Immunohistochemical detection of global epigenetic DNA modifications using antibodies to 5-methylcytosine and 5-hydroxymethylcytosine.
Impact of antigen retrieval protocols.

doi:10.21203/rs.3.rs-2234522/v1

Download PDF

Method Article

Immunohistochemical detection of global epigenetic DNA modifications using antibodies to 5-methylcytosine and 5-hydroxymethylcytosine. Impact of antigen retrieval protocols.

https://doi.org/10.21203/rs.3.rs-2234522/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

The aim of this study was to compare three different pretreatment protocols for the detection of 5-methylcytosine (5-mC) and 5-hydroxymethylcytosine (5-hmC) in nuclear DNA. Several types of human biological samples were analyzed, including formalin fixed, paraffin embedded (FFPE) normal squamous epithelium, ethanol fixed cultured cells and metaphase chromosomes. The antigen retrieval methods included the low pH citrate and high pH Tris/EDTA protocols, as well as a method including a pepsin pretreatment step combined with an HCl DNA denaturation step. We saw a gradual increase in the detection levels of 5-mC and 5-hmC when going from citrate via Tris/EDTA to pepsin/HCl retrieval. While the citrate retrieval protocol was the least efficient for the immunochemical detection of 5-mC and 5-hmC, it did preserve nuclear morphology and enabled the visualization of intra- and internuclear differences in FFPE tissue and cell culture samples. By simultaneous fluorescent detection, we assessed the differences in distribution patterns for both 5-mC and 5-hmC. In addition, (hydroxy)methylation levels in FFPE material were quantified using confocal and non-confocal microscopic imaging. As a result, we observed a significant heterogeneity, as well as differences between the levels for 5-mC and 5-hmC, within and between nuclei in the different compartments of normal squamous epithelium.

Molecular Biology

methylation

hydroxymethylation

uterine cervix

squamous epithelium

anti 5-methylcytosine

anti 5-hydroxymethylcytosine

Methylation of deoxycytydine moietes is one of the major processes for the epigenetic regulation of mammalian DNA activity. It plays a critical role in the regulation of gene activity during fertilization, early embryogenesis and development, X chromosome inactivation and genomic imprinting, as well as in the organization of chromatin. Also, alterations in the methylation status have an impact on the activation or inhibition of oncogenes or tumour-suppressor genes. In general, DNA methylation turns off the activity of specific genes, while demethylation induces reactivation and expression of these genes (Paulsen and Ferguson-Smith 2001; Esteller 2007; Handy et al. 2011; Miller and Grant 2013; Moore et al. 2013).

DNA methylation occurs predominantly on CpG dinucleotides, with 60–90% of these dinucleotides being methylated in the mammalian genome. One of the most common epigenetic modifications involves DNA methyltransferase, which is responsible for methylation at the 5' position of cytosine residues of CpG dinucleotides, creating 5-methylcytosine (5-mC) (see for example Lu et al. 2020). Reactivation of epigenetically silenced genes is initiated by the TET (ten-eleven translocation) enzyme family, which can convert 5-mC to 5- hydroxymethylcytosine (5-hmC) and further cytosine derivatives, and as such plays a central role in the DNA demethylation process (Tahiliani et al. 2009; Yang et al. 2013; Pfeifer et al. 2014). He et al. (2021) recently published convincing evidence that 5hmC serves as a fundamental regulatory element affecting tissue-specific gene expression programs and functions.

Detection methods for gene-specific 5-mC and 5-hmC DNA modifications are mostly based on methylation specific polymerase chain reaction (PCR) assays or sequencing analyses of isolated DNA (Rein et al. 1998; Storebjerg et al., 2018). The overall hyper- and hypomethylation status of an individual cell can be determined by using antibodies directed against 5-mC and 5-hmC in immunohisto(cyto)chemical procedures applying fluorescence or brightfield microscopy (Singh et al. 2018; Kato et al., 2020). Immunohistochemical studies targeting 5-mC and 5-hmC have been performed on different types of biological samples, including tumour tissues (Szulwach et al. 2011; Singh et al. 2018; Kato et al. 2020). Discrepancies in staining patterns, the presence of positive nuclei adjacent to negative ones and the impact of immunocytochemical pretreatment protocols on the detection sensitivity of 5-mC and 5-hmC have been reported (Pendina et al. 2017; Wang et al. 2019; Estévez-Torres and Baigl 2011). The detectability of these modified nucleotides in tissues and cells may differ significantly with different DNA and protein compactness in different areas of the nucleus or tissue sample (Estévez-Torres and Baigl 2011), which may result in artificial negative results. In this respect, it is crucial to apply appropriate retrieval methods for an optimal availability of the antigenic sites for antibody binding.

The aim of our study was to compare and revisit three different pretreatment protocols to (simultaneously) detect 5-mC and 5-hmC. These retreival protocols are based on widely applied protein retrieval methods such as microwave treatment in a low pH citrate buffer or a high pHTris-EDTA buffer solution. A third retrieval protocol that was applied in this study, the Pepsin/HCl pretreatment procedure, was initially developed to detect DNA incorporated BrdU in formaline fixed, paraffin embedded (FFPE) tissues. These pretreatment methods exhibit significant differences in their efficacy to expose DNA and remove proteins, which is needed to enable antibody penetration and binding to the modified nucleotides. Several types of biological material were analyzed, including FFPE normal squamous epithelium tissue from the uterine cervix, ethanol fixed cell lines (HeLa, CaSki, SiHa) and human metaphase chromosomes from normal lymphocytes and the cancer cell cultures. Special attention was payed to the levels of 5-mC and 5-hmC in the different cellular compartments of the squamous epithelium, including the basal/parabasal (progenitor) cells and the differentiating epithelium. Detection of 5-mC and 5-hmC was performed by using single colour brightfield microscopy, while fluorescence detection enabled the simultaneous visualization and quantification of 5-mC and 5-hmC levels.

