Title: Method for detecting a pathological defect in a cell, an antibody and an assay kit.
The present invention relates to a method for detecting a pathological defect in a cell of a particular type, wherein the cell is contacted with an antibody or a functional part or derivative thereof.
For detecting pathological defects in cells, frequently use is made of antibodies that specifically recognize a determinant which in normal cells of that type is not present, or is present in a significantly lesser amount or, conversely, in a larger amount. For leukemia, for this purpose, often use is made of antigenic determinants on the cell surface (outer membrane proteins, including antigenic determinants of modified outer membrane proteins). It is also known to prepare a cell in order to make parts present within the cell membrane accessible to an antibody, and to contact the thus prepared cell with an antibody in order to detect pathological defects in a cell. Detection of a pathological defect can serve not only for making a diagnosis, but also for monitoring a therapeutic treatment. There exists a continuous need for detection methods with which pathological defects in a cell can be detected, preferably in an early stage, with a high specificity and with a high sensitivity.
The object of the present invention is to achieve one or more of the above-mentioned objectives. To that end, the present invention provides a method according to the preamble, which is characterized in that, said antibody or functional part or derivative thereof is specific for i) a protein of a Polycomb group complex; ii) a corepressor or a coactivator of a Polycomb group complex; or iii) a Boundary element-binding protein, the method further comprising detecting the presence of said antibody or functional part or derivative thereof.
In a preferred embodiment said method further comprises determining the extent of binding of the antibody or functional part or
derivative thereof and comparing said extent of binding with a reference value determined for a normal cell of that type. Thus in this preferred embodiment the invention provides a method according to the preamble, which is characterized in that, said antibody or functional part or derivative thereof is specific for i) a protein of a Polycomb group complex; ii) a corepressor or a coactivator of a Polycomb group complex; or iii) a Boundary element-binding protein, the method further comprising detecting the presence of said antibody or functional part or derivative thereof and comparing the extent of binding of the antibody or functional part or derivative thereof with a reference value determined for a normal cell of that type.
It has been found that in proteins as defined here a deviant concentration thereof is a very good indicator for detecting a pathological defect. In the present application, an antibody is understood to mean any protein selectively binding the defined protein, such as a polyclonal or monoclonal antibody or a selectively binding fragment thereof obtained by proteolysis, a peptide or peptoid coding for the binding part of an antibody, a specifically recognizing peptide expressed through phage display technique, and so forth. A functional part of an antibody is a part comprising at least the same selective binding characteristics in kind not necessarily in amount. A functional derivative of an antibody comprises amino-acid substitutions, deletions and/or insertions or is otherwise altered without altering the antigen binding specificity in kind, not necessarily in amount. It is of course entirely possible to use an antibody, for instance a bi-or more specific antibody with a specificity for more than one of the referred to protein(s) of a Polycomb group complex, corepressor(s), coactivator(s) of a Polycomb group complex or a boundary element-binding protein(s). An antibody or functional part or derivative thereof as used in the invention does not have to be raised against the protein it is intended to detect in a method of the
invention. It is entirely possible that the antibody or functional part or derivative thereof, is raised against for instance a consensus Polycomb protein. This antibody can still be capable of specifically binding to a different Polycomb protein in a method of the invention. When in the present application reference is made to a cell, this is understood to be a cell of a mammal, in particular a human, whilst the cell being examined for a defect and the normal cell are both of the same cell type.
When the present application refers to a reference value of a normal cell, this is understood to mean a reference value of a cell which is in the same, or a corresponding, differentiation stage as, or has been stimulated correspondingly to, the cell to be examined for a pathological defect, which, as set out above, is of the same cell type. This is to say that, depending on the differentiation stage or a stimulation, normal cells can exhibit a varying reference value. The reference value is, for instance, a value of a parameter proportional to the concentration (including the concentration itself) of a protein as defined, such as fluorescence intensity. The reference value can also be a ratio of values for the extent of binding (i.e., the concentration). It will be self-evident that upon selection of the protein to be detected (or the proteins to be detected), it is advantageous to choose those reference values that are independent of the differentiation stage, or are not attained in the stimulation of a normal cell. In demonstrating a defective cell among a large population of normal cells, those normal cells can exhibit a varying concentration of a defined protein, e.g. ring 1, depending, for instance, on the age of the cell, stimulation if any, nutrients, previous contact with other cells which involved exchange of substances, etc. Then it may well be that a malignant cell contains an increased concentration of a protein as defined herein in comparison with cells that have the same history, the same conditions, etc., while that concentration is still below the
concentration of a non-malignant cell with a different history (and hence is below that reference value). By normalizing with another defined protein, a reference value can be found which varies in a manner not dependent, or less dependent, on the circumstances, so that in case of a pathological defect, the measured value is significant.
