Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervix

Jobran M Moshi; Monique Ummelen; Frank Smedts; Frans C S Ramaekers; Anton H N Hopman

doi:10.1371/journal.pone.0297008

. 2024 Apr 18;19(4):e0297008. doi: 10.1371/journal.pone.0297008

Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervix

Jobran M Moshi ^1,², Monique Ummelen ¹, Frank Smedts ³, Frans C S Ramaekers ¹, Anton H N Hopman ^1,^*

Editor: Ricardo Ney Oliveira Cobucci⁴

PMCID: PMC11025792 PMID: 38635731

Abstract

Methylation and hydroxymethylation of cytosine moieties in CpG islands of specific genes are epigenetic processes shown to be involved in the development of cervical (pre)neoplastic lesions. We studied global (hydroxy)methylation during the subsequent steps in the carcinogenic process of the uterine cervix by using immunohistochemical protocols for the detection of 5-methylcytosine (5-mC) and 5-hydroxymethylcytosine (5-hmC) in paraffin-embedded tissues of the normal epithelia and (pre)malignant lesions. This approach allowed obtaining spatially resolved information of (epi)genetic alterations for individual cell populations in morphologically heterogeneous tissue samples. The normal ectocervical squamous epithelium showed a high degree of heterogeneity for both modifications, with a major positivity for 5-mC in the basal and parabasal layers in the ectocervical region, while 5-hmC immunostaining was even more restricted to the cells in the basal layer. Immature squamous metaplasia, characterized by expression of SOX17, surprisingly showed a decrease of 5-hmC in the basal compartments and an increase in the more superficial layers of the epithelium. The normal endocervical glandular epithelium showed a strong immunostaining reactivity for both modifications. At the squamocolumnar junctions, a specific 5-hmC pattern was observed in the squamous epithelium, resembling that of metaplasia, with the typical weak to negative reaction for 5-hmC in the basal cell compartment. The reserve cells underlying the glandular epithelium were also largely negative for 5-hmC but showed immunostaining for 5-mC. While the overall methylation status remained relatively constant, about 20% of the high-grade squamous lesions showed a very low immunostaining reactivity for 5-hmC. The (pre)malignant glandular lesions, including adenocarcinoma in situ (AIS) and adenocarcinoma showed a progressive decrease of hydroxymethylation with advancement of the lesion, resulting in cases with regions that were negative for 5-hmC immunostaining. These data indicate that inhibition of demethylation, which normally follows cytosine hydroxymethylation, is an important epigenetic switch in the development of cervical cancer.

Introduction

Genomic methylation is involved in the regulation of a plethora of epigenetic processes in both prokaryotes and eukaryotes. In prokaryotes, adenine is the most important methylation target, while such modifications in mammalian cells mostly involve cytosine methylation of CpG residues to create 5-methylcytosine (5-mC). CpG dinucleotides are found across the entire genome and are often clustered in islands where the cytosine is methylated. CpG DNA methylation significantly changes during normal growth and differentiation, but also during ageing and cancer formation. Active demethylation has been shown to be mediated by oxidation of 5-mC by ten-eleven translocation (TET) proteins, which convert 5-mC into 5-hydroxymethylcytosine (5-hmC) [1, 2].

The presence of 5-mC has been described in several normal tissues such as muscle, lung, kidney, heart, brain, gastric tract and embryonic cells [3–7]. Numerous genes have been found to undergo hypermethylation in cancer. The genes that are susceptible are the genes involved in cell cycle regulation (e.g. p16^INK4a, Rb), genes associated with DNA repair (e.g. BRCA1, apoptosis (DAPK, TMS1), drug resistance, detoxification, differentiation, angiogenesis, and metastasis. Although certain genes such as RASSF1A and p16 are commonly methylated in a variety of cancers, other genes are methylated in specific cancers [8, 9].

Usually, during carcinogenesis, the methylation level around specific genes is increased. Still, global methylation of the genome is low because cells have a natural tendency to normalize methylation via the active demethylation mechanism described above [10]. Several studies have examined 5-mC as a global DNA methylation marker in different cancer types such as cervix, colon and oral cavity malignancies [11, 12]. In addition, 5-hmC has been detected in different tissues [13] and cancer types such as those from liver, lung, pancreas, breast, brain, prostate and ovary [4, 5, 14].

To date, several publications have indicated that 5-mC and 5-hmC may become diagnostic and prognostic biomarkers for cervical cancer [15–17]. In addition, several studies have reported that the methylation and demethylation processes can impact the origin and progression of premalignant lesions, especially those in the cervix [11, 12]. Cervical cancer, with human papillomavirus (HPV) infections being the primary causal factor, is the second most frequent gynecological cancer among women globally. In 2018 the prevalence of cervical cancer rate was 570,000 new cases, which increased to reach 604,000 new cases in 2020 despite the introduction of the HPV vaccination program in several countries worldwide [18, 19]. One model for the carcinogenic process involves HPV infection of metaplastic squamous epithelium or the cervical reserve cells, which are believed to be a progenitor cell population capable of generating squamous-type epithelium. Besides that, these cells can differentiate into columnar cells, thereby replenishing the endocervical glandular epithelium. Upon persistent HPV infection, these different cell and tissue types can produce cervical intraepithelial neoplasms (CIN lesions) and finally invasive squamous cell carcinoma or adenocarcinoma in situ (AIS) and finally adenocarcinoma [20, 21].

The stepwise process of carcinogenesis can be ideally studied in the uterine cervix, where subsequent stages of metaplastic, dysplastic and (micro)invasive changes can be easily accessed by biopsies of the transformation zone (TZ) and then be studied in a histo-morphological context. Since several studies have shown that multiple genes are hypermethylated during the process of cervical carcinogenesis (for an overview see [4, 22–24]), we wondered whether the detection of the global methylation and hydroxymethylation status could be used as indicators of epigenetic switches occurring during cervical carcinogenesis. Using an immunohistochemical approach we obtained spatially resolved information that allowed us to correlate the (hydroxy)methylation status to the position of individual cells in morphologically heterogeneous tissue samples derived from normal cervix and its (pre)malignant derivatives.

Materials and methods

Tissues

In this retrospective study of archived samples, all data were fully anonymized before the study was started. The ethics committee waived the requirement for informed consent. Formalin-fixed and paraffin-embedded (FFPE) tissue samples of the uterine cervix, obtained after colposcopy of 60 patients with cytological indications of a (pre)malignant lesion (Table 1), were selected from the archives of the Departments of Pathology of the SSZOG (Stichting Samenwerking Ziekenhuizen Oost Groningen) and of the Reinier de Graaf Hospital Delft. These tissues are used in the context of routine diagnostics, after which they are archived. Samples were collected between 2008 and 2020. Research on these anonymized archival tissue samples was performed in accordance with the Declaration of Helsinki and with the Code for Proper Secondary Use of Human Tissue in the Netherlands (http://www.federa.org/, update 2011: Human Tissue and Medical Research).

Table 1. Selected formalin-fixed and paraffin embedded tissue samples of (pre)malignant lesions of the uterine cervix.

	Patient diagnosis^*	Nr of patients	Analyzed histological areas	Areas analyzed
			Normal squamous epithelium^#	73
			Normal glandular epithelium^#	26
			Normal SCJ	8
Squamous lesions	CIN 1	6	CIN 1	8
	CIN 2	3	CIN 2	7
			CIN 1⁺	1
	CIN 3	25	CIN 3	55
			<CIN 3⁺	22
Glandular lesions	AIS	12		21
Glandular lesions	AdC	14		18
	Total	60		239

Open in a new tab

*: Patients were diagnosed on basis of the most severe CIN grade present in the lesion.

^#: No healthy controls were examined, normal histological areas present in the biopsies were analyzed. These areas were present in the patient samples and recognized by the pathologist and by molecular analysis. The normal areas showed to be HPV and p16 negative.

⁺: coinciding areas with less severity. CIN: cervical intraepithelial lesion; AIS: adenocarcinoma in situ; AdC: adenocarcinoma.

SCJ: normal squamocolumnar junction (transition between squamous epithelium and glandular/columnar epithelium), no preneoplasia present in the junction.

Tissue processing for immunohistochemistry

Four μm thick paraffin sections were collected on Superfrost Plus microscope slides (Thermo Fisher Scientific, Waltham, Massachusetts, United States), dewaxed in xylol, rehydrated in a descending alcohol series and processed further using the citrate retrieval protocol. For comparison of staining intensities, a few slides of squamous and glandular epithelia were subjected to the Tris/EDTA pH 9.0 (TE) or Pepsin/HCl antigen retrieval protocol as described before [25]. For the citrate protocol, the specimens were boiled for 20 minutes in 10 mM Na-citrate buffer (pH 6.0) in a microwave oven and then left at room temperature for 30 minutes to cool down. Subsequently, slides were 3 times dip washed with demineralized water. Before starting with the immunohistochemical incubations the specimens were washed with PBS/0.05% Tween-20 (PBST) for 5 min.

Antibodies and immunohistochemical protocols

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.

These two antibodies were diluted 1:100 and detected in single- and double-label immunofluorescence protocols as detailed in S1 Table (S1A), as well as in a brightfield immunohistochemical protocol as detailed in S1 Table (S1B).

As negative controls, incubations were performed with only the secondary antibodies.

Additional antibodies directed against SOX2, SOX17, p16, cytokeratin 7, cytokeratin 17 and Ki-67, as well as optimized detection protocols used for the bright field immunohistochemical characterization of the different cell and tissue types in (pre)malignancies are summarized in S1 Table.

Fluorescence and bright field microscopy

Fluorescence microscopy: After subsequent incubations with the primary and fluorochrome-conjugated secondary antibodies as described before [25], the tissue sections were washed 2 times 5 minutes with PBST and once with PBS for 5 minutes, then 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).

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 single band pass (BP) filters for single color analysis of DAPI (excitation BP340-380; emission BP450-490), FITC (excitation; BP460-500; emission BP512-490), Texas Red (excitation BP540-580; emission BP593-668) and TRITC, used to view the Alexa 555 conjugate (excitation BP510-560; emission BP572-647). Images were obtained using an automatic integration time, allowing quantitative evaluation, and using the full dynamic range of the camera without signal intensity saturation [25].

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%, the 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 excited at 488nm (emission filter 500–594 nm), while Texas Red and Alexa 555 were 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 the reconstruction of the sections.

Bright field microscopy: After antibody incubations as detailed in S1 Table (S1B) for bright field microscopy, the sections were washed in PBS 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). The bright field sections were scanned with a Ventana iScan HT slide scanner (Ventana Medical Systems, Inc. Tucson, Arizona, USA), and images were viewed and selected using Image Viewer Software (Ventana MS).

Results

For the visualization of global methylation and hydroxymethylation of DNA, the citrate pH 6.0 pretreatment protocol was applied as previously described [25]. In this previous study we have also shown that a wide range of fluorescence intensities was typical for the immunohistochemical detection of 5-mC and 5-hmC in normal squamous epithelium. This was observed and quantified by fluorescence intensity measurements in cells belonging to the different compartments of the epithelium, i.e. the basal/parabasal, intermediate and superficial cell layers. Together with the result from simultaneous detection of 5-mC and 5-hmC within individual nuclei, these observations suggested that immunohistochemical detection levels of methylation and hydroxymethylation can not only differ between different cellular compartments but even between neighboring cells within a single cell compartment. In the underlying study we were confronted with the observation that not only differences in signal intensities but also typical differences in staining patterns can be found in both normal and premalignant lesions. Because of this heterogeneous immunofluorescence reactivity noticed for 5-mC and 5-hmC, with cells showing staining for both or only one of the two, 5-mC and 5-hmC were mostly stained simultaneously and are presented in single- and merged-color images for visual inspection. We classified the immunofluorescence intensities for 5-mC and 5-hmC as negative, weak, intermediate or intense/strong, with staining for 5-hmC in stromal tissue always being intense.

5-mC and 5-hmC immunostaining levels in normal squamous epithelium

Fig 1A depicts a H&E-stained stretch of normal squamous ectocervical epithelium with underlying stroma. In the right part of the epithelium, a gland is recognized close to the basement membrane (see Fig 1B). In Fig 1C and 1D an overview of the immunofluorescence detection of 5-mC and 5-hmC is shown at low magnifications, while higher resolution immunofluorescence images for the lower compartments of the epithelium are shown in Fig 1E–1K. Immunoreactivity for 5-mC is seen from the basal/parabasal layer up to the superficial layer and small differences in staining intensity were noticed over the length of the stretch of epithelium. Negative and weakly stained nuclei were seen to be intermingled with strongly stained nuclei (Fig 1E, 1H, 1K and 1N).

For 5-hmC, however, differences in intensities were recognized throughout the different layers of the epithelium. Intense staining is frequently seen in the basal cell layer which differs in intensity from the layers above up to the intermediate cell layer (Fig 1C and 1F; this pattern was seen in 21 out of 73 normal areas analyzed). These intensity differences were also found when other pretreatment retrieval procedures were applied to the slides (results not shown). The epithelium in areas more closely located to the squamo-columnar junction (SCJ) showed a more intense 5-hmC staining in the intermediate cell layer as compared to the basal/parabasal layers (Fig 1I and 1L). This weaker immunostaining of 5-hmC in the basal cell compartment of the squamous epithelium, mainly recognized in the areas where glands were underlying the epithelium, instigated us to study in more detail the topological distribution of 5-hmC at the position of the SCJ. Fig 2A and 2B show the H&E-stained junction between squamous and columnar epithelium, with the squamous epithelium adjacent to the SCJ showing intense 5-mC staining from the basal/parabasal compartment up to the superficial layer of the epithelium (Fig 2C and 2G), while for 5-hmC a weak staining is seen in the basal/parabasal layer compared to the superficial layers (Fig 2D and 2F). Fig 2A also shows that two histologically different types of normal epithelia can be recognized distal from the SCJ (areas 1 and 2), the right part of which represents metaplastic squamous epithelium, which is supported by the presence of a gland underlying the epithelium, indicating a new SCJ. We defined metaplasia as normal epithelium since it represents the transition of endocervical glandular epithelium into squamous epithelium, a normal physiological process that can be detected in cervical biopsies. Metaplastic epithelium characterizes the transformation zone and bridges the epithelium between the original and new SCJ. Fig 3A shows a stretch of (immature) metaplastic epithelium with the typical co-expression of SOX2 and SOX17 (Fig 3B and 3C, respectively), and at the same time a negative p16 staining (Fig 3D), indicating that the epithelium was not infected by HPV. For an overview showing the transition in these immunostaining patterns between normal and metaplastic epithelium we refer to S1 Fig. Also typical for metaplastic squamous epithelium, cytokeratin7 was expressed in the basal/parabasal and superficial compartments, while cytokeratin17 was found solely in the basal/parabasal layer. Immunostaining for 5-hmC was mainly seen in the intermediate and upper compartment and was weak to negative in the basal/parabasal layers (Fig 3G and 3H), similar to what was observed in most of the SCJ areas.

