Patent Application
20160024578
The present invention provides an epigenetic haemogram, also referred to as an epigenetic blood cell count that identifies the quantitative, comprehensive picture of cellular composition in a biological sample, wherein advantageously a normalization standard is used. The normalization standard is a nucleic acid molecule comprising at least one marker-region being specific for each of the blood and/or immune cells to be detected, and at least one control-region being cell-unspecific, wherein said regions are present in the same number of copies on said molecule and/or a natural blood cell sample of known composition. Furthermore, the present invention relates to a kit and the use of a kit for performing the epigenetic assessment of comprehensive, quantitative cellular composition of a biological sample. The biological sample is derived from e.g. a mammalian body fluid, including peripheral, capillary or venous blood samples or subfractions thereof, such as peripheral blood mononuclear cells or peripheral blood monocytes, or a tissue sample, organ sample, or from frozen, dried, embedded, stored or fresh body fluids or tissue samples.
A “blood count”, “complete blood count”, or “blood cell profile” commonly designates a set of tests to determine the number, ratio and appearances of blood cells and/or their cellular subgroups (e.g., neutrophils, eosinophils, basophils, CD19 or CD3 cells, and their subgroups, such as CD3+CD4+ and/or CD3+/CD8+ cells). Such a blood count is used in clinical diagnostics as a broad screening test for disorders or a determination of the general health status of an individual. In general, a “blood count” includes assays directed at hematocrit, quantification of hemoglobin, total blood cells, and red blood cell index (e.g. mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, red blood cell distribution).
White blood cells (also referred to as leukocytes) are part of the cellular immune system (we explicitly define all immune cells, including B-cells as cellular immune system) and play a key role in defending an mammals from pathological effects caused either by foreign organisms (in particular for example: viruses, bacteria, parasites, etc.), but also from aberrations of diseased self-cells, such as tumor cells. In addition, immune cells are themselves subject to diseases, either as primary (congenital) immune diseases, such as the IPEX syndrome or as secondary (acquired) immune diseases, such as for example AIDS, HIV. In the former, the immune system itself is impaired, whereas in the latter external factors (such as virus infections, radiation, chemotherapies or environmental factors) lead to a weakening of the immune system. Several types of leukocytes exist and they either derive from the myeloid lineage—e.g. neutrophil, eosinophil and basophil granulocytes, mast cells and macrophages—or derive from the lymphoid lineage including all lymphocyte subpopulations—such as for example T-cells, B-cells, NK cells. Since the composition of the immune system, and its cellular members, has been subjected to many analyses, aberrations of this normal immune cell count (or ratio) can be recognized easily, is used diagnostically, and may be used for clinical decision-making. Thus, the ratio and count of these cells are regularly analyzed in clinical settings—both as routine diagnostic or analytical tool as well as in clinical research or trials—in order to detect any abnormalities or apparent changes that may be caused by a disease or a disease treatment or other internal or external factors. For example, blood counts are used to diagnose the onset/occurrence of leuko- or lymphopenias or leuko- or lymphocytosis, such as granulocytosis. Furthermore, blood counts are taken to monitor the treatment success of all diseases that result from, cause or whose treatment may result in changes of the overall or specific leuko- or lymphocyte counts. For example, for diagnosing or monitoring infections, anemia; leukemia or the effects of chemotherapies, a so-called “differential” whole blood count is used in order to analyze and identify immune cells and subpopulations thereof. In some primary and secondary immune disorders, this procedure may be the only available diagnostic tool. The differential blood count includes assays directed at a quantification of total white blood cells, neutrophil granulocytes, lymphocytes, monocytes, eosinophil granulocytes, and basophil granulocytes.
Routinely, for soluble cells, i.e., mainly blood but also solubilized tissues or body fluids such a specific immune cell profile is measured by flow cytometry, or by immunohistochemistry (IHC) for solid tissues. Both technologies work on the basis of protein epitopes exposed on cell membranes that are specific for each subtype of cell subpopulation. Recently, research focused on the biological role of leukocyte subpopulations, and this results in a strong demand for clinical as well as for research applications allowing to identifying such populations.
Technically, in routine diagnostics, hematocrit, hemoglobin as well as total white blood counts are determined by an automatic cell counter based on light detection and electrical impedance. A differential white blood count, including neutrophil, eosinophil, basophil granulocytes, monocytes and mast cells are determined either via manual microscopic counting or automatic counting of blood smears.
Additional methods, allowing for the detection of T cell populations are MHC multimetric analyses, the Cytokine-Capture Assay, individual T cell detections (ELISPOT-Assay) or the merely qualitative detection and localization of immune cells (immunohistochemical analyses). Like flow cytometry, these assays are based on a detection of proteins; no specific expression level-independent markers are used. It is noteworthy that all of these assays as well as all assays based on the detection of mRNA, vary from cell to cell. This is because even cells that are undoubtedly positive for a certain protein present time wise varying amounts of protein. Hence, a threshold for “positivity” has to be determined for each and every protein marker depending on the affinity and unspecific binding properties of the given antibody as well as on the average amount of surface expression of the target protein.
Even though almost all cells in an individual contain the exact same complement/composition of DNA code, higher organisms must impose and maintain different patterns of gene expression in the various types of tissue. Most gene regulation is transitory, depending on the current state of the cell and changes in external stimuli. Persistent regulation, on the other hand, is a primary role of epigenetics—heritable regulatory patterns that do not alter the basic genetic coding of the DNA. DNA methylation is the archetypical form of epigenetic regulation; it serves as the stable memory for cells and performs a crucial role in maintaining the long-term identity of various cell types. Recently, other forms of epigenetic regulation were discovered. In addition to the “fifth base” 5-methylcytosine (mC), a sixth (5-hydroxymethylcytosine, hmC), seventh (5-formylcytosine, fC) and eighth (5-carboxycytosine, cC) can be found (Michael J. Booth et al. Quantitative Sequencing of 5-Methylcytosine and 5-Hydroxymethylcytosine at Single-Base Resolution Science 18 May 2012, Vol. 336 no. 6083 pp. 934-937). The primary target of mentioned DNA modifications is the two-nucleotide sequence Cytosine-Guanine (a ‘CpG site’); within this context cytosine (C) can undergo a simple chemical modification to become formylated, methylated, hydroxymethylated, or carboxylated. In the human genome, the CG sequence is much rarer than expected, except in certain relatively dense clusters called ‘CpG islands’. CpG islands are frequently associated with gene promoters, and it has been estimated that more than half of the human genes have CpG islands (Antequera and Bird, Proc Natl Acad Sci USA 90: 11995-9, 1993).
For one of the recently described modification of cytosine, 5-hydroxymethylation, the utility of oxidative bisulfite sequencing to map and quantify 5hmC at CpG islands was shown (Michael J. Booth et al. Quantitative Sequencing of 5-Methylcytosine and 5-Hydroxymethylcytosine at Single-Base Resolution Science 18 May 2012, Vol. 336 no. 6083 pp. 934-937).
In the context of the present invention, the term “bisulfite convertible chromatin” shall mean a chromatin structure (e.g. a sufficiently opened structure) that allows bisulfite to chemically modify cytosines. Consequently, the term “DNA bisulfite convertibility” relates to the extent of cytosine bases in said chromatin and/or the respective nucleic acid that is part of said chromatin, that can be (or have been) converted using a bisulfite treatment. The term also relates to the extent of cytosine bases in a reference nucleic acid (such as a plasmid) that can be (or have been) converted using a bisulfite treatment. In turn, the term “non-bisulfite convertible chromatin” or “non-bisulfite convertible nucleic acid” relates to the extent of cytosine bases that cannot be (or could not been) converted using a bisulfite treatment.
As mentioned above, recently three new cytosine modifications were discovered. Therefore, it is expected that future scientific findings will lead to a more precise interpretation of epigenetic patterns of bisulfite convertibility described in the past. These past result of cytosine modification encompass bisulfite convertible (non-methylated, non-modified) and non-convertible (methylated, modified) cytosine. Both termini need to be reinterpreted, as described. According to the novel scientific findings (i) non-bisulfite convertible cytosine encompasses 5-methylcytosine (mC) and 5-hydroxymethylcytosine (hmC), and (ii) bisulfite convertible cytosine encompasses 5-formylcytosine (fC), 5-carboxycytosine (cC) as well as non-modified cytosine.
Additionally, earlier inventions are based on (i) the ratio of bisulfite convertible cytosine to whole amount of chromatin (cell-type independent, 100% bisulfite convertible DNA locus) or (ii) on the ratio of bisulfite convertible cytosine (fC, cC, non-modified cytosine) to non-bisulfite convertible cytosine (hmC and mC). These ratios are used to characterize cell type, cell differentiation, cell stage as well as pathological cell stages. Therefore, new techniques will result in novel, more specific ratios and might supplement current cell specific, cell state specific as well as pathological patterns of epigenetic modifications and therefore, define potential novel biomarkers. Novel ratios to be discovered as biomarkers can be defined as:
Biomarker Ratio=a/b
a=Σ(C and/or mC and/or hmC and/or fC and/or cC)
b=Σ(C and/or mC and/or hmC and/or fC and/or cC),
whereby a and b differs from each other by one to four kinds of modifications. Discovery of novel DNA modifications will certainly broaden this enumeration.
