DNA hypomethylation and aberrant expression of the human endogenous retrovirus ERVWE1/syncytin-1 in seminomas
- Martina Benešová†1,
- Kateřina Trejbalová†1Email author,
- Denisa Kovářová1,
- Zdenka Vernerová2,
- Tomáš Hron1,
- Dana Kučerová1 and
- Jiří Hejnar1Email authorView ORCID ID profile
© The Author(s) 2017
Received: 21 November 2016
Accepted: 3 March 2017
Published: 17 March 2017
Syncytin-1 and 2, human fusogenic glycoproteins encoded by the env genes of the endogenous retroviral loci ERVWE1 and ERVFRDE1, respectively, contribute to the differentiation of multinucleated syncytiotrophoblast in chorionic villi. In non-trophoblastic cells, however, the expression of syncytins has to be suppressed to avoid potential pathogenic effects. Previously, we have shown that the transcriptional suppression of ERVWE1 promoter is controlled epigenetically by DNA methylation and chromatin modifications. In this study, we describe the aberrant expression of syncytin-1 in biopsies of testicular germ cell tumors.
We found efficient expression and splicing of syncytin-1 in seminomas and mixed germ cell tumors with seminoma component. Although another fusogenic gene, syncytin-2 was also derepressed in seminomas, its expression was significantly lower than that of syncytin-1. Neither the transcription factor GCM1 nor the increased copy number of ERVWE1 were sufficient for this aberrant expression of syncytin-1 in seminomas. In accordance with our recent finding of the highly increased expression of TET1 dioxygenase in most seminomas, the ERVWE1 promoter was significantly hypomethylated in comparison with the matched controls. In contrast, 5-hydroxymethylcytosine levels were not detectable at the ERVWE1 promoter. We further describe that another endogenous retroviral element adjacent to ERVWE1 remains transcriptionally suppressed and two additional HERV-W family members are only slightly upregulated in seminomas.
We conclude that DNA demethylation of the ERVWE1 promoter in seminomas is a prerequisite for syncytin-1 derepression. We propose the spliced syncytin-1 expression as a marker of seminoma and suggest that aberrant expression of endogenous retroviruses might be a correlate of the hypomethylated genome of seminomas.
KeywordsHuman endogenous retrovirus ERVWE1 Germ cell tumor Seminoma Promoter DNA methylation 5-Hydroxymethylcytosine Transcription RNA splicing
Human endogenous retroviruses (HERVs) represent the remnants of the past retroviral infections in the human lineage that penetrated the germ line leaving behind the heritable proviral copies in the human genome. With a few exceptions, HERVs have become defective by gradual accumulation of inactivating mutations in their open reading frames. Transcription of most HERV copies is silenced by epigenetic mechanisms such as DNA methylation and association with histone molecules marked by suppressive modifications.
Although endogenous retroviruses are widely employed in genome evolution as a source of regulatory sequences and interfere with new exogenous infections, there are few documented captures of intact retrovirus-encoded proteins for cellular functions. Syncytin-1 and -2 are retroviral envelope glycoproteins whose original receptor-mediated virus-to-cell fusion capacity evolved towards cell-to-cell fusion [1–3]. This fusogenic potential unfolds in all respects in placenta where syncytins are indispensable for syncytiotrophoblast differentiation. Only recently, the contribution of fusogenic syncytins to myogenesis and osteoclast fusion was suggested for mouse and human [1, 4]. On the other hand, the expression of syncytins must be suppressed in non-syncytial tissues, and, therefore, the capture of fusogenic retroviral envelope glycoproteins has to be accompanied by the gain of tissue-specific regulation. The importance of tissue-specific regulation of syncytin-1 is even underscored by the ubiquitous expression of its specific receptors, the human sodium-dependent neutral amino acid transporters type 1 and 2 (ASCT1, also known as SLC1A4, and ASCT2/SLC1A5) [5, 6].
Syncytin-1 is encoded by the env gene of the ERVWE1 provirus (NCBI accepted name ERVW1), a prototype member of the HERV-W family, localized on chromosome 7 [7, 8]. Both cis and trans regulatory circuits have been described [9, 10]; however, the tissue-specificity of syncytin-1 expression is controlled epigenetically. We and others demonstrated that the ERVWE1 transcription was regulated by DNA methylation and trimethylation of H3K9 of the ERVWE1 5′ LTR [11–14]. Furthermore, we showed that the splicing of ERVWE1 mRNA occurrs in trophoblastic but not in non-placental cells  and serves as an additional control mechanism. Syncytin-2 encoded by the env gene of a unique member of the HERV-FRD family, ERVFRDE1 (NCBI accepted name ERVFRD1), is also important for the fusion of human cytotrophoblast . Like ERVWE1, ERVFRDE1 is also regulated epigenetically [14, 16], but the role of DNA methylation and splicing is less understood.