Tissues and cells

Formalin fixed and paraffin embedded (FFPE) tissue samples of normal squamous epithelium of the uterine cervix, obtained after colposcopy of 5 patients, were selected from the archives of the Pathology Department of the Maastricht University Medical Center, Maastricht, the Netherlands. Research on these tissue samples has been performed in accordance with the Code for Proper Secondary Use of Human Tissue in the Netherlands (http://www.federa.org/, update 2011: Human Tissue and Medical Research: Code of conduct for responsible use) and has been approved by the board of the Maastricht Pathology Tissue Collection at the Maastricht University Medical Centre (Registration Number MPTC 2011-05).

Four µm thick paraffin sections were collected on Superfrost Plus microscope slides (Thermo Fisher Scientific, Waltham, Massachusetts, United States). They were dewaxed in xylol, rehydrated in a descending alcohol series, and processed further using three different antigen retrieval pretreatment protocols (see below).

Three HPV-infected human epidermoid cervical cancer cell lines (HPV 16-positive CaSki and SiHa cells; HPV18-positive HeLa cells) were purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA). The cells were maintained in Dulbecco’s modified minimal essential medium (DMEM; Life Technologies/GIBCO BRL, Paisley, Scotland) supplemented with 10% fetal bovine serum and gentamycin. After culturing the adherent cells were trypsinized with Trypsin EDTA solution (Merck, T40049), washed twice with phosphate buffered saline (PBS; Janssen Chimica, Beerse, Belgium), centrifuged at 350xg for 5 minutes, the cell pellet resuspended in 70% ethanol (10⁶ cells/ml) and stored at -20°C. Ten µl of the cell suspensions were spotted on Superfrost Plus microscope slides and heated at 80⁰C for 15 minutes. After heating the slides were processed using three antigen retrieval pretreatment protocols (see below).

Metaphase preparations were prepared from the CaSki cell line and from peripheral blood lymphocyte samples. Ten ml of a cell suspension with 10⁶ cells/ml were added to 100 ml Roswell Park Memorial Institute (RPMI) 1640 Medium supplement with 10% fetal calf serum, antibiotics, and phytohemagglutinin M (Thermo-Fisher, Cat no 10576015). The cells were cultured for 72h, after which 80 µl of KaryoMax Colcemid Solution (GIBCO 15210-040) was added, incubated for 2h at 37°C and centrifuged for 5 minutes at 200 relative centrifugal force (rcf). Lymphocytes were treated with a hypotonic solution of 0.075 M KCl for 20–30 minutes at 37°C and then prefixed with 3:1 methanol: acetic acid for 5 min. Cells were collected by centrifugation and washed four times with 3:1 methanol: acetic acid. After the final centrifugation step, the supernatant was removed completely, and the cell pellet was resuspended in 10 ml of final fixative (3:1 methanol: acetic acid). The cell-suspension was stored at -20⁰C. Ten µl of the cell suspension was spotted onto Superfrost Plus microscopy slides to obtain the methaphase spreads and heated for 15 minutes at 80°C on a plate heater. Thereafter the slides were processed using three different antigen retrieval protocols (see below).

Antigen retrieval protocols

We used three different pretreatment steps for the visualization of 5-methylcytosine-modified and 5-hydroxymethylcytosine-modified nucleotides.

For the Citrate protocol the specimens were boiled for 10 minutes in 10 mM Na-citrate buffer (pH 6.0) in a microwave oven and then incubated for 20–30 minutes at room temperature to cool down.

For the Tris/EDTA (TE) protocol the specimens were boiled for 10 minutes in 10 mM Tris/HCl, 1mM ethylenediaminetetraacetic (EDTA) buffer (pH 9.0) in a microwave oven and then incubated for 20–30 minutes at room temperature to cool down.

For the Pepsin/HCl protocol the specimens were incubated in 100 µg/ml pepsin ( SIGMA P7000-25gr) in 0.01M HCl at 37°C (for 15 minutes for paraffin 2mg/ml pepsin in 0.1M HCl), and thereafter the specimens were incubated in 2M HCl at 37°C for 30 minutes. Subsequently slides were 3 times dip washed with PBS and with PBS/ 0.05% Tween-20 (PBT) for 5 min.

Primary anti 5-mC and 5-hmC antibodies

The following primary antibodies to 5-mC and 5-hmC were applied in this study:

The mouse monoclonal OptimAb anti 5-methylcytosine (anti 5-mC), (BI-MECY-0100, clone 33D3; Eurogentec, Seraing, Belgium). Clone 33D3 recognizes the modified base 5-methylcytidine in DNA and specifically distinguishes it from its normal DNA base counterpart.

The recombinant rabbit monoclonal anti-5-hydroxymethylcytosine antibody (anti 5-hmC) (clone RM236, ab214728; Abcam, Cambridge, UK). The antibody ab214728 reacts to 5-hydroxymethylcytosine in both single-stranded and double-stranded DNA. No cross reactivity with non-methylated cytosine and methylcytosine in DNA is reported.

The optimal detection levels of 5-mC and 5-hmC were determined by titration to reach an optimized antibody concentration for the different sample types. These dilutions were in the range of those recommended by the suppliers and as published in previous reports (Hewitt et al; 2014). As a negative control only incubation with the secondary antibody was performed.