For the purpose of the present application, epigenetic modifications in the DNA sequence is referred to by the terminology of (i) bisulfite convertible cytosine (5-formylcytosine, (fC) and/or 5-carboxycytosine (cC)) and (ii) non-bisulfite convertible cytosine ((including 5-methylcytosine (mC), 5-hydroxymethylcytosine, (hmC)). As both kinds of methylation, mC and hmC are not bisulfite convertible it is not possible to distinguish between these two. Likewise, IC, cC as well as non-modified cytosine are bisulfite convertible and can also not be distinguished from each other as well.
Furthermore, apart from the modifications of DNA also histones undergo posttranslational modifications that alter their interaction with DNA and nuclear proteins. Modifications include methylation, acetylation, phosphorylation, ubiquitination, sumoylation, citrullination, and ADP-ribosylation. The core of the histones H2A, H2B, and H3 can also be modified. Histone modifications act in diverse biological processes such as gene regulation, DNA repair, chromosome condensation (mitosis) and spermatogenesis (meiosis). Also for these modifications a specific pattern of modification is specific for different cell types, cell stages, differentiation status and such a pattern can be analyzed for bisulfite convertibility or similar methods in order to identify certain cells and cell stages. The present invention also encompasses a use of these modifications.
It is expected that further variants of DNA modifications will be discovered in future. Each type of modification will be either bisulfite-convertible or not. These novel modifications can also be used as biomarker readout. Additionally, it is expected that novel methods for bisulfite modification will be established, resulting in a different set of convertible and non-convertible DNA.
The variety of indications for which reporting of the cellular immune status is clinically or analytically helpful is very large. For almost every disease the cellular immune status is either directly relevant or—such as in cancer—becomes relevant due to the impact of drugs that may cause secondary immunological disorders and aberrations. This broad significance of the overall immune status in diseases settings results in a significant demand for methods to measure these parameters, i.e., the leukocyte subtypes and subpopulations.
The current way of addressing this demand is by flow cytometric and immunohistochemical methods, which are well-established and have been developed into high throughput systems for hospital use, are standard procedures in reference laboratories and are, for more simple applications, made available for practitioners. However, certain problems and requirements limit the applicability of flow cytometry and immunohistochemistry.
a) For flow cytometry, cells need to be intact. This means that the blood sample has to be measured in a “fresh” state, any delay in measurement may lead to deviation of results. As a rule of thumb, samples should be measured within 8 hours, since after that time frame granulocytes (one main cellular fraction in the blood) begin to disintegrate. As an alternative to fresh handling, it is possible to cryopreserve blood samples, but there are significant issues associated with respect to performance and reproducibility. As a consequence, flow cytometry in clinical routine is avoided and many potentially meaningful analyzes are omitted, whereas in clinical trials, where immune markers are prime biomarker candidates for treatment predictions, are often left out, or if required by regulations extra facilities need to be set up.
b) Antigen expression is not a digital (on-off), but an analog (low, medium, high) process. Therefore, thresholds defining positive versus negative signals must be determined. For certain markers, this is unproblematic, for others thresholds are very difficult and imprecise.
c) For flow cytometry, it also poses problems that many cell types are not simply identified by a surface (cluster of differentiation—CD) molecule, but some cell types are characterized by intra- or extracellular soluble proteins, e.g. transcription factors or cytokines. Current markers for Tfh, Th1, Th2 cells, and Tregs belong to this category of cell types—the application of fully standardized procedures is even more difficult. This is because the cell-type specific markers need to be captured in order to be associated to the cell.
d) Furthermore, flow cytometry is dependent on the solubility of the analyzed substrate (cell suspensions). With respect to this, tissue cells may be solubilized by enzymatic digestion, but this often leads to the loss of their surface molecules—rendering the CD markers, as prime targets for flow cytometric analysis useless.
e) Often, neither surface- nor intra- or extracellular markers are 100% cell-type specific. “Leaky” expression of certain gene products has been reported (Wiezcorek et al., Cancer Res. 2009 Jan. 15; 69(2):599-608), rendering the quantification somewhat imprecise.
f) Since immunohistochemistry is based on the same principle as flow cytometry, specificity problems overlap. However, the main problem with this technology is that it is considered only semiquantitative. In particular, a particular problem is that an overall cell counting is not feasible due to the presence of various different cell layers, which are difficult to distinguish and count correctly.
As far as aspect e) is concerned, the inventors have previously published a publication proving that flow cytometry detects expressed surface epitopes, but it cannot distinguish between cell-type specific epitope expression and cell-type independent induction of epitope expressions as well as it cannot detect specific-cells that currently not express or less express certain surface markers. In vitro stimulation of CD4+CD25+CD45RA+ T cells, for example, leads to a high expression level of FOXP3 whereby the FOXP3 gene is still methylated and therefore inactivated (Baron et al., Epigenetics. 2006 January-March; 1(1):55-60). Additionally, for in vitro differentiated Th17 cells no demethylation of IL-17A promotor was observed despite high levels of IL-17A transcripts (Janson P. C. J. et al. Profiling of CD4+ T cells with epigenetic immune lineage analysis. The Journal of Immunology. 2010, 92-102). On the other side it is disclosed that methylation is connected with marker expression (Hamerman, Page, Pullen. Distinct methylation states of the CD8β gene in peripheral T cells and Intraepithelial Lymphocytes. The Journal of Immunology 1997, P1240-1246; Janson P. C. J. et al. Profiling of CD4+ T cells with epigenetic immune lineage analysis. The Journal of Immunology. 2010. 92-102; Melvin et al. Hypomethylation in IFN-Gamma Gen correlates with expression of IFN-G, including CD8 cells., Eur J Immunol. 1995 February; 25(2):426-30; Landolfi M M et al. CD2−CD4−CD8− lymph node T lymphocytes in MRL lpr/lpr mice are derived from a CD2+CD4+CD8+ thymic precursor J Immunol. 1993 Jul. 15; 151(2):1086-96; and Carbone A M et al. Demethylation in CD8 suggests that CD4+ derives from CD8+ cells. Role of methylation pattern during cell development. Science. 1988 Nov. 25; 242(4882):1174-6).
In view of the above mentioned demands in both clinical diagnostics and pharmaceutical research, a new method to provide a precise and comprehensive quantification of a variety of cell types in a sample is desired, in order to establish a more precise and thus markedly improved haemogram. Further objects and advantages will become apparent to the person of skill upon reading the present disclosure, and particularly the examples below.
In a first aspect thereof, this object is solved by the present invention by a method for producing an epigenetic haemogram, comprising the steps of epigenetically detecting blood cells in a biological sample, and quantifying said blood cells as detected using a normalization standard, wherein said normalization standard is a nucleic acid molecule comprising at least one marker-region being specific for each of the blood cells to be detected, and at least one control-region being cell-unspecific, wherein said regions are present in the same number of copies on said molecule and/or a natural blood cell sample of known composition.
Key and basis of the present invention is the use of a variety of different cell-type specific bisulfite-convertible DNA marker. These markers are employed for the identification and quantification of a single blood and immune cell types.
In principle, it was previously shown how a quantification of cell types and blood cell counting based on known epigenetic procedures is performed ((Wiezcorek et al., Cancer Res. 2009 Jan. 15; 69(2):599-608, Sehouli et al. Epigenetics: 2011 February; 6(2):236-46.). In brief, either a cell type specifically modified gene region is specifically (amplified and) counted and hence quantitated along with the opposite species of the cell type specific gene region. To provide for an independent quantification, these two measurements arc then put into relation to provide the percentile part of the cell type in the given (blood) sample:
Copy number of bisulfite convertible DNA of cell-type specific genomic region/(copy number of bisulfite convertible DNA of cell-type specific genomic region)+(Copy number of non-bisulfate convertible DNA of cell-type specific genomic region)=% cell type
Alternatively, the number of copies of bisulfite convertible DNA of a cell-type specific gene region is measured and divided by the copy number of bisulfite-convertible DNA of a cell-type non-specific gene region in the given sample. The latter can be determined by measuring all DNA copies using a completely bisulfite-convertible, cell-unspecific gene region or a region that is known to be uniformly bisulfite-unconvertible or bisulfite-convertible in all cell types.
Copy number of bisulfite convertible DNA of cell-type specific genomic region/Copy number of a bisulfite convertible DNA of cell-type unspecific genomic region=% cell type
Hence, when a single specific bisulfite-convertible genomic marker is known, the previously established system allows the relative (percentile (%)) quantification of any one cell type in a given sample. For this, any given standardization of copy numbers or copy equivalents can be used. The resulting percentile share of the cell type in question correlates with the share of cells measured with a different method. Here, “correlating” means that—according to Spearman correlation—the lowest share measured by the epigenetic technology corresponds to the lowest share measured by—for example—flow cytometry. Such system has been shown to be very stable, technically robust and reliable. Therefore, whenever there is a highly cell-specific bisulfite convertible DNA marker achieved, in theory it should be possible to make an accurate and precise determination of the amount of those cells that own the specific bisulfite convertible genomic marker region.