Fig 2 — A) H&E-stained stretch of squamous epithelium with SCJ, higher magnification in B (arrow). C-E) SCJ area shown in box in A, with arrows pointing at the junction between squamous and columnar epithelium. In (C) red immunofluorescence of 5-mC, (D) 5-hmC in green and (E) merged images of 5-mC, 5-hmC and blue DAPI counterstaining. F) Higher magnification of 5-hmC immunostaining as shown in D. G) Merged images for 5-mC (red) and 5-hmC (green) with DAPI counterstaining in another area showing the SCJ. The lines in E, F and G mark the position of the basement membrane. Note also the intense 5-hmC immunostaining in the stromal cell compartment. Scale bars indicate 500 μm in A, 100 μm in C (same magnifications in D and E), and 20 μm in B, F and G.

Fig 3 — A) H&E staining of metaplastic epithelium with B) immunostaining for SOX2, C) SOX17, D) p16, E) cytokeratin 7, and F) cytokeratin 17. G) Immunofluorescent staining of 5-hmC in green and DNA with DAPI in blue. H) Higher magnification of the boxed area indicated in G. The lines mark the position of the basement membrane. The scale bar indicates 100 μm in A (same magnifications in B-G) and 20 μm in H.

5-mC and 5-hmC immunostaining levels in the normal columnar epithelium and reserve cells

Intense immunofluorescence staining could be observed for both 5-mC and 5-hmC in normal columnar epithelium (Fig 4). The epithelial nuclei showed a significant heterogeneity in staining along the trajectory of the columnar epithelium, with an even more extensive range in 5-hmC intensities compared to 5-mC. This holds for the columnar epithelium close to and further away from the SCJ.

Reserve cells present in the endocervix beneath the columnar epithelium are considered progenitor cells for the squamous epithelium and can differentiate into (immature) metaplastic squamous epithelium. Fig 5A–5F illustrate the phenotype of these cells with a positive expression profile for SOX17, cytokeratin 7 (CK7) and 17 (CK17) but negative for p16 and SOX2. The columnar epithelial cells are SOX17 and cytokeratin 7 (CK7) positive, but negative for cytokeratin 17 (CK17), p16 and SOX2. Occasionally Ki-67 positive cells could be recognized in the bilayer of columnar and reserve cells. These reserve cells are strongly positive for 5-mC (Fig 5G), but only weak or even negative for 5-hmC (Fig 5H). The merged image of 5-mC, 5-hmC and DAPI clearly shows the methylation and hydroxymethylation level differences in the reserve cells and the overlying columnar cells (Fig 5I).

Fig 5 — A-F) Bright field immunostaining of p16 (A), SOX2 (B), SOX17 (C), cytokeratin 17 (CK17) (D), cytokeratin 7 (CK7) (E) and Ki-67 (F). Note the intense staining of the reserve cells for CK17 in D, while columnar epithelium and the reserve cells are SOX17 and CK7 positive. G-I) Immunofluorescence staining of 5-mC (G), 5-hmC (H) and merged images of 5-mC and 5-hmC with counterstaining of DNA with DAPI (I). The line in I marks the position of the basement membrane. Scale bar in A indicates 20 μm (same magnifications in B-I). The columnar epithelial cells and reserve cells are indicated as CC and RC, respectively.

Complex immunostaining patterns for 5-mC and 5-hmC in squamous intraepithelial lesions

Fig 6 summarizes different epithelial immunofluorescence distribution patterns that were recognized for 5-mC and 5-hmC in areas marked by the pathologist as CIN1, CIN2 and CIN3. We want to stress that CIN1 lesions exhibited similar patterns for 5-mC and 5-hmC as recorded for the normal squamous epithelium (Fig 6A–6D; see also S2 Fig), with intense (3 out of 8 areas analyzed) to weak (5 out of 8 analyzed) staining in the basal compartment and positive immunostaining in the intermediate and superficial layers. On top of these distributions both 5-mC and 5-hmC positive nuclei were intermingled with weakly stained and negative nuclei recognized in 5 out of 8 CIN1 areas analyzed. These differences were most pronounced for 5-hmC as compared to 5-mC. The distribution pattern as shown for CIN1 in Fig 6A–6D, exhibiting the low intensity for 5-hmC in the basal compartment, was frequently observed in patients diagnosed with a higher CIN grade, as illustrated in Fig 6E–6H, showing weak staining in the lower compartment, but a higher reactivity in the upper layers. This pattern was recognized in 30 out of 84 areas graded higher than CIN1 (see Table 1: the 84 areas comprised 55 CIN3, 7 solitary CIN2 and 22 coinciding CIN2 areas in CIN3).

The CIN2 and CIN3 areas showed also for both 5-mC and 5-hmC immunostaining throughout the entire epithelium, next to the low intensity staining for 5-hmC in the lower compartment of the epithelium (Fig 6I and 6L; 34 out of 84 areas). Finally, in about 20% of the patients diagnosed with CIN3 (5 out of 25 patients), a (very) weak staining for 5-hmC was observed throughout the entire epithelium (Fig 6O), while most nuclei were strongly positive for 5-mC (Fig 6N). Quantitative data were obtained to support the different distribution patterns as illustrated in Figs 1 and 6 (and S2 Fig). As shown in S3 Fig these quantitative analyses support the semiquantitative observations that largely three different 5-hmC distribution patterns can be distinguished in CIN lesions: 1) a pattern with a high intensity staining for 5-hmC in the basal cell layer, a reduced intensity in the parabasal cell layer and increasing intensity in the direction of differentiation of the epithelium to the more upper layers of the epithelium seen in CIN1; 2) a pattern with a low intensity staining for 5-hmC in the lower compartment of the epithelium including the bc, pbc layers and lower compartment of the icl seen in CIN1-3; 3) a complex staining pattern for 5-mC and 5-hmC throughout the epithelium, with a strong staining at the bc layer for both 5-mC and 5-hmC seen in CIN2-3.

In summary, CIN lesions showed a heterogeneous staining pattern for 5-hmC. When studying CIN3 cases in more detail, combinations of the patterns described above were found next to each other (20 out of 55 areas; see also S4 Fig for examples). Also, when performing confocal microscopy, CIN lesions were found to show areas where 5-mC and 5-hmC immunostainings were mutually exclusive, with nuclei exhibiting intense staining for 5-hmC, while being negative for 5-mC and vice versa (see also S5 Fig).

Immunostaining results of 5-mC and 5-hmC in premalignant endocervical glandular lesions

Cervical adenocarcinoma in situ (AIS) lesions, previously analyzed for SOX17 gene expression and methylation [26] showed a strong immunostaining for 5-mC in all glandular structures, both normal and neoplastic. Strikingly, compared to the normal glandular structures, a drastic decrease in 5-hmC immunostaining was observed in nearly all regions with AIS, which were recognized by p16 positivity and partial SOX17 negativity (Fig 7). Small areas of normal (p16 negative, SOX17 positive) glandular epithelium inside the AIS did, however, show the typical, strong 5-hmC immunoreactivity.

Fig 7 — Merged image of 5-mC in red and 5-hmC in green with blue DNA counterstaining using DAPI. Note the dramatic decrease in 5-hmC immunofluorescence in the AIS structures (gland 1 normal columnar epithelium, glands 2 and 3 containing AIS structures). The arrows in A and B point to normal columnar epithelium (nce) adjacent to AIS cells, the latter being identified by p16 positivity (B) and SOX17 positivity (C). D, E and F, higher magnifications of boxed areas in A, B and C. Normal columnar epithelium is characterized by the absence of immunostaining for p16 and SOX17. Note that the images in A, B and C, although being derived from the same area in the AIS lesion, do not represent consecutive sections, resulting in slight differences in cellular distribution patterns. G-J) Brightfield images of 5-hmC immunostaining patterns of a cervical AIS, showing mutual exclusive staining for 5-hmC and p16 in consecutive tissue sections. G and I) show 5-hmC positivity only in the normal columnar epithelium (nce); H and J) show p16 positivity only in the AIS region. I and J, higher magnifications of boxed areas in G and H. Scale bars indicate 30 μm.

Higher magnification images in the separated colors as depicted in Fig 8 illustrate the differences in staining intensity for 5-mC and 5-hmC in more detail. When studying AIS lesions from different patients some variation is seen for 5-hmC staining intensity, but in all 12 AIS cases a drastic decrease of hydroxymethylation was obvious. Also, when using TE and pepsin/HCl as antigen retrieval methods, to exclude that the observed reduction of staining for 5-hmC resulted from increased DNA compactness of the 5-hmC domains, such a decrease was still clearly visible (data not shown).

Immunostaining results for 5-mC and 5-hmC in cervical adenocarcinomas

Also, in nearly all 14 adenocarcinoma cases studied a dominant fluorescence for 5-mC was observed, while 12 out of 14 patients showed a greatly reduced immunostaining for 5-hmC, indicating a strong reduction of hydroxymethylation in cervical adenocarcinomas. Fig 9A–9C illustrate this phenomenon at a low magnification, but a decrease of hydroxymethylation is even more evident at higher magnification (Fig 9D and 9E). 5-mC levels were not altered in all adenocarcinomas when comparing fluorescence intensities in normal columnar, malignant and stromal nuclei. In a minor fraction of the adenocarcinoma cases (2 out of 14), however, an intense staining for 5-hmC was seen (Fig 9H and 9K), while 3 out of 14 adenocarcinoma cases showed heterogeneity for the loss of hydroxymethylation levels (See S6 Fig).

Fig 9 — A, D, G and J) 5-mC; B, E, H and K) 5-hmC; C, F, I and L) merged images of 5-mC and 5-hmC with DAPI counterstaining. Note the weak/negative staining for 5-hmC in B and E and intense staining in H and K. Scale bars indicate 100 μm in A and G (same magnifications in B, C, H and I) and 20 μm in D and J (same magnifications in E, F, K and L).

Discussion

The transformation zone of the uterine cervix is an ideal tissue to study the stepwise process of carcinogenesis because subsequent stages of metaplastic, dysplastic and (micro)invasive changes can be easily accessed and molecular switches that take place during this process can be studied in a histo-morphological context. Triggered by the many gene specific methylation studies in cervical lesions (see Introduction) we decided to study the process of (hydroxy)methylation at a more global scale in the normal and (pre)malignant lesions using an immunohistochemical approach. Contradictory results exist in the literature concerning the methylation and hydroxymethylation status of the genome in cervical cancer. Several authors state that the methylation level of cytosine (5-mC) is reduced in cervical cancers when compared to normal epithelium [11, 12], while others provide evidence for an increase of 5-mC [27]. Downregulation of 5-hmC was shown to be associated with the progression of CIN lesions [28], with 5-hmC being significantly decreased in cervical squamous cell carcinoma as compared to CIN lesions and normal cervical tissues [11, 29]. Finally, most of the studies into gene methylation and hydroxymethylation in the cervix have been performed with ectocervical squamous lesions, while endocervical glandular lesions, such as AIS and adenocarcinoma, have not been studied to a great extent. In the underlying study we show that the global 5-hmC levels decrease to a certain extent with progression of CIN, but that particularly AIS and adenocarcinoma lesions exhibit a steep decrease in global hydroxymethylation, while 5-mC remained largely detectable in these (pre)malignancies to a similar extent as in their normal counterparts. The major advantage of using an immunohistochemical approach, as opposed to the more integrated approaches used by others, is that we obtained spatially resolved information that allowed us to correlate the epigenetic switches to the position of individual cells in morphologically heterogeneous tissue samples. The observed (hydroxy)methylation patterns in the normal epithelia and their progenitor cells provide clues about their relation and differentiation status, while the differences in the hydroxymethylation process between squamous and glandular uterine cervical lesions can partially be explained in light of mechanisms that characterize the different carcinogenic pathways, including the physical status of HPV, genomic alterations and instability, gene-specific methylation profiles and the activity of TET proteins.

Dynamics in 5-mC levels

Only minimal differences in fluorescence intensity for 5-mC were observed when comparing normal (including metaplastic) squamous epithelium, normal glandular epithelium and (pre)neoplastic lesions. A high degree of heterogeneity in staining intensity was, however, observed in the different histological compartments, with intensely and weakly stained nuclei intermingled and significant differences in immunofluorescence intensity between these nuclei. This heterogeneous staining pattern was not only seen throughout the basal/parabasal, intermediate and superficial compartments of the squamous epithelium, also the normal columnar epithelium with underlying reserve cells showed intermingled intensely and weakly stained nuclei in the different stretches of this single-layered epithelium.

The observed random distribution of intensely and weakly stained nuclei in the normal tissues follows a pattern similar to the random distribution of maternal or paternal inactivation of the X chromosome as a result of DNA methylation [30]. This X chromosome inactivation occurs early during embryo development and is maintained during life. It remains to be seen whether or not a correlation exists between the process of X chromosome inactivation and the heterogeneity of 5-mC immunoreactivity.

The intermingled staining pattern as observed in the premalignant lesions, and the observation that nuclei with intense or weak staining were not clustered, is unexpected based on the assumption that most of these lesions result from clonal expansion. Our data may, however, be in line with those of Ueda et al. (2003), who showed that most CIN lesions are monoclonal in composition, but that CIN cases in which the HPV genome was present in episomal form were polyclonal [31]. This may imply that HPV viral integration into the host genomic DNA is associated with progression from a polyclonal to monoclonal status in CIN. As a result, also the 5-mC staining patterns may vary in such CIN lesions based on the physical status of the virus.

Several authors have described changing levels of 5-mC during carcinogenesis, but their results are interpreted differently. Zhang et al. (2016) reported a significant difference in staining intensities when comparing normal to cancer samples [29]. However, since the presented images showed no staining for 5-mC in the stromal compartment, these data should be interpreted with care. Wang et al. (2019) visualized 5-mC utilizing brightfield microscopy and showed intensely staining nuclei in both control and cancer samples [27]. In their samples the ratio of intensely stained nuclei intermingled with nuclei showing no staining altered during progression. Kato (2020) showed data suggesting an abrupt loss in 5-mC methylation with progression from CIN2 to CIN3 and cancer [28]. The corresponding plots of immunofluorescence showed, however, a wide range of intensities rather than an abrupt change. Our data align with those of Bhat et al. (2017) who measured a two-fold difference for 5-mC when comparing DNA isolated from normal epithelium and cervical squamous cell carcinoma [11]. This integrated approach does, however, not show the different levels of 5-mC in individual cells and tissue compartments, which did become clearly apparent from our immunohistochemical approach.