In the present application, Boundary elements are understood to be DNA protein complexes which define the limits of chromatin domains and which are capable of preventing the repression of gene expression by chromatin-associated repressors. According to an important embodiment, for detecting a pathological defect, the cell is a cancer cell.
There exist a great many types of cancer, each in turn subdivided into different subtypes. For making a diagnosis, determining the prognosis, and in particular also determining and monitoring a therapeutic treatment, it is important to know which variant one is dealing with and what, for a particular variant, the value measured in an individual is in comparison with the reference value. Manufacturing antibodies against the various (sub)types of cancer is a very laborious job. By contrast, the method according to the invention leaves open the possibility that for very widely diverging types of cancer and/or variants, one or just a few antibodies against as many of the defined proteins suffices or suffice.
A method of the invention can of course be combined with one or more other methods for detecting a pathological defect in a cell. The method may for instance be combined with a method to determine the type of normal cell. This can be achieved by scrutinizing morphological characteristics through for instance microscopic analysis, alternatively, antibodies may be used that can identify a normal cell of a certain type. A person skilled in the art will be able to arrive at other suitable possibilities.
According to a first embodiment, the protein of a Polycomb complex comprises HPH, BMI1, RING 1, HPC, EZH, EED, or YY1, or parts or derivatives thereof.
According to an alternative embodiment, the corepressor comprises CtBP or HDAC
According to a preferred embodiment, a second antibody or functional part or derivative is used, which is specific for a second protein comprising of l) a protein of a Polycomb group complex; π) a co-repressor or co-activator of a Polycomb group complex; or m) a Boundary element- binding protein, the ratio of the extent of binding of the antibody, or functional part or derivative thereof, and the second antibody, or functional part or derivative thereof, is determined and the thus obtained ratio value is compared with a reference value determined for the normal cell This method can be used for increasing the accuracy of the detection Thus, for instance, for a pathological defect which involves a slight decrease of one protein as defined herein and a slight increase of another protein as defined herein, a more reliable detection can be achieved with fewer false positive cells when the ratio of the measured values is used as a reference value. This embodiment is also suitable in those cases where, for the normal cell type m question, the concentration varies strongly depending on the differentiation stage and/or a stimulation, and the detection of the extent of binding of the second protein is used for performing a normalization.
With advantage, the determination of the extent of binding is performed by measuring fluorescence or light absorption The latter can be done, for instance, by means of a histochemical staining, a technique well known in the art.
The method according to the invention is suitable for detecting a pathologically defective cell in a biopsy Advantageously, the cell is a cell present in a smear, such as a cervical cell. According to an alternative favorable embodiment, the cell is a cell present in a cytospm preparation. A
cytospin preparation is typically produced by spinning suspended cells (usually blood cells) onto a solid surface, for microscopic analysis. In both favorable embodiments, obtaining material can be carried out quickly and in a manner little burdening the patient. In all cases, it holds that it is very simple to obtain the cell, and preparative techniques for contacting with an antibody can be performed rapidly and thus enable large-scale screening. The use of a technique based on cytospin further has the advantage that any detached cancer cells that have entered the blood stream can be detected by virtue of the highly sensitive technique according to the invention, even if they are present in very small numbers (in comparison with blood cells).
The invention also relates to an antibody directed against a Boundary element-binding protein of a mammal. Finally, the invention relates to an assay kit for detecting a pathological defect in a cell, which kit comprises an antibody or a functional part or derivative thereof directed against, i) a protein of a Polycomb group complex; ii) a corepressor or a coactivator of a Polycomb group complex; or iii) a Boundary element-binding protein, and the extent of binding of the antibody or functional part or derivative is compared with a reference value determined for a normal cell of that type.
The present invention is illustrated by the following non-limiting examples. In the examples two different types of cancer cells are characterized utilizing deregulated expression of Polycomb group proteins. These cancer cells are subsequently diagnosed by specific antibodies against these Polycomb group proteins.