It is known that the efficiency and performance of Real time (RT-)PCR systems differ depending on the RT-PCR components, including primers, probes, and the purity of DNA. Therefore, standards are employed in order to account for the problem to know at which Cp (crossing point) or Ct (threshold cycle) value a given (known) amount of standard DNA can be detected. A dilution series of said standard DNA gives a standard curve, and allows for the normalization of differently performing/efficient RT-PCR systems. Since the quantification is performed on an equivalent system, differences in performance are normalized. However, the problem addressed concerns an (RT-)PCR that is performed on DNA aiming at the detection of biologically and/or chemically altered DNA. The complexity of this biologically and/or chemically altered DNA differs from normal/natural genomic DNA (starting by the simple fact that the complexity of the DNA molecules differ, since a plasmid consists of double stranded DNA of four bases (CTGA), whereas genomic DNA consists of double stranded five bases (CTGACm), and bisulfite converted DNA merely consists of only three single stranded bases (TGA)). Thus, the efficiency of amplification differs between the target DNA (i.e., human chromosomal genomic or bisulfite-converted DNA) and the standard DNA, if the standard is a plasmid or genomic DNA, but more importantly, the “amplification efficiency difference” between (plasmid) standard and the target DNA differs from amplification target to amplification target. (i.e., primer pair, probe etc.). This leads to a number of observations, when qPCR is performed on bisulfite treated and amplified DNA, such as, for example:
When different blood cells in a sample shall be measured, independently of method as used, the total cell number should be equivalent. However, in a given sample that is equally distributed for the performance of different qPCR-assays, despite the use of individual standards for each reaction the total number of copies as detected is different. This leads to the following problem (here shown with CD3 as an example) in case of (e.g.) blood samples that are measured using different RT-PCR systems:
As indicated in Table 1, calculated overall CD3+ DNA copy numbers differ between the two used standardization systems: bisulfite-converted vs. non-converted DNA and bisulfite-converted CD3+ marker region to bisulfite-converted GAPDH (overall cell) marker region. For the first blood sample (WBL02), number of CD3+ DNA copies calculated via number of GAPDH bisulfite converted DNA (1420 copies) is smaller than calculated via bisulfite converted added to non-bisulfite converted CD3+ DNA copy numbers (1678 copies). For the second sample (WBL03) the situation is similar. Differences become more obvious when using these calculated copy numbers for quantification of CD3+ cells within these two blood samples. For sample WBL02, quantification via bisulfite converted to non-converted DNA copy numbers results in 36.6% CD3+cells, whereas quantification via bisulfite converted CD3+ DNA copy numbers to bisulfite converted GAPDH DNA copy numbers results in 50.1% CD3+ cells. Both results and methods differ strongly.
As mentioned, even if normalization on a bisulfite-converted plasmid standard is performed, the different performances/efficiencies of the different assays do not lead to the same copy number.
This problem becomes particularly apparent, when purified cell types are measured with “their” specific epigenetic cell type markers, and compared to the total amount of cells in the sample (as measured by an cell-type unspecific marker (GAPDH)) as well as measured by non-bisulfite convertible DNA of a cell-type specific marker region (here FOXP3).
CN-BC × 100
As can be seen from table 2, again, results for both of the quantification methods differ strongly (97% vs. 102% and 97% vs. 106%).
Finally, when different cell fractions, e.g. blood leukocytes, are measured that, when added up, should make up all cells in the sample as present, the above problem makes it impossible to provide for a correct “complete blood count”. As an example for this, for two blood samples the leukocytes were quantified (Table 3, sample 04 and sample 08). Here, the term leukocytes summarize all the five types of white blood cells: granulocytes, monocytes, B-lympocytes, natural killer cells, and CD3+ T-lymphocytes. Accordingly, it was expected that the single cell counts sum up to 100%, representing a (complete) leukocytogram. However, when using epigenetic qPCR analyses, this is often not the case (see Table 3). The sum of individual quantities of leukocytes often differs from 100%.
CN-BC × 100
CN-BC × 100
When summarizing the above mentioned problems of epigenetic cell quantification, a precise blood counting tool provides the following:
1. allows for the assessment of a precise, comprehensive blood and immune cell count,
2. overcomes differences in assay performance and/or efficiency between standards as used and the biological sample to be analyzed,
3. is independent of membrane integrity of cells to be counted (intact or non-intact cells), and
4. is independent of type of cell containing sample (fresh, frozen, embedded, stored, fluids, solid tissues).
The present invention provides such a tool, and respective methods. According to the present invention, assessing the epigenetic haemogram comprises measurement of the absolute amount of cells by normalization of qPCR results on a bisulfite-unconverted or -converted normalization standard. The normalization standards consist of a nucleic acid molecule comprising at least one marker-region being specific for each of the blood cells to be detected, and at least one control-region being cell-unspecific, wherein said regions are present in the same number of copies, on said molecule and/or a natural blood cell sample of known composition.
In a first step of a preferred embodiment of the method, qPCR assay-specific correction factors are determined to achieve normalization and comparability of all qPCR assays as well as to correct for differences in assay efficiencies. In a second step, DNA of biological sample is isolated, purified and bisulfite treated. This is followed by qPCR specific for bisulfite-converted cell-type specific and/or cell-type unspecific genomic marker regions. The qPCR amplification results are then normalized with said normalization standard, which represents the relative amount of copies of marker DNA, and therefore the relative amount of specific cells. The normalization standard contains bisulfite-converted genomic marker regions or contains native, bisulfite-unconverted, marker regions. Before starting the qPCR, in the latter case the nucleic acid will be bisulfite treated in parallel to the biological sample as analyzed is treated. In a next step, following qPCR, the normalized relative amount of copies of marker DNA is corrected by an assay specific correction factor as described herein in order to correct for differences in assay efficiencies indicating the absolute amount of cells.
The present method allows for a quantification of non-intact but also intact blood cells in biological samples, such as, for example, dried, frozen, embedded, stored as well as fresh body fluids, dried blood spots, blood clots and tissue samples. The sample does not contain purified or enriched cells. Furthermore, the method of the present invention provides for a blood count, wherein the identity and quantity of cells is based on a clear yes/no information on the genomic level that is independent from protein expression levels.
The present invention thus provides a blood and/or immune cell count to be used as an analytical and diagnostic tool for medical use and as a basis for decisions in therapy.
Preferred is a method according to the present invention, furthermore comprising the step of obtaining a comprehensive blood picture, based on said detecting and quantifying. The blood cell count thus identifies the comprehensive picture of the cellular composition based on a number of epigenetic parameters. The combination of these epigenetic parameters is used to identify the cell composition of a blood or tissue sample, i.e. an epigenetic haemogram, and said epigenetic haemogram is provided based on the analysis of the bisulfite convertibility of cell-specific genomic regions.
Preferably, said epigenetic haemogram resembles a leukocytogram and/or a T-lymphocytogram and/or a granulocytogram and/or a monocytogram and/or a B-lymphocytogram and/or a NK cytogram.
Preferably, the method according to the present invention furthermore comprises the use of a bisulfite-unconverted or -converted normalization standard for the normalization, e.g. of the qPCR results. The term “bisulfite-unconverted” normalization standard encompasses natural DNA molecules containing the original/primary biologic modifications, such as formylation, carboxylation, methylation, or hydroxymethylation and that is not bisulfite-treated, and therefore bisulfite-unconverted. The term “bisulfite-converted” normalization standard encompasses DNA molecules containing (genomic) marker sequences corresponding to already bisulfite-converted cell-type specific and unspecific marker regions.
The bisulfite-unconverted or bisulfite-converted nucleic acid molecule is preferably selected from a plasmid, a yeast artificial chromosome (YAC), human artificial chromosome (HAC), P1-derived artificial chromosome (PAC), a bacterial artificial chromosome (BAC), and a PCR-product. Bisulfite-converted normalization standard is a plasmid, yeast artificial chromosomes (YAC), human artificial chromosome (HAC), P1-derived artificial chromosome (PAC), bacterial artificial chromosome (BAC) or a PCR-product.
The natural blood cell sample preferably is a blood sample of known cellular composition, and/or of known composition of blood cell types, and is preferably produced in advance, i.e. the amount and number blood cell types as combined is pre-determined.
In a preferred embodiment of the method according to the invention, the normalization standard, i.e. the plasmid, YAC, HAC, PAC, BAC, and PCR-product, contains cell-specific and unspecific genomic marker regions (to be analyzed in accordance with the epigenetic haemogram) in the same known number of copies on said molecule. In one embodiment, each of these standards is a single molecule containing the same number of all cell-type specific and unspecific genomic marker regions of interest in the epigenetic haemogram to be established. The natural blood cell sample (preferably mammalian, such as human) used as the bisulfite-unconverted normalization standard contains cells in a known composition and quantity, whereby cells can be pre-purified and pre-mixed to obtain a sample of known composition, that is also pre-determined.
During analytical processing, the bisulfite-unconverted normalization standard is bisulfite-treated in parallel and in the same fashion than the bisulfite treatment of the biological sample to be analyzed.