Dynamics in 5-hmC levels

For 5-hmC significant differences in fluorescence intensity were seen in the different compartments of the normal epithelia and even more so during the development of the two types of (pre)neoplasia.

While normal ectocervical squamous epithelium often shows a strong 5-hmC staining in (part of) the basal cells, this compartment is largely negative when located adjacent to the SCJ or within (im)mature metaplastic epithelium. Also, the reserve cells underlying the glandular epithelium, which are the progenitor cells for metaplastic squamous epithelium, show only low levels of 5-hmC or are completely negative. The low level of 5-hmC in reserve cells aligns with the published observations that stem cells or progenitor cells frequently show low levels of 5-hmC [10, 32, 33]. The formation of immature metaplasia results from proliferation of reserve cells [26] and the immature metaplastic epithelium subsequently matures into a squamous epithelium that is indistinguishable from the original ectocervical squamous epithelium. This process occurs in the transformation zone with the new SCJ as the border between squamous and columnar epithelium. In contrast to the basal cells in the original ectocervical squamous epithelium and the columnar epithelium in which 5-hmC levels are high, the parabasal cells in the squamous epithelium showed a lower level of 5-hmC. This difference suggests that these highly proliferative progenitor cells responsible for the renewal of the epithelium (the so-called transient amplifying compartment) have a low level of 5-hmC. An increase in 5-hmC levels was observed during the subsequent process of epithelial differentiation, resulting in relatively high 5-hmC immunofluorescence levels in the terminally differentiated cells in the more superficial layers. These observations align with those of Haffner et al. (2011) who showed that in the basal part of the human adult colon the level of 5-hmC is low, while on top of the epithelium the highest 5-hmC levels were recorded [10]. Also, the cervical columnar epithelium showed a high 5-hmC level, suggesting that these cells can be regarded as terminally differentiated cells with a very low proliferation activity.

When studying CIN lesions, grade-specific 5-hmC immunostaining patterns were recognized. CIN1 lesions showed the highest 5-hmC levels in the basal/parabasal compartment, comparable to the pattern seen in normal squamous epithelium. In contrast, CIN2 lesions often showed an intense 5-hmC immunostaining from basal/parabasal to the upper layers, a pattern shared with CIN3. And finally, part of the CIN3 lesions showed a reduced staining throughout the dysplastic epithelium. A typical pattern for the higher grade CIN lesions was the low 5-hmC level in the basal/parabasal compartment. The grade-specific patterns align with non-disturbed (low-grade CIN) and disturbed differentiation (high grade CIN) of the epithelium upon infection with HPV (see also S1 Fig). In the literature reduced 5-hmC levels were reported when cancer was compared with normal epithelium, but again data with respect to the spatial distribution within the tissues and levels in the different grades of CIN were lacking so far [11, 27–29].

In the underlying study we report for the first time on the 5-hmC levels in (pre)malignant glandular cervical lesions. In nearly all premalignant glandular (AIS) lesions, as well as adenocarcinomas, a (very) low 5-hmC level was detected, certainly when compared to the normal columnar epithelium. A straightforward explanation for this finding could be that changes in TET2 activity underly the reduced hydroxymethylation activity. Such reduced expression or activity levels of TET2 and resulting low 5-hmC levels have been reported for many solid tumor types as well as hematopoietic disorders [14, 15, 17, 34–36]. Studies are in progress to correlate the 5-hmC immunostaining patterns to the expression patterns of TET proteins in normal cervical epithelia and the (pre)malignancies derived therefrom.

Influence of HPV infection and integration on 5-mC and 5-hmC levels

HPV infection of metaplastic epithelium or reserve cells in the TZ can result in a squamous intraepithelial lesion. The viral proteins E6 and E7 are expressed and uncouple cell growth arrest, apoptosis and differentiation of the host cell primarily through the inactivation of p53 and pRb, respectively. Also, HPV E7 binds to DNA (cytosine-5)-methyltransferase 1 (DNMT1) and stimulates its activity, while at the same time transcription of DNMT1 is activated through the pRb/E2F pathway and HPV E6 upregulates DNMT1 by suppression of p53 [37]. This increased DNA methylation activity then (de)regulates the expression of specific genes.

Several studies have shown that the frequency of methylation of candidate genes increases with increasing severity of the cervical lesion, suggesting that these changes already occur early in cancer development. DNA methylation is found to be more common in invasive cervical carcinoma and CIN3 as compared to CIN1/2 [38]. This can to a certain extent be explained by the decreased hydroxymethylation status as detected in the underlying study for the higher grade CIN lesions. Kato et al. (2020) recently showed that 5-hmC levels are decreased between CIN2 and CIN3 through the TP53-A3B pathway [28]. These authors also propose that, since A3B (apolipoprotein B mRNA editing enzyme) could impair genome stability, 5-hmC loss might increase the chances of accumulating mutations and of progressing from CIN3 to cervical cancer.

Besides the locus-specific methylation events, cell cycle deregulation disturbs the coordination between DNA replication and activity of DNA methyltransferases, leading to hypomethylation of repetitive regions in the genome, finally resulting in genomic instability.

Integration of HPV DNA into the host cell genome is a key event in HPV-mediated carcinogenesis [39]. In CIN1 and CIN2 the virus is mostly present in an episomal form which allows viral replication, during which differentiation is moderately altered. Upon integration into the human genome, which occurs relatively late in the development of CIN3, differentiation is interrupted. The AIS lesions examined in the underlying study, however, contained integrated virus in nearly all cases [40]. The cases with low 5-hmC levels in our study therefore align very well with the cases in which the virus was shown to be integrated into the host genome. To explain this direct correlation between the integration of the virus on the one hand and the low 5-hmC levels at the other, we hypothesize this to be the result of TET2 deregulation. The 5-hmC moieties resulting from TET2 activity are used as a tag at stalled replication forks, formed during replication stress in order to recruit enzymes for the base excision repair during the cell cycle. When TET2 is inactive as a result of mutations or not expressed as a result of promotor CpG methylation, the stalled replication forks are not degraded but stabilized. These conditions contribute to genomic instability [41, 42], in particular within the so-called fragile sites of the genome. McBride and Warburton (2017) hypothesized that HPV replicates its DNA by applying the host DNA response machinery adjacent to areas in the genome that are susceptible to replication stress [43]. These fragile sites align with transcriptionally active regions (enriched for repetitive elements and CpG dinucleotides) and are candidate regions for HPV integration. We hypothesize therefore that deregulated TET2 activity is the cause of accidental viral integration into the host genome, leading to abrogation of the viral life cycle.

Clinical utility

Our results seem to indicate that inhibition of global demethylation, that will normally follow upon cytosine hydroxymethylation, is an important epigenetic switch in the development of cervical cancer. As a consequence, not only the methylation status of specific genes can be used as an indicator for the degree of malignant transformation of the cervical squamous and glandular lesions [15–17], but also the detection of global hydroxymethylation by means of immunohistochemical staining of 5-hmC. In case of glandular lesions an inhibition of global demethylation was frequently recognized in AIS, the pre-stage of adenocarcinoma, and may become of clinical utility in combination the conventional H&E staining. Such a correlation between the loss of 5-hmC levels and the degree of malignant progression was less evident for the squamous premalignant lesions. For that reason, the potential clinical utility of 5-mC and 5-hmC immunostaining is limited for these squamous lesions. In diagnostically challenging clinical CIN samples, however, the 5-hmC staining patterns, showing either positivity throughout the full thickness of the epithelium or a negative epithelium, are indications for a high-grade lesion in addition to the morphological criteria.

When interpreting the result of such an immunohistochemical analysis one has to realize, however, that in addition to 5-hmC, the Tet proteins can generate 5-formylcytosine (5-fC) and 5-carboxylcytosine (5-caC) from 5-mC. A potential alternative 5-mC demethylation mechanism has therefore been suggested by Ito et al [2], in which the Tet proteins oxidize 5-mC not only to 5-hmC, but also to its aldehyde form 5-fC and the carboxylic acid form 5-caC.

Supporting information

S1 Fig. Direct correlation of immunostaining results for 5-hmC, HPV, SOX2 and SOX17 in tissue sections of normal squamous epithelium, metaplastic epithelium and CIN3.

A) H&E-stained stretch of squamous epithelium. B) Immunostaining of immature metaplastic epithelium by SOX17 and C) immunostaining of p16 for HPV-infected CIN3. D-F) Higher magnifications of boxed area 3 indicated in A, showing SOX2 immunostaining (D), HPV detection by chromogenic in situ hybridization (E) and 5-hmC immunofluorescence (F). Straight lines in D-F mark a collision area with the left area being SOX2 negative, HPV negative and 5-hmC weak to negative in the basal/parabasal compartment of the immature metaplastic epithelium. In the right area SOX2 stained positive, HPV is detected and 5-hmC shows an overall weak to negative staining and scattered positive nuclei. The lines in F-I indicate the position of basement membrane. G-I) Higher magnifications of boxed areas 1, 2 and 4, respectively, showing 5-hmC immunostaining. In the area distal (box 1) from the infected TZ (box 3) the basal cells showed a high fluorescence for 5-hmC (compare Fig 1D). In the area in box 4, the dysplastic epithelium showed a weak staining for 5-hmC (compare Fig 6K). Scale bars indicate 500 μm in A (same magnifications in B and C), 100 μm in D (same magnifications in E-F), and 20 μm in G (same magnifications in H and I).

(TIF)

pone.0297008.s001.tif^{(7.6MB, tif)}

S2 Fig. Complex immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions.

A-C) CIN1 area in CIN1 patient; D-F) CIN1 area in CIN3 patient; G-I) CIN3 area diagnosed in CIN3 patient; J-L) CIN3 area in CIN3 patient. Images were captured using a 40 times oil objective and are higher magnification images of cases shown in Fig 6. A, D, G and J) 5-mC; B, E, H and K) 5-hmC; C, F, I and L) merged images of 5-mC and 5-hmC with DAPI counterstaining. Note the different staining patterns for 5-hmC with a variation in the basal/parabasal layers being strongly positive to weakly positive or even negative. Scale bar in A indicates 20 μm (same magnifications in B-L). The lines in C, F, I and L mark the position of the basement membrane (bm). The stromal cell layer, basal and parabasal cells and intermediate cell layers are indicated as sc, bc, pbc and icl respectively.

(TIF)

pone.0297008.s002.tif^{(7.4MB, tif)}

S3 Fig. Quantitative analysis of 5-mC and 5-hmC immunofluorescence intensity (Arbitrary Units, AU) distributions in normal squamous epithelium, CIN1 and CIN3.

A, D, and G) Merged immunofluorescence images of 5-mC (red), 5-hmC (green) and DAPI (blue) in normal squamous epithelium, CIN1 and CIN3. Lines depicted in A, D and G show the immunofluorescence scan track for the three individual colors from the lower to the higher compartment of the epithelium. B, E and H) Intensity plots for DAPI and 5-mC. C, F and H Intensity plots for 5-hmC. C shows an example of the pattern with a high intensity staining for 5-hmC in the basal cell layer (bc layer) with a reduced intensity in the parabasal cell layer (pbc layer). The intensity for 5-hmC increases in the direction of differentiation of the epithelium to the more upper layer of the epithelium (intermediate cell layer; icl). F) shows an example of the pattern with a low intensity staining for 5-hmC in the lower compartment of the epithelium including the bc, pbc layers and lower compartment of the icl. Compare the difference in fluorescence intensity distribution between 5-mC and 5-hmC. I) shows an example of the pattern with a complex staining for 5-mC and 5-hmC throughout the epithelium, with a strong staining at the bc layer for both 5-mC and 5-hmC, and quantitatively supporting the observed inter- and intra-nuclear differences in immunostaining for 5-mC and 5-hmC. Note that the stromal compartment always shows cells with a high nuclear intensity staining for 5-hmC (see also Figs 1 and 2), which are localized underlying the epithelial basal cell compartment (see S3D Fig). These stromal cells have not been included in this quantitative analysis (see S3E and S3F Fig).

(TIF)

pone.0297008.s003.tif^{(4.4MB, tif)}

S4 Fig. Immunofluorescence staining patterns for 5-mC and 5-hmC in CIN3 lesions with overlying columnar epithelium, showing extensive heterogeneity for methylation and hydroxymethylation.

A-G) Merged images of 5-mC in red and 5-hmC in green with DAPI counterstaining in blue. Two patient samples are shown, one in A-C and the other in D-G. Scale bar in A indicates 20 μm (same magnifications in B-G). The lines mark the position of the basement membrane. In A and D arrowheads point to normal columnar epithelium and columnar epithelium on top of the dysplastic squamous epithelium. In C and E +: indicates areas with an increased 5-hmC immunostaining as compared to * indicating areas with a decrease of 5-hmC immunostaining.

(TIF)

pone.0297008.s004.tif^{(6.3MB, tif)}

S5 Fig. Confocal images of double-label immunofluorescence staining patterns for 5-mC and 5-hmC in a CIN3 lesion.

A) Merged image of 5-mC in red and 5-hmC in green with blue DAPI counterstaining; B) Higher magnification of boxed area in A; C) 5-mC with blue DAPI counterstaining. D) 5-hmC with blue DAPI counterstaining. Note the mutually exclusive staining patterns for 5-mC and 5-hmC. Nuclei with dominant 5-hmC staining indicated with 1 and 5-hmC dominant staining indicated with 2. Scale bar in A indicates 10 μm and 20 μm in B (same magnifications in C and D).

(TIF)

pone.0297008.s005.tif^{(7.6MB, tif)}

S6 Fig. Immunofluorescence staining results for 5-mC and 5-hmC showing heterogeneity for 5-hmC in cervical adenocarcinoma.

A) H&E-stained dysplastic glands in adenocarcinoma. B) 5-mC; C) 5-hmC; D) merged mages of 5-mC and 5-hmC with DAPI counterstaining. Scale bar in A indicates 100 μm (same magnifications in B-D).

(TIF)

pone.0297008.s006.tif^{(7.1MB, tif)}

S1 Table

A Antibody characteristics and optimized detection methods applied for the single- and double-label immunofluorescence analyses of 5-mC and 5-hmC. B Antibody characteristics and optimized detection methods applied for the bright field immunohistochemical analyses.

(PDF)

pone.0297008.s007.pdf^{(122.4KB, pdf)}

Acknowledgments

We thank Dr. Jack Cleutjens (Dept of Pathology, Maastricht University Medical Center, Maastricht, The Netherlands) for help with scanning of the tissue slides. We are grateful to Frans Kwaspen (PanPath, Budel, The Netherlands) for providing HPV probes for in situ hybridization.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

All relevant data are within the paper and its Supporting Information files.

Funding Statement

Jobran Moshi: His work was supported by a grant from Jazan University, Jazan, Saudi Arabia (grant number 1067568103). This study was part of the PhD study of Jobran Moshi. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y, et al. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science. 2009;324(5929):930–5. doi: 10.1126/science.1170116 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Ito S, Shen L, Dai Q, Wu SC, Collins LB, Swenberg JA, et al. Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5-carboxylcytosine. Science. 2011;333(6047):1300–3. doi: 10.1126/science.1210597 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Szulwach KE, Li X, Li Y, Song CX, Han JW, Kim S, et al. Integrating 5-hydroxymethylcytosine into the epigenomic landscape of human embryonic stem cells. PLoS Genet. 2011;7(6):e1002154. doi: 10.1371/journal.pgen.1002154 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Chen Y, Damayanti NP, Irudayaraj J, Dunn K, Zhou FC. Diversity of two forms of DNA methylation in the brain. Front Genet. 2014;5:46. doi: 10.3389/fgene.2014.00046 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Pfeifer GP, Xiong W, Hahn MA, Jin SG. The role of 5-hydroxymethylcytosine in human cancer. Cell Tissue Res. 2014;356(3):631–41. doi: 10.1007/s00441-014-1896-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Wang KC, Kang CH, Tsai CY, Chou NH, Tu YT, Li GC, et al. Ten-eleven translocation 1 dysfunction reduces 5-hydroxymethylcytosine expression levels in gastric cancer cells. Oncol Lett. 2018;15(1):278–84. doi: 10.3892/ol.2017.7264 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Wang Z, Du M, Yuan Q, Guo Y, Hutchinson JN, Su L, et al. Epigenomic analysis of 5-hydroxymethylcytosine (5hmC) reveals novel DNA methylation markers for lung cancers. Neoplasia. 2020;22(3):154–61. doi: 10.1016/j.neo.2020.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Das PM, Singal R. DNA methylation and cancer. J Clin Oncol. 2004;22(22):4632–42. doi: 10.1200/JCO.2004.07.151 [DOI] [PubMed] [Google Scholar]
9.Vrba L, Futscher BW. A suite of DNA methylation markers that can detect most common human cancers. Epigenetics. 2018;13(1):61–72. doi: 10.1080/15592294.2017.1412907 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Haffner MC, Chaux A, Meeker AK, Esopi DM, Gerber J, Pellakuru LG, et al. Global 5-hydroxymethylcytosine content is significantly reduced in tissue stem/progenitor cell compartments and in human cancers. Oncotarget. 2011;2(8):627–37. doi: 10.18632/oncotarget.316 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Bhat S, Kabekkodu SP, Varghese VK, Chakrabarty S, Mallya SP, Rotti H, et al. Aberrant gene-specific DNA methylation signature analysis in cervical cancer. Tumour Biol. 2017;39(3):1010428317694573. doi: 10.1177/1010428317694573 [DOI] [PubMed] [Google Scholar]
12.Zhu H, Zhu H, Tian M, Wang D, He J, Xu T. DNA Methylation and Hydroxymethylation in Cervical Cancer: Diagnosis, Prognosis and Treatment. Front Genet. 2020;11:347. doi: 10.3389/fgene.2020.00347 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Li W, Liu M. Distribution of 5-hydroxymethylcytosine in different human tissues. J Nucleic Acids. 2011;2011:870726. doi: 10.4061/2011/870726 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Storebjerg TM, Strand SH, Høyer S, Lynnerup AS, Borre M, Ørntoft TF, et al. Dysregulation and prognostic potential of 5-methylcytosine (5mC), 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC) levels in prostate cancer. Clin Epigenetics. 2018;10(1):105. doi: 10.1186/s13148-018-0540-x [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Rogeri CD, Silveira HCS, Causin RL, Villa LL, Stein MD, de Carvalho AC, et al. Methylation of the hsa-miR-124, SOX1, TERT, and LMX1A genes as biomarkers for precursor lesions in cervical cancer. Gynecol Oncol. 2018;150(3):545–51. doi: 10.1016/j.ygyno.2018.06.014 [DOI] [PubMed] [Google Scholar]
16.van Leeuwen RW, Oštrbenk A, Poljak M, van der Zee AGJ, Schuuring E, Wisman GBA. DNA methylation markers as a triage test for identification of cervical lesions in a high risk human papillomavirus positive screening cohort. Int J Cancer. 2019;144(4):746–54. doi: 10.1002/ijc.31897 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Yuan LY, Qin X, Li L, Zhou J, Zhou M, Li X, et al. The transcriptome profiles and methylation status revealed the potential cancer-related lncRNAs in patients with cervical cancer. J Cell Physiol. 2019;234(6):9756–63. doi: 10.1002/jcp.27661 [DOI] [PubMed] [Google Scholar]
18.Gatti A, Haruyama R, Elit L, Howard SC, Lam CG, Ilbawi A, et al. How to reduce the impact of cervical cancer worldwide: Gaps and priority areas identified through the essential cancer and primary care packages: An analysis of effective interventions. Cancer. 2020;126(21):4697–705. doi: 10.1002/cncr.33116 [DOI] [PubMed] [Google Scholar]
19.Brisson M, Kim JJ, Canfell K, Drolet M, Gingras G, Burger EA, et al. Impact of HPV vaccination and cervical screening on cervical cancer elimination: a comparative modelling analysis in 78 low-income and lower-middle-income countries. Lancet. 2020;395(10224):575–90. doi: 10.1016/S0140-6736(20)30068-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Bosch FX, Broker TR, Forman D, Moscicki AB, Gillison ML, Doorbar J, et al. Comprehensive control of human papillomavirus infections and related diseases. Vaccine. 2013;31 Suppl 7(Suppl 7):H1-31. doi: 10.1016/j.vaccine.2013.10.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Woodman CB, Collins SI, Young LS. The natural history of cervical HPV infection: unresolved issues. Nat Rev Cancer. 2007;7(1):11–22. doi: 10.1038/nrc2050 [DOI] [PubMed] [Google Scholar]
22.Kremer WW, Steenbergen R, Heideman D, Kenter GG, Meijer C. The use of host cell DNA methylation analysis in the detection and management of women with advanced cervical intraepithelial neoplasia: a review. BJOG. 2021;128(3):504–14. doi: 10.1111/1471-0528.16395 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Steenbergen RD, Ongenaert M, Snellenberg S, Trooskens G, van der Meide WF, Pandey D, et al. Methylation-specific digital karyotyping of HPV16E6E7-expressing human keratinocytes identifies novel methylation events in cervical carcinogenesis. J Pathol. 2013;231(1):53–62. doi: 10.1002/path.4210 [DOI] [PubMed] [Google Scholar]
24.von Knebel Doeberitz M, Prigge ES. Role of DNA methylation in HPV associated lesions. Papillomavirus Res. 2019;7:180–3. doi: 10.1016/j.pvr.2019.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Moshi JM, Ummelen M, Broers JLV, Ramaekers FCS, Hopman AHN. Impact of antigen retrieval protocols on the immunohistochemical detection of epigenetic DNA modifications. Histochem Cell Biol. 2023;159(6):513–26. doi: 10.1007/s00418-023-02187-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Moshi JM, Hoogduin KJ, Ummelen M, Henfling MER, van Engeland M, Wouters KAD, et al. Switches of SOX17 and SOX2 expression in the development of squamous metaplasia and squamous intraepithelial lesions of the uterine cervix. Cancer Med. 2020;9(17):6330–43. doi: 10.1002/cam4.3201 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Wang J, Su Y, Tian Y, Ding Y, Wang X. Characterization of DNA hydroxymethylation profile in cervical cancer. Artif Cells Nanomed Biotechnol. 2019;47(1):2706–14. Epub 2019/07/05. doi: 10.1080/21691401.2019.1634578 . [DOI] [PubMed] [Google Scholar]
28.Kato M, Onoyama I, Kawakami M, Yoshida S, Kawamura K, Kodama K, et al. Downregulation of 5-hydroxymethylcytosine is associated with the progression of cervical intraepithelial neoplasia. Plos One. 2020;15(11). ARTN e0241482 doi: 10.1371/journal.pone.0241482 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Zhang LY, Han CS, Li PL, Zhang XC. 5-Hydroxymethylcytosine expression is associated with poor survival in cervical squamous cell carcinoma. Jpn J Clin Oncol. 2016;46(5):427–34. doi: 10.1093/jjco/hyw002 [DOI] [PubMed] [Google Scholar]
30.Panning B, Jaenisch R. DNA hypomethylation can activate Xist expression and silence X-linked genes. Genes Dev. 1996;10(16):1991–2002. doi: 10.1101/gad.10.16.1991 [DOI] [PubMed] [Google Scholar]
31.Ueda Y, Enomoto T, Miyatake T, Ozaki K, Yoshizaki T, Kanao H, et al. Monoclonal expansion with integration of high-risk type human papillomaviruses is an initial step for cervical carcinogenesis: association of clonal status and human papillomavirus infection with clinical outcome in cervical intraepithelial neoplasia. Lab Invest. 2003;83(10):1517–27. doi: 10.1097/01.lab.0000092234.68751.83 [DOI] [PubMed] [Google Scholar]
32.Ko M, Bandukwala HS, An J, Lamperti ED, Thompson EC, Hastie R, et al. Ten-Eleven-Translocation 2 (TET2) negatively regulates homeostasis and differentiation of hematopoietic stem cells in mice. Proc Natl Acad Sci U S A. 2011;108(35):14566–71. doi: 10.1073/pnas.1112317108 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Moran-Crusio K, Reavie L, Shih A, Abdel-Wahab O, Ndiaye-Lobry D, Lobry C, et al. Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation. Cancer Cell. 2011;20(1):11–24. doi: 10.1016/j.ccr.2011.06.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Boudra R, Woappi Y, Wang D, Xu S, Wells M, Schmults CD, et al. Regulation of 5-Hydroxymethylcytosine by TET2 Contributes to Squamous Cell Carcinoma Tumorigenesis. J Invest Dermatol. 2022;142(5):1270–9 e2. doi: 10.1016/j.jid.2021.09.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A, et al. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell. 2010;18(6):553–67. doi: 10.1016/j.ccr.2010.11.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Kudo Y, Tateishi K, Yamamoto K, Yamamoto S, Asaoka Y, Ijichi H, et al. Loss of 5-hydroxymethylcytosine is accompanied with malignant cellular transformation. Cancer Sci. 2012;103(4):670–6. doi: 10.1111/j.1349-7006.2012.02213.x [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Au Yeung CL, Tsang WP, Tsang TY, Co NN, Yau PL, Kwok TT. HPV-16 E6 upregulation of DNMT1 through repression of tumor suppressor p53. Oncol Rep. 2010;24(6):1599–604. doi: 10.3892/or_00001023 [DOI] [PubMed] [Google Scholar]
38.Wilting SM, Steenbergen RDM. Molecular events leading to HPV-induced high grade neoplasia. Papillomavirus Res. 2016;2:85–8. doi: 10.1016/j.pvr.2016.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Bodelon C, Vinokurova S, Sampson JN, den Boon JA, Walker JL, Horswill MA, et al. Chromosomal copy number alterations and HPV integration in cervical precancer and invasive cancer. Carcinogenesis. 2016;37(2):188–96. doi: 10.1093/carcin/bgv171 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Hopman ANH, Moshi JM, Hoogduin KJ, Ummelen M, Henfling MER, van Engeland M, et al. SOX17 expression and its down-regulation by promoter methylation in cervical adenocarcinoma in situ and adenocarcinoma. Histopathology. 2020;76(3):383–93. doi: 10.1111/his.13980 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Kafer GR, Li X, Horii T, Suetake I, Tajima S, Hatada I, et al. 5-Hydroxymethylcytosine Marks Sites of DNA Damage and Promotes Genome Stability. Cell Rep. 2016;14(6):1283–92. doi: 10.1016/j.celrep.2016.01.035 [DOI] [PubMed] [Google Scholar]
42.Kharat SS, Ding X, Swaminathan D, Suresh A, Singh M, Sengodan SK, et al. Degradation of 5hmC-marked stalled replication forks by APE1 causes genomic instability. Sci Signal. 2020;13(645). doi: 10.1126/scisignal.aba8091 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.McBride AA, Warburton A. The role of integration in oncogenic progression of HPV-associated cancers. PLoS Pathog. 2017;13(4):e1006211. doi: 10.1371/journal.ppat.1006211 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0297008.r001

Decision Letter 0

Ricardo Ney Oliveira Cobucci

28 Jun 2023

PONE-D-23-13996Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervixPLOS ONE

Dear Dr. Hopman,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Aug 12 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ricardo Ney Oliveira Cobucci, Ph.D

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following financial disclosure:

“Jobran Moshi: His work was supported by a grant from Jazan University, Jazan, Saudi Arabia (grant number 1067568103). This study was part of the PhD study of Jobran Moshi.”

Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: N/A

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Jobran et al. used the citrate pH 6.0 retrieval protocols to detect the levels of 5-mC and 5-hmC in FFPE tissues from normal cervix, precervical lesions, and cervical cancer patients. They concluded that the inhibition of demethylation, which normally follows cytosine hydroxymethylation, is an important epigenetic switch in the development of cervical cancer. In principle, this is a very interesting piece of work. Their findings provide new evidence for understanding the mechanisms of cervical disease progression. However, before the article can be formally published, there are still several questions that need to be explained by the authors.

1.“Figure 6. Immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions”. Please provide distribution maps of CIN-associated mild, moderate, or severe atypical hyperplasia cells and cancer cells in situ, and explain whether the levels of 5-mC and 5-hmC are expressed in these abnormal cells？

2."Figure 7: Immunostaining patterns of 5-mC and 5-hmC in normal glandular epithelium and adenocarcinoma in situ (AIS)." (1) Please provide distribution maps of the cells of adenocarcinoma in situ (AIS), and explain whether the levels of 5-mC and 5-hmC are expressed in these abnormal cells？ (2) In addition, the three images (A-B-C) in Figure 7 seem to display different tissues rather than consecutive sections of the same tissue. It is unclear whether these images represent the same region or area within the tissue. The authors should provide a clear explanation regarding the different of these images.

3.Figure 6, it appears that the labels D-F should represent CIN2 rather than CIN1-3, as indicated in the paper text. Figure 7, the Line 4 of the figure legend mentions ‘E’, but there is no corresponding E figure in Figure 7. Please clarify and correct this inconsistency in the figures.

4.In the "Clinical utility" section of the Discussion, "Our results seem to indicate that inhibition of global demethylation, which will normally follow upon cytosine hydroxymethylation, is an important epigenetic switch in the development of cervical cancer.” It is well known that the demethylation process is complex and involves multiple stages, including hydroxymethylation, aldehyde methylation, and carboxymethylation. Therefore, the authors should provide additional explanations about the view point.

5.To evaluate differences in immunofluorescence staining results, it is recommended to use quantitative or semi-quantitative methods such as image analysis software instead of relying on visual inspection. Quantitative methods provide a more objective and reliable analysis of stain strength.

Reviewer #2: Numerous genomic modifications, including epigenetic changes that result in abnormal gene expression as well as altered DNA sequences, have an impact on cancers progression. The DNA methylation and demethylation have recently received a lot of interest. Together with the traditional histo-pathological characteristics, epigenetic profiles may improve the clinical treatment of patients.

The manuscript by Moshi et al. discusses the potential prognostic significance of 5-cytosine epigenetic modifications for progression of cancer precursor lesions in the uterine cervix. The authors discovered a substantial association between the distribution and expression of 5-mC and 5-hmC and the pathologic stages of cervical cancers.

The manuscript is generally well-written and interesting to the field. High-quality figures that clearly show the spatial distribution of 5-mC and 5-hmC across cervical carcinoma grading are its strength. The article generally fulfills the requirements for publication in PLOS ONE.

Please find below some of my concerns:

Introduction

The introduction provided background for the concepts and discussed the study objectives. However, in the sentence: „Usually, during carcinogenesis, the methylation level around specific genes is increased” (page 3), it would be helpful to include the examples of aforementioned genes as well as to clarify whether they are associated to any particular types of cancer or are more general.

Materials and Methods

This section of the manuscript was well prepared, including multiple details.

The points for revision:

- It was not specified how many healthy controls were examined, despite the experiments involving 65 cervical patients. (page 5)

- Since the text and the table exhibited differences (65 vs. 60 respectively), the number of cervical samples needs to be clarified. (page 5)

- The cervical grade abbreviations in Table 1 would be easier to understand if they were explained within the legend. (page 5)

- In my opinion, it was not required to reconsider the effects of antigen retrieval protocols for the discovery of epigenetic DNA modifications. The authors have previously stated the benefit of citrate epitope unmasking in preserving cell nucleus morphology during immunofluorescence detection of 5-mC and 5-hmC (Moshi et al., Histochem Cell Biol 2023). (page 5-6)

- Ensure that the reagent names are accurate, please double-check them, e.g., buffer PBT or PBST (page 6)

Results

The points for revision:

- As it was discussed above, authors have previously stated the benefit of citrate epitope unmasking in preserving cell nucleus morphology during immunofluorescence detection of 5-mC and 5-hmC (Moshi et al., Histochem Cell Biol 2023).

Therefore, I do not think it is essential to include this chapter and supplemental figure S1 (chapter: Choice for the citrate pH 6.0 retrieval protocol for the detection of 5-mC and 5-hmC in FFPE tissue sections. – page 8, Supplemental Figure S1 – page 29) as well as the text: “Also, when using TE and pepsin/HCl as antigen … was still clearly visible” and supplemental figure S4 (chapter: Immunostaining results of 5-mC and 5-hmC in premalignant endocervical glandular lesions – page 18, Supplemental Figure S4 – page 32) in the manuscript.

- The headline of the chapter:“5-mC and 5-hmC immunostaining levels in normal ectocervical squamous epithelium (Figures 1-3)” is not accurate since authors included information about the metaplastic epithelium. The benign condition known as squamous metaplasia of uterine cervix presents no risk of malignancy. However, it cannot be defined as a normal state. (chapter: 5-mC and 5-hmC immunostaining levels in normal ectocervical squamous epithelium (Figures 1-3) – page 8)

- The squamous epithelium is formed by multiple layers, such as: basal, parabasal, intermediate and superficial. To make the results more readable, I would consider marking the layer distribution within stained tissue. (Figure 1 – page 9-10)

- The results for 5-hmC expression in Figures 1D and 1G are confusing. Similar layer organization and, hence, corresponding 5-hmC labeling patterns should be found in the squamous epithelium represented by boxes 1 and 2. (Figure 1 – page 9-10)

- What was the rationale behind selecting and presenting samples from four separate patients. (Figure 4 – page 13)

- I would suggest adding arrows to indicate where 5-mC and 5-hmC signals are seen/ not seen within reserve cells. (Figure G, H, J – page 14)

- The signatures of 5-mC/5-hmC/DAPI are missing within Figure 7A. (Figure 7A – page 17)

- The headlines in the results chapters might be altered to eliminate the phrase “figure”. (page 8-18)

Reviewer #3: The Authors of the study performed image analysis of DNA methylation/demethylation in cervical cancer FFPE tissues, using immunostymulator method for 5mC and 5hmC detection, and confocal microscopy for visualization. The study comprised samples from 60 patients.

The major concern of this study is lack of any statistical analysis. The Authors just made a simle comparison of localization of the epigenetic changes in the selected samples, without any quantitative analysis. The pictures are nice, but they lack significance without showing results of the real data analysis (retrieved possibly from optical density values).

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2024 Apr 18;19(4):e0297008. doi: 10.1371/journal.pone.0297008.r002

Author response to Decision Letter 0

31 Aug 2023

Reply to suggestions and comments of Reviewer #1:

1. “Figure 6. Immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions”. Please provide distribution maps of CIN-associated mild, moderate, or severe atypical hyperplasia cells and cancer cells in situ, and explain whether the levels of 5-mC and 5-hmC are expressed in these abnormal cells？

Reply: Because of the complexity and heterogeneity of the premalignant lesions, particularly in the higher-grade lesions, which contain often combinations of areas with different grades of preneoplasia (CIN), it is difficult to provide a simple correlation between the degree of progression of the CIN lesions on the one hand and the immunostaining results for 5-hmC on the other. For this reason we have included several examples of such complex combinations of immunostaining results and different patterns in the spectrum of CIN lesions. For clarification we have amended the text and legend (see below), and explain the staining patterns in a more general fashion.

Complex immunostaining patterns for 5-mC and 5-hmC in squamous intraepithelial lesions

The patient diagnosis as shown in Table 1 was based on the highest histological CIN grade observed in the tissue sample, but virtually all samples contained areas with different grades of CIN. This was especially obvious in the higher-grade lesions. Using H&E staining of the tissue sections of the different patient samples as a guide to recognize specific areas with different grades of CIN, a correlation between the CIN stages with the different immunohistochemical (hydroxy)methylation patterns was made. From these analyses it became clear that with progression of the lesion the immunostaining patterns became more complex. Fig 6 summarizes these different immunostaining patterns recognized for 5-mC and 5-hmC in areas marked by the pathologist as CIN1, CIN2 and CIN3. In the CIN1 lesions we observed similar patterns for 5-mC and 5-hmC as recorded in the normal squamous epithelium (Fig 6A-C), with intense to weak staining in the basal/parabasal compartment and lesser staining in the intermediate and superficial layers. Both 5-mC and 5-hmC positive nuclei intermingled with weakly stained and negative nuclei. A pattern that was observed in nearly all higher grade CIN2-3 lesions, several of which also contained areas with a CIN1 morphology, is illustrated in Figs 6D-F, exhibiting weak to negative staining in the lower compartment, but more reactivity in the upper layers. Most CIN3 areas showed this pattern with a weak staining in the lower compartment of the epithelium, but areas diagnosed as CIN2/3 or CIN3 occasionally showed overlapping immunostaining for both 5-mC and 5-hmC in all epithelial layers (Figs 6 G and H). Finally, in about 20% of the patients diagnosed with CIN3 (5 out of 25 patients), a (very) weak staining for 5-hmC was observed throughout the entire epithelium (Fig 6K), while most nuclei were positive for 5-mC (Fig 6J). In summary, CIN lesions showed a heterogeneous staining pattern for 5-hmC. When studying CIN3 cases in more detail, combinations of the patterns described above were found next to each other (see S2 Fig for examples). Also, when performing confocal microscopy, CIN lesions were found to show areas where 5-mC and 5-hmC immunostainings were mutually exclusive, with nuclei exhibiting intense staining for 5-hmC, while being negative for 5-mC and vice versa (see S3 Fig).

Fig 6. Complex immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions

A-C) CIN1 area in CIN1 patient ; D-F) CIN1 area in CIN3 patient; G-I) CIN3 area diagnosed in CIN3 patient ; J-L) CIN3 area in CIN3 patient. A, D, G and J) 5-mC; B, E, H and K) 5-hmC; C, F, I and L) merged images of 5-mC and 5-hmC with DAPI counterstaining. Note the different staining patterns for 5-hmC with a variation in the basal/parabasal layers being strongly positive to weakly positive or even negative. Scale bar in A indicates 20 μm (same magnifications in B-L). The lines in C, F, I and L mark the position of the basement membrane (bm). The stromal cell layer, basal and parabasal cells and intermediate cell layers are indicated as sc, bc, pbc and icl respectively. The superficial cell layer is not shown in the fluorescence images.

2. "Figure 7: Immunostaining patterns of 5-mC and 5-hmC in normal glandular epithelium and adenocarcinoma in situ (AIS)." (1) Please provide distribution maps of the cells of adenocarcinoma in situ (AIS) and explain whether the levels of 5-mC and 5-hmC are expressed in these abnormal cells. In addition, the three images (A-B-C) in Figure 7 seem to display different tissues rather than consecutive sections of the same tissue. It is unclear whether these images represent the same region or area within the tissue. The authors should provide a clear explanation regarding the different of these images.

Reply: We have modified the Figure 7 and changed the legend by indicating the different cellular structures to help reading the figure. Furthermore, we have performed new immunohistochemical stainings in bright field microscopy of consecutive sections to show the direct correlation between the downregulation of 5-hmC in p16 positive areas of AIS. We have included additional images of these results in Figs 7G-J.

Fig 7. Immunostaining patterns of 5-mC and 5-hmC in normal glandular epithelium and adenocarcinoma in situ (AIS)

A) Merged image of 5-mC in red and 5-hmC in green with blue DNA counterstaining using DAPI. Note the dramatic decrease in 5-hmC immunofluorescence in the AIS structures (gland 1 normal columnar epithelium, glands 2 and 3 containing AIS structures). The arrows in A and B point to normal columnar epithelium (nce) adjacent to AIS cells, the latter being identified by p16 positivity (B) and SOX17 positivity (C). D, E and F, higher magnifications of boxed areas in A, B and C. Normal columnar epithelium is characterized by the absence of immunostaining for p16 and SOX17. Note that the images in A, B and C, although being derived from the same area in the AIS lesion, do not represent consecutive sections, resulting in slight differences in cellular distribution patterns. G-J) Brightfield images of 5-hmC immunostaining patterns of a cervical AIS, showing mutual exclusive staining for 5-hmC and p16 in same area in the AIS lesion (no consecutive tissue sections). G and I) show 5-hmC positivity only in the normal columnar epithelium (nce); H and J) show p16 positivity only in the AIS region. I and J, higher magnifications of boxed areas in G and H. Note in I and J an alternating area with staining patterns for nce and ACIS. Compare with panels A and B depicting normal columnar epithelial cells trapped in ACIS area. Scale bars indicate 30 μm.

3. Figure 6, it appears that the labels D-F should represent CIN2 rather than CIN1-3, as indicated in the paper text. Figure 7, the Line 4 of the figure legend mentions ‘E’, but there is no corresponding E figure in Figure 7. Please clarify and correct this inconsistency in the figures.

Reply: We have modified the text in the legends of Figs 6 and 7 to be consistent with the paper text. In the Results section, we have changed the text and explained that the Figure illustrates patterns recognized in the different preneoplastic regions; the areas shown in the panels have a single histological CIN characteristic as indicated by the pathologist.

Fig 6. Complex immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions

Figure 7: We corrected the legend: E should be C.

4. In the "Clinical utility" section of the Discussion, "Our results seem to indicate that inhibition of global demethylation, which will normally follow upon cytosine hydroxymethylation, is an important epigenetic switch in the development of cervical cancer.” It is well known that the demethylation process is complex and involves multiple stages, including hydroxymethylation, aldehyde methylation, and carboxymethylation. Therefore, the authors should provide additional explanations about the viewpoint.

Reply: Indeed, as also mentioned and cited in the introduction, the demethylation process is a multistep process in which the enzymatic modification of 5-mC into 5-hmC is an initial step towards demethylation. We modified the text in the "Clinical utility" section of the discussion.

“When interpreting the result of such an analysis one has to realize, however, that the demethylation process not only involves a single step of hydroxymethylation but is a complex cascade that involves multiple stages, including aldehyde methylation and carboxymethylation, which could have a similar effect on this demethylation process.

5. To evaluate differences in immunofluorescence staining results, it is recommended to use quantitative or semi-quantitative methods such as image analysis software instead of relying on visual inspection. Quantitative methods provide a more objective and reliable analysis of stain strength.

Reply: In our study we did not only rely on visual inspection alone. After tuning of the immunofluorescence protocols to obtain an optimal specificity and sensitivity, images were collected using a fixed integration time, allowing semiquantitative evaluation, and using the camera’s full dynamic range without signal intensity saturation. These images were then analyzed by visual inspection. In a previous study, we compared the result of visual inspection with quantitative microscopy of 5-hmC immunostaining patterns in normal cervical epithelium (Histochem Cell Biol…) and showed that differences between levels for 5-hmC within and between nuclei in tissue context could be analyzed. However, the complexity of the immunostaining patterns, particularly in CIN lesions, and the often scattered positive cells (as indicated above), only enabled a descriptive rather than a (semi)quantitative approach.

Reply to suggestions and comments of Reviewer #2:

1. Usually, during carcinogenesis, the methylation level around specific genes is increased” (page 3), it would be helpful to include the examples of aforementioned genes as well as to clarify whether they are associated to any particular types of cancer or are more general.

Reply: In the Introduction on page 3 we have included two new references [8,9} focusing on the role of gene methylation in several types of cancer.

“Numerous genes have been found to undergo hypermethylation in cancer. The genes that are susceptible are the genes involved in cell cycle regulation (e.g. p16INK4a, Rb), genes associated with DNA repair (e.g. BRCA1, apoptosis (DAPK, TMS1), drug resistance, detoxification, differentiation, angiogenesis, and metastasis. Although certain genes such as RASSF1A and p16 are commonly methylated in a variety of cancers, other genes are methylated in specific cancers [8.9]”.

2. It was not specified how many healthy controls were examined, despite the experiments involving 65 cervical patients.

Reply: No healthy controls were examined, normal histological areas present in the biopsies were analyzed. These areas were present in the patient samples and recognized by the pathologist and by molecular analysis. The normal areas showed to be HPV and p16 negative. We have added this text in the Materials and Methods under Table 1.

3. Since the text and the table exhibited differences (65 vs. 60 respectively), the number of cervical samples needs to be clarified. (page 5)

Reply: 60 patients were analyzed, this was adapted in the text.

4. The cervical grade abbreviations in Table 1 would be easier to understand if they were explained within the legend. (page 5)

Reply: We added CIN: cervical intraepithelial lesion, AIS: adenocarcinoma in situ; AdC: adenocarcinoma, SCJ: normal squamocolumnar junction (transition between squamous epithelium and glandular/columnar epithelium, no preneoplasia present in junction.

5. In my opinion, it was not required to reconsider the effects of antigen retrieval protocols for the discovery of epigenetic DNA modifications. The authors have previously stated the benefit of citrate epitope unmasking in preserving cell nucleus morphology during immunofluorescence detection of 5-mC and 5-hmC (Moshi et al., Histochem Cell Biol 2023). (pages 5-6)

Reply: We have now omitted Supplemental Figure S1 and Supplemental Figure S4 (old figure numbers) and the antigen retrieval protocols for TE and pepsin/HCl in the Materials & Methods section because these have indeed been described earlier.

6. Ensure that the reagent names are accurate, please double-check them, e.g., buffer PBT or PBST (page 6)

Reply: We have corrected and checked the document for PBS and PBST.

7. As it was discussed above, authors have previously stated the benefit of citrate epitope unmasking in preserving cell nucleus morphology during immunofluorescence detection of 5-mC and 5-hmC (Moshi et al., Histochem Cell Biol 2023).

Reply: As stated above these Figures and the text in the Materials and Methods sections has been modified.

8. The headline of the chapter: “5-mC and 5-hmC immunostaining levels in normal ectocervical squamous epithelium and metaplastic (Figures 1-3)” is not accurate since authors included information about the metaplastic epithelium. The benign condition known as squamous metaplasia of uterine cervix presents no risk of malignancy. However, it cannot be defined as a normal state. (chapter: 5-mC and 5-hmC immunostaining levels in normal ectocervical squamous epithelium (Figures 1-3) – page 8)

Reply: We have changed the title of the chapter into 5-mC and 5-hmC immunostaining levels in normal squamous epithelium. As now indicated in the legend we defined metaplasia as normal epithelium since metaplasia is the transition of endocervical epithelium into squamous epithelium, a normal process that can be detected in cervical biopsies. We have modified the text as follows:

“We defined metaplasia as normal epithelium since it represents the transition of endocervical glandular epithelium into squamous epithelium, a normal physiological process that can be detected in cervical biopsies. Metaplastic epithelium characterizes the transformation zone and bridges the epithelium between the original and new SCJ. Fig 3A shows a stretch of (immature) metaplastic epithelium with the typical co-expression of SOX2 and SOX17 (Figs 3B and C, respectively), and at the same time a negative p16 staining (Fig 3D), indicating that the epithelium was not infected by HPV. For an overview showing the transition in these immunostaining patterns between normal and metaplastic epithelium we refer to S1 Fig.”

9. The squamous epithelium is formed by multiple layers, such as the basal, parabasal, intermediate and superficial. To make the results more readable, I would consider marking the layer distribution within stained tissue. (Figure 1 – page 9-10)

Reply: We agree and modified the figures by indicating the different cellular structures to help reading the images.

10. The results for 5-hmC expression in Figures 1D and 1G are confusing. Similar layer organization and, hence, corresponding 5-hmC labeling patterns should be found in the squamous epithelium represented by boxes 1 and 2. (Figure 1 – page 9-10)

Reply: As indicated in the text referring to Figure 1 the distance between the squamous epithelium in boxes 1 and 2 and the squamocolumnar junction apparently has an impact on the immunohistochemical distribution patterns, especially in the basal compartment. For that reason, we also studied normal squamous epithelium close to the SCJ (Figure 2), and two types of 5-hmC patterns can indeed be recognized.

11. What was the rationale behind selecting and presenting samples from four separate patients. (Figure 4 – page 13)

Reply: Figure 4 has been reduced in size and now illustrates the most frequently occurring immunostaining patterns for 5-mC and 5-hmC in normal glandular epithelium.

12. Would suggest adding arrows to indicate where 5-mC and 5-hmC signals are seen/ not seen within reserve cells. (Figure G, H, J – page 14)

Reply: We modified the Figures and added RC and CC.

13. The signatures of 5-mC/5-hmC/DAPI are missing within Figure 7A. (Figure 7A – page 17).

Reply: We added the signatures of 5-mC/5-hmC/DAPI in the figure. Figure 7 was reorganized and labels were included for clarification.

14. The headlines in the results chapters might be altered to eliminate the phrase “figure” (page 8-18).

Reply: We removed the phrase “figure” in all Chapter headings.

Reply to suggestions and comments of Reviewer #3:

The Authors just made a simple comparison of localization of the epigenetic changes in the selected samples, without any quantitative analysis. The pictures are nice, but they lack significance without showing results of the real data analysis (retrieved possibly from optical density values).

Reply: In our study we could not rely on quantitative analyses because of the complexity of the immunostaining patterns, particularly in CIN lesions, and the often scattered positive cells (as indicated above). This enabled only a descriptive rather than a (semi)quantitative approach. The text in the results section, in particular the part for the CIN lesions, has been amended to better indicate the heterogeneity of these tissues and the complexity of the immunostaining patterns. Also, we have included new data from bright field analyses of AIS showing a clear down regulation of 5-hmC in these lesions (see Fig 7).

Complex immunostaining patterns for 5-mC and 5-hmC in squamous intraepithelial lesions

PLoS One. doi: 10.1371/journal.pone.0297008.r003

Decision Letter 1

Ricardo Ney Oliveira Cobucci

11 Sep 2023

PONE-D-23-13996R1Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervixPLOS ONE

Dear Dr. Hopman,

Please submit your revised manuscript by Oct 26 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Ricardo Ney Oliveira Cobucci, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments:

Dear authors, one of the reviewers still requests that you review the manuscript and we ask that you pay attention to the recommendations made.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: No

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: Jobran et al.'s response addressed some of the questions, but there are still certain questions that require further clarification.

1. In Figure 1, in the normal tissue section, whether in the 5-mC staining group or the 5-hmC staining group, we can observe two types of cells: a group of cells with strong staining and a group of cells with weak staining. In Figure 6 and 7, in the abnormal tissue section (CIN), whether in the 5-mC staining group or the 5-hmC staining group, we can also observe groups of cells with strong and weak staining. Therefore, we are curious about the relationship between "staining intensity" and normal or abnormal cells. Additionally, the authors introduce the concept of "staining pattern". They descripted that "Intense staining is observed in the basal/parabasal layers, which differs in intensity from the upper layers in the part of the epithelium shown in box 1 (Fig 1D). The epithelium represented in box 2 and box 3 showed more intense 5-hmC staining in the intermediate and superficial cell layers compared to the basal/parabasal layers (Figs 1G and J)" (Page 10). We think that the description of "staining pattern" carries a strong subjectivity from the authors themselves. Can the "staining pattern" be further divided into different types? Is the "staining pattern" consistent across all samples (60 samples)? To what extent can the "staining pattern" replace the results of “HE staining”? We understand that scientific research requires quantitative descriptions. Results with excessive subjectivity make it difficult to accurately classify different stages of diseases. The authors need to delve into these important questions in order to answer the relationship between the "staining intensity" and "staining pattern" of 5-mC and 5-hmC with the cell types of the disease and disease progression.

2. In the figure legend of Fig 6, it is mentioned that the four groups of images correspond to "CIN1 area", "CIN1 area," "CIN3 area", and "CIN3 area". However, in page 15 the authors refer to "CIN2" and "CIN2/3" when describing Figure 6. The figure legend of Fig 6 does not correspond with the text of the manuscript. Further clarification is required to resolve this discrepancy.

3. Page 15. The text mentions "Using H&E staining of the tissue sections of the different patient samples as a guide to recognize specific areas with". The authors should provide H&E staining pathology images and describe the percentage of atypical cells.

4. Page 15. In the text, Figure 6A-C is described as "In the CIN1 lesions, we observed similar patterns for 5-mC and 5-hmC as recorded in the normal squamous epithelium (Fig 6A-C), with intense to weak staining in the basal/parabasal compartment and lesser staining in the intermediate and superficial layers", in which "basal/parabasal compartment, intermediate and superficial layers" correspond to basal cells (bc), parabasal cells (pbc), intermediate cell layers (icl) in figure legend, respectively.

However, the author mentioned "lesser staining in the superficial layers" in the text, while the figure legend states "The superficial cell layer is not shown in the fluorescence images." Please provide additional images that include the "superficial cell layer." This will help ensure consistency between the text and the figure legend and provide a more complete representation of the study's findings.

5. Page 15. Please clarify how to obtain the result that "Most CIN3 areas showed this pattern with weak staining in the lower compartment of the epithelium" of Figs 6 G and H. And the authors should display the immunostaining results for all 5 patients diagnosed with CIN3. Please label the images clearly to indicate their source from each of the five patients. This will help readers understand and assess the consistency of the observed staining patterns across the CIN3 cases.

6. Page 15. The author mentioned, "while most nuclei were positive for 5-mC (Fig 6J)", but it is not clear from Figure 6J that the nuclei of the cells are positive. The authors should provide clearer and more representative images or data that demonstrate the positive staining of nuclei for 5-mC. This will help ensure that the findings are accurately represented and can be readily interpreted by readers.

7. Page 26. In the "Clinical utility" section of the Discussion, the authors modified the text as "When interpreting the result of such an analysis one has to realize, however, that the demethylation process not only involves a single step of hydroxymethylation but is a complex cascade that involves multiple stages, including aldehyde methylation and carboxymethylation, which could have a similar effect on this demethylation process [2]".

As mentioned in the reference you cited, it might be more accurate to state that "5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC)" compared to "cytosine hydroxymethylation", "aldehyde methylation" and "carboxymethylation".

Reviewer #2: My suggestions have been implemented. The manuscript discusses the potential prognostic significance of (hydroxy)methylation for progression of cancer precursor lesions in the uterine cervix. I believe the work is worthy of publishing.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

PLoS One. 2024 Apr 18;19(4):e0297008. doi: 10.1371/journal.pone.0297008.r004

Author response to Decision Letter 1

26 Oct 2023

Rebuttal to the Reviewer’s Comments

First of all, we would like to thank the reviewer for the meticulous assessment of our data and our interpretation of the results. We have therefore made every effort to amend the parts of the manuscript that were commented on. Several new figures now replace the former ones, some of which are included in the Supplemental information paragraph. Also, the text has been adapted and extended to respond to the questions and suggestions of this reviewer.

We can reply to the reviewer’s comments as follows:

Question 1A: In Figure 1, in the normal tissue section, whether in the 5-mC staining group or the 5-hmC staining group, we can observe two types of cells: a group of cells with strong staining and a group of cells with weak staining. In Figure 6 and 7, in the abnormal tissue section (CIN), whether in the 5-mC staining group or the 5-hmC staining group, we can also observe groups of cells with strong and weak staining. Therefore, we are curious about the relationship between "staining intensity" and normal or abnormal cells.

Answer: We have now performed quantitative analyses of the immunostaining in nuclei from normal and neoplastic cervical epithelium to compare the staining intensities of normal and abnormal cells. These data have been included in the Supplemental information paragraph as S3 Fig and the results have been described in the legend to the figure and summarized in the paragraph dealing with the squamous intraepithelial lesions. It should be realized that comparison of staining intensities amongst different tissues (normal versus neoplastic) are difficult because of the different fixation conditions of the tissues, the accessibility of the antigens, etc. That is why these quantitative analyses can only be used to compare the staining intensities between cells in the same tissue region. The quantitative data can, however, be used to support the semiquantitative observations that largely three different 5-hmC distribution patterns can be distinguished in normal and neoplastic cervical epithelium.

Question 1B: Additionally, the authors introduce the concept of "staining pattern". They descripted that "Intense staining is observed in the basal/parabasal layers, which differs in intensity from the upper layers in the part of the normal (?) epithelium shown in box 1 (Fig 1D). The epithelium represented in box 2 and box 3 showed more intense 5-hmC staining in the intermediate and superficial cell layers compared to the basal/parabasal layers (Figs 1G and J)" (Page 10). We think that the description of "staining pattern" carries a strong subjectivity from the authors themselves. Can the "staining pattern" be further divided into different types? Is the "staining pattern" consistent across all samples (60 samples)? To what extent can the "staining pattern" replace the results of “HE staining”? We understand that scientific research requires quantitative descriptions. Results with excessive subjectivity make it difficult to accurately classify different stages of diseases. The authors need to delve into these important questions in order to answer the relationship between the "staining intensity" and "staining pattern" of 5-mC and 5-hmC with the cell types of the disease and disease progression.

Answer: We have taken several steps to refute the image of subjectivity when it comes to the “staining pattern” terminology.

- Firstly, at the beginning of the Results section we have now included a short paragraph to explain the difference and relationship between the terms "staining intensity" and "staining pattern" of 5-mC and 5-hmC in normal and preneoplastic tissues.

- Secondly, we refer to our earlier study (reference 25) that describes the staining and distribution patterns of 5-mC and 5-hmC in normal cervical squamous epithelium. This is a study in which we tested several retrieval protocols for an optimal detection of 5-mC and 5-hmC in paraffin sections and performed already quantitative analyses of the staining intensities of cells in the different layers of the normal cervical epithelium.

- Thirdly, in the Results paragraph dealing with the squamous intraepithelial lesions the 3 major types of 5-hmC staining patterns (now designated as distribution patterns) are specified.

- Fourthly, these distribution patterns are not consistent across all 60 patient samples and therefore we have included the number of tissue areas that show the individual patterns. This also replies to the reviewer’s comment urging for a quantitative description to answer the relationship between the staining pattern of 5-mC and 5-hmC with the cell types of the disease and disease progression.

The H&E staining cannot be replaced by the 5-mC or 5-hmC staining pattern. In our study the H&E staining is used as the golden standard for the diagnosis of the CIN lesions in combination with more conventional immunomarkers such as p16, Ki-67, Cytokeratins, SOX2 and SOX17.

Question 2: In the figure legend of Fig 6, it is mentioned that the four groups of images correspond to "CIN1 area", "CIN1 area," "CIN3 area", and "CIN3 area". However, in page 15 the authors refer to "CIN2" and "CIN2/3" when describing Figure 6. The figure legend of Fig 6 does not correspond with the text of the manuscript. Further clarification is required to resolve this discrepancy.

Answer: We have now included a new Figure 6 into the body of the text and moved the former Figure 6 to the Supplemental information paragraph as S2 Fig. The new Figure 6 shows H&E and immunofluorescence staining images for 5-mC and 5-hmC in a CIN1, a CIN2 and two different patterns for CIN3. Also the body of the text and the legend to figure 6 has been adapted to clarify the apparent discrepance that existed in the former version of the manuscript.

Question 3: Page 15. The text mentions "Using H&E staining of the tissue sections of the different patient samples as a guide to recognize specific areas with". The authors should provide H&E staining pathology images and describe the percentage of atypical cells.

Answer: H&E staining images are now incorporated into the new Figure 6 for a CIN1, a CIN2 and two different patterns for CIN3. The number of atypical cells, as well as the dysplastic appearance increases with the CIN grade and was used as a diagnostic criterium by the pathologist.

Question 4: Page 15. In the text, Figure 6A-C is described as "In the CIN1 lesions, we observed similar patterns for 5-mC and 5-hmC as recorded in the normal squamous epithelium (Fig 6A-C), with intense to weak staining in the basal/parabasal compartment and lesser staining in the intermediate and superficial layers", in which "basal/parabasal compartment, intermediate and superficial layers" correspond to basal cells (bc), parabasal cells (pbc), intermediate cell layers (icl) in figure legend, respectively.

Answer: We have now included a new Figure 6 that contains lower magnifications of immunofluorescence staining patterns for 5-mC and 5-hmC for the different grades of CIN. These images now also include the superficial cell layers. The former Figure 6 with the higher magnification images (and therefore missing the superficial cell layers) is now S2 Fig in the Supplemental information paragraph.

Also, in Figure 1 we have now included lower magnifications of immunofluorescence staining patterns for 5-mC and 5-hmC for the normal cervical epithelium.

Question 5: Page 15. Please clarify how to obtain the result that "Most CIN3 areas showed this pattern with weak staining in the lower compartment of the epithelium" of Figs 6 G and H. And the authors should display the immunostaining results for all 5 patients diagnosed with CIN3. Please label the images clearly to indicate their source from each of the five patients. This will help readers understand and assess the consistency of the observed staining patterns across the CIN3 cases.

Answer: We believe that there exists some confusion about the number of patients diagnosed with CIN3. As can be seen from Table 1, of the 60 patients included in our study a total of 25 patients was diagnosed with CIN3. In the different neoplastic regions within the tissue sections different 5-hmc immunostaining patterns were observed. The observed numbers of these different patterns in relation to the total number of regions examined have now been included throughout the text. Only 5 of the 25 CIN3 lesions showed the pattern where all layers of the epithelium were negative for 5-hmC pattern. We see no use in depicting specifically these 5 cases with similar staining patterns in a separate image. In Figure 6 and in supplemental figure S2 Fig we have depicted the major 5-mC and 5-hmC staining patterns observed in the different CIN lesions, including an example of these negative cases.

Also, the text in the paragraph dealing with the squamous intraepithelial lesions has been adapted and extended, the quantitative data have been included and the different patterns have been more clearly described.

Question 6: Page 15. The author mentioned, "while most nuclei were positive for 5-mC (Fig 6J)", but it is not clear from Figure 6J that the nuclei of the cells are positive. The authors should provide clearer and more representative images or data that demonstrate the positive staining of nuclei for 5-mC. This will help ensure that the findings are accurately represented and can be readily interpreted by readers.

Answer: We have now included a new figure 6 with representative images that demonstrate the positive staining of nuclei for 5-mC and its relation to the 5-hmC and DAPI staining patterns. The low magnification images provide a good overview of the total tissue area and should now be easier to interpret by the readers.

The former Figure 6 is now supplementary figure S2 Fig. To show the overlap with the 5-mC staining we have now also slightly increased the intensity of the DAPI image in what was the former Figure 6L and is now supplementary figure S2L Fig.

Question 7: Page 26. In the "Clinical utility" section of the Discussion, the authors modified the text as "When interpreting the result of such an analysis one has to realize, however, that the demethylation process not only involves a single step of hydroxymethylation but is a complex cascade that involves multiple stages, including aldehyde methylation and carboxymethylation, which could have a similar effect on this demethylation process [2]".

Answer: We have now amended the text in the Clinical utility paragraph as follows:

“When interpreting the result of such an analysis one has to realize, however, that in addition to 5-hmC, the Tet proteins can generate 5-formylcytosine (5-fC) and 5-carboxylcytosine (5-caC) from 5-mC. A potential alternative 5-mC demethylation mechanism has therefore been suggested by Ito et al [2], in which the Tet proteins oxidize 5-mC not only to 5-hmC, but also to its aldehyde form 5-fC and the carboxylic acid form 5-caC.“

Attachment

Submitted filename: Responses to reviewer.docx

pone.0297008.s008.docx^{(21.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0297008.r005

Decision Letter 2

Ricardo Ney Oliveira Cobucci

8 Nov 2023

PONE-D-23-13996R2Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervixPLOS ONE

Dear Dr. Hopman,

One of the reviewers even requests corrections to the manuscript.

Please submit your revised manuscript by Dec 23 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Ricardo Ney Oliveira Cobucci, Ph.D

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

Dear authors,

one of the reviewers even requests corrections to the manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: No

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: Jobran et al.'s responses addressed some of the questions, but there are still certain questions that require further clarification.

� In the "Clinical utility" section, "As a consequence... the detection of global hydroxymethylation by means of immunohistochemical staining of 5-hmC may potentially become an important diagnostic tool in the investigation of glandular lesions of the cervix ", The authors concluded that immunohistochemical staining of 5-hmC can be used as an important diagnostic tool for cervical glandular lesions.

The 5-hmC staining technique used in this study can be divided into strong staining and weak staining in principle. In addition, the samples in this study were divided into 5 groups according to HE staining results, including normal squamous epithelium, normal columnar epithelium and reserve cells, squamous intraepithelial lesions, premalignant endocervical glandular lesions, cervical adenocarcinomas. The authors then described 5-hmC staining patterns (staining intensity and staining distribution) in these samples, including negative, weak, intermediate or intense/strong (see the attached figure). The authors also agree that the 5-hmC staining patterns cannot replace HE staining results. Therefore, the authors believe that immunohistochemical staining of 5-hmC can be used as a diagnostic tool for cervical glandular lesions, an important conclusion that more evidence is needed to support. For example: (1) the proportion of 5-hmC staining pattern in different disease samples (the attached figures showed that the proportion of 5-hmC staining strong was not high in some disease samples); (2) 5-hmC staining pattern sensitivity and specificity of different diseases using ROC/AUC value comparing with the golden standard HE method; (3) Individual differences in staining patterns judged by different experimenters. Therefore, it is recommended that authors need to add explanations to this issue or revise their conclusion to prevent misleading readers. Or add reasonable explanations in the discussion section to explain the shortcomings of this research work, so as to facilitate readers to understand the innovation points and potential application directions of this research work.

� In addition, in "S3 fig R2, D-E-F", the bottom green signal is not measured, so there is no signal peak on the left in figure F. Please confirm whether the author missed this important signal?

Reviewer #2: All of my suggestions and comments have already been incorporated into the manusript.

Nonetheless, I believe the presented modifications substantialy improved the paper's quality.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

Attachment

Submitted filename: Figure_integration.pdf

pone.0297008.s009.pdf^{(16.3MB, pdf)}

PLoS One. 2024 Apr 18;19(4):e0297008. doi: 10.1371/journal.pone.0297008.r006

Author response to Decision Letter 2

21 Dec 2023

Rebuttal to the Reviewer’s Comments

We would like to thank the reviewer for the meticulous assessment of our revision. The pdf entitled: Figure_integration.pdf provides an excellent overview of the 5-hmC staining patterns in relation to the type of (premalignant) epithelia. We would appreciate that the peer review history would be included with our manuscript, including the Figure_integration.pdf from the reviewer.

We can reply to the editor’s and reviewer’s comments as follows:

Question: Please review your reference list to ensure that it is complete and correct.

Answer: We have reviewed the reference list to ensure that it is complete and correct.

Question Reviewer #1: Jobran et al.'s responses addressed some of the questions, but there are still certain questions that require further clarification.

In the "Clinical utility" section, "As a consequence... the detection of global hydroxymethylation by means of immunohistochemical staining of 5-hmC may potentially become an important diagnostic tool in the investigation of glandular lesions of the cervix ", The authors concluded that immunohistochemical staining of 5-hmC can be used as an important diagnostic tool for cervical glandular lesions……. Therefore, it is recommended that authors need to add explanations to this issue or revise their conclusion to prevent misleading readers. Or add reasonable explanations in the discussion section to explain the shortcomings of this research work, so as to facilitate readers to understand the innovation points and potential application directions of this research work.

Answer: We have adapted the text in the “Clinical utility” section and revised our conclusion as follows.

As a consequence, not only the methylation status of specific genes can be used as an indicator for the degree of malignant transformation of the cervical squamous and glandular lesions [15-17], but also the detection of global hydroxymethylation by means of immunohistochemical staining of 5-hmC. In case of glandular lesions an inhibition of global demethylation was frequently recognized in AIS, the pre-stage of adenocarcinoma, and may become of clinical utility in combination the conventional H&E staining. Such a correlation between the loss of 5-hmC levels and the degree of malignant progression was less evident for the squamous premalignant lesions. For that reason, the potential clinical utility of 5-mC and 5-hmC immunostaining is limited for these squamous lesions. In diagnostically challenging clinical CIN samples, however, the 5-hmC staining patterns, showing either positivity throughout the full thickness of the epithelium or a negative epithelium, are indications for a high-grade lesion in addition to the morphological criteria.

Question Reviewer #1: In addition, in "S3 fig R2, D-E-F", the bottom green signal is not measured, so there is no signal peak on the left in figure F. Please confirm whether the author missed this important signal?

Answer: We included a note in the legend of Figure S3: Note that the stromal compartment always shows cells with a high nuclear intensity staining for 5-hmC (see also Figs 1 and 2), which are localized underlying the epithelial basal cell compartment (see S3 Fig D). These stromal cells have not been included in this quantitative analysis (see S3 Figs E and F).

For the reader we also have added this note to the legends of Figs 1 and 2: Note also the intense 5-hmC immunostaining in the stromal cell compartment.

Attachment

Submitted filename: Responses to reviewer Revision 21_12_2023.docx

pone.0297008.s010.docx^{(17.2KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0297008.r007

Decision Letter 3

Ricardo Ney Oliveira Cobucci

28 Dec 2023

Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervix

PONE-D-23-13996R3

Dear Dr. Hopman,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ricardo Ney Oliveira Cobucci, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Congratulations.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: No

**********

6. Review Comments to the Author

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

**********

PLoS One. doi: 10.1371/journal.pone.0297008.r008

Acceptance letter

Ricardo Ney Oliveira Cobucci

25 Jan 2024

PONE-D-23-13996R3

PLOS ONE

Dear Dr. Hopman,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

PROFESSOR Ricardo Ney Oliveira Cobucci

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Direct correlation of immunostaining results for 5-hmC, HPV, SOX2 and SOX17 in tissue sections of normal squamous epithelium, metaplastic epithelium and CIN3.

(TIF)

pone.0297008.s001.tif^{(7.6MB, tif)}

S2 Fig. Complex immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions.

(TIF)

pone.0297008.s002.tif^{(7.4MB, tif)}

S3 Fig. Quantitative analysis of 5-mC and 5-hmC immunofluorescence intensity (Arbitrary Units, AU) distributions in normal squamous epithelium, CIN1 and CIN3.

(TIF)

pone.0297008.s003.tif^{(4.4MB, tif)}

S4 Fig. Immunofluorescence staining patterns for 5-mC and 5-hmC in CIN3 lesions with overlying columnar epithelium, showing extensive heterogeneity for methylation and hydroxymethylation.

(TIF)

pone.0297008.s004.tif^{(6.3MB, tif)}

S5 Fig. Confocal images of double-label immunofluorescence staining patterns for 5-mC and 5-hmC in a CIN3 lesion.

(TIF)

pone.0297008.s005.tif^{(7.6MB, tif)}

S6 Fig. Immunofluorescence staining results for 5-mC and 5-hmC showing heterogeneity for 5-hmC in cervical adenocarcinoma.

A) H&E-stained dysplastic glands in adenocarcinoma. B) 5-mC; C) 5-hmC; D) merged mages of 5-mC and 5-hmC with DAPI counterstaining. Scale bar in A indicates 100 μm (same magnifications in B-D).

(TIF)

pone.0297008.s006.tif^{(7.1MB, tif)}

S1 Table

(PDF)

pone.0297008.s007.pdf^{(122.4KB, pdf)}

Attachment

Submitted filename: Responses to reviewer.docx

pone.0297008.s008.docx^{(21.8KB, docx)}

Attachment

Submitted filename: Figure_integration.pdf

pone.0297008.s009.pdf^{(16.3MB, pdf)}

Attachment

Submitted filename: Responses to reviewer Revision 21_12_2023.docx

pone.0297008.s010.docx^{(17.2KB, docx)}

Data Availability Statement

All relevant data are within the paper and its Supporting Information files.

[pone.0297008.ref001] 1.Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y, et al. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science. 2009;324(5929):930–5. doi: 10.1126/science.1170116 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref002] 2.Ito S, Shen L, Dai Q, Wu SC, Collins LB, Swenberg JA, et al. Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5-carboxylcytosine. Science. 2011;333(6047):1300–3. doi: 10.1126/science.1210597 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref003] 3.Szulwach KE, Li X, Li Y, Song CX, Han JW, Kim S, et al. Integrating 5-hydroxymethylcytosine into the epigenomic landscape of human embryonic stem cells. PLoS Genet. 2011;7(6):e1002154. doi: 10.1371/journal.pgen.1002154 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref004] 4.Chen Y, Damayanti NP, Irudayaraj J, Dunn K, Zhou FC. Diversity of two forms of DNA methylation in the brain. Front Genet. 2014;5:46. doi: 10.3389/fgene.2014.00046 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref005] 5.Pfeifer GP, Xiong W, Hahn MA, Jin SG. The role of 5-hydroxymethylcytosine in human cancer. Cell Tissue Res. 2014;356(3):631–41. doi: 10.1007/s00441-014-1896-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref006] 6.Wang KC, Kang CH, Tsai CY, Chou NH, Tu YT, Li GC, et al. Ten-eleven translocation 1 dysfunction reduces 5-hydroxymethylcytosine expression levels in gastric cancer cells. Oncol Lett. 2018;15(1):278–84. doi: 10.3892/ol.2017.7264 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref007] 7.Wang Z, Du M, Yuan Q, Guo Y, Hutchinson JN, Su L, et al. Epigenomic analysis of 5-hydroxymethylcytosine (5hmC) reveals novel DNA methylation markers for lung cancers. Neoplasia. 2020;22(3):154–61. doi: 10.1016/j.neo.2020.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref008] 8.Das PM, Singal R. DNA methylation and cancer. J Clin Oncol. 2004;22(22):4632–42. doi: 10.1200/JCO.2004.07.151 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref009] 9.Vrba L, Futscher BW. A suite of DNA methylation markers that can detect most common human cancers. Epigenetics. 2018;13(1):61–72. doi: 10.1080/15592294.2017.1412907 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref010] 10.Haffner MC, Chaux A, Meeker AK, Esopi DM, Gerber J, Pellakuru LG, et al. Global 5-hydroxymethylcytosine content is significantly reduced in tissue stem/progenitor cell compartments and in human cancers. Oncotarget. 2011;2(8):627–37. doi: 10.18632/oncotarget.316 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref011] 11.Bhat S, Kabekkodu SP, Varghese VK, Chakrabarty S, Mallya SP, Rotti H, et al. Aberrant gene-specific DNA methylation signature analysis in cervical cancer. Tumour Biol. 2017;39(3):1010428317694573. doi: 10.1177/1010428317694573 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref012] 12.Zhu H, Zhu H, Tian M, Wang D, He J, Xu T. DNA Methylation and Hydroxymethylation in Cervical Cancer: Diagnosis, Prognosis and Treatment. Front Genet. 2020;11:347. doi: 10.3389/fgene.2020.00347 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref013] 13.Li W, Liu M. Distribution of 5-hydroxymethylcytosine in different human tissues. J Nucleic Acids. 2011;2011:870726. doi: 10.4061/2011/870726 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref014] 14.Storebjerg TM, Strand SH, Høyer S, Lynnerup AS, Borre M, Ørntoft TF, et al. Dysregulation and prognostic potential of 5-methylcytosine (5mC), 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC) levels in prostate cancer. Clin Epigenetics. 2018;10(1):105. doi: 10.1186/s13148-018-0540-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref015] 15.Rogeri CD, Silveira HCS, Causin RL, Villa LL, Stein MD, de Carvalho AC, et al. Methylation of the hsa-miR-124, SOX1, TERT, and LMX1A genes as biomarkers for precursor lesions in cervical cancer. Gynecol Oncol. 2018;150(3):545–51. doi: 10.1016/j.ygyno.2018.06.014 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref016] 16.van Leeuwen RW, Oštrbenk A, Poljak M, van der Zee AGJ, Schuuring E, Wisman GBA. DNA methylation markers as a triage test for identification of cervical lesions in a high risk human papillomavirus positive screening cohort. Int J Cancer. 2019;144(4):746–54. doi: 10.1002/ijc.31897 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref017] 17.Yuan LY, Qin X, Li L, Zhou J, Zhou M, Li X, et al. The transcriptome profiles and methylation status revealed the potential cancer-related lncRNAs in patients with cervical cancer. J Cell Physiol. 2019;234(6):9756–63. doi: 10.1002/jcp.27661 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref018] 18.Gatti A, Haruyama R, Elit L, Howard SC, Lam CG, Ilbawi A, et al. How to reduce the impact of cervical cancer worldwide: Gaps and priority areas identified through the essential cancer and primary care packages: An analysis of effective interventions. Cancer. 2020;126(21):4697–705. doi: 10.1002/cncr.33116 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref019] 19.Brisson M, Kim JJ, Canfell K, Drolet M, Gingras G, Burger EA, et al. Impact of HPV vaccination and cervical screening on cervical cancer elimination: a comparative modelling analysis in 78 low-income and lower-middle-income countries. Lancet. 2020;395(10224):575–90. doi: 10.1016/S0140-6736(20)30068-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref020] 20.Bosch FX, Broker TR, Forman D, Moscicki AB, Gillison ML, Doorbar J, et al. Comprehensive control of human papillomavirus infections and related diseases. Vaccine. 2013;31 Suppl 7(Suppl 7):H1-31. doi: 10.1016/j.vaccine.2013.10.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref021] 21.Woodman CB, Collins SI, Young LS. The natural history of cervical HPV infection: unresolved issues. Nat Rev Cancer. 2007;7(1):11–22. doi: 10.1038/nrc2050 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref022] 22.Kremer WW, Steenbergen R, Heideman D, Kenter GG, Meijer C. The use of host cell DNA methylation analysis in the detection and management of women with advanced cervical intraepithelial neoplasia: a review. BJOG. 2021;128(3):504–14. doi: 10.1111/1471-0528.16395 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref023] 23.Steenbergen RD, Ongenaert M, Snellenberg S, Trooskens G, van der Meide WF, Pandey D, et al. Methylation-specific digital karyotyping of HPV16E6E7-expressing human keratinocytes identifies novel methylation events in cervical carcinogenesis. J Pathol. 2013;231(1):53–62. doi: 10.1002/path.4210 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref024] 24.von Knebel Doeberitz M, Prigge ES. Role of DNA methylation in HPV associated lesions. Papillomavirus Res. 2019;7:180–3. doi: 10.1016/j.pvr.2019.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref025] 25.Moshi JM, Ummelen M, Broers JLV, Ramaekers FCS, Hopman AHN. Impact of antigen retrieval protocols on the immunohistochemical detection of epigenetic DNA modifications. Histochem Cell Biol. 2023;159(6):513–26. doi: 10.1007/s00418-023-02187-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref026] 26.Moshi JM, Hoogduin KJ, Ummelen M, Henfling MER, van Engeland M, Wouters KAD, et al. Switches of SOX17 and SOX2 expression in the development of squamous metaplasia and squamous intraepithelial lesions of the uterine cervix. Cancer Med. 2020;9(17):6330–43. doi: 10.1002/cam4.3201 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref027] 27.Wang J, Su Y, Tian Y, Ding Y, Wang X. Characterization of DNA hydroxymethylation profile in cervical cancer. Artif Cells Nanomed Biotechnol. 2019;47(1):2706–14. Epub 2019/07/05. doi: 10.1080/21691401.2019.1634578 . [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref028] 28.Kato M, Onoyama I, Kawakami M, Yoshida S, Kawamura K, Kodama K, et al. Downregulation of 5-hydroxymethylcytosine is associated with the progression of cervical intraepithelial neoplasia. Plos One. 2020;15(11). ARTN e0241482 doi: 10.1371/journal.pone.0241482 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref029] 29.Zhang LY, Han CS, Li PL, Zhang XC. 5-Hydroxymethylcytosine expression is associated with poor survival in cervical squamous cell carcinoma. Jpn J Clin Oncol. 2016;46(5):427–34. doi: 10.1093/jjco/hyw002 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref030] 30.Panning B, Jaenisch R. DNA hypomethylation can activate Xist expression and silence X-linked genes. Genes Dev. 1996;10(16):1991–2002. doi: 10.1101/gad.10.16.1991 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref031] 31.Ueda Y, Enomoto T, Miyatake T, Ozaki K, Yoshizaki T, Kanao H, et al. Monoclonal expansion with integration of high-risk type human papillomaviruses is an initial step for cervical carcinogenesis: association of clonal status and human papillomavirus infection with clinical outcome in cervical intraepithelial neoplasia. Lab Invest. 2003;83(10):1517–27. doi: 10.1097/01.lab.0000092234.68751.83 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref032] 32.Ko M, Bandukwala HS, An J, Lamperti ED, Thompson EC, Hastie R, et al. Ten-Eleven-Translocation 2 (TET2) negatively regulates homeostasis and differentiation of hematopoietic stem cells in mice. Proc Natl Acad Sci U S A. 2011;108(35):14566–71. doi: 10.1073/pnas.1112317108 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref033] 33.Moran-Crusio K, Reavie L, Shih A, Abdel-Wahab O, Ndiaye-Lobry D, Lobry C, et al. Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation. Cancer Cell. 2011;20(1):11–24. doi: 10.1016/j.ccr.2011.06.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref034] 34.Boudra R, Woappi Y, Wang D, Xu S, Wells M, Schmults CD, et al. Regulation of 5-Hydroxymethylcytosine by TET2 Contributes to Squamous Cell Carcinoma Tumorigenesis. J Invest Dermatol. 2022;142(5):1270–9 e2. doi: 10.1016/j.jid.2021.09.026 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref035] 35.Figueroa ME, Abdel-Wahab O, Lu C, Ward PS, Patel J, Shih A, et al. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell. 2010;18(6):553–67. doi: 10.1016/j.ccr.2010.11.015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref036] 36.Kudo Y, Tateishi K, Yamamoto K, Yamamoto S, Asaoka Y, Ijichi H, et al. Loss of 5-hydroxymethylcytosine is accompanied with malignant cellular transformation. Cancer Sci. 2012;103(4):670–6. doi: 10.1111/j.1349-7006.2012.02213.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref037] 37.Au Yeung CL, Tsang WP, Tsang TY, Co NN, Yau PL, Kwok TT. HPV-16 E6 upregulation of DNMT1 through repression of tumor suppressor p53. Oncol Rep. 2010;24(6):1599–604. doi: 10.3892/or_00001023 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref038] 38.Wilting SM, Steenbergen RDM. Molecular events leading to HPV-induced high grade neoplasia. Papillomavirus Res. 2016;2:85–8. doi: 10.1016/j.pvr.2016.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref039] 39.Bodelon C, Vinokurova S, Sampson JN, den Boon JA, Walker JL, Horswill MA, et al. Chromosomal copy number alterations and HPV integration in cervical precancer and invasive cancer. Carcinogenesis. 2016;37(2):188–96. doi: 10.1093/carcin/bgv171 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref040] 40.Hopman ANH, Moshi JM, Hoogduin KJ, Ummelen M, Henfling MER, van Engeland M, et al. SOX17 expression and its down-regulation by promoter methylation in cervical adenocarcinoma in situ and adenocarcinoma. Histopathology. 2020;76(3):383–93. doi: 10.1111/his.13980 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref041] 41.Kafer GR, Li X, Horii T, Suetake I, Tajima S, Hatada I, et al. 5-Hydroxymethylcytosine Marks Sites of DNA Damage and Promotes Genome Stability. Cell Rep. 2016;14(6):1283–92. doi: 10.1016/j.celrep.2016.01.035 [DOI] [PubMed] [Google Scholar]

[pone.0297008.ref042] 42.Kharat SS, Ding X, Swaminathan D, Suresh A, Singh M, Sengodan SK, et al. Degradation of 5hmC-marked stalled replication forks by APE1 causes genomic instability. Sci Signal. 2020;13(645). doi: 10.1126/scisignal.aba8091 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0297008.ref043] 43.McBride AA, Warburton A. The role of integration in oncogenic progression of HPV-associated cancers. PLoS Pathog. 2017;13(4):e1006211. doi: 10.1371/journal.ppat.1006211 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Inhibition of cytosine 5-hydroxymethylation during progression of cancer precursor lesions in the uterine cervix

Jobran M Moshi

Monique Ummelen

Frank Smedts

Frans C S Ramaekers

Anton H N Hopman

Roles

Abstract

Introduction

Materials and methods

Tissues

Table 1. Selected formalin-fixed and paraffin embedded tissue samples of (pre)malignant lesions of the uterine cervix.

Tissue processing for immunohistochemistry

Antibodies and immunohistochemical protocols

Fluorescence and bright field microscopy

Results

5-mC and 5-hmC immunostaining levels in normal squamous epithelium

Fig 1. Double-label immunostaining of 5-mC and 5-hmC in normal squamous epithelium.

Fig 2. Double-label immunostaining results for 5-mC and 5-hmC in the squamo-columnar junction (SCJ).

Fig 3. Immunostaining for 5-hmC in (immature) metaplastic squamous epithelium.

5-mC and 5-hmC immunostaining levels in the normal columnar epithelium and reserve cells

Fig 4. Immunostaining results for 5-mC and 5-hmC in normal endocervical columnar epithelium.

Fig 5. Phenotypic profile and 5-mC and 5-hmC immunostaining results of normal glandular epithelium and underlying reserve cells.

Complex immunostaining patterns for 5-mC and 5-hmC in squamous intraepithelial lesions

Fig 6. Complex immunofluorescence staining patterns for 5-mC and 5-hmC in CIN lesions.

Immunostaining results of 5-mC and 5-hmC in premalignant endocervical glandular lesions

Fig 7. Immunostaining patterns of 5-mC and 5-hmC in normal glandular epithelium and adenocarcinoma in situ (AIS).

Fig 8. Immunostaining of 5-mC and 5-hmC in adenocarcinoma in situ (AIS).

Immunostaining results for 5-mC and 5-hmC in cervical adenocarcinomas

Fig 9. Immunostaining results of 5-mC and 5-hmC in adenocarcinomas.

Discussion

Dynamics in 5-mC levels

Dynamics in 5-hmC levels

Influence of HPV infection and integration on 5-mC and 5-hmC levels

Clinical utility

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Ricardo Ney Oliveira Cobucci

Roles

Author response to Decision Letter 0

Decision Letter 1

Ricardo Ney Oliveira Cobucci

Roles

Author response to Decision Letter 1

Decision Letter 2

Ricardo Ney Oliveira Cobucci

Roles

Author response to Decision Letter 2

Decision Letter 3

Ricardo Ney Oliveira Cobucci

Roles

Acceptance letter

Ricardo Ney Oliveira Cobucci

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases