EP2059906A2

EP2059906A2 - Method for the quantitative determination of the colocalization of molecular markers in tissue sections

Info

Publication number: EP2059906A2
Application number: EP07788341A
Authority: EP
Inventors: Ansgar Pommer; Bernd Bonnekoh; Raik BÖCKELMANN; Harald Gollnik; Lars Phillipsen
Original assignee: Skinsystec GmbH
Current assignee: Tpa Biotech GmbH
Priority date: 2006-08-30
Filing date: 2007-08-09
Publication date: 2009-05-20
Also published as: WO2008025662A3; WO2008025662A2; DE102006040513A1

Abstract

The invention relates to a method for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, especially in sections of cutaneous and mucosal tissue. The method of the invention is characterized by the following procedural steps: generation of a digital image of the tissue section for each of the at least two molecular markers for visualization of the respective molecular marker, superimposition of the digital images to give a summed image, pixelwise quantization of the molecular marker signal for each of the at least two molecular markers in the summed image, generation of a matrix, where a vector whose components are the quantized molecular marker signals of the at least two molecular markers is assigned to each pixel, and tissue-specific standardized alignment of the matrix. The invention further relates to an apparatus for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, especially in sections of cutaneous and mucosal tissue, and to the use of the method of the invention.

Description

Method for the quantitative determination of colocalization of molecular markers in tissue sections

The invention relates to a method for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, in particular in skin and mucosal tissue sections.

To study the correlative distribution and arrangement of different molecular species in tissue sections, various techniques and procedures are currently used in the fields of biology, biochemistry and medicine. The aim of applying such methods is to obtain diagnostic data on specific clinical pictures from the knowledge gained in this way for molecular arrangement and distribution, or to gain insights into the organization of cells or cell systems at the molecular level.

The methods known from the prior art and used as standard in biology, biochemistry and medicine are largely based on manual methods. In these, the tissue section to be examined is mixed with a solution containing a specific reagent, eg. As a dye or an antibody contains. This marks a molecule group to be examined by attaching to or reacting with it, for example, to form a specific complex. Subsequently, the tissue section containing the molecular groups marked in this way is replaced by different, mostly optical Investigated methods, wherein a signal distribution pattern reflecting the arrangement and distribution of the molecular group of interest is recorded.

A proven method here is the

Fluorescence microscopy, in which the fluorescence of with a fluorophore, for example, fluorescein iso-thiocyanate or phycoerythrin, labeled molecules is measured. This is a

Fluorescence signal distribution pattern recorded, which reflects the distribution and arrangement of the labeled molecule in the microscope object. In order to determine the colocalization of several labeled molecular species, the procedure described above must be repeated frequently using different fluorophores. The fluorescence signal distribution patterns obtained in each case for the different molecular species are subsequently to be evaluated together to determine the correlative distribution of the molecular species studied.

A method for automating repeated fluorescence labeling for the purpose of determining the colocalization of molecular species in a cell or tissue sample is described in DE 197 09 348 C2. This automated method, termed "automatic multi-epitope ligand mapping method" (MELK), uses a device that includes a pipetting system, a 3D handling system, and an optical measuring system. Through the automatic pipetting system, the cell or tissue sample to be examined will gradually with different incubation solutions, each one or more Fluorophore-conjugated reagents, with a fluorescence distribution pattern being recorded by a fluorescence microscope image acquisition device, such as a CCD camera, for each incubation cycle. The fluorescence images obtained in this way, the number of which corresponds to the number of different incubation solutions, can now be superimposed by computer assistance, so that from the various individual patterns a resulting complex combination pattern results for the common representation of the distribution and arrangement of the different labeled molecular species. With this method, it is consequently possible to obtain information about the molecular arrangement or distribution of individual molecular species in a cell or tissue sample with comparatively little manual preparative effort. However, the method is limited to purely qualitative statements on the colocalization of different molecular species.

On this basis, the invention is therefore based on the object of specifying a method which also allows reliable quantitative statements on the colocalization of molecular markers in tissue sections, in particular in skin and mucosal tissue sections.

This object is achieved according to the invention by a method which is characterized by the following method steps:

Generation of a digitized image of the

Tissue section for each of the at least two molecular markers to represent the respective molecular marker, - superimposition of the digitized images into a summation image,

Pixel-wise quantization of the molecular marker signal for each of the at least two molecular markers in the summation image and

Generation of a matrix, wherein each pixel is assigned a vector whose components are the quantized molecular marker signals of the at least two molecular markers.