Examples
Example 1 Lymphnodes from healthy individuals or from patients with
Hodgkin's Disease were obtained by surgery. Biopsies were obtained from patients at the time of diagnosis and were immediately frozen or fixed in 10% formalin and embedded in paraffin. 3μ sections were cut, deparaffinized, and endogenous peroxidase was inhibited by incubation of the tissue sections for 30 minutes at room temperature with 0.3% H2O2 in methanol. Antigens were retrieved by boiling for 10 minutes in citrate buffer (pH=6), followed by successive rinses in PBS containing 0.5% Triton and PBS only . Slides were then incubated for ten minutes in 0.1 M glycine (diluted in PBS), and rinsed in PBS only. Before application of the primary antiserum or antibody, sections were incubated for 10 minutes in normal swine serum (diluted 1:10 in PBS + 1% BSA) or normal rabbit serum (diluted 1:50 in PBS + 1% BSA). Primary antibodies against the Polycomb group proteins BMI1 and EZH2 were applied following antigen retrieval and pre -incubation with goat serum. BMI1, RING1, EZH2 and EED Polycomb group proteins were detected with the 6C9 mouse monoclonal, K320, K358 and K365 rabbit polyclonal antisera, respectively (Sewalt R.G.A.B., J. van der Nlag, M.J. Gunster, K.M. Hamer, J.L. den Blaauwen, D.P.E. Satijn, T. Hendrix, R. van Driel and A.P. Otte. 1998. Characterization of interactions between the mammalian Polycomb -group proteins Enxl/EZH2 end EED suggests the existence of different mammalian Polycomb -group protein complexes. Mol. Cell. Biol. 18, 3586- 3595). Secondary antisera were biotinylated goat-anti-mouse or biotinylated swine-anti-rabbit. Immunostaining was performed with 3-amino-9- ethylcarbazole (AEC) using the streptavidin-biotin complex/horse radish peroxidase (sABC-HRP) method and tyramine intensification. Sections were
counterstained with hematoxylin. When employing double immunofluorescence labelling, BMIl was detected by anti-mouse antiserum and streptavidin-cy3 (Jackson, USA) and EZH2 was detected by swine-anti rabbit Ig-FITC (Dako, Denmark).
RESULTS
The human germinal center (GC) architecture of lymph nodes reflects different B-cell differentiation stages. In resting B-cells in the mantle zone (M, Figure 1) BMIl and RINGl expression is high and EZH2 and EED expression is absent. During the transition of resting mantle B-cells to rapidly dividing follicular centroblasts in the dark zone (DZ, Figure 1), BMIl and RINGl expression becomes low and EZH2 and EED expression becomes high. Thus, the extent of binding of the BMIl and RINGl antibodies is highest in the mantle zone (M) and lowest in the dark zone (DZ). The extent of binding of the EZH2 and EED antibodies is lowest in the mantle zone (M) and highest in the dark zone (DZ). Therefore, the reference value for BMIl and EZH2 in the normal mantle zone of healthy persons is much higher (higher than 10) and the reference value for BMIl and EZH2 in the normal dark zone of healthy persons is much lower (lower than 0.1). (Raaphorst, F.M., van Kemenade, F.J., Fieret, E., Hamer, CM., Satijn, D.P.E., Otte, A.P., & Meijer, C.J.L.M. 2000. Polycomb gene expression patterns reflect distinct B cell differentiation stages in human germinal centers. The Journal of Immunology, 164, 1-4).
Reed-Sternberg tumor cells of Hodgkin's disease are descendants of germinal center B-cells. In non-malignant cells (Figure 2, indicated with stars) the extent of binding of BMIl is high and the extent of binding of EZH2 is low. As a consequence the non-malignant cells remain red in the far right panel (the red BMIl signal only). Thus, the reference value for
BMIl and EZH2 in the non-malignant, normal cells is much higher than 10 (as in Figure 1). In Reed-Sternberg tumor cells the extent of binding of both BMIl and EZH2 antibodies is high (Figure 2, the arrows indicate Reed- Sternberg tumor cells in which the red BMIl signal overlaps with the green EZH2 signal). Thus, the ratio value of the extent of binding of BMIl and EZH2 in Reed-Sternberg tumor cells is approximately 1. This ratio value differs significantly from the reference value in the normal, non-malignant cells and this difference can be used to diagnose Reed-Sternberg tumor cells of Hodgkin's disease. (Raaphorst, F.M., van Kemenade, F.J., Blokzijl, T., Fieret, E., Hamer, CM., Satijn, D.P.E., Otte, A.P., & Meijer, C.J.L.M. 2000. Co-expression of BMIl and EZH2 Polycomb-group genes in Reed-Sternberg cells of Hodgkin's disease. The American Journal of Pathology, 157, 709- 715).
EXAMPLE 2
Human Mantle Cell Lymphoma (MCL) cells were obtained from blood or pleural fluid. After Ficoll-Isopaque (1077 g/ml) density gradient centrifugation MCL cells were cryopreserved in 10% dimethylsulfoxiode until further study. The diagnosis was established by cytomorphological assessment of the blood; histological examination of lymph nodes and bone marrow; immunophenotyping; molecular and cytogenetic analysis. After thawing, residual T cells were removed with help of magnetic immunobeads. 3T6 cells transfected with the human CD40L were added in flat bottom 24-well culture plates. One million T cell depleted malignant B cells were added in a final volume of 1 ml. Human recombinant (HuR) IL-10 was added at a final concentration of 20 ng/ml. Cells were cultured in a fully humidified 5% CO2 atmosphere at 37 °C for 7 days. Coverslips were coated with 3-aminopropyl-tri-ethoxisilan (APTS) by applying 50 μl 2% APTS in ethanol and letting dry to the air. Subsequently the coverslips were rinsed three times with water and again dried to the air.
MCL cells were rinsed with PBS and cytospun on APTS coated coverslips for 10 min at 700 rpm (Cytospin, Shandon, Pittsburgh, PA, USA) and dried to the air for 10 min. Attached MCL cells were rinsed with 400 μl phosphate- buffered saline (PBS) and incubated with 2% (wt/vol) paraformaldehyde in PBS for 15 min at room temperature. After fixation, cells were rinsed three times with PBS and permeabilized with 0.5% (wt/vol) Triton X-100 (Sigma) for 5 min at room temperature. Cells were subsequently rinsed twice with PBS, incubated in PBS containing 100 mM glycine for 10 min, and incubated twice in PBG (PBS containing 0.5% bovine serum albumin and 0.05% gelatin from cold-water fish skin [Sigma]) for 10 min. Fixed cells were incubated for 2 h at room temperature with polyclonal rabbit antibodies against EZH2 or BMIl diluted in PBG. Subsequently, cells were washed in PBG and incubated with biotinylated donkey anti-rabbit IgG diluted in PBG for 1 h at room temperature. After labeling, cells were subsequently washed in PBG and incubated with streptavidine-FITC diluted in PBG for 30 min. Cells were washed in PBG and PBS. Images of labeled cells were produced on a Zeiss confocal laser scanning microscope.
RESULTS Mantle cells from healthy persons do express the BMIl Polycomb group protein and they do not express the EZH2 Polycomb group protein (see Figure 1). Thus, the extent of binding of BMIl to mantle cells from healthy persons is high and the extent of binding of EZH2 to these cells is low. Therefore, the reference value for BMIl and EZH2 in the mantle cells is much higher than 10. CD40 and IL-10 stimulate the transition of resting mantle B-cells to dividing follicular centroblasts. In follicular centroblasts from healthy persons the extent of binding of BMIl is low and the extent of binding of EZH2 to these cells is high (Figure 1, Dark Zone). Thus, the reference value for BMIl and EZH2 in normal follicular centroblasts is much lower than 0.1 (see Figure 1). In MCL cells that respond to CD40 plus
IL-10, we detect appearance of EZH2 expression, thus a high extent of EZH2 binding (Figure 3 A,B,C). However, in MCL cells that respond to CD40 plus IL-10, also the extent of binding of BMIl remains high (Figure 3 D,E,F). Therefore, the ratio value of the extent of binding of BMIl and EZH2 in these responding MCL cells is approximately 1. This irregular ratio value differs significantly with the reference value in the normal, non- malignant cells and can be used to diagnose MCL cells. (Visser, H. P. J., Gunster, M. J., Kluin-Nelemans J. C, Manders, E. M. M., Raaphorst, F. M., Meijer, C. J. L. M., Willemze, R. and Otte, A. P. 2001. The Polycomb group protein EZH2 is up-regulated in proliferating, cultured human Mantle Cell Lymphoma. British Journal of Haematology, in press).
Brief description of the drawings
Figure 1 Lymphnodes from healthy persons have been incubated with antibodies against the Polycomb group proteins BMIl, RINGl, EZH2 or EED. Shown is an overview of lymph node germinal center (original magnification: x 200). Immunostaining was performed with horseradish peroxidase, visualizing the nuclei with a brown colour.
Figure 2 Lymphnodes from patients with Hodgkin's Disease were incubated with antibodies against BMIl and EZH2. The figure shows nuclei of distinct follicular lymphocytes (original magnification: x 400), visualized by immunofluorescence. In Reed-Sternberg tumor cells both BMIl and EZH2 Polycomb group proteins are expressed. The arrows indicate the Reed-
Sternberg tumor cells in which the red BMIl signal overlaps with the green EZH2 signal. In contrast, as in Figure 1, expression of BMIl and EZH2 proteins is clearly separated in normal, non-malignant follicular B-cells (indicated with stars). As a consequence the non-malignant cells remain red in the most right panel (the red BMIl signal only).
Figure 3 Mantle cells from patients with Human Mantle Cell Lymphoma (MCL) were incubated with antibodies against BMIl and EZH2. Shown are high magnifications of single nuclei (original magnification x 1000), visualized by immunofluorescence.. CD40 and IL-10 stimulate the transition of resting mantle B-cells to dividing follicular centroblasts. In MCL cells that respond to CD40 plus IL-10, we detect appearance of EZH2 expression (Figure 3 A,B,C). In these MCL cells also the extent of binding of BMIl remains high (Figure 3 D,E,F).