Bright field microscopy

The FFPE tissue sections, pretreated as described above, were incubated with the primary antibodies mouse monoclonal anti 5-mC diluted 1:500 (OptimAb Anti-5-Methylcytosine (33D3) Eurogentec BI-MECY-0100)) or rabbit monoclonal anti 5-hmC diluted 1:500

(Recominant Anti-5-hydroxymethylcytosine antibody [RM236] (ab214728)) at 37°C for 30 minutes, washed 3 times 5 minutes with PBT and incubated at 37°C for 30 minutes with Biotinylated Horse anti-Mouse Ig, (Vector Laboratories, BA-2001 Burlington, Canada) or Biotinylated Goat anti rabbit IgG, (Vector Laboratories, BA-1000 Burlington, Canada) was applied in a 1:200 dilution in PBT with 5% normal goat serum. After washing with 2 times 5 minutes PBT and 1 time 5 minutes with PBS, samples were incubated at 37°C for 30 minutes with ABC complex, diluted 1:50 in PBS, (Vector Laboratories, PK-6010 Burlingame, CA, USA). Sections were washed and incubated with freshly prepared diaminobenzidine (Liquid DAB + substrate chromogen system; DAKO, K3467) for 7 minutes at room temperature and subsequently washed with MilliQ water. The sections were counterstained with hematoxylin (Hematoxylin Solution modified according to Gill II, Merck, 1.05175.0500), dehydrated in an ascending ethanol series and mounted with Entallan (Merck 1.07961.0100)

Fluorescence microscopy

All three sample types were incubated with the primary antibodies, either separately or simultaneously at 37°C for 30 minutes The dilutions used for the anti 5-mC were as follows: FFPE tissue sections 1:100, cell lines 1:500, metaphases 1:100. The dilutions used for the anti 5-hmC were1:100 for all three sample types. The samples were then washed 3 times 5 minutes in PBT and incubated at 37°C for 30 minutes with the secondary antibodies. The following secondary antibodies were applied:

Polyclonal Goat anti-Mouse conjugated with FITC, 1:100 dilution (Southern Biotech; Goat F(ab')2 Anti-Mouse Ig, Human ads-FITC; cat nr. 1012-02).

Polyclonal Goat anti-Rabbit conjugated with FITC, 1:100 dilution (Southern Biotech; Goat Anti-Rabbit Ig, Human ads-FITC; cat nr. 4010-02).

Polyclonal Goat anti-Mouse conjugated with Texas Red, 1:100 dilution (Southern Biotech; Goat Anti-Mouse IgG, Human ads-TXRD; cat nr.1030-07).

Polyclonal Goat anti-Rabbit conjugated with Texas Red, 1:100 dilution (Southern Biotech; Goat Anti-Rabbit Ig, Human ads-TXRD; cat nr. 4010-07).

Combinations of Goat anti-Mouse FITC/Goat anti-Rabbit Texas Red and Goat anti-Mouse Texas Red/Goat anti-Rabbit FITC were applied in the double label immunofluorescence studies, and were shown to provide similar images in the simultaneous detection of 5-mC and 5-hmC.

All the secondary antibodies were diluted in PBT, containing 5% normal goat serum. After washing 2 times 5 minutes with PBT and 1 time 5 minutes with PBS, the tissue sections and cell samples were dehydrated, and mounted with 90% glycerol in 0.02M Tris/HCl (pH 8.0), containing 0.02% NaN3, 2% DABCO and 0.5 µg/ml 4', 6-diamidino-2-phenylindole (DAPI; Merck 10236276001).

Imaging of bright field and fluorescence immunostaining

The bright field sections were scanned with a Ventana iScan HT slide scanner (Ventana Medical Systems, Inc. Tucson, Arizona, USA) and semi-quantitatively scored for staining levels of 5-mC and 5-hmC. Images were viewed and selected using Image Viewer Software (Ventana MS).

For non-confocal imaging the fluorescent signals were recorded with the Metasystems Image Pro System (black and white CCD camera; Sandhausen, Germany) mounted on top of a Leica DM-RE fluorescence microscope, equipped with Texas Red, FITC and DAPI single band pass filters for single color analysis. Images were obtained using an automatic integration time, allowing semi quantitative evaluation, and using the full dynamic range of the camera without signal intensity saturation. For the total fluorescence intensity of 5-mC and 5-hmC, as well as the ratio in fluorescence intensity per nucleus, the blue DNA staining (DAPI image) was used to set the nuclear contour for the measurement. The background was subtracted from the total fluorescence intensity measured. For a comparison of the fluorescent intensity between the different specimens a fixed integration time was applied, set at the automated integration time for the specimen with the highest intensity.

For confocal imaging the sections were imaged with a laser scanning confocal microscope (Leica SPE confocal) using LAS AF software (Leica Application Suite Advanced Fluorescence, version 2.7.3.9723). Imaging was done in acquisition mode xyz, with an ACS APO 63.0x1.30 oil immersion lens, a gain of 800-1000V, offset between − 0.5 and 0%, format of 1024x1024 pixels, speed of 400 Hz and frame average of 3. For DAPI excitation, a laser wavelength of 405 nm was used. FITC was excitated at 488nm in acquisition mode xyz. (emission filter 500–594 nm). TxRd was excited at 532 nm, emission filter 575–737 nm) Image J (https://imagej.nih.gov/ij/docs/menus/analyze.html; NIH, Bethesda, Maryland, USA) was used for further image analysis, processing and merging of the fluorescent images for reconstruction of the sections. An intensity curve was used to set the background correction. The background was subtracted from the total fluorescence intensity measured. For the total fluorescence intensity of 5-mC and 5-hmC, as well as the ratio in fluorescence intensity per nucleus, the blue DNA staining (DAPI image) was used to set the nuclear contour for the measurement (Schindelin et al. 2012).