In the method according to the invention, an image which represents the signal distribution pattern of the respective molecular marker is first recorded and digitized for each of the at least two molecular markers. The images can be generated by different analysis methods. Fluorescence microscopy has proved to be particularly suitable as an analytical method which has been used successfully for many years in biology and medicine. In this case, fluorophores are used as molecular markers whose fluorescence signals are displayed in a fluorescence image. However, other methods of analysis may be used to generate the digitized images of the tissue section, such as autoradiography, in which the digitized images are read by reading a radiation detector which registers the radioactive radiation emitted by isotope-labeled molecules, or by digitizing one by the radiation blackened photographic film are generated. The digitized images are then superimposed according to the invention to form a summation image, wherein it depends particularly on a preparation point or pixel-precise overlay to obtain exact information about the relative positions of the individual molecule markers to each other. For this purpose, it is advantageous to also record a phase contrast image in a manner known per se in addition to the individual images for the representation of the respective molecule markers.

According to the invention, the signals for each of the at least two molecular markers in the summation image are quantized pixel by pixel after the image has been overlaid. This can be done by different types of coding, for example by a Trinär-Codierung (Basis 3) or in general by an encoding to the base of a number n.

According to the teachings of the invention, a matrix is finally generated, wherein each pixel of the summation image is assigned a vector whose components correspond to the quantized molecular marker signals of the at least two molecular markers. As a result, it is now possible to quantitatively accurately evaluate the qualitative measurement data obtained in the methods known from the prior art for the colocalization of molecular markers. This allows, inter alia, a valid comparability of measurement results obtained, for example, on different samples. According to an advantageous embodiment of the invention, it is provided that the tissue section is aligned anatomically standardized before the production of the digitized images. For example, in the case of a skin tissue section, the standardized orientation may be such that the basal membrane zone (BMZ) of the skin tissue in each of the digitized images is disposed substantially horizontally. Alternatively, the digitized images can be aligned anatomically standardized, so that an alignment of the tissue section itself is not required.

According to a further advantageous teaching of the invention, immediately before the superimposition of the images into a summation image, each of the images is checked for non-informative contents, in particular with regard to possible artifacts, image areas occupied with non-informative contents are marked and these image areas are removed from all images. This ensures that the digitized images superimposed on a summation image no longer contain any image content that falsifies the quantitative analysis of the molecular marker colonization. The removal of the image areas from all digitized images occupied in at least one image with non-informative contents makes it possible for the summation image subsequently generated by superimposition to contain usable information on all molecular markers in each pixel. Examples of test criteria are the integrity of the tissue section or the contamination of the tissue section by particles. Obviously unnaturally high signal values also justify an exclusion of the relevant image area. According to a further advantageous teaching of the invention, the vectors associated with each pixel are normalized to a defined horizontal width of a Hautgewebeschnitts. This can be done by vectorial addition of the signal vectors of the pixels arranged side by side on the defined horizontal width of the Hautgewebeschnitts. This takes into account the biological situation that the skin is a nonhomogeneous, stratified epithelial tissue that is permanently differentiating perpendicular to the BMZ line, so that normalization only follows along the BMZ line, ie in the horizontal direction, and not perpendicular to it makes sense. The defined horizontal width of the Hautgewebeschnitts may be for example 100 microns.

The analysis of the quantized molecular marker signals represented as a vector for each pixel may relate to the entire tissue section, but it is also possible that colocalization of certain molecular markers, e.g. B. certain T-cell subpopulations, toposelective in certain microcompartments, such as the epidermis, in the blood or lymph vessel system or in the dermis to determine quantitatively. Likewise, the quantitative determination of the colocalization of the molecular marker can be carried out in at least one focal field of the tissue section. As a focus field, any, for example, a rectangular, circular or any polygonal area to be determined in the tissue section to be examined. Due to the high automation potential of the method according to the invention, it makes sense to perform this while executing a computer program on a data processing system.

The invention is further based on the object of providing a device for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, in particular in skin and mucosal tissue sections, which is composed of commercially available standard components and thus can be realized comparatively inexpensively.

The object is achieved with a device according to claim 15.

Advantageously, the unit for generating digitized images is designed as a fluorescence microscope and comprises a CCD camera.

In the following, the invention will be explained in more detail with reference to an illustrative drawing illustrating further advantages of the invention. Show it

1 shows a skin tissue section with a reference line along the basal membrane zone (BMZ), Fig. 2 shows a number of fluorescence images of

Hautgewebeschnitts of FIG. 1 and a phase contrast image with horizontally oriented BMZ line,

Fig. 3, the determination of one for all

Fluorescence images of valid image areas excluding non-informative image content,

4 shows the pixel-wise quantization of

Signal values for each of the molecular markers used,

Fig. 5, the quantitative analysis in

Microcompartments of the image area from FIG. 3 in a summation image formed from the fluorescence images of FIG. 2,

Fig. 6a - d the quantitative analysis in individual

Focus fields of the image area of Fig. 3,

Fig. 7 the analysis of the immediate

Pixel neighborhood of a selected pixel,

8 shows the analytics of

Pixel neighborhood to a selected higher order pixel, Fig. 9 is a graphical representation of a bivariate analysis for the quantitative determination of the colocalization of a key molecule marker with each one of the remaining molecular marker and

Fig. 10, the use of the invention

Method for cell recognition and characterization.

In Fig. 1, the microscopic image of a skin tissue section is shown. Visible is the upper layer of the horny layer 1 (stratum corneum) and below the epidermis 2 (epidermis). Also recognizable are the basal membrane zone 3 (BMZ), the dermis 4 (dermis) and the subcutis 5 (subcutis). In the area of the basal membrane zone 3 and the dermis 4 different blood and lymph vessels 6 are shown in cross section.

To prepare the inventively provided generation of digitized images of the tissue section to represent the molecular marker used in each case, the microscopic image of the skin tissue section is first aligned anatomically standardized in a suitable manner. This is preferably done in such a way that the basement membrane zone 3 runs substantially horizontally in the reference coordinate system of the image acquisition apparatus. The mean course of the basement membrane zone 3 is designated in FIG. 1 as BMZ line.

FIG. 2 now shows a total of four digitized images of the skin tissue section from FIG. 1, this being different in the context of a test series with n Molecular markers was added. All images were taken by fluorescence microscopy and show the fluorescence signals of the molecular marker shown in each case as a hatched area. In Fig. 2, these carry the reference numeral 7. Further, in Fig. 2, a phase contrast image of the Hautgewebeschnitts is shown. The inclusion of a phase contrast image facilitates the preparation point exact superposition of the fluorescence images in the context of the method according to the invention. As can also be seen from FIG. 2, all fluorescence images as well as the phase contrast image are already anatomically standardized by the horizontal arrangement of the basal membrane zone.

As shown in FIG. 3, a uniform image area (general information field) is defined for all fluorescence images (the fluorescence signals are not shown in FIG. 3 for the sake of clarity) and the phase contrast image. For this purpose, according to an advantageous embodiment of the method according to the invention immediately before the superimposition of the fluorescence images into a summation image, all the fluorescence images are checked for non-informative contents, in particular artifacts. In the present example, a technical artifact 8 uniquely detectable by an unnatural fluorescence signal value or a particle is identified in the first fluorescence image. The image area occupied by the artifact 8 is marked and removed from the line through the upper edge of the skin tissue section and a line drawn parallel to it in the area of the subcutaneous tissue and laterally delimited by vertical lines in all fluorescence images. As a result, at all Fluorescence images only the image area of the skin tissue section, which is shown in FIG. 5 and is designated as a general information field, is considered. FIG. 5 simultaneously represents the summation image from the individual fluorescence images. Here, the colocalization of the individual molecule markers in the tissue section can already be determined on a qualitative level.

FIG. 4 now shows a further essential step of the method according to the invention. After the specimen or pixel-by-pixel superimposition of the digital fluorescence images into a summation image, the signals of all molecular markers in the summation image are quantized pixel by pixel. In the present case, they are trinely coded, i. H. into the three categories "not available" (value 0), "weakly available" (value 1) and "strongly present" (value 2). Thus, according to the invention, for each pixel of the summation image, a vector whose dimension corresponds to the number of examined molecular markers and thus to the number of different fluorescence images is formed and arranged in a matrix corresponding to the pixel grid of the digitized images. In other words, the components of the vectors formed for each pixel of the pixel grid correspond to the quantized ones

Molecular marker signals of the n analyzed molecular markers. These individual vectors can also be referred to as Molecular Phenotype (MOLP). In FIG. 4, a 4 × 4 pixel block and specifically the pixel P ₃₂ (3rd row, 2nd column) is considered concretely. This provides no signal for the molecular marker 1, since the fluorescent signal X recognizably does not protrude sufficiently far into the 4 × 4 pixel block. The situation is different with the molecular marker 2. Here, the weak signal Y becomes represented by the value 1 in the context of quantization. The signal Z (molecular marker 3) is a strong signal. Accordingly, this is quantized with the value 2. As can be seen from the corresponding fluorescence image, no signal is registered for the molecular marker n (ie quantized value 0).

It proves expedient to normalize the signal vectors obtained for each pixel to a defined horizontal width of the Hautgewebeschnitts. The normalization is preferably carried out by vectorial addition of the signal vectors of juxtaposed pixels on the defined horizontal width of Hautgewebeschnitts, which in the present case is suitably set to 100 microns. The standardization takes into account the biological situation that the skin is a permanently differentiating epithelial tissue perpendicular to the BMZ line, so that a standardization only along the BMZ line, ie. H. makes sense in the horizontal direction. This normalization is also referred to as "normalized number of pixels (NORP) with respect to a defined horizontal skin width".

According to FIG. 5, the analysis of the quantized molecular marker signals may relate to the entire tissue section, but it is also possible for a toposelective colocalization of certain molecular markers in certain microcompartments of the skin tissue section. As shown by way of example in FIG. 5, these may be the epidermis 2, blood or lymphatic vessels 6 or also the dermis 4.

As shown in FIGS. 6a to 6d, the colocalization of the analyzed molecular markers can also be determined by analyzing individual focal fields of the tissue section make. These can be of any geometrical shape, for example as a square (FIG. 6a), as circles (FIG. 6b) or else as any surface (FIG. 6d).

As shown in FIGS. 7 and 8, the colocalization of the molecular markers studied can be determined more closely by different search strategies. Fig. 7 shows the examination of the immediate pixel neighborhood to a central pixel. In this search strategy, referred to as "pixel neighborhood search", the eight neighboring pixels laterally or diagonally adjacent to the central pixel are examined in order to recognize the distribution patterns of the molecular markers.

Fig. 8 shows a higher order examination of adjacent pixels for the purpose of detecting molecular marker distribution patterns. In this principle, referred to as "closest-pixels centrifuge", the pixels adjacent to the central pixel in the first, second, third or nth order are also analyzed, this being the case in the entire general information field (see FIG Microcompartments (see Fig. 5) or focus fields (see Fig. 6) can take place. For example, in a quality unmatched by the prior art, the molecular neighborhood of T-cell subpopulations in a tissue context can be mapped and analyzed.

The signal vectors for each pixel that have been quantitatively evaluated with different search strategies in different subareas of the examined tissue section can now be used for the most diverse biomathematical Evaluation methods are subjected. For example, a monovariate analysis of the signal vectors can be performed. Due to the practically unlimited number of analysable molecular markers when using the method according to the invention, a previously unachievable gain in information, for example in the immunodermahistological diagnosis of cutaneous lymphoma, is possible with the methods known from the prior art.

Figure 9 shows the bivariate analysis of the signal vectors for the purpose of quantifying the colocalization of a key molecular marker with each one of the remaining molecular markers analyzed. This makes possible the analysis of the dual-radial relationship of a key molecular marker, a so-called stroke marker, to the remaining molecular markers of a network. It goes without saying that in addition to monovariate and bivariate analysis of the signal vectors, a multivariate analysis can also be carried out as the most complex form of data evaluation. So can be z. For example, the frequency of a particular lymphocyte subpopulation characterized by the positive appearance of two or more molecular markers within the microcompartment of the dermis can be accurately detected.

Another aspect of the invention will be finally explained in connection with FIG. Due to the lipid content of the cell membranes and certain fixation steps that dissolve lipids from the tissue section, cell recognition with respect to intact cell membranes is generally not possible. However, with the method according to the invention, it is possible to determine an automatic cell nucleus-related To achieve cell recognition. The starting point for this is the use of a suitable nuclear marker, with the help of which the surface of the cell nucleus 9 can be defined in a calibration measurement. Based on this, using the above-described search strategy, taking first and higher order into a central pixel or pixel block of adjacent pixels ("dosest pixel-centrifuge" principle), it is possible to analyze the cytoplasmic environment by molecular marker.

10, the analysis of the cytoplasmic environment of the nucleus 9 begins in its center by analysis of the pixels adjacent to a central pixel in the first-order first order (see Fig. 7) and then successively in a higher order (see Fig. 8). This is done e.g. until the molecular marker signal corresponding to the cytoplasm 10 breaks off in the signal vectors associated with the individual pixels.

Overall, the method according to the invention thus allows the analysis of the absolute and relative spatial position of selected cells in relation to one another or to the extracellular matrix.

Claims

PATENT APPLICATIONS

1. A method for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, in particular in skin and

Mucosal tissue sections, but not by the following process steps:

Generation of a digitized image of the tissue section for each of the at least two molecular markers to represent the respective molecular marker,

- superimposition of the digitized images into a summation image,

pixel-wise quantization of the molecular marker signal (7, X, Y, Z) for each of the at least two molecular markers in the summation image and

- Generation of a matrix, wherein each pixel is assigned a vector whose components are the quantized molecular marker signals (7, X, Y, Z) of the at least two molecular markers.

2. Method according to claim 1, wherein the at least two molecular markers are fluorophores, which are each displayed in a fluorescence image by means of fluorescence microscopy.

LO / hs 061206WO

3. Method according to claim 1 or 2, characterized in that the tissue section is anatomically standardized prior to the generation of the digitized images.

4. Method according to claim 1 or 2, characterized in that the digitized images are aligned anatomically in a standardized manner.

5. The method of claim 3 or 4, wherein the tissue section is a skin tissue section and the orientation of the tissue section or digitized images is such that the basal membrane zone (3, BMZ) of the skin tissue is substantially horizontal in each of the images.

6. Method according to one of claims 1 to 5, wherein, immediately before the superimposition of the images to form a summation image, the following additional method steps are carried out:

- Checking each of the images for non-informative content, especially artifacts

(8th),

- Marking of image areas occupied by non-informative content and

- Removal of image areas from all images.

LO / hs 061206WO

7. Method according to one of claims 1 to 6, characterized in that the pixel-wise quantization of the

Molecular marker signal (7, X, Y, Z) by a Trinär- coding or coding to the base of the number n.

8. Method according to one of claims 1 to 7 and claim 5, characterized in that the vectors associated with each pixel are normalized to a defined horizontal width of the skin tissue section.

9. The method according to claim 8, wherein the standardization is carried out by vectorial addition of the signal vectors of the pixels arranged next to one another on the defined horizontal width of the cutaneous tissue section.

10. Method according to claim 8 or 9, wherein the defined horizontal width of the skin tissue cut is 100 μm.

11. Method according to one of claims 1 to 10, wherein a quantitative determination of the colocalization of the at least two molecular markers in at least one microcompartment of the tissue section is carried out.

LO / hs 061206WO

12. Method according to claim 11, wherein the tissue section is a skin tissue section and the quantitative determination of the colocalization of the at least two molecular markers takes place in the epidermis (2), in the blood or lymphatic vessel system (6) or in the dermis (4).

13. Method according to one of claims 1 to 10, wherein a quantitative determination of the colocalization of the at least two molecular markers in at least one focal field of the tissue section is carried out.

14. The method according to any one of claims 1 to 13, d a d u r c h e c e n e c e s, e s t s the method under execution of a

Computer program is performed on a data processing system.

15. A computer program for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, in particular in skin and mucosal tissue sections, the execution of which causes at least one processor to carry out a method according to one of claims 1 to 13.

16. A device for the quantitative determination of the colocalization of at least two molecular markers in tissue sections, in particular in skin and mucosal tissue sections, comprising a unit for generating digitized images of the tissue section to represent the respective

LO / hs 061206WO Molecular marker and a data processing system, characterized in that the device is arranged for carrying out the method according to one of claims 1 to 13.

17. Device according to claim 16, wherein the unit for generating digitized images is designed as a fluorescence microscope.

18. Device according to claim 17, characterized in that the fluorescence microscope comprises a CCD camera.

19. Use of the method according to one of claims 1 to 13 for cell recognition in tissue sections, but at least one of the at least two molecular markers is a cell nucleus marker.

LO / hs 061206WO