Aberrant expression of Syncytin-1 has been reported in multiple tumor types, but epigenetically-based evidence is available just for endometrial carcinomas [17, 18] and several samples of testicular tumors without exact characterization [13, 14]. Testicular germ cell tumors (GCT) originate from embryonic primordial germ cells (PGC) or gonocytes as in situ neoplasias and transform into seminomas. Non-seminomas require further development, probably with reprogramming and dedifferentiation steps . GCTs maintain, to a various extent, the epigenetic characteristics of their PGC precursors, i.e., DNA hypomethylation and low levels of H3K9 trimethylation [20, 21]. Comparison of seminomas and non-seminomas showed a lower level of genome methylation in seminomas and increased methylation in non-seminomas [22–26]. This again suggests that seminomas and non-seminomas arise in distinct periods of the PGC development with different degrees of cell differentiation. Recently published analyses of the transcriptional and epigenetic landscape during human PGC development revealed progressive erasure of DNA methylation not only from the global genome, but also from the transposable elements [27–29].
The deeply hypomethylated genome of seminoma cells has been recently correlated with elevated expression of the TET1 enzyme in GCTs . The DNA demethylation activity of TET dioxygenases proceeds through 5-hydroxymethylcytosine (5-hmC) intermediate , which subsequently converts to 5-formylcytosine (5-fC), 5-carboxycytosine and unmodified cytosine (C) [32, 33]. At least in mouse, Tet1 and Tet2 expression has been observed in late PGCs  and together with the repression of de novo DNA methyltransferases  create the hypomethylated germ line genome.
Based on the knowledge of DNA hypomethylation in GCTs and the observation of syncytin-1 expression in testicular tumors [13, 14], we explored the ERVWE1 expression systematically within a panel of GCTs with particular respect given to discrimination between seminomas and non-seminoma GCTs. In addition, we included several samples of lymphomas and endometrial carcinomas in our analysis because of (1) the presence of multinuclear giant Reed-Sternberg cells in Hodgkin lymphomas , (2) the detection of full-length ERVWE1 mRNA in endometrial carcinomas [17, 18, 37], and (3) the recently described increased expression of DNA demethylating dioxygenases TET2 and TET3 in seminomas, lymphomas and endometrial carcinomas  that could contribute to ERVWE1 transcriptional derepression. We estimated, for the first time, the levels of both 5-methylcytosine (5-mC) and 5-hmC modifications at the ERVWE1 promoter within the 5′LTR. We also examined the transcription level of four other endogenous retroviruses, including ERVFRDE1, in the GCTs.
Testicular samples were collected from patients who were surgically treated at the Institute of Urology, University Hospital Kralovske Vinohrady, and Third Faculty of Medicine, Charles University between years 2011–2015. Histological classification and TNM staging were done in accordance with WHO classification . Lymphoma samples were acquired from the Department of Otorhinolaryngology, University Hospital Kralovske Vinohrady, and Department of Surgery, University Hospital Kralovske Vinohrady. Endometrial carcinomas and six healthy term placental samples were obtained from the Department of Obstetrics and Gynecology, University Hospital Kralovske Vinohrady. Immediately after the surgery, the samples were frozen and stored in −80 °C.
Our extensive set of post-pubertal testicular malignant GCT samples contained 30 pure seminomas, one scar after the seminoma, and 17 non-seminomas. The non-seminomas included 12 mixed GCTs, two pure embryonal carcinomas, two pure teratomas, and one pure yolk sac tumor. For the majority of cancer biopsies, the tumor-matched controls were collected from the adjacent tissue not affected macroscopically with the GCT. In some cases, two biopsies from distinct parts of the tumor were analyzed. The set also contained ten samples of the testes not diagnosed with the GCTs (termed non-GCT controls throughout the text). The non-GCT controls included five samples of atrophic testes, one sample with testicular necrotic tissue, two testis after ischemia–reperfusion injury, and two testis without GCT. Our set of malignant human tumor samples further contained tumors of non-germ cell origin, specifically three Hodgkin lymphomas and five non-Hodgkin lymphomas, and seven endometrial carcinomas, including six endometrioid and one non-endometrioid. Description of tumor samples is indicated in Additional file 1: Table S1.
Cell culture procedures
Choriocarcinoma BeWo cells were maintained in F-12 and MEM-D media mixed 1:1 (Sigma) supplemented with 1% NaHCO3, 10% FBS. Seminoma TCam-2 cells were maintained in RPMI-1640 (Sigma) supplemented with 10% FBS and 0.3 mg/ml l-glutamine. A mix of penicillin and streptomycin (0.1 mg/ml each) was added to both cell lines in culture. Cells were kept at 37 °C in humidified atmosphere of 5% CO2.