Then, qPCR on the unknown biological sample as well as on the (now) bisulfite-treated bisulfite-unconverted normalization standard is performed using specific primers that help to detect cell-type specific or unspecific bisulfite-converted genomic regions. In contrast, the bisulfite-converted normalization standard will (obviously) not be bisulfite-treated, as it already contains specific marker sequences that correspond to bisulfite-converted marker sequences recognized by qPCR primers that are specific for bisulfite-converted genomic regions.
In a preferred embodiment, the normalization standard comprises a predetermined amount of blood cells of the types to be detected and analyzed according to the haemogram. Preferably, a normalization standard is used consisting of a defined copy number and same stoichiometric amount of specific cells and/or of cell-type specific and/or cell-type unspecific marker regions. Preferred is a single plasmid containing the same copy number and/or stoichiometric amount of cell-type specific and/or cell-type unspecific marker regions for all cell types of interest for the haemogram.
A preferred embodiment of the method according to the invention furthermore comprises the step of correcting said epigenetic haemogram as produced with an assay specific correction factor. Said assay-specific correction factor (for the cells as detected and analyzed) is determined by comparing the known quantitative amount of cells in said mammalian natural cell sample as provided with the relative amount of copy numbers of bisulfite-converted cell-type specific marker DNA of said mammalian natural cell sample assessed by the qPCR using the normalization standard. Using this approach, the present method allows for an accurate quantification of cells, as any assay-specific variations that may have occurred are taken into account. Depending from the kind of normalization standard as used, the assay specific correction factors can differ. The more the normalization standard and its analytical processing are adapted to the biological sample and its processing, the more the assay specific correction factors will approach 1, or even can be neglected. In a preferred embodiment, a bisulfite-unconverted normalization standard is used as it resembles the complexity and impureness of natural cell samples, and therefore the qPCR efficiency between a biological sample and standard should be aligned. Most preferred is the use of a mammalian natural cell sample of known cell composition and quantity as described herein.
The method according to the present invention then comprises the step of determining the relative amount (of copies) of cell-type specific and unspecific DNA within the biological sample of unknown composition. This is achieved by qPCR on isolated, purified and bisulfite-converted DNA of said biological sample under the use of primers specific for bisulfite-converted cell-type specific and unspecific DNA marker sequences. qPCR amplification results for all target cell types are the normalized on said bisulfite-unconverted or converted standard indicating the relative amount of target cells. According to standards and assays used, specific assay correction factors are applied on relative amount of target cells to receive the absolute amount and percentage of the content of cells according to said haemogram as established. Thereby, the absolute, comprehensive cellular composition in said biological sample is determined. Depending from the normalization standard used, the assay correction factor differs from 1, or is approximately 1, and then can be neglected. Other methods for determining the relative amount (of copies) of cell-type specific and unspecific DNA comprise a method selected from specific enzymatic digests or dye exclusion technologies, bisulfate sequencing, next generation sequencing, nanopore sequencing, single molecule real-time sequencing, analyses of epigenetic modifications in promoter regions, using primers specific for bisulfite-converted DNA, using blocking oligonucleotides specific for bisulfite-converted DNA, using fluorescence-labeled, quenched oligonucleotide probes, using primers for single nucleotide primer extension specific for bisulfate-converted DNA, digital or quantitative PCR analysis, and specific selective (nucleic acid and/or chromatin) precipitation.
Preferred is a method according to the present invention, wherein the determination of the relative amount of target cells is based on comparing the amounts of copies of said bisulfite-converted cell-specific regions as determined with the amounts of copies of the bisulfate-converted regions that are unspecific for a cell-type as determined, thereby identifying the relative amount of a specific cell type in relation to all cells present in the sample.
In one embodiment according to the present invention, the relative amount of target cells is determined based on comparing the amounts of copies of said bisulfate-converted cell-specific regions as determined with the amounts of copies of bisulfite-unconverted cell-specific regions as determined, thereby identifying the relative amount of target cells in relation all other cells present in the sample.
In a preferred embodiment of the method according to the invention, further a knowledge base comprising information about cell-specific assay-correction factors estimated/calculated during previous assessments of epigenetic assays is generated. These values may be advantageously used in order to select particularly suitable normalization standards.
In a particularly preferred embodiment of the method according to the present invention, cell-type marker regions are detected that discriminate a specific cell type and/or at least one specific subpopulation of cells from other cells of a leukocytogram, a T-lymphocytogram, a granulocytogram, a monocytogram, a B-lymphocytogram and/or a NK-cytogram. Preferably, a) the leukocytogram consists of T-lymphocytes, natural killer cells, B-lymphocytes, monocytes and/or granulocytes, b) the T-lymphocytogram consists of CD3+CD4+, CD3+CD8+, CD8−CD4−, and/or CD8+CD4+ c) the granulocytogram consists of basophilic, eosinophilic, neutrophilicgranulocytes, and/or granulocytic myeloid-derived suppressor cells, d) the monocytogram consists of CD14+ monocytes, CD 14− monocytes, macrophages, monocytic myeloid-derived suppressor cells, plasmacytoid dendritic cells, myeloid dendritic cells, and/or overall dendritic cells, c) the B-lymphocytogram consists of naïve B cells, pre-B cells, memory B cells, transitional B cells and/or immature B cells, and f) the NK cytogram consists of CD56dim and/or CD56bright NK cells.
Preferably, within the haemogram as determined sub-haemograms (or subpopulations) can be determined. Preferred is a T-helper-cytogram comprising, e.g., Th1, Th2, Th9, Th17, Th19, Th 21, Th22, Tfh, CD4+ natural killer cells (NKT), naïve CD4+, memory CD4+, effector CD4+ cells, and/or CD4+ regulatory T cells, or a T-cytotoxogram comprising, e.g., naïve CD8+, effector CD8+, memory CD8+, CD8+ natural killer cells (NKT), and/or CD8+ regulatory T cells. Furthermore, sub-populations of monocytes can be determined, comprising classical monocytcs (CD 14−), intermediate monocytes (CD 14+) and/or non-classical monocytes (CD 14++) or a dendritogram comprising myeloid dendritic cells, and plasmacytoid dendritic cells. Future scientific studies may discover and identify yet unknown blood cells and leukocyte subgroups and may will assign new functions to certain blood cells and/or will assign known blood cells to different leukocyte subpopulations.
To determine the relative amount of bisulfite-convertible and/or non-bisulfite convertible DNA or nucleic acid comprises a method selected from specific enzymatic digests or dye exclusion technologies, bisulfite sequencing, next generation sequencing, nanopore sequencing, single molecule real-time sequencing, analyses of epigenetic modifications in promoter regions, using primers specific for bisulfite-converted DNA, using blocking oligonucleotides specific for bisulfite-converted DNA, using fluorescence-labeled, quenched oligonucleotide probes, using primers for single nucleotide primer extension specific for bisulfite-converted DNA, digital or quantitative PCR analysis, and specific selective (nucleic acid and/or chromatin) precipitation.
Further preferred is a method according to the present invention, wherein said normalization standard is bisulfite-unconverted and contains at least one bisulfite-convertible CpG position.
Further preferred is a method according to the present invention, wherein said quantifying of cell types in said biological sample is based on the normalization of the relative amount of cell-type specific and unspecific chromatin using the bisulfite-unconverted normalization standard or using the bisulfite-converted normalization standard.
Even further preferred is a method according to the present invention, wherein said normalization using the bisulfite-unconverted normalization standard is indicative for the absolute amount and/or percentage of content of cells within said biological sample
Even further preferred is a method according to the present invention, wherein said biological sample is a sample of unknown cellular composition.
The biological sample as analyzed in the context of the present invention is any sample that contains cells to be analyzed, i.e. cells of the blood and/or immune system, such as cells of a leukocytogram, selected from T-lymphocytes, natural killer cells, B lymphocytes, monocytes, and/or granulocytes, and combinations thereof; a T-lymphocytogram, selected from CD3+CD4+, CD4+ memory, CD4+ effector cells, CD4+ naïve, CD3+CD8+, CD8+ memory, CD8+ effector cells, CD8+ naïve, CD3+CD8−CD4−, CD3+CD8+CD4+, NKT cells, iTreg, Treg, Tfh, Th1, Th2, TH9, Th17, Th19, Th21, Th22, memory and/or effector T helper cells, and combinations thereof, a granulocytogram, selected from basophilic, eosinophilic, neutrophilic, overall neutrophil granulocytes, and/or granulocytic myeloid-derived suppressor cells, and combinations thereof, a monocytogram, selected from CD14+ monocytes, CD14− monocyes, macrophages, plasmacytoid dendritic cells, monocytic myeloid-derived suppressor cells, intermediate monocyets, classical monocytes, non-classical monocytes, and/or overall dendritic cells, and combinations thereof, a B-lymphocytogram, selected from naïve B cells, pre B cells, memory B cells, transitional B cells and/or immature B cells, and combinations thereof, and a NK cytogram, selected from CD56dim and/or CD56bright NK cells.
The term “cell-specific region(s)” herein shall mean genetic regions in the genome of cells and/or nucleic acids that are selected to discriminate on an epigenetic level one cell type and/or subpopulations of cells from all other cell types and/or subpopulations of cells. These regions include the genes of certain markers (such as, for example, certain protein markers), such as 5′ untranslated regions, promoter regions, introns, exons, intron/exon borders, 3′ regions, CpG islands, and in particular include specific regions as amplified after bisulfite treatment (amplicons) that are “informative” about the one cell type and/or subpopulations of cells. Examples for these cell-specific regions are known from the literature, such as, for example, the gene CD3 γ, δ and ε (WO 2010/069499); the granulysine gene (WO 2010/125106); the CCR6 gene (WO 2011/135088); the FOXP3 gene (WO 2004/050706 and Wieczorek et al. Quantitative DNA methylation analysis of FOXP3 as a new method for counting regulatory T cells in peripheral blood and solid tissue. Cancer Res. 2009 Jan. 15; 69(2):599-608.)
Cell-specific marker region usually are DNA regions that contain single CpGs or CpG islands that are bisulfite-convertible only in a specific cell type and therefore indicative for the specific cell type. Additionally, these cell-specific marker regions discriminate one cell type from all other blood cells as well as other tissue cells.
According to the present invention, cells of the epigenetic haemogram are identified and quantified by analyzing the bisulfite convertibility of at least on CpG position in said cell-specific genomic regions.
Thus, preferred is a method according to the present invention, wherein a bisulfite conversion of at least one CpG position within a region as listed in the following table 4 is indicative for the respective blood cell type as listed in said table. These are e.g. the following genomic marker regions for the given cell types:
In table 4, regions that contain CpGs that are specific for the blood cell types granulocytes, monocytes, CD4+ cells, cytotoxic T-cells, B-cells, Natural Killer-cells, and Natural Killer T-cells are listed, as well as their SEQ ID NOs for the so-called “discovery fragment” (preferred region) and the discriminative “region of interest” (more preferred region). The discovery fragments comprise at least one CpG that is specific for the cell type as indicated, and thus suitable to distinguish this cell type from all other cell types of the haemogram. The discriminative region of interest (ROT) sequences are regions that are positioned around the discovery region, and which form the basis for the design of the specific assay for a specific cell type as indicated, and contain additional relevant CpGs, that is, a sequence of CpGs that can also be used in order to distinguish between the call types as indicated.
In table 4A to 4J, regions that contain CpGs that are specific for the respective blood cell types as shown in each table header are listed. The sequence provided in the column “discovery fragment” is the preferred region and comprises at least one CpG that is specific for the cell type of the respective table (identifiable by the shown data). Also comprised in the context of the various embodiments and aspects of the invention is a region 500 base pairs upstream and downstream of (therefore “around”) the sequence of the “discovery fragment” in the human genome for each marker. The region 500 base pairs upstream and downstream of the “discovery fragment” are the discriminative ROI of the marker of the tables 4A to 4J.
The present invention therefore also pertains to a bisulfite conversion of at least one CpG position within any one of the “discovery fragments” or ROI (500 bp up and downstream for each “discovery fragment” in the human genome) of any one of the Tables 4 and 4A to 4J as shown above, which is indicative for a respective cell type as listed in the tables 4. The T-lymphocytogram in all of the embodiments and aspects of the invention may therefore contain any of the cell types listed in the above tables 4 and 4A to 4J, and any combinations of these cell types.
An additional region for neutrophilic granulocytes (nGRC) is derived from the Lipocalin-2, neutrophil gelatinase-associated lipocalin (LCN2) genomic region (Ensembl-ID: ENSG00000148346); herein designated AMP1730. The AMP 1730 genomic sequence and the discriminative ROI 1132 are SEQ ID NOs: 686 and 685 respectively. See also
Additional regions for eosinophilic granulocytes (eGRC) arc derived from the proteoglycan 2 (PRG2) genomic region (Ensembl-ID: ENSG00000186652), herein designated as AMP 2034 and 2035, respectively. The AMP 2034 and 2035 genomic sequences, and the discriminative ROI1403 are SEQ ID NOs: 687, 688, and 689, respectively. See also
Preferably, the cell-specific gene regions as described herein are selected to discriminate one cell type or subpopulation of cells from all other cell types, such as the leukocytogram, T-lymphocytogram, granulocytogram, monocytogram, B-lymphocytogram and/or NK cytogram as described herein. Thus, highly specific cell-type markers are used as a basis for identification and quantification that are not based on protein expression levels but on cell type-specific epigenetic information. The method provides a clear yes/no information and is independent of thresholding as the cell-specific CpG-rich genomic region is bisulfite convertible or not, is detectable by qPCR or not as well as genomic copies do not vary. The method also detects and identifies as well as quantifies a potentially unlimited number of subpopulations of cells, and the detection limit for, for example, regulatory T cells is at 0.3%.
Preferred is a method according to the present invention, wherein the cells that are detected and thus for the epigenetic haemogram are selected from a leukocytogram, and/or a T-lymphocytogram, and/or a granulocytogram, and/or a monocytogram, and/or a B-lymphocytogram, and/or a NK cytogram.
Preferably, said marker regions as analyzed are specific for the cells of a pre-selected haemogram, and these cells are preferably selected from T-lymphocytes, natural killer cells, B-lymphocytes, monocytes, granulocytes, and combinations thereof, for a leukocytogram, selected from CD3+CD4+, CD4+ memory, CD4+ effector cells, CD4+ naïve, CD3+CD8+, CD8 positive, CD8+ memory, CD8+ effector cells, CD8+ naïve, CD3+CD8−CD4−, CD3+CD8+CD4+, NKT cells, iTreg, Treg, Tfh, Th1, Th2, TH9, Th17, Th19, Th21, Th22, memory and effector T helper cells, and combinations thereof, for a T-lymphocytogram, selected from basophilic, eosinophilic, neutrophilic granulocytes, and/or granulocytic myeloid-derived suppressor cells, and combinations thereof, for a granulocytogram, selected from CD14+ monocytes, CD14− monocytes, macrophages, plasmacytoid dendritic cells, myeloid-dendritic cells, intermediate monocytes, classical monocytes, non-classical monocytes, and/or overall dendritic cells, and combinations thereof, for a monocytogram, selected from naïve B cells, pre B cells, memory B cells, transitional B cells and/or immature B cells, and combinations thereof, for a B cell cytogram, and selected from CD56dim and/or CD56bright NK cells for an NK cytogram.
In contrast to the term “cell-specific regions”, the term “cell-unspecific regions” herein shall mean genetic regions in the genome of cells and/or nucleic acids that are selected to be unspecific, i.e. arc specific for more than one, preferably all, cell type and/or subpopulation of cells. These cell-unspecific regions also include the genes of certain markers (such as, for example, certain protein markers), such as 5′ untranslated regions, promoter regions, introns, exons, intron/exon borders, 3′ regions, CpG islands, and in particular include specific regions as amplified after bisulfite treatment (amplicons) that are “informative” for more than one cell type and/or subpopulation of cells. Examples for these cell-unspecific regions are known from the literature, and are selected from, for example regions comprising a housekeeping gene, such as GAPDH, ACTB (beta-actin), UBC (ubiquitin C), ribosomal proteins (e.g. RPS27A, RPS20, RPL11, RPL38, RPL7, RPS11, RPL26L1), CALR (calreticulin), ACTG1 (gamma actin) RPS20 (ribosomal protein S20), HNRPD (ribonucleoprotein D), NACA (nascent poly-peptide-associated complex subunit alpha), NONO (octamer-binding protein), PTMAP7 (prothymosin), GFRA4 (GDNF receptor alpha-4), CDC42 (GTP-binding protein), EIF3H (translation initiation factor), UBE2D3 (ubiquitin-conjugating enzyme), and genes as described in, for example, She et al. (Definition, conservation and epigenetics of housekeeping and tissue-enriched genes. BMC Genomics. 2009 Jun. 17; 10:269.), and PCT/EP2011/051601.
The method according to the present invention generally identifies the quantitative cellular composition of a biological sample. Preferred is a method according to the present invention, wherein said biological sample is a sample of unknown cellular composition. Nevertheless, also samples of known cellular composition, or even partially known composition can be quantified.
Biological samples to be analyzed can be stored fresh-frozen, paraffin-embedded or Heparin, Citrate or EDTA-stabilized as cells in samples do not need to be intact. The present method is very robust and allows, in contrast to flow cytometry, a parallel, independent assessment of cell identity and quantity as well as sample composition. A very good correlation to FACS is provided, too.
The biological sample to be analyzed can be any sample comprising one or more type(s) of cells or that is suspected of comprising one or more type(s) of cells that are to be quantified. Preferred materials/biological samples are selected from a blood sample, in particular peripheral, capillary or venous blood samples, blood clots, or samples that are considered to contain blood cells as e.g. synovial fluid, lymph fluid, sputum, urine, tumor samples, as well as other fluid and tissue samples, histological preparations, DBS, artificially generated cells and mixtures thereof (e.g. cell culture samples).
Yet another aspect of the present invention then relates to a method according to the present invention, further comprising the step of concluding on the immune status of a mammal based on said epigenetic haemogram as produced.
Yet another aspect of the present invention then relates to a method according to the present invention, further comprising the step of monitoring said cellular composition in said biological sample as identified by comparing said composition and/or haemogram as identified with the composition in an earlier biological sample taken from the same mammal, and/or with the composition in a control sample. In this aspect, for example, modifications and changes of the cellular composition in a patient can be monitored during a medical treatment.
Yet another aspect of the present invention then relates to a method for diagnosing a disease or a predisposition for a disease, comprising a method according to the present invention as described above, and the step of concluding on the disease or a predisposition for said disease based on the cellular composition in said biological sample as identified. In this aspect, for example, modifications and changes of the cellular composition in a patient can be used for diagnosing a disease or a predisposition for a disease, in particular when the sample is compared to a sample of a healthy subject or to medical reference ranges. Preferably, said biological sample is a blood sample, in particular a whole or peripheral blood sample, and said cell-specific regions in the genome of cells in said sample are selected from regions specific for blood cell types. The disease to be diagnosed can be selected from the group consisting of immune diseases or conditions, transplant rejections, infection diseases, cancer, neurological diseases, allergy, primary and secondary immune deficiencies and hematologic malignancies such as, for example, lymphatic neoplasms, mature B-cell neoplasms, mature T- and NK-cell neoplasms, Hodgkin lymphomas, lympho-proliferative processes after transplantations, HIV and AIDS, Graft versus Host disease, rheumatoid arthritis, lupus erythematosus, breast cancer, colorectal cancer, esophageal cancer, stomach cancer, leukemia/lymphoma, lung cancer, prostate cancer, uterine cancer, skin cancer, endocrine cancer, kidney cancer, urinary cancer, pancreatic cancer, other gastrointestinal cancers, ovarian cancer, cervical cancer, head and neck cancer, adenomas, birth defects, myopathies, mental retardation, obesity, diabetes, gestational diabetes, multiple sclerosis, and asthma.
In one preferred embodiment of the present invention, the diagnostic use of the epigenetic haemogram is also based on the use of ratios of different populations and/or/to different subpopulations (subhaemograms) and/or/to of cells belonging to one subhaemogram according to the said epigenetic haemogram. Such ratios are e.g. but are not limited to, population of regulatory T cells in relation to CD3+ T-lymphocytes, or regulatory T cells in relation to population of CD4+ T-lymphocytes, or regulatory T cells in relation to population of CD8+ T-lymphocytes, or CD3+ T-lymphocytes to CD4+ T-helper cells, or CD3+ T-lymphocytes to CD8+ cytotoxic T cells, or CD4+ T-helper cells to CD8+ cytotoxic T-cells, or Th1 to Th2, or Th1 to Th17, or Th2 to Th17, or memory or naïve CD4+ T-helper cells to CD3+ T-lymphocytes, or memory CD8+ cytotoxic T-cells to CD3+ T-lymphocytes, all as subpopulations of the T-lymphocytogram; or CD3+ T-lymphocytes related to neutrophilic granulocytes, or macrophages to CD4+ T-helper cells; CD4+ T-lymphocytes related to neutrophilic granulocytes, or CD8+ T-lymphocytes related to neutrophilic granulocytes all as relations between cells of different subhaemograms; or CD3+ T-lymphocytes related to granulocytes, or B-lymphocytes to CD3+ T-lymphocytes, or monocytes to CD3+ T-lymphocytes, or monocytes to B-lymphocytes all as ratios out of populations of the leukocytogram; or CD3+ T-lymphocytes or monocytes or B-lymphocytes, or granulocytes or NK cells related to overall leukocytes. But also other ratios of subpopulations assessed according to the present invention and according to the epigenetic haemogram can be used as a diagnostic method. The disease can be selected from the group consisting of immune diseases or conditions, transplant rejections, infection diseases, cancer, neurological diseases, allergy, primary and secondary immune deficiencies and hematologic malignancies such as, for example, lymphatic neoplasms, mature B-cell neoplasms; mature T- and NK-cell neoplasms, Hodgkin lymphomas, lympho-proliferative processes after transplantations, HIV and AIDS, Graft versus Host disease, rheumatoid arthritis, lupus erythematosus, breast cancer, colorectal cancer, esophageal cancer, stomach cancer, leukemia/lymphoma, lung cancer, prostate cancer, uterine cancer, skin cancer, endocrine cancer, kidney cancer, urinary cancer, pancreatic cancer, other gastrointestinal cancers, ovarian cancer, cervical cancer, head and neck cancer, adenomas, birth defects, myopathies, mental retardation, obesity, diabetes, gestational diabetes, multiple sclerosis, and asthma. The diagnostic use encompasses but is not limited to the diagnosis of a disease and/or the follow-up of a disease and/or the predisposition for a disease and/or the monitoring of an effect of a chemical or biological substance.
The epigenetic haemogram of the invention is in another embodiment used for the assessment of the risk to develop a disease in a patient, therefore for diagnostic purposes. In one preferred embodiment of the present invention, the use of the epigenetic haemogram for the assessment of the risk to develop a disease is also based on the use of ratios of different populations and/or/to different subpopulations (subhaemograms) and/or/to of cells belonging to one subhaemogram according to the said epigenetic haemogram. Such ratios are e.g. but are not limited to, population of regulatory T cells in relation to CD3+ T-lymphocytes, or regulatory T cells in relation to population of CD4+ T-lymphocytes, or regulatory T cells in relation to population of CD8+ T-lymphocytes, or CD3+ T-lymphocytes to CD4+ T-helper cells, or CD3+ T-lymphocytes to CD8+ cytotoxic T cells, or CD4+ T-helper cells to CD8+ cytotoxic T-cells, or Th1 to Th2, or Th1 to Th17, or Th2 to Th17, or memory or naïve CD4+ T-helper cells to CD3+ T-lymphocytes, or memory CD8+ cytotoxic T-cells to CD3+ T-lymphocytes, all as subpopulations of the T-lymphocytogram; or CD3+ T-lymphocytes related to neutrophilic granulocytes, or macrophages to CD4+ T-helper cells; CD4+ T-lymphocytes related to neutrophilic granulocytes, or CD8+ T-lymphocytes related to neutrophilic granulocytes all as relations between cells of different subhaemograms; or CD3+ T-lymphocytes related to granulocytes, or B-lymphocytes to CD3+ T-lymphocytes, or monocytes to CD3+ T-lymphocytes, or monocytes to B-lymphocytes all as ratios out of populations of the leukocytogram; or CD3+ T-lymphocytes or monocytes or B-lymphocytes, or granulocytes or NK cells related to overall leukocytes.
But also other ratios of subpopulations as assessed in accordance with the present invention and according to the epigenetic haemogram can be used to assess the risk for developing a disease. The disease for the herein described embodiment can be selected from the group consisting of immune diseases or conditions, transplant rejections, infection diseases, cancer, neurological diseases, allergy, primary and secondary immune deficiencies and hematologic malignancies such as, for example, lymphatic neoplasms, mature B-cell neoplasms, mature T- and NK-cell neoplasms, Hodgkin lymphomas, lympho-proliferative processes after transplantations, HIV and AIDS, Graft versus Host disease, rheumatoid arthritis, lupus erythematosus, breast cancer, colorectal cancer, esophageal cancer, stomach cancer, leukemia/lymphoma, lung cancer, prostate cancer, uterine cancer, skin cancer, endocrine cancer, kidney cancer, urinary cancer, pancreatic cancer, other gastrointestinal cancers, ovarian cancer, cervical cancer, head and neck cancer, adenomas, birth defects, myopathies, mental retardation, obesity, diabetes, gestational diabetes, multiple sclerosis, and asthma. The diagnostic use encompasses but is not limited to the diagnosis of a disease and/or the follow-up of a disease and/or the predisposition and/or the assessment of a risk for a disease and/or the monitoring of an effect of a chemical or biological substance.
As indicated, the above mentioned ratios as assessed in accordance with the present invention bear the potential to indicate e.g. the risk to develop a certain disease during the life time of a subject. A clinical role in risk assessment was found for the ratio of regulatory T-lymphocytes to CD3+T-lymphocytes. Particularly preferred in the context of the present invention is that an increase in the ratio of regulatory T-lymphocytes to CD3+T-lymphocytes indicates a risk to develop cancer (cancerous disease) during life time. The cancer is selected from but not limited to the list as provided herein above, wherein a high impact of an increased ratio of regulatory T-lymphocytes to CD3+T-lymphocytes is expected for the development of lung cancer, which is particularly preferred. Furthermore, ratios bear the potential to predict the development of Graft versus Host Disease wherein an increased ratio of regulatory T-lymphocytes to CD4+T-lymphocytes within the first two weeks after stem cell transplantation predicts the development of a graft versus host disease.
Yet another aspect of the present invention then relates to a method for identifying the effect of a chemical or biological substance or drug on the composition of cells, comprising performing the method according to the present invention as described above, preferably on a blood sample obtained from a mammal treated with or exposed to said substance, and comparing the composition of cells in said sample with the composition of samples before treatment or with the composition of an untreated sample. The mammal to be treated with said chemical or biological substance or drug might be healthy or suffers from a disease selected from the group consisting of immune diseases or conditions, transplant rejections, infection diseases, cancer, neurological diseases, allergy, primary and secondary immune deficiencies and hematologic malignancies such as, for example, lymphatic neoplasms, mature B-cell neoplasms, mature T- and NK-cell neoplasms, Hodgkin lymphomas, lympho-proliferative processes after transplantations, HIV and AIDS, Graft versus Host disease, rheumatoid arthritis, lupus erythematosus, breast cancer, colorectal cancer, esophageal cancer, stomach cancer, leukemia/lymphoma, lung cancer, prostate cancer, uterine cancer, skin cancer, endocrine cancer, kidney cancer, urinary cancer, pancreatic cancer, other gastrointestinal cancers, ovarian cancer, cervical cancer, head and neck cancer, adenomas, birth defects, myopathies, mental retardation, obesity, diabetes, gestational diabetes, multiple sclerosis, and asthma.
Yet another aspect of the present invention then relates to a diagnostic kit and its use, comprising materials for performing the method according to the invention as described herein, optionally with instructions for use. The diagnostic kit particularly contains oligonucleotides (e.g. for producing amplicons) specific for regions of interest, bisulfite reagents, and/or components for PCR. The diagnostic kit and its use encompasses but is not limited to the diagnosis of a disease and/or the follow-up of a disease and/or the predisposition and/or the assessment of a risk for a disease and/or the monitoring of an effect of a chemical or biological substance.
As mentioned above, currently, in both, clinical diagnostics and research, and drug development, a new method to provide a precise and comprehensive quantification of leukocytes and their subpopulations is desired even if biological samples are not intact anymore. The present invention, overcomes most problems of current, routinely used quantitative methods, flow cytometry and immune histochemistry, but more importantly, overcomes several biochemical and technical problems of qPCR in regard to absolute quantification of target cells. The present invention thus provides a method to effectively detect and quantify the different cell populations. In particular, the present method for the first time allows for an expression-independent method for the assessment of a comprehensive blood cell picture. Moreover, the present invention enables flexible time framing which is not dependent on quick sample processing but rather allows long term sample storage and individual coordination between sample collecting and sample processing.
The present invention will now be explained further in the following examples and figures, nevertheless, without being limited thereto. For the purposes of the present invention, all references as cited herein are incorporated by reference in their entireties.
SEQ ID No. 1 to 689 show sequences as used in the context of the present invention.
The present examples have been performed on a sample of known and unknown leukocyte and T-lymphocyte compositions. The person of skill will understand how to modify the experiments in order to identify and quantify other cell types, in particular blood cells in the context of an epigenetic haemogram, without undue burden and/or the need to become inventive.
The inventors provided a human blood sample of known leukocyte and T-lymphocyte composition. The composition of this blood sample was analyzed via flow cytometry. The sample contained 61% granulocytes, 12% monocytes, 3% B-lymphocyte, 4% natural killer cells, and 19% T-lymphocytes (Table 5). The T cell population consisted of 13% of CD4+ T helper cells, 1.4% regulatory T cells, 5% CD8+ cytotoxic cells, and 2% naïve CD8+ cells.
In a next step, this sample of known leukocyte and T-lymphocyte composition was analyzed for the relative amount of bisulfite convertible chromatin in cell-type specific gene regions, resulting in a unique, discriminating cell-type specific pattern of bisulfite convertible chromatin, e.g. for granulocytes a region in the gene for neutrophil gelatinase-associated lipocalin, for monocytes a region in the leukocyte immunoglobulin-like receptor gene, for B cells in a region of the gene for the low-affinity receptor for IgE, for natural killer cells a region in the gene for oxysterol-binding protein-like protein 5 isoform a, for T-lymphocytesin a region in the CD3D/G gene, for CD4+ T helper cells in a region in the CD4 gene, for regulatory T cells in a region in the FOXP3 gene, for CD8+ cytotoxic T cells in a region in the CD8A/B gene, for naïve CD8+ cells a region in the endosialin gene. Analyses were performed by qPCR using a bisulfite-converted normalization standard indicating the relative amount of numbers of gene copies containing mentioned unique, said cell-type specific pattern of bisulfite convertibility. These relative numbers of cell-specific gene copies indicate the relative amount of said specific cells.
This relative number of specific cells (said leukocytes and T-lymphocytes) was compared with the result of flow cytometry. Both results were set in relation, and a correction factor was determined (Table 1). Flow cytometry revealed 61% and qPCR 91.6% of granulocytes, and therefore the cell-specific granulocyte assay-correction factor was 1.502.
Correction factors were determined separately for each set of assessments as well as are incorporated into data base for assay-specific correction factors. In addition to the individual and separate determination of correction factors (for each set of assessments), the average of past correction factors can be used as well.
Human blood samples of unknown leukocyte and T-lymphocyte composition of healthy volunteers were obtained for assessment of absolute leukocyte and T-lymphocyte composition via qPCR. As for Example 1, DNA of blood samples were isolated, bisulfite converted and relative amount of bisulfite converted DNA assessed via qPCR under the use of Bisulfite-converted normalization standards. Amount of bisulfite convertible DNA in cell-specific gene regions was set in relation to bisulfite-convertible DNA of cell-unspecific DNA region (always, cell independent, constant pattern of bisulfite-convertibility) to obtain relative amount of assessed cells.
Cell-specific assay-correction factors were determined in a parallel experimental set for assays of granulocytes, monocytes, B-lymphocytes, natural killer cells, T-lymphocytes, CD4+ T helper cells, regulatory T cells, and CD8+ cytotoxic T cells using flow cytometry on a human blood sample (methodology see example 1, human blood sample differs for Example 2 compared to Example 1). Relative amounts of assessed cells as obtained were corrected using the cell-specific assay correction factors. E.g., qPCR for monocytes patient sample S04 gave a relative amount of monocytes of 7.94%, but the correction revealed an absolute cell amount of 3.69% monocytes.
One would expect the sum of cells belonging to a leukocytogram to be 100%, and the sum of cells belonging to a T-lymphocytogram to have exactly the same amount of cells as determined for T-lymphocytes in the leukocytogram. It is known that even the flow cytometry quantification is not without limitations, as described above.
Flow cytometry measurement errors are reflected in qPCR corrections. On the other hand, the epigenetics based qPCR, as described herein, detected cell types independently of marker expression. Even if a cell-specific marker is expressed at a very low amount, or is not present at all, epigenetic-qPCR can detect these cells (e.g. as found for Th17 cells, see above). In addition, certain cells do express cell-specific markers, even if these cells did not enter a specific cellular state known to be associated with the marker expression (e.g. as found for regulatory T cells, see description above). Such cells are not detected by epigenetic-based qPCR. Additionally, for this example, the selection of T-lymphocytes (CD4+ T helper cells, CD8+ cytotoxic cells) does not represent the complete T-lymphocyte set (see
Human blood samples of unknown leukocyte and T-lymphocyte composition of auto-immune diseased volunteers were obtained for assessment of absolute leukocyte and T-lymphocyte composition via qPCR. As for Example 1, DNA of blood samples were isolated, bisulfite converted and relative amount of bisulfite converted DNA assessed via qPCR. Amount of bisulfite convertible DNA in cell-specific gene regions was set in relation to bisulfite convertible DNA of cell-unspecific DNA region (always, cell independent, constant pattern of bisulfite convertibility) to obtain relative amount of assessed cells.
Cell-specific assay-correction factors were determined in a parallel experimental set for assays of granulocytes, monocytes, B-lymphocytes, natural killer cells, T-lymphocytes, CD4+ T helper cells, regulatory T cells, and CD8+ cytotoxic T cells using flow cytometry on a human blood sample (methodology see example 1, human blood sample differs for Example 3 compared to Example 1). Obtained relative amounts of assessed cells were corrected using these cell-specific assay correction factors. E.g., qPCR for T-lymphocytes assessed a relative amount of T-lymphocytes of 8.49% for patient M06 and 23.94% for patient M10. Correction revealed an absolute cell amount of 5.4% and 15.3% T cells, respectively.
In comparison to data from healthy patients, see Example 2, for auto-immune diseased patient M06 an obvious decrease in 4 of the 5 subtypes of leukocytes within the leukocytogram was observed. For patient M10 an obvious decrease in absolute number of only B-lymphocytes and monocytes was observed.
Additionally, also for T-lymphocyte subtypes, differences, between both patients were observed. qPCR analysis of three subtypes of T-lymphocytes for patient M06 revealed a strong decrease of CD4+T helper cells as well as CD8+ cytotoxic cells whereas the decrease in level of regulatory T cells was less pronounced. For patient M10 all three cell levels decreased simultaneously by about 50-60% compared to the average of the two healthy patients in Example 2.
All these differences might be related to e.g. a different medication and/or disease stage of these both patients and offer a clinical routine instrument for disease diagnosis, prediction as well as accompanying monitoring.
Matrix indicating bisulfite-inconvertibility in cell-type specific genomic marker regions. Different cell types were analyzed indicating that CpGs within genomic region AMP2034 and 2035 arc, in contrary to other cell types given, convertible by bisulfite to a high extent and indicative for this specific cell-type (see
The inventors developed non-bisulfite converted, genomic plasmid standards as a normalization standard. One of these genomic plasmid standards comprises marker regions being specific for stable regulatory T cells (TSDR region)(Treg cells) as well as marker regions being cell-type unspecific (GAPDH, housekeeping gene, detecting all cells, 100% of cells). This plasmid standard is used to determine the Treg-specific assay correction factor that allows assessing the absolute amount of stable Tregs within an unknown blood sample.
In a first step, a human blood sample of unknown composition was provided, DNA isolated, and bisulfite treated. Following, the amount of bisulfite converted TSDR copies and GAPDH copies were assessed (Table 8, section 2). These qPCR analyses were performed using a bisulfite-converted normalization standard (Table 8, section 1) indicating the number of bisulfite-converted DNA copies containing the TSDR marker region as well as the GAPDH marker region (Table 8 section 2). The relative amount of stable Tregs is calculated as number of bisulfite converted TSDR copies related to bisulfite converted GAPDH copies in percent.
no. bisulfite-converted TSDR copies/no. bisulfite-converted GAPDH copies×100=% Treg
67.70/6026.67×100=1.123%
The cell-type specific region for stable regulatory T cells, TSDR, is located on the X-chromosome. For women an epigenetic silencing of one allele of the X-chromosome is known. This affect is deduced by using a factor 2 when calculating relative amount of stable Tregs (final result=2.25% stable Tregs)(Table 8, section 2).
In a second step, Treg-specific assay-correction factor based on said genomic plasmid standard was assessed. Said plasmid standard was bisulfite converted and number of plasmid copies assessed by qPCR using primers specific for bisulfite-converted marker regions for Treg cells and for GAPDH. These qPCR analyses were also performed using the bisulfite-converted normalization standard (Table 8, section 1). The efficiency of qPCR for Treg cells and GAPDH should be equal as the novel genomic, non-bisulfite converted plasmid standard (the substrate) contains an equimolar amount of Treg cell-specific and GAPDH-specific genomic copies. Therefore, assessed deviation of Treg copy numbers from GAPDH copy numbers corresponds to differences in assay effiCiencies.
Treg(TSDR) copy numbers=6760 vs. GAPDH copy numbers=6273,33
This deviation defines the cell-type assay-specific correction factor. E.g.:
Treg(TSDR)copy numbers/GAPDH copy numbers/100=6760/6273.33=1.077.
For Treg cells an assay correction factor of 1.1 (average, n=3) was assessed (Table 8, section 3). Correcting the relative amount of Treg cells by factor 1.1 results in an absolute amount of 2.05% Treg cells within the unknown blood sample WB01.
relative amount of Treg cells/specific assay-correction factor=absolute amount of Treg
2.25%/1.1=2.05% Treg cells
DNA from the purified neutrophil granulocytes (neutrophils), monocytes, CD4+ cells CD8+ cells, B cells, NK-cells, and NKT cells was bisulfite-treated and bisulfite converted DNA analyzed at various CpG dinucleotide motifs. The inventors then compared the bisulfite convertibility (finding C as for Cytosine that was methylated in the original (genomic) sequence versus T for cytosine that was unmethylated in the original sequence) of these CpG dinucleotides (see Table 4, position 259).
Surprisingly, it was found that specific areas in the genomic region of lipocalin-2 were differentially methylated in neutrophil granulocytes compared to all other blood cell types tested. These areas were defined as discovery fragments, such as e.g. SEQ ID 517 for neutrophils (Table 4, position 259).
Then, upon finding of the differential bisulfite convertibility, the inventors analyzed larger genomic regions by means of bisulfite sequencing. This latter procedure served for exploring and extending the discovered, differentially methylated areas and was conducted, for example with the differentially bisulfite converted discovery fragment, SEQ ID 517, within the gene lipocalin-2 as disclosed herein (see Table 4, SEQ ID 517 discovery fragment and 518 discriminative region of interest (ROI)).
Within the discriminative ROI defined as SEQ ID 518 a preferred region of interest including preferable CpG positions to be analyzed was identified (amplicon (AMP) 1730, see
Development of Cell-Type Specific qPCR Assay:
In AMP 1730, a detailed analysis was performed in order to develop a highly specific qPCR assay based on the use of amplification primers and probes. Amplification primers (forward and reverse) for bisulfite converted neutrophils specific AMP 1730 as well as probes were designed and tested (data not shown).
In order to develop a particularly preferred “perfect” primer system for the assay, primers were developed that do not correspond 100% to the original bisulfite converted sequence but include specific mismatches that surprisingly increased the specificity. Mismatches in the primer sequence are underlined and bold.
The technical specificity of the TpG-specific PCR-system was tested based on test-templates (see
The biological specificity of the neutrophils-specific qPCR-system was tested using certain sorted cell fractions as well as using whole blood samples (see Table 9). The established qPCR assay was found to be highly specific for neutrophils.
The relative amount of neutrophils in the sample is calculated from the number of bisulfite converted, neutrophil-specific marker copies and the sum of bisulfite converted and non-bisulfite converted neutrophil-specific marker copies in the sample as follows:
% neutrophils=no. of bisulfite converted neutrophil copies/no. of non-bisulfite converted neutrophil copies×100;
% neutrophils=253.67/(253.67+152.67)×100=62.43
The present assay is special in the sense that the amplification of the bisulfite-converted neutrophils-target-DNA using “common” fitted primers and standard PCR-protocols does not provide a sufficient result. Only after using amplification primers that were designed having a mutation (a “mismatch”) at strategic sites as identified herein, together with the use of a much higher Mg2+-concentration in the PCR allows for the efficient amplification of the neutrophils-target region.
In a next step a genomic plasmid standard can be designed and cell-specific assay-correction factor can be assessed (see Example 6).
The inventors developed non-bisulfite converted, genomic-plasmid standards as a normalization standard. One of these genomic plasmid standards comprises a marker region being specific for T-lymphocytes as well as a marker region being cell-type unspecific (GAPDH, housekeeping gene, detecting all cells, 100% of cells). Each single plasmid contains the same number of copies of these two marker regions (equimolar); two of these plasmids correspond to the number of DNA copies per one single immune cell and are therefore counted as one single cell. A stock solution containing defined numbers of said genomic plasmid molecules is used to determine the T-lymphocyte-specific assay-correction factor as well as to assess the absolute number of T-lymphocytes per microliter within an unknown blood sample.
In a first step, DNA of four human blood samples of unknown composition was isolated. This isolated DNA as well as the genomic plasmids of genomic plasmid standard were bisulfite treated. Following, the amount of copies of bisulfite converted T-lymphocyte-specific and GAPDH-specific marker regions were assessed by qPCR (Table 10, section B, C). These qPCR analyses were performed using a bisulfite-converted normalization standard (Table 10, section A) indicating the relative number of bisulfite-converted DNA as well as relative number of genomic plasmid copies containing the T-lymphocyte-specific marker region and the GAPDH marker region (Table 10 section B, C).
The relative amount of T-lymphocytes in percent within unknown blood samples is calculated as number of bisulfite converted T-lymphocyte-specific marker copies related to bisulfite converted GAPDH copies (Table 10, section B).
In a next step, T-lymphocyte-specific assay-correction factor based on said genomic plasmid standard was assessed (Table GR, section C). As described above, said genomic plasmid standard was bisulfite converted and number of plasmid copies assessed by qPCR using primers specific for bisulfite-converted marker regions for T-lymphocytes and for GAPDH. These qPCR analyses were also performed using the bisulfite-converted normalization standard (Table 10, section A). The efficiency of qPCR for T-lymphocytes and GAPDH should be equal as the novel genomic, non-bisulfite converted plasmid standard contains an equimolar amount of copies T-lymphocyte-specific and GAPDH-specific marker regions. Therefore, assessed deviation of genomic T-lymphocyte copy numbers from GAPDH copy numbers corresponds to differences in qPCR assay efficiencies.
e.g. Mean T-lymphocyte copy numbers=6058 vs. mean GAPDH copy numbers=5483
This deviation defines the cell-type assay-specific correction factor.:
Mean T-lymphocytes copy numbers/GAPDH copy numbers=6058/5483=1.1.
For T-lymphocytes an assay correction factor of 1.1 (average, n=2) was assessed (Table 10, section C). Correcting the relative amount of T-lymphocytes by factor 1,1 results in an absolute amount, e.g., of 26.24% T-lymphocytes within the unknown blood sample RD260314 (Table 10, section D).
absolute amount of T-lymphocytes=relative amount of T-lymphocytes/specific assay-correction factor
e.g.:28.87%/1.1=26.24% Treg cells
Additionally, the absolute number of T-lymphocytes per microliter within unknown blood samples was assessed (Table 10, section E). As described above, said genomic plasmid standard (stock solution of 6250 copies per microliter) was bisulfite converted and number of plasmid copies assessed by qPCR using primers specific for bisulfite-converted marker region for T-lymphocytes (section C). These qPCR was performed using the bisulfite-converted normalization standard (section A).
The amount of T-lymphocytes per microliter within unknown blood samples is calculated from relation of known, initial number of genomic plasmids of stock solution (6250 copies) and qPCR assessed number of copies of T-lymphocyte-specific marker within unknown blood samples (see section B) to qPCR assessed number of copies of genomic plasmid standard (see section C).
(See Table 10 below.)
| Number | Date | Country | Kind |
|---|---|---|---|
| 13173442.8 | Jun 2013 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2014/058087 | 4/22/2014 | WO | 00 |
| Number | Date | Country | |
|---|---|---|---|
| 61813802 | Apr 2013 | US |