Effect of the pretreatment protocols on the detection of 5-mC and 5-hmC in FFPE tissue sections of normal cervical squamous epithelium.

The effect of the three pretreatment methods on the staining intensities for 5-mC and 5-hmC in FFPE tissue sections of normal squamous cervical epithelium, using brightfield detection and the same primary antibody dilution (1:500) to enable a comparison between the protocols, is illustrated in Fig. 1. Immunostaining without any pretreatment did not result in a positive signal with both antibodies. A heterogeneous pattern of stained nuclei was recognized when applying the Citrate protocol, with concomitant strongly and weakly immunostained nuclei (Figs. 1A, D and F). With the TE and Pepsin/HCl protocol a stronger immunostaining reactivity was observed, but the inter-nuclear staining differences were less obvious and the staining intensities appeared to reach a maximum under the conditions applied (see Figs. 1B, E, H and Figs. 1C, F, I for TE and Pepsin/HCl retreival, respectively). Overall, the normal histological areas in the series of 5 tissue biopsies, studied with the three pretreatment protocols, revealed areas that exhibited this inter-nuclear staining difference, with the most homogeneous and strongest staining intensity being observed with the Pepsin/HCl pretreatment protocol.

The observation that for both 5-mC and 5-hmC a heterogenous staining pattern was recognized in the normal squamous epithelium and stretches of nuclei were incidentally recognized in the basal compartment of the epithelium that showed an alternating pattern of strongly and weakly stained nuclei, made us wonder whether individual nuclei could be detected that exhibit a mutually exclusive staining pattern for 5-mC and 5-hmC or even no staining at all for both modifications. Using a single label immunofluorescence approach for 5-mC (Figs. 2A-C), and 5-hmC (Figs. 2D-F) with the three different retrieval methods revealed differences in staining intensities between individual nuclei which ran parallel with the bright field staining patterns and heterogenous patterns could again be recognized. The Citrate pretreatment resulted in the weakest signal, while the intensity was higher after TE and Pepsin/HCl treatment, as concluded from visual inspection of non-confocal images. Similar results were obtained when using double-label immunofluorescence imaging for the simultaneous detection of 5-mC and 5-hmC in the tissue sections (data not shown).

Figure 3 shows a typical example of the simultaneous immunofluorescent detection of 5-mC and 5-hmC using Citrate as pretreatment protocol and non-confocal imaging. Colour differences between nuclei are shown in Fig. 3B, and on the basis of DAPI staining the contours for the nuclei or nuclear fragments are plotted in Fig. 3C. Since overlapping nuclei are not easily separated for quantification of immunofluorescence (see below), confocal imaging was applied which partly solved this problem by the improved resolution of the DAPI, FITC and Texas Red signals. The numbers in Figs. 3B and 3D-G refer to nuclei with variable combinations of Texas Red and FITC fluorescence detecting 5-mC and/or 5-hmC, respectively: 1: strong positivity for both 5-mC and 5-hmC (Red and Green equally positive), 2: 5-mC > 5-hmC (Red > Green), 3: 5-hmC > 5-mC (Green > Red); 4: both weak/no staining for 5-mC and 5-hmC (only DAPI positive). These images clearly illustrate again the heterogeneity of immunostaining for both types of epigenetic modifications in the squamous epithelium.

Quantification of 5-mC and 5-hmC immunofluorescence in normal squamous epithelium.

For a quantitative comparison of the immunostaining levels obtained with the different retrieval protocols, the immunofluorescence intensities for 5-mC and 5-hmC were measured within the DAPI contours. Table 1 provides a summary of the average fluorescence intensities for 5-mC and 5-hmC, measured using non-confocal imaging in 4 areas for each individual pretreatment protocol. The average fluorescence intensity for 5-mC increased gradually when going from Citrate via TE to Pepsin/HCl retreival. With Pepsin/HCl, the fluorescence intensity is a factor of 5 times higher as compared to Citrate pretreatment. For 5-hmC, the difference in average intensity between Citrate on the one hand and TE and Pepsin/HCl on the other is about a factor 3.5. Negative nuclei for 5-mC and 5-hmC were recognized after using the Citrate and TE protocol, next to nuclei negative for both 5-mC and 5-hmC.

The distributions of the fluorescence signal intensities measured in the individual nuclei for 5-mC and 5-hmC are plotted in Figs. 4A and 4B. These curves illustrate the difference in intensities for the three pretreatment protocols, clearly showing the lowest intensities for the Citrate and highest for the Pepsin/HCl pretreatment protocol. A wide range of intensities was observed, the widest distribution being observed with Citrate and the smallest after Pepsin/HCl.

The ratio between 5-mC and 5-hmC per individual nucleus is plotted in Figs. 4C and 4D. These curves show the dominant 5-hmC (5-hmC > 5-mC) staining in the left part of the curve (compare Fig. 3, nuclei indicated with number 3) and dominant 5-mC (5-mC > 5-hmC ) in the right part of the curve (compare Fig. 3, nuclei indicated with number 2). In the middle part of the curve, the intensities are more balanced (compare Fig. 3, nuclei indicated with number 1). Nuclei that showed low immunostaining for both 5-mC and 5-hmC and those exhibiting a low intensity for 5-mC or 5-hmC were excluded for these ratio measurements because they would strongly affect the plots and interpretation. When measured by confocal imaging the range of the fluorescence intensity ratios between 5-mC and 5-hmC appears to be smaller as compared to the non-confocal imaged fluorescence value (compare Figs. 4C and 4D). This quantitative difference was partly attributed to by the out-of-focus nuclei normally seen in an FFPE section. Both onfocal and non-confocal imaging showed, however, that within the squamous epithelium nuclei were present with a balanced, as well as intra- and internuclear fluorescence variation for 5-mC and 5-hmC.

Effect of the pretreatment protocols on 5-mC and 5-hmC detection in cervical cell lines.

The effect of the three pretreatment methods on ethanol-fixed CaSki cells is illustrated in Fig. 5. The highest fluorescence intensity for 5-mC and 5-hmC was again obtained with the Pepsin/HCl protocol (see inserts in Figs. 5A-F). The impact of the retrieval protocol was most evident for 5-mC, with increasing fluorescence intensity from Citrate via TE to Pepsin/HCl., For 5-hmC the highest signal intensity increase was found with the TE protocol as compared to the Citrate protocol. Supplemental Figure S1 illustrates the same analyses for the cervical cell lines HeLa and SiHa, which led to largely the same conclusions as for the analyses of the CaSki cell line.

Not only the fluorescence intensity, but also the staining patterns changed when comparing the three different retrieval protocols. The fluorescence staining pattern for 5-mC and 5-hmC showed a clear speckled pattern with the TE and Pepsin/HCL protocol (see for example Figs. 5C and 5F and Supplemental Figures S1C, F, I, L), while also a concomitant diffuse staining throughout the nuclei was obtained with all three procedures. This diffuse staining pattern was reduced for representation purposes in the optimized images shown in Fig. 5 and Supplemental Figure S1. To study the localization of 5-mC and 5-hmC relative to each other, in particular the speckled pattern, the modified nucleotides were detected simultaneously in a double-label immunofluorescence approach. Confocal microscopy was used to collect the speckled patterns in a single projection by merging the different confocal levels. Figure 6 illustrates the simultaneous detection of 5-mC and 5-hmC in CaSki using Citrate as retrieval method. The fluorescence detection of 5-mC (Fig. 6A, red) showed small and large speckles in the nucleus, occasionally being localized close to the periphery of the nucleus (arrowheads in Fig. 6A). For 5-hmC (Fig. 6B, green) mainly small, discrete speckles were observed, with a more uniform spot size and being randomly distributed throughout the nucleus. The merged images for 5-mC and 5-hmC (Figs. 6C, D) demonstrate that the speckles are not overlapping, as concluded from visual inspection and the absence of mixed-coloured (orange) speckles. Supplemental Figure S2 shows that a similar staining pattern as in CaSki cells can be seen in normal human lymphocytes.

Effect of the pretreatment protocols on 5-mC and 5-hmC detection in chromosome preparations.

The three pretreatment protocols were applied to metaphase chromosomes after a hypotonic shock and acetic acid treatment of G2M arrested cultured CaSki cells. During this preparation step of the metaphase chromosomes also interphase cells are collected on the slides. The subsequent retrieval methods significantly impacted the morphology of the nuclei and metaphase chromosomes. An acceptable morphology for interphase nuclei and metaphase chromosomes could only be obtained with the Citrate protocol. Both the TE and Pepsin/HCl pretreatment protocols partly disrupted the morphology of the interphase cells and the chromosomes. With the TE protocol chromosomes sometimes puffed out, while the Pepsin/HCl protocol resulted in the detachment of part of the metaphase chromosomes from the microscope slides. Furthermore, the Pepsin/HCl treatment diminished DNA staining with DAPI. Figure 7 illustrates staining patterns observed for 5-mC and 5-hmC in such interphase cells and metaphase chromosomes using the Citrate retrieval protocol. A speckled pattern is again seen in interphase cells (Figs. 7A and 7D) and metaphase chromosomes showed an inhomogeneous distribution pattern for both 5-mC and 5-hmC, with alternating positive and negative areas (Figs. 7B, C, E, F). Centromeric and telomeric regions often seemed to exhibit a stronger immunofluorescence staining than the rest of the chromosomes. The 5-mC staining showed to be symmetrically distributed over both sister chromatids within a chromosome (see Fig. 7C, F and I; arrows). For 5-hmC, however, we frequently noted that only one of two sister chromatids was labelled, indicating that this epigenetic modification is assymetricaly distributed over chromosomes immediately after metaphase (Figs. 7F and 7I). In the merged images (Figs. 7G, H, I), the speckled pattern is clearly seen for both 5-mC and 5-hmC in the interphase nuclei, while 5-hmC staining is clearly seen to be present on only one of the sister chromatids. To confirm that this phenomenon is not only seen in cancer derived cells, we also analysed metaphase chromosomes obtained from normal human lymphocytes. Supplemental Figure S3 shows that 5-mC staining is again strong in the centromeric and telomeric regions of some of the chromosomes in such lymphocyte preparations from a healthy individual (Supplemental Figure S3C; arrowhead and arrow, respectively). The number of speckles and also the staining intensity seemed reduced as compared to what was seen in the cancer cell line. For 5-hmC also in the lymphocyte chromosomes staining was seen in only one of the two chromatids (Supplemental Figures S3F and I).

Epigenetic modifications, resulting in the activation or silencing of specific genes or gene clusters often involve the methylation and/or demethylation of deoxycytidine moieties in promotor regions in the DNA. Such modifications have been found in a vast number of genes and have been correlated with a large variety of normal physiological processes, such as cell cycle regulation and cell differentiation. Changes in such methylation and/or demethylation processes have, for example, also been found during aging and (pre)neoplastic progression (Pappalardo and Barra, 2021). Since an increasing number of such gene specific methylation and demethylation events have been published in recent years, we wondered whether or not the immunohisto/cytochemical study of overall methylation and hydroxymethylation in tissues and cells can be provide information about, for example, the differentiation status of a specific tissue or cell type. An advantage of such an immunostaining approach is that a heterogeneous distribution of these epigenetic modifications in a given tissue or cell population can be easily recognized, allowing a correlation between the modification and its histo-morphological distribution. This is in general not possible with a more integrative test, such as PCR, ELISA or sequencing approach.

When detecting these epigenetic modifications by means of an immunostaining approach, optimal availability of the antigenic sites for antibody binding is a prerequisit. DNA compactness, proteins binding to the CpG islands and the fixation procedure may be factors that influence the epitope availability. Parameters like low salt, high temperature, acid/alkaline pH and enzymatic digestion of nuclear proteins may favour the denaturation of DNA and opening of the nuclear structure for epitope recognition.

A comparison of the three retrieval protocols used in the underlying study in bright field and fluorescent detection revealed that the Pepsin/HCl protocol was the most efficient in making 5-mC and 5-hmC available for detection with antibodies and the highest signal intensities were obtained with this method. However, this harsh protocol did affect the nuclear morphology, detectability of DNA using DAPI and diminished inter- and intracellular differences observed with the other two protocols. The Citrate (pH6.0) and to a lesser extent TE (pH9.0) protocols on the other hand left the nuclear organization intact and were superior in visualizing inter- and intranuclear differences in methylation and hydroxymethylation levels. The immunofluorescence intensity obtained with the TE (pH 9.0) protocol was higher than when using the Citrate (pH6.0) method. This difference could be the result of a more efficient opening of the nuclear structure under the mild alkaline conditions during microwave treatment.

The Pepsin/HCl protocol was initially developed for the detection of BrdU incorporated into the DNA during S-phase. To make the labeled DNA optimally accessible for the antibody that recognizes this substituted thymidine in single-stranded DNA, the tissues or cells were treated with DNAse or acid denaturation with 2M HCL, combined with enzymatic digestion using pepsin. These acidic conditions result in condensation of DNA based on protonation of nuclear phosphate groups, thus limiting enzymatic digestion using pepsin (Dover and Patel 1994; Schutte et al.1987). As mentioned before, in our study the Pepsin/HCl method resulted in DNA loss as detected by reduced DNA staining. Loss of morphology was particularly evident in ethanol fixed cells and metaphase preparations. The reduced DNA staining by the intercalating DNA dye DAPI was observed in all specimens, probably due to hydrolysis of the double stranded DNA, leading to denaturation and blocking the intercalation of DAPI.

Nuclear distribution of 5-mC and 5-hmC fluorescence.

The nuclei of FFPE squamous epithelium, of the cell lines and the intact nuclei in the metaphase preparations showed a punctate 5-mC fluorescent nuclear distribution pattern, occasionally localized at the periphery of the nuclei. The metaphase chromosomes showed often accumulations of 5-mC fluorescent staining in the centromeric and telomeric regions of specific chromosomes. On top of these intensely stained areas, a diffuse pattern throughout the nucleus and chromosome was observed. These observations support the high level of methylation in heterochromatin (containing silenced genes) and lower levels of methylation in euchromatic areas in the nucleus (Pappalardo and Barra, 2021). The localization of 5-hmC, which has not yet been described in detail in the literature, showed that the fluorescence signals for 5-hmC are smaller as compared to those for 5-mC, they are more evenly distributed over the nuclei and their frequency is higher. Our study showed that the fluorescent patterns for 5-mC and 5-hmC are not overlapping, which does, however, not exclude that at the molecular level within small chromosomal domains, the 5-mC and 5-hmC can be colocalized.

In the individual metaphase chromosomes differences in staining intensity for 5-mC and 5-hmC are recognized over the length of the chromosome, which cannot be directly correlated to DNA compactness. The methylation (5-mC) pattern is copied from the parent strand to the daughter strand during somatic cell division and is maintained during DNA replication by conversion of hemi-methylated sites back to symmetrically methylated molecules by DNA methyltransferase DNMT1, to prevent loss of the mark (Petryk et al. 2021). The 5-mC distribution therefore is stable with chromosomes being methylated in R-bands and the heterochromatic regions 1q12, 9q12, and 16q11.2. In our study, we clearly recognized the intense 5-mC staining of 1q12, the AT-rich DAPI positive centromere of chromosome 1. In cancer cells, this pattern is altered, and methylation in repetitive elements can even influence cell division (Nishiyama and Nakanishi 2021). Over-methylation during the cell cycle has been reported and in order to compensate for such hypermethylation, the cell actively demethylates the genome during G0/G1 in order to restore the methylation patterns to the correct state (He et al. 2021). The TET enzymes that are responsible for the formation of 5-hmC play a role in this demethylation process. The 5-hmC pattern is not simply copied during cell division (Shen and Zhang, 2013). Earlier studies (Kubiura et al. 2012) revealed a sharp difference in the levels of 5-hmC in the two opposite DNA strands of a given chromosome and a chromosome-wide cell-to-cell variability in mammalian cells. This asymmetric distribution and difference in chromosome hydroxymethylation is also recognized in our study of metaphases from the HPV positive cancer cell line and from healthy lymphocytes.

Inter- and intranuclear differences for 5-mC and 5-hmC.

A fluorescence intensity range of several magnitudes was typical for both 5-mC and 5-hmC in FFPE normal squamous epithelium. This large range resulted mainly from the intensity differences detected in cells belonging to the different compartments of the epithelium, e.g. the basal/parabasal, intermediate and superficial layers. Also, the methylation and hydroxymethylation levels can alternate in neighbouring cells within a single compartment or between different cellular compartments. It can be debated whether or not these differences are a direct consequence of differences in methylation and hydroxymethylation levels or result from differences in DNA compactness, in which case the pretreatment protocol dictates the availability of the modified nucleotides for immunohistochemical detection. Since heterogeneity is detected with all protocols, it is most likely that intrinsic differences in methylation and hydroxymethylation levels cause these observed heterogeneous staining patterns.

The observed heterogeneity makes a comparison with biochemical assays measuring methylation and hydroxymethylation levels difficult. Mixing of hyper(hydroxy)methylated cellular DNA with hypo(hydroxy)methylated cellular DNA will be averaged out in a biochemical assay. Only an immunocytological or immunohistological analysis enables the correlation of such a heterogeneous distribution with histo-morphological features of a tissue..

Acknowledgements

We thank Prof. Dr. Steffen Hauptmann (Dept of Pathology, Maastricht University Medical Center, Maastricht, The Netherlands) for providing paraffin sections of cervical tissues. Dr. John Engelen (Dept of Genetics and Cell Biology, Maastricht University Medical Center, Maastricht, The Netherlands) is acknowledged for providing the metaphase chromosome preparations of normal human lymphocytes. We are greatful to Dr. Jack Cleutjens (Dept of Pathology, Maastricht University Medical Center, Maastricht, The Netherlands) for help with scanning of the tissue slides. We would also like to acknowledge the input of Prof. Dr. Leendert Looijenga (Princess Máxima Center for Pediatric Oncology, Utrecht, The Netherlands) in the initial set up of the experiments.

Data Availability Statement

The study includes original data, A.H.N. Hopman confirms that he had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Competing interest statement

The author(s) declare no competing interests.

Dover R, Patel K (1994) Improved methodology for detecting bromodeoxyuridine in cultured cells and tissue sections by immunocytochemistry. Histochemistry 102 (5):383–387. doi:10.1007/BF00268909
Esteller M (2007) Cancer epigenomics: DNA methylomes and histone-modification maps. Nat Rev Genet 8 (4):286–298. doi:10.1038/nrg2005
Estévez-Torres A, Baigl D (2011) DNA compaction: fundamentals and applications. Soft Matter 7 (15):6746–6756. doi:10.1039/C1SM05373F
Handy DE, Castro R, Loscalzo J (2011) Epigenetic modifications: basic mechanisms and role in cardiovascular disease. Circulation 123 (19):2145–2156. doi:10.1161/circulationaha.110.956839
He B, Zhang C, Zhang X, Fan Y, Zeng H, Liu J, Meng H, Bai D, Peng J, Zhang Q, Tao W, Yi C (2021) Tissue-specific 5-hydroxymethylcytosine landscape of the human genome. Nat Commun 12 (1):4249. doi:10.1038/s41467-021-24425-w
Hewitt SM, Baskin DG, Frevert CW, Stahl WL, Rosa-Molinar E. Controls for immunohistochemistry: the Histochemical Society's standards of practice for validation of immunohistochemical assays. J Histochem Cytochem. 2014;62(10):693–697. doi:10.1369/0022155414545224
Kato M, Onoyama I, Kawakami M, Yoshida S, Kawamura K, Kodama K, Hori E, Cui L, Matsumura Y, Yagi H, Asanoma K, Yahata H, Itakura A, Takeda S, Kato K (2020) Downregulation of 5-hydroxymethylcytosine is associated with the progression of cervical intraepithelial neoplasia. PLoS One 15 (11):e0241482. doi:10.1371/journal.pone.0241482
Kubiura M, Okano M, Kimura H, Kawamura F, Tada M (2012) Chromosome-wide regulation of euchromatin-specific 5mC to 5hmC conversion in mouse ES cells and female human somatic cells. Chromosome Res 20 (7):837–848. doi:10.1007/s10577-012-9317-9
Lu Y, Chan YT, Tan HY, Li S, Wang N, Feng Y (2020) Epigenetic regulation in human cancer: the potential role of epi-drug in cancer therapy. Mol Cancer 19 (1):79. doi:10.1186/s12943-020-01197-3
Miller JL, Grant PA (2013) The role of DNA methylation and histone modifications in transcriptional regulation in humans. Subcell Biochem 61:289–317. doi:10.1007/978-94-007-4525-4_13
Moore LD, Le T, Fan G (2013) DNA Methylation and Its Basic Function. Neuropsychopharmacology 38 (1):23–38. doi:10.1038/npp.2012.112
Nishiyama A, Nakanishi M (2021) Navigating the DNA methylation landscape of cancer. Trends Genet 37 (11):1012–1027. doi:10.1016/j.tig.2021.05.002
Pappalardo XG, Barra V (2021) Losing DNA methylation at repetitive elements and breaking bad. Epigenetics Chromatin 14 (1):25. doi:10.1186/s13072-021-00400-z
Paulsen M, Ferguson-Smith AC (2001) DNA methylation in genomic imprinting, development, and disease. J Pathol 195 (1):97–110. doi:10.1002/path.890
Pendina AA, Efimova OA, Tikhonov AV, Chiryaeva OG, Fedorova ID, Koltsova AS, Krapivin MI, Parfenyev SE, Kuznetzova TV, Baranov VS (2017) Immunofluorescence Staining for Cytosine Modifications Like 5-Methylcytosine and Its Oxidative Derivatives and FISH. In: Liehr T (ed) Fluorescence In Situ Hybridization (FISH): Application Guide. Springer Berlin Heidelberg, Berlin, Heidelberg, pp 337–346. doi:10.1007/978-3-662-52959-1_35
Petryk N, Bultmann S, Bartke T, Defossez PA (2021) Staying true to yourself: mechanisms of DNA methylation maintenance in mammals. Nucleic Acids Res 49 (6):3020–3032. doi:10.1093/nar/gkaa1154
Pfeifer GP, Xiong W, Hahn MA, Jin SG (2014) The role of 5-hydroxymethylcytosine in human cancer. Cell Tissue Res 356 (3):631–641. doi:10.1007/s00441-014-1896-7
Rein T, DePamphilis ML, Zorbas H (1998) Identifying 5-methylcytosine and related modifications in DNA genomes. Nucleic Acids Res 26 (10):2255–2264. doi:10.1093/nar/26.10.2255
Shen L, Zhang Y (2013) 5-Hydroxymethylcytosine: generation, fate, and genomic distribution. Curr Opin Cell Biol 25 (3):289–296. doi:10.1016/j.ceb.2013.02.017
Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012 Jun 28;9(7):676 – 82. doi: 10.1038/nmeth.2019.
Schutte B, Reynders MM, Bosman FT, Blijham GH (1987) Effect of tissue fixation on anti-bromodeoxyuridine immunohistochemistry. J Histochem Cytochem 35 (11):1343–1345. doi:10.1177/35.11.3116075
Singh RK, Diaz PE, Binette F, Nasonkin IO (2018) Immunohistochemical Detection of 5-Methylcytosine and 5-Hydroxymethylcytosine in Developing and Postmitotic Mouse Retina. J Vis Exp (138). doi:10.3791/58274
Storebjerg TM, Strand SH, Høyer S, Lynnerup AS, Borre M, Ørntoft TF, Sørensen KD (2018) Dysregulation and prognostic potential of 5-methylcytosine (5mC), 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC) levels in prostate cancer. Clin Epigenetics 10 (1):105. doi:10.1186/s13148-018-0540-x
Szulwach KE, Li X, Li Y, Song CX, Han JW, Kim S, Namburi S, Hermetz K, Kim JJ, Rudd MK, Yoon YS, Ren B, He C, Jin P (2011) Integrating 5-hydroxymethylcytosine into the epigenomic landscape of human embryonic stem cells. PLoS Genet 7 (6):e1002154. doi:10.1371/journal.pgen.1002154
Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y, Agarwal S, Iyer LM, Liu DR, Aravind L (2009) Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 324 (5929):930–935. doi:10.1126/science.1170116
Wang J, Su Y, Tian Y, Ding Y, Wang X (2019) Characterization of DNA hydroxymethylation profile in cervical cancer. Artif Cells Nanomed Biotechnol 47 (1):2706–2714. doi:10.1080/21691401.2019.1634578
Yang H, Liu Y, Bai F, Zhang JY, Ma SH, Liu J, Xu ZD, Zhu HG, Ling ZQ, Ye D, Guan KL, Xiong Y (2013) Tumor development is associated with decrease of TET gene expression and 5-methylcytosine hydroxylation. Oncogene 32 (5):663–669. doi:10.1038/onc.2012.67

Table 1 is available in the Supplementary Files section.

floatimage1.png
Table 1 Quantification of immunofluorescence intensities (Arbitrary Units) for 5-mC and 5-hmC nuclei of 4 FFPE squamous epithelium samples, pretreated with the 3 different retrieval protocols. In the Pepsin/HCl pretreatement, the contour plotting for immunofluorescence measurement is hampered by the weak DAPI staining of the nuclei, so the number of negative nuclei can not be assessed. Nuclei with a fluorescence intensity < 5% of the average intensity were not included in the list of values to calculate the final average fluorescence intensity.
Supplementalfigure1.jpg
Comparison of immunofluorescent staining of 5-mC and 5-hmC in HeLa and SiHa cells using the Citrate, TE and Pepsin/HCl retrieval methods. A-L) Detection of 5-mC and 5-hmC after non-confocal imaging and optimization of fluorescence intensity. The fluorescence images were captured using a fixed integration time of 1 sec (inserts) and similar secondary antibody dilutions.
Supplementalfigure2.jpg
Simultaneous detection of 5-mC and 5-hmC in ethanol fixed normal human lymphocytes using confocal microscopy, pretreated with the Citrate protocol. A) 5-mC in green (FITC), B) 5-hmC in red (Texas Red), C) merged image of 5-mC and 5-hmC and D) merged image with DAPI included. Note the difference in spot size between 5-mC and 5-hmC and the difference in the frequency of fluorescent spots for both modifications. The arrowhead points to a large fluorescent spot for 5-mC (A) and the arrow to a small spot for 5-hmC (B); the spots are not overlapping but close to each other (C).
Supplementalfigure3.jpg
Simultaneous immunofluorescent detection of 5-mC and 5-hmC in TE-treated lymphocyte metaphase preparations. A, D and E) lymphocyte interphase nuclei and B, E and H metaphase chromosomes showing 5-mC in green (FITC; A-C, G-I), 5-hmC in red (Texas Red, D-I ) and merged images with blue DNA staining using DAPI in all images. In C, F and I selected metaphase chromosomes (see arrowheads in B and E) show the telomere (arrows) and centromere (arrowheads) staining for 5-mC in both chromatids (C). For 5-hmC the positive staining of only one of the chromatids is seen in F and I (see parallel arrows).

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Immunohistochemical detection of global epigenetic DNA modifications using antibodies to 5-methylcytosine and 5-hydroxymethylcytosine. Impact of antigen retrieval protocols.

Archived Versions:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Tissues and cells

Antigen retrieval protocols

Primary anti 5-mC and 5-hmC antibodies

Fluorescence microscopy

Imaging of bright field and fluorescence immunostaining

Results

Discussion

Declarations

References

Table 1

Supplementary Files

Archived Versions:

Version 1