In order to confirm the results normalized to POLR2A, we additionally normalized expressions of the full-length ERVWE1, spliced syncytin-1, full-length ERVFRDE1 and spliced syncytin-2 in all samples also to the TATA-box binding protein (TBP) gene. The calibration curve derived from serial ten-fold dilutions of the known quantity of TBP molecules was used to calculate the absolute numbers of mRNA molecules. The results were proportional to the POLR2A normalization (see Additional file 3: Fig. S1 (A) to (D)) corresponding to the primer efficiency of 93 and 95% for POLR2A and TBP, respectively.
Droplet digital PCR
To analyze the copy number variation and expression of different human endogenous retroviruses, we performed absolute quantification by means of droplet digital PCR (ddPCR). For the copy number variation we used total chromosomal DNA, for the expression analysis we used the synthetized cDNA. The PCR reaction contained ddPCR™ Supermix for Probes and Droplet Generation Oil for Probes and was performed by means of QX100™ Droplet Digital™ PCR System (all BioRad). The protocol for the copy number analysis consisted of: 95 °C 10 min, 95 °C 30 s–60 °C 60 s—40 cycles. For the qRT-PCR analysis, ddPCR protocol was following: 95 °C 10 min, 95 °C 30 s–57 °C 30 s–72 °C 30 s—43 cycles. The data were analyzed with QuantaSoft™ software. All reactions were run in duplicate and the average concentrations were used for quantification. The negative control contained water instead of DNA or cDNA. For both analyses, the same PCR reaction contained two pairs of primers and two probes in parallel to quantify the gene of interest and the gene used for normalization at the same time. The copy number variation of ERVWE1 was normalized according to the RPP30 (Ribonuclease P/MRP Subunit P30) copy number. The absolute expression of human endogenous retrovirus HERV-H was normalized according to the POLR2A expression, and respective RT-minus controls reaching up to 5% of their RT-plus counterparts were subtracted. The specificity of primers and probes was verified by sequencing the amplified products. Sequences of primers and probes used for ddPCR analysis are indicated in Additional file 2: Table S2.
Chromosomal DNA isolation
Samples of total genomic DNA from the tissues, TCam-2, and BeWo cell lines were isolated using Proteinase K, RNase A, phenol–chloroform extraction and ethanol precipitation.
Bisulfite and oxidative bisulfite sequencing
Simple bisulfite treatment does not distinguish between 5-mC and 5-hmC, and keeps both 5-mCs and 5-hmCs as Cs while converting all unmodified Cs into uracils (U). On the other hand, oxidative DNA treatment ensures quantitative conversion of 5-hmC bases into 5-fCs. The bisulfite treatment of oxidized DNA converts the 5-fCs and the non-modified Cs into Us, while only the 5-mCs are kept as non-converted Cs. Therefore, to identify the percentage of 5-hmC in a specific sequence, the percentage of 5-mC obtained after oxidative treatment plus bisulfite sequencing must be subtracted from the percentage of 5-modified C obtained after the simple bisulfite treatment. One µg of genomic DNA was used for simple bisulfite conversion or oxidative treatment plus bisulfite conversion by means of TrueMethyl® Seq kit (CEGX, Cambridge Epigenetix) applied according to the protocol recommended by the manufacturer. Bisulfite-treated DNA was amplified by PCR specific for 5′ LTR in a 50-µl reaction mixture using ERVWE1-BIS-FW and ERVWE1-BIS-RV primers. See Additional file 2: Table S2 for the primer sequences. The bisulfite-specific sense primers contained thymine and the antisense primers adenine instead of C in positions complementary to non-methylable C (i.e., C out of CpG dinucleotides). One 50-μl reaction contained: 2 μl of bisulfite/oxidative bisulfite-treated product, 2.5 mM MgCl2, 2 μg HotStart-IT® Binding Protein (Affymetrix), 0.32 μM primers, 2 U Platinum Taq polymerase (ThermoFisher Scientific) and 0.2 mM dNTPs (Promega). The following PCR program was applied: 95 °C 5 min, 95 °C 50 s–58 °C 2 min–68 °C 1 min 30 s—25 cycles; 95 °C 45 s–54 °C 2 min–68 °C 1 min 30 s—15 cycles and 68 °C 10 min. At least three primary PCRs were performed for each sample to exclude amplification of one template molecule. Non-converted DNA did not provide bands. Several non-template controls were included in each bisulfite PCR reaction. Amplification products were cloned in the pGEM-T-EasyVector System (Promega) and sequenced. Analysis was performed using the Quma (Quantification tool for Methylation Analysis) software (http://quma.cdb.riken.jp/). Only PCR clones with at least 95% conversion of cytosines outside CpGs were taken into account. When more converted molecules with identical sequences were obtained, only one was used for calculation of the methylated CpG percentage to minimize the bias originating from the preferential amplification of one molecule.
The efficiency of 5-hmC to 5-fC oxidation was controlled using the 5-hmC-containing synthetic probe. The corresponding region of the ERVWE1 promoter was amplified by PCR from genomic DNA using the dNTP mix with dhmCTP replacing the dCTP (see Additional file 2: Table S2 for the primer sequences ERVWE1-nonBIS-FW and ERVWE1-nonBIS-RV). The amplified product of 538 bp containing 155 5-hmCs was isolated from the agarose gel and 100 ng of the 5-hmC-containing fragment was subjected to oxidative bisulfite treatment by means of TrueMethyl® Seq kit (CEGX, Cambridge Epigenetix). Thereafter, PCR was performed using the bisulfite-specific ERVWE1 primers (see Additional file 2: Table S2, primers ERVWE1-BIS-FW and RV). Amplification products of three primary PCRs were cloned in the pGEM-T-Easy vector and sequenced. Analysis was performed using the Quma software and all PCR clones were taken into account regardless of the C conversion rate. Overall 5-hmC to U conversion levels of 97.9 to 99.3% were observed in individual molecules of the oxidation controls indicating efficient oxidation.
For statistical analysis of data GraphPad Prism software, version 5.04, was used and two-tailed Mann–Whitney test was applied. Significance was assigned as follows: **** for P-values <0.0001, *** for P-values <0.001, ** for P-values <0.01, * for P-values <0.05.
Derepression of ERVWE1 transcription in seminoma and other GCTs
In order to assess the ERVWE1 expression in the GCTs with particular respect to seminomas, we employed qRT-PCR and quantified the absolute levels of full-length ERVWE1 RNA (Fig. 1) in our panel of tumor samples and in tumor-matched controls (Additional file 1: Table S1). In total, 30 pure seminomas, one scar after the seminoma, and 17 non-seminoma GCTs, mostly mixed GCTs, were examined.
On the other hand, the analyzed samples of endometrial carcinomas and lymphomas displayed a significantly lower level of full-length ERVWE1 RNA (median 1 and 10% of POLR2A, respectively) than the seminomas (Fig. 2a). No significant differences in ERVWE1 RNA levels were observed between Hodgkin and non-Hodgkin lymphomas (median 14 and 5% of POLR2A, respectively, Additional file 4: Fig. S2). Unexpectedly, placenta samples also displayed a low level of full-length ERVWE1 RNA (median 13% of POLR2A, Fig. 2a).
In conclusion, seminomas displayed significantly higher levels of non-spliced ERVWE1 RNA in comparison with the seminoma-matched controls and also with the non-seminomas or endometrial carcinomas and lymphomas.
Efficient splicing of ERVWE1 RNA in seminoma
Fusogenic Syncytin-1 can be translated only from spliced ERVWE1 env mRNA (Fig. 1). Therefore, we quantified the absolute levels of spliced syncytin-1 mRNA in GCTs employing the splice-specific qRT-PCR described previously .
We detected a very high level of syncytin-1 mRNA in seminomas (median 49% of POLR2A), BeWo cells (80% of POLR2A), and particularly placenta (median 820% of POLR2A, Fig. 2b). BeWo cells and placentas validated our splice-specific qRT-PCR assay because Syncytin-1 has been shown to be placenta-specific  and Syncytin-1-mediated cell-to-cell fusion was described in BeWo cells. Syncytin-1 mRNA expression in TCam-2 cell line reached 32% of POLR2A. The level of syncytin-1 mRNA in seminomas was significantly higher than in the seminoma-matched controls (5% of POLR2A) and non-seminoma GCTs (2% of POLR2A, Fig. 2b). Importantly, two mixed GCTs with the most abundant expression of full-length ERVWE1 RNA, T33 and T32, also displayed high levels of spliced syncytin-1 mRNA. Both tumors contained a large proportion of seminoma. In contrast, T43 with the highest level of full-length ERVWE1 RNA displayed only a low level of spliced syncytin-1 mRNA (3% of POLR2A). This mixed GCT lacked any seminoma component. Only negligible levels of the spliced syncytin-1 mRNA were found in the non-GCT testes, endometrial carcinomas and lymphomas (Fig. 2b). No significant differences in syncytin-1 mRNA levels were observed between Hodgkin and non-Hodgkin lymphomas (Additional file 4: Fig. S2).
We further determined the splicing efficiency as a ratio of spliced syncytin-1 mRNA to the sum of spliced and full-length ERVWE1 RNA in the same sample. Despite significantly higher amounts of ERVWE1 RNA and syncytin-1 mRNA in seminomas versus seminoma-matched controls, the efficiency of splicing in seminomas and seminoma-matched controls did not differ significantly (Fig. 2c). However, the efficiency of splicing in seminomas (median 0.63) was significantly higher in comparison with the non-seminomas (median 0.34, Fig. 2c). The 0.99, 0.79 and 0.77 splicing efficiencies were found in placentas, TCam-2 and BeWo cell lines, respectively. On the other hand, the lowest efficiency of splicing was detected in the non-GCT testes (median 0.13), endometrial carcinomas (median 0.21) and lymphomas (median 0.00).
Our results showed that, similarly to the non-spliced ERVWE1 RNA, the levels of the spliced syncytin-1 mRNA were upregulated in seminomas over the seminoma-matched controls and, again, over the non-seminomas. In seminomas, the ERVWE1 RNA was transcribed at high intensity, and the majority of this RNA was efficiently spliced into the syncytin-1 mRNA.
Derepression of ERVFRDE1 expression in seminoma
ERVFRDE1 is the second endogenous retrovirus encoding placenta-specific, fusogenic glycoprotein Syncytin-2. Therefore, we examined the expression of both full-length and spliced ERVFRDE1 RNAs in our panel of seminomas and non-seminoma GCTs.
We further compared the efficiency of ERVFRDE1 splicing and found no significant differences between the seminomas and non-seminomas (Fig. 3c). Nevertheless, the efficiency of ERVFRDE1 splicing was lower in seminomas than in the seminoma-matched controls or the non-GCT testes, endometrial carcinomas or lymphomas (Fig. 3c). As in the case of ERVWE1, placentas displayed very high efficiency of ERVFRDE1 RNA splicing (median 0.99). Interestingly, the splicing efficiency of ERVFRDE1 RNA in both seminomas and non-seminomas was higher than that of ERVWE1 RNA (Additional file 5: Fig. S3).
Overall, elevated levels of the full-length ERVFRDE1 RNA and spliced syncytin-2 mRNA were found in seminomas in comparison to non-seminoma GCTs. When comparing the levels of the full-length and spliced forms of ERVWE1 versus ERVFRDE1 in the non-seminomas, the spliced syncytin-2 mRNA was slightly elevated over the spliced syncytin-1 mRNA (P > 0.05). On the other hand, in the seminomas, the levels of both non-spliced ERVWE1 RNA and spliced syncytin-1 mRNA were significantly higher than the levels of the non-spliced ERVFRDE1 RNA and spliced syncytin-2 mRNA, respectively (P > 0.0001 in both cases, Additional file 6: Fig. S4A and B). These results underscore the significance of ERVWE1 and syncytin-1 overexpression in seminomas.
ASCT1 and ASCT2, the receptors of Syncytin-1, are not overexpressed in GCTs
Copy number of ERVWE1 in seminoma
Next, in order to determine the mechanism of ERVWE1 overexpression in seminomas, we quantified the ERVWE1 copy number using droplet-digital PCR. We examined the genomic DNA of five seminomas and four seminoma-matched controls, six non-seminomas and the corresponding matched controls, one non-GCT testis, one lymphoma, one endometrial carcinoma and one placenta sample.
In conclusion, we detected increased ERVWE1 copy numbers in several GCTs. However, the higher ERVWE1 copy number did not correspond to the expression level of ERVWE1 RNA or syncytin-1 mRNA. Therefore, the increased ERVWE1 copy number was not sufficient to explain ERVWE1 derepression in GCTs.
GCM1 is not overexpressed in GCTs or seminomas
Low level of CpG methylation and hydroxymethylation of ERVWE1 promoter in seminoma
We have shown previously that DNA hypermethylation suppresses transcription from the ERVWE1 promoter in non-placental tissues [12, 13] and we assume that demethylation is a prerequisite for ERVWE1 derepression. Recently, we have described elevated expression of TET enzymes in GCTs, particularly seminomas . We therefore explored the levels of 5-mC and 5-hmC at the ERVWE1 promoter using bisulfite sequencing and oxidative bisulfite modification. The combination of these techniques allows discrimination between 5-mC and 5-hmC.
Percentage of 5-mCpG and 5-hmCpG on the ERVWE1 promoter
Total 5-mCpG + 5-hmCpGa
Oxidative bisulfite sequencing
The results showed that 5-hmC levels at the ERVWE1 promoter in the GCTs were undetectable despite the high expression of TET1-3 mRNAs . These findings pointed out the importance of 5-hmC turnover kinetics. Importantly, seminomas displayed low levels of 5-mC modification at the ERVWE1 promoter that allowed transcription from the ERVWE1 promoter to occur.
Expression of additional endogenous retroviruses in GCTs
Finally, to assess whether derepression of endogenous retroviruses in seminomas is a more common phenomenon or whether it is ERVWE1-specific, we explored transcription of the HERV-H endogenous retroviral element co-localized with ERVWE1 on chromosome 7 (Fig. 1), in randomly selected samples of GCTs using ddPCR. We further quantified RNA levels of two additional non-pseudogenic HERV-W elements, localized on chromosomes 4 and 21 (Fig. 1), using the RT-qPCR approach. We selected these HERV-W loci because they were previously found overexpressed in testicular tumors using RNA microarray analysis .
In conclusion, the expression of HERV-H located upstream to ERVWE1 and two additional HERV-W loci do not reach by far the levels of either full-length or spliced ERVWE1 RNA in seminomas (compare Figs. 2a, b and 8, see also the Additional file 6: Fig. S4). These results indicate that ERVWE1 is regulated independently of the other analyzed endogenous elements located either in close proximity to ERVWE1 or in remote genomic regions.
In this study, we have described elevated syncytin-1 mRNA expression in seminomas and mixed GCTs with a seminoma component. Our previous studies showed that ERVWE1 is suppressed by epigenetic mechanisms in non-placental tissues [12, 13]. This is further corroborated in this study showing the importance of ERVWE1 5′ LTR hypomethylation for its expression. Another mechanism indispensable for this aberrant upregulation in the seminoma was efficient splicing of ERVWE1 RNA providing the template for syncytin-1 translation. In summary, we show that the full-length and particularly the spliced form of ERVWE1 RNA is a hallmark of seminoma.
Although there are many examples of aberrant HERV expression in cancer (reviewed in ), ERVWE1 and ERVFRDE1 are exceptional in several aspects: expression of their envelope glycoproteins could have functional consequences for the cell and is, therefore, tightly controlled in non-placental tissues [11–14]. Second, only spliced transcripts can be used for envelope glycoprotein translation, which adds a further step to the complex control mechanisms . Splicing efficiency has not been studied at all in cancer-specific aberrant expression of other HERV families. Third, the trophoblastic expression of ERVWE1 is controlled from a juxtaposed trophoblast-specific enhancer localized in MaLR LTR  and requires stimulation by certain tissue-specific transcription factors, such as GCM1 [10, 39], whose activity could be studied in germ line tumors. In neither type of tumors described in this study, we observed GCM1 expression comparable with placenta or BeWo cells, which points to the critical importance of promoter demethylation as a prerequisite for ERVWE1 transcriptional de-repression.
We did not observe elevated expression of ERVWE1 RNA in endometrial carcinomas as it was reported in . Our collection of endometrial cancers was less extensive, we examined only seven endometrial carcinomas, including six endometrioid and one clear cell endometrial carcinomas of endometrial and myometrial localization, all in stage T1. Lymphomas showed slightly elevated levels of the full-length ERVWE1 RNA but only the background levels of the spliced syncytin-1 mRNA indicating very low splicing efficiency of the ERVWE1 RNA (Fig. 2). Comparison of the Hodgkin and non-Hodgkin lymphomas is shown in Additional file 4: Fig. S2.
The fusogenic capacity of Syncytin-1 and Syncytin-2 imposes the idea of syncytial activity in germ cell tumors with increased levels of spliced syncytin mRNA. We observed syncytial cells with low numbers of nuclei in the seminoma cell line TCam-2 as well as multinucleated syncytia in the BeWo cell line correlating well with the high expression of Syncytin-1 receptors, ASCT1 and ASTC2 in these cell lines. On the other hand, we did not observe systematic overexpression of ASCT2 in the GCTs. However, both seminomatous and, to a lesser extent, mixed GCTs are known to contain neoplastic giant cells of various types including mono- or multinuclear elements  with increased hCG expression . It remains to be studied whether the presence of syncytia in GCTs requires syncytin-1 and -2 expression.
Interestingly, the methylation patterns of ERVWE1 promoter in several analyzed seminomas and placentas were bimodal showing either completely demethylated molecules or heavily methylated molecules. Only the T5 seminoma displayed a patchy methylation pattern of CpGs (Additional file 9: Fig. S7). This observation could underline the importance of promoter DNA demethylation for transcriptional derepression. A similar “all or nothing” pattern has also been observed in promoters of latent or activated HIV-1 proviruses . Alternatively, different cellular types present in the samples could produce contrasting methylation patterns. The technical procedure of bisulfite sequencing and analysis used in this study was designed to avoid biased results by preferential amplification of single molecules and already adjusted for extremely rare targets .
In our recent work , we documented the hypomethylated genome of seminomas and associated this hypomethylation with strikingly elevated levels of TET dioxygenases, enzymes with DNA demethylase activities. Any decrease in DNA methylation affects particularly the repetitive part of the genome, as up to 90% of methylated CpGs are found in repetitive elements including the endogenous retroviruses . Accordingly, there are reports on activation of HERVs and retrotransposons in 5-azacytidine-treated cells [44–46], embryonic stem cells , and induced pluripotent stem cells [48, 49]. ERVWE1, and to a lesser extent ERVFRDE1 derepression could be the direct consequence of the genome hypomethylation in seminoma cells. This, however, does not apply generally to all HERVs, and other proviral copies localized either in close proximity to ERVWE1 or on chromosomes 4 and 21 are not efficiently derepressed.
There are several reports of HERV-K expression in GCTs (e.g. ) and inducible expression of HERV-K rec gene in mice led to a disturbance of germ cell development reminiscent of early seminoma . Genome-wide comparison of HERV expression in testicular cancer and normal testicular tissue has been done using copy-specific microarray and several HERV copies of families E, H, K, and W were found to be differentially expressed in testicular tumors. The expression of six HERV-W copies including ERVWE1 was confirmed by quantitative RT-PCR and the prevalence of spliced ERVWE1 transcripts was demonstrated. In comparison to non-testicular tumors and control healthy tissues, normal testicular tissue was not exceptional and expressed several ubiquitously represented HERV-E, H, and K copies.
Our finding of TET1 overexpression in seminomas  implicates the role of 5-hmC in transcriptional derepression. We, however, observed a very low or undetectable level of 5-hmC at the ERVWE1 promoter in seminomas examined in this study. Although unexpected, this paucity of 5-hmC is not at odds with previous reports. E.g., 5-hmC was detectable in early human PGCs but get gradually depleted later on . The genic regions in the human PGCs showed a low but significant levels of 5-hmC whereas the retroelements showed very low levels of 5-hmC . Because seminoma maintains multiple features of PGCs, this differential 5-hmC distribution could explain also the lack of 5-hmC on the ERVWE1 promoter.
The ERVFRDE1 expression was always lower than that of ERVWE1 resembling the low ERVFRDE1 level in placenta and BeWo cells (this study and ). Also, the level of DNA methylation at ERVFRDE1 5′LTR in healthy matched controls was rather low in comparison with ERVWE1 [13, 14]. On the other hand, splicing of ERVFRDE1 RNA was quite efficient despite low expression levels. Further research of the ERVFRDE1 promoter epigenetic status in the context of its transcriptional activation would improve our understanding of Syncytin-1 and Syncytin-2 contribution to cell-to-cell fusion.
In summary, we analyzed ERVWE1 and ERVFRDE1 expression and splicing in a wide and well characterized set of germ cell tumors. Particularly, we correlated the ERVWE1 expression, low CpG methylation and, for the first time, also low CpG hydroxymethylation status of the ERVWE1 promoter with the whole genome hypomethylation phenotype of seminoma. We propose the spliced syncytin-1 expression as a marker of seminoma or seminomatous component of non-seminoma GCTs. As seminoma and non-seminoma treatments differ diametrically, we assume that detection of the spliced syncytin-1 mRNA could be useful for differential diagnosis and individualized therapy of GCTs. The elevated ERVWE1 expression without significantly increased copy number or increased level of GCM underscores the importance of epigenetic control of HERV-Ws. Identification of other HERV loci transcriptionally active in different types of GCTs would improve our understanding of endogenous retrovirus regulation.
endogenous retrovirus WE1
endogenous retrovirus FRDE1
alanine/serine/cysteine/threonine transporter 1
alanine/serine/cysteine/threonine transporter 2
glial cells missing homolog 1
ten-eleven translocation methylcytosine dioxygenase 1
ten-eleven translocation methylcytosine dioxygenase 2
ten-eleven translocation methylcytosine dioxygenase 3
germ cell tumor
primordial germ cell
KT, MB, ZV, JH, have conceived and designed the experiments. MB, KT, DK, DK, ZV, have performed the experiments. KT, MB, ZV, DK, TH, and JH have analyzed the data. ZV, TH, MB, DK, DK have contributed reagents/materials/analysis. KT, MB, ZV, JH have written the paper. All authors read and approved the final manuscript.
We thank Petr Klézl, M.D., Ph.D., from the Department of Urology, Robert Gürlich, M.D., Ph.D., from the Department of Surgery, Aleš Čoček, M.D., Ph.D., from the Department of Otorhinolaryngology, and Pavla Líbalová, M.D., Ph.D., from the Department of Obstetrics and Gynecology, University Hospital Kralovske Vinohrady for their help with obtaining clinical samples. We also thank Lubomíra Pecnová for her technical assistance with molecular analyses.
The authors declare that they have no competing interests.
Availability of data and materials
The data supporting this work’s conclusions is included within the manuscript and its additional files.
Ethics approval and consent to participate
The study was conducted in accordance with the principles of WMA Declaration of Helsinki. The study was approved by the Ethics Committee of the University Hospital Kralovske Vinohrady (Reference Number EK-VP/06/2012); each patient provided informed written consent to participate in this study. After intervention, all samples were kept and processed at the Institute of Molecular Genetics under the guidance and regulation of the internal Committee for Ethics, Manipulation with Recombinant DNA and Clinical Research of the Institute of Molecular Genetics. The study was approved by the Ethics Committee of the University Hospital Kralovske Vinohrady under the Reference Number EK-VP/06/2012. Each patient provided informed written consent to participate in this study.
This work was supported by Grant No. 13-3700S awarded by the Czech Science Foundation to JH and Grant No. NT14601-3/2013 awarded by Grant Agency of Ministry of Health of the Czech Republic to JH and ZV. The work was also institutionally supported by RVO: 68378050 and the research project PRVOUK—Oncology P27, awarded by Charles University in Prague to ZV, PK and AS.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Blaise S, de Parseval N, Benit L, Heidmann T. Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc Natl Acad Sci USA. 2003;100(22):13013–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Frendo JL, Olivier D, Cheynet V, Blond JL, Bouton O, Vidaud M, et al. Direct involvement of HERV-W Env glycoprotein in human trophoblast cell fusion and differentiation. Mol Cell Biol. 2003;23(10):3566–74.View ArticlePubMedPubMed CentralGoogle Scholar
- Mi S, Lee X, Li X, Veldman GM, Finnerty H, Racie L, et al. Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature. 2000;403(6771):785–9.View ArticlePubMedGoogle Scholar
- Moller AMJ, Delaisse J-M, Soe K. Osteoclast fusion: time-lapse reveals involvement of CD47 and syncytin-1 at different stages of nuclearity. J Cell Physiol. 2016;232(6):1396–140.View ArticlePubMedGoogle Scholar
- Lavillette D, Marin M, Ruggieri A, Mallet F, Cosset FL, Kabat D. The envelope glycoprotein of human endogenous retrovirus type W uses a divergent family of amino acid transporters/cell surface receptors. J Virol. 2002;76(13):6442–52.View ArticlePubMedPubMed CentralGoogle Scholar
- Marin M, Lavillette D, Kelly SM, Kabat D. N-linked glycosylation and sequence changes in a critical negative control region of the ASCT1 and ASCT2 neutral amino acid transporters determine their retroviral receptor functions. J Virol. 2003;77(5):2936–45.View ArticlePubMedPubMed CentralGoogle Scholar
- Grandi N, Cadeddu M, Blomberg J, Tramontano E. Contribution of type W human endogenous retroviruses to the human genome: characterization of HERV-W proviral insertions and processed pseudogenes. Retrovirology. 2016;13(1):67.View ArticlePubMedPubMed CentralGoogle Scholar
- Pavlicek A, Paces J, Elleder D, Hejnar J. Processed pseudogenes of human endogenous retroviruses generated by LINEs: their integration, stability, and distribution. Genome Res. 2002;12(3):391–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Prudhomme S, Oriol G, Mallet F. A retroviral promoter and a cellular enhancer define a bipartite element which controls env ERVWE1 placental expression. J Virol. 2004;78(22):12157–68.View ArticlePubMedPubMed CentralGoogle Scholar
- Yu C, Shen K, Lin M, Chen P, Lin C, Chang GD, et al. GCMa regulates the syncytin-mediated trophoblastic fusion. J Biol Chem. 2002;277(51):50062–8.View ArticlePubMedGoogle Scholar
- Gimenez J, Montgiraud C, Oriol G, Pichon JP, Ruel K, Tsatsaris V, et al. Comparative methylation of ERVWE1/syncytin-1 and other human endogenous retrovirus LTRs in placenta tissues. DNA Res. 2009;16(4):195–211.View ArticlePubMedPubMed CentralGoogle Scholar
- Matouskova M, Blazkova J, Pajer P, Pavlicek A, Hejnar J. CpG methylation suppresses transcriptional activity of human syncytin-1 in non-placental tissues. Exp Cell Res. 2006;312(7):1011–20.View ArticlePubMedGoogle Scholar
- Trejbalova K, Blazkova J, Matouskova M, Kucerova D, Pecnova L, Vernerova Z, et al. Epigenetic regulation of transcription and splicing of syncytins, fusogenic glycoproteins of retroviral origin. Nucleic Acids Res. 2011;39(20):8728–39.View ArticlePubMedPubMed CentralGoogle Scholar
- Gimenez J, Montgiraud C, Pichon JP, Bonnaud B, Arsac M, Ruel K, et al. Custom human endogenous retroviruses dedicated microarray identifies self-induced HERV-W family elements reactivated in testicular cancer upon methylation control. Nucleic Acids Res. 2010;38(7):2229–46.View ArticlePubMedPubMed CentralGoogle Scholar
- Vargas A, Moreau J, Landry S, LeBellego F, Toufaily C, Rassart E, et al. Syncytin-2 plays an important role in the fusion of human trophoblast cells. J Mol Biol. 2009;392(2):301–18.View ArticlePubMedGoogle Scholar
- Liang CY, Wang LJ, Chen CP, Chen LF, Chen YH, Chen HW. GCM1 regulation of the expression of syncytin 2 and its cognate receptor MFSD2A in human placenta. Biol Reprod. 2010;83(3):387–95.View ArticlePubMedGoogle Scholar
- Strissel PL, Ruebner M, Thiel F, Wachter D, Ekici AB, Wolf F, et al. Reactivation of codogenic endogenous retroviral (ERV) envelope genes in human endometrial carcinoma and prestages: emergence of new molecular targets. Oncotarget. 2012;3(10):1204–19.View ArticlePubMedPubMed CentralGoogle Scholar
- Strick R, