 2 hours in order to block non-saturated binding sites. The surface is subsequently washed twice with PBS and dried, and can be used immediately in this form or else stored, j Coating of microtiter plates with recombinant fusion protein containing an antigen 25 As in the suitable monoclonal mouse antibody which is directed against the N-terminal region of 6-galactosidase |is diluted to a concentration of 0.1 pg of protein per cavity in physiological phosphate buffer (PBS) and applied to the ELISA plate. After adsorption of the antibody, the coating solution is removed, and the surface is washed and incubated with a 0.1 to 5 strength solution of protective protein, fer example lactoferrin, human serum albumin, bovine serum albumin or else detergents, for example Tween, deoxycholate, amine oxide etc, for 45 minutes to 2 hours in order to block I Il-- 8 non-saturated binding sites. The surface is subsequently washed twice with PBS, and the fusion protein or mixture is applied in suitable dilution. The reaction of the fusion protein with the monoclonal antibody is complete after incubation for 2 hours. The nature of the antigen is immaterial in this connection as long as the epitope part, which is relevant for the monoclonal antibody, of the N-terminal p-galactosidase fragment is common to the fusion proteins employed. Differences in the physicochemical properties of the proteins employed, for example their charge, hydrophobicity etc, are unimportant in the case of mixtures of recombinant proteins because the reactive component is contributed not by the foreign portion of the hybrid protein but by the common p-galactosidase component. This results in a uniform binding S. behavior which has unambiguously fixed binding kinetics and which is determined solely by the p-galactosidase •i part of the hybrid proteins. Since the monoclonal antibody has equal affinity for the individual proteins, the 20 binding of the antigen to the antibody is to be regarded as a process determined purely by statistics. In contrast to coating with heterologous components such as, for example, synthetic peptides from different regions of a protein or homogenates from cells, there is no risk in this case of individual components being overrepresented because of their exceptionally favorable binding properties, and other essential components having been displaced by competition. Tab. 1 demonstrates on the basis of 37 sera (9 CMV negative, 28 CMV positive) the funda- S 30 mental possibility of implementing the assay design.

6. Defined coating of microtiter plates with a mixture Sof two antigens A microtiter plate is coated in a volume of 150 pl containing 0.1 pg of monoclonal anti-p-galactosidase antibody by adsorption overnight and, after washing twice with 50 mM tris-HCl aid 50 mM EDTA (pH coated in each case with pXP1 protein (Fig. 2 and Fig. 3) as first 9 recombinant fusion protein and pSEML3 protein (Fig. 2) as second recombinant protein. The coating is carried out in such a way that a total of 50 ng of a fusion protein mixture are employed, with the percentage content of the pXP1 protein decreasing from 100 percent in steps to 0 percent while the content of the pSEML3 protein in the mixture increased in the same way. After incubation at 37 0 C for one hour, the cavities are emptied, rinsed twice with the above washing solution and incubated with two human sera with preference for, in .ch case, one of the abovementioned recombinant proteins in a dilution of 1:50 for one hour. The serum sample is subsequently removed from the cavity again and, after washing twice with the abovementioned buffer, a suitable anti-human inmmunoglobulin G-peroxidase conjugate is *added.

a The immune reaction is subsequently detected by photometry using a suitable substrate, for example tetramethylbenzidine. The specific signal which is recorded owing to the seroreactivity decreases to the same extent as the content of one recombinant protein in the mixture j W decreases. The other serum behaves in a correspondingly a contrary way. It is possible in this way at the intercept of the two titration plots

S

1 25 a) to determine the optimal coating ratio of the fusion proteins, b) to understand the actual coating behavior of antigen mixtures on microtiter plates, c) and to ensure reproducible coating quality when the sera employed are standardized.

7. Checks on the coating of ELISA microtiter plates When ELISA plates are coated with recombinant fusion proteins, the coating concentration is adjusted via the protein concentration. It is possible with a suitable murine monoclonal antibody against p-galactosidase to determine the actual antigen concentration of the coating

I

1~

I

I

A

f 10 *4 00 o too.

at,, eta.

*6 S a, -g S It to I 00 1 It I 00

'III

(lIt with the aid of a calibration plot as reference and thus to optimize the coating concentration. It is furthermore possible to detect signal losses or similar losses in quality of stored plates as part of quality control measures and to reject unusable production batches. This means that it is likewise possible to carry out checks for long-term quality monitoring of delivered batches of microtiter plates because of to the standardization of the test methods. This procedure is described hereinafter taking the example of the recombinant CMV protein pXPl: a microtiter plate is coated with a measurement solution of the recombinant protein whose exact concentration is known in defined dilution steps in analogy to the process used in production. The dilution steps are adjusted so that they are at least one order of magnitude below and above the concentration employed in production.

The values found in the subsequent EI 0 ISA procedure are related a blank which is carried out at the same time (corresponds to a coating procedure with the coating 20 buffer without the test antigen); it is suitable to carry out multiple determinations for statistical validation.

The extinctions found in the ELISA evaluation correspond to the course of a dilution calibration plot which can now be used as concentration reference system for unknown samples. A microtiter plate to be checked can be examined for a correct coating concentration by interpolation of the values found after development in the ELISA. It is furthermore possible for the uniformity of the coating process, the quality of the reagents employed on the basis of standard solutions and their stability and storability to be examined.

11 Table 1 Evaluation Serum No. CMV pXPl 56642 28 56646 29 56941 30 282904 34 107 56668 38 61 286132 41 31 5.947 48 12 56949 53 23 81776 108 103 285544 279 619 310990 577 557 240516 752 2016 282852 817 1225 286574 942 1476 2B6139 1001 909 311041 1086 2500 282900 1099 2008 2 405 11 1123 2500 e' '56640 1378 2500 310955 1382 2136 56643 1386 1173 tt.1286602 1394 1079 81841 1405 566 270091 1467 1173 285557 1474 2393 270114 1488 587 56645 1541 1628 81873 1575 2500 C286174 1580 1991 270099 1595 1468 56961 1730 1536 311031 1779 2004 56963 1956 1336 56641 2148 1650 t283754 2192 1740 600164 2234 2048 600069 2500 .2500 12 Legend to Fig. 1 Expression vector for fusion proteins from Escherichia coli The E. coli plasmid vectors pSEM-1, pSEM-2 and pSEM-3 are used for the expression of protein domains in bacteria.

These vectors contain a rudimentary form of the E. coli lacZ gene (lacZ') which encodes only the N-terminal amino acids 1 to 375 of p-galactosidase (p-Gal). A polylinker region which is attached thereto and has several cloning sites in the three possible reading frames makes it possibl to construct fusion proteins based on lacZ'. The transcription of the fusion gene is controlled by derepression of the lac promoter (plac) and stopped by JI the rrnB transcription terminators T1 and T2 proximal to the polylinker region. In addition, translation stops have been inserted in all three reading frames immedi- I, ately behind the polylinker region Knapp, M. Broker, E. Amann, Biotechniques 8, 280-281 [1990]).

Legend to Fig. 2 20 Diagrammatic structure of lacZ' fusion proteins All the said fusion proteins contain amino acids 1 to 375 S'r, *of p-galactosidase (lacZ') and, attached thereto, additional amino acids which are encoded by the polylinker region, and optionally the relevant cloned protein 25 domains. Mentioned here as example are constructs of human cytomegalovirus with foreign portions from different gene segments: 1. pSEM-1/2/3 basic construct without foreign portion; 2. pXP1 foreign portion nucleotide sequence 1632 2118 of the HCMV gene ppl50, corresponds to 162 amino acids; 3. pSEML3 foreign portion nucleotide sequence 2686 2946 of the HCMV gene ppl50, corresponds

I:

13 4. pSEMLlV2 5. pHE68-1 6. pPS28 to 87 amino acids; foreign portion nucleotide sequence 873 1514 of the HCMV gene ppl50, corresponds to 213 amino acids; foreign portion nucleotide sequence 548 1136 of the HCMV gene pp65, corresponds to 196 amino acids; foreign portion nucleotide sequence 40 540 of the HCMV gene pp28, corresponds to 167 amino acids.

Legend to Fig.

P 9 It I II 400 ft fl 15 aoo I i i t t lI t Diagrammatic structure of the pXP1 protein The recombinant pXP1 protein contains, besides amino acids 1 to 374 of the rudimentary fi-galactosidase (lacZ'; E. coli) and the amino acids encoded by a polylinker region, a protein domain from the ppl50 protein of human cytomegalovirus, whose DNA fragment (base pairs 1632- 2118; 162 amino acid residues) has been cloned as C-terminal gene fusion to the vector fragment.

I-I I I C IltI alt, I I

'LI'

It C It a it in i

Claims (7)

1. A solid support material coated with a monoclonal or polycclonal trapping antibody and at least two different fusion proteins, each of which comprises a constant fusion partner and a variable antigen part, characterized in that the trapping antibody is directed against the constant fusion partner and the variable antigen parts are derived from at least two different antigens of diagnostic interest.

2. A solid support material as claimed in claim 1, wherein the solid support materials are the cavities of microtiter plates, beads, microbeads, filters, films or magnetic particles.

3. A solid support material as claimed in claim 2, wherein said support material is a microtiter plate whcse cavities are coated with at least 0.05 0.5 ltg of a poly- or monoclonal antibody per cavity.

4. A solid support material as claimed in claim 2, wherein said support material means magnetic particles which are coated with at least 0.005 ltg to [g of a monoclonal or polyclonal antibody per cm2. The use of solid support material as claimed in any of claims 1 to 4 for detecting specific antibodies which are directed against i recombinant fusion protein or a mixture of recombinant fusion proteins.

6. The use as claimed in claim 5, wherein the analyte employed is whole blood, plasma, serum, cerebrospinal fluid, sputum or urine.

7. A process for coating a solid support material with at least two different antigens of diagnostic interest, wherein a) a monoclonal or polyclonal trapping antibody is adsorbed or ~-~q~2~covalently bound to the solid support material, and I( 1 te-e i (;1TIIF I in this method that the components acuaL.y a=j- u.u. the required concentration to the solid phase. I I I i r- I i i I~ L i IliL I I 1 aE :I ii /i b) at least two different fusion proteins are bound to the trapping antibody through an antigen-antibody binding, characterized in that the trapping antibody binds to the constant part of the fusion protein and a substance of diagnostic interest can be bound to the variable antigen part of the fusion protein. DATED this 17th day of July, 1995. BEHRINGWERKE AKTIENGESELLSCHAFT i I I *1 it II( 4 a WATERMARK PATENT TRADEMARK ATTORNEYS 290 BURWOOD ROAD HAWTHORN VICTORIA 3122 AUSTRALIA (DOC 04 AU1107892.WPC DBM/KJS:KP) (I

411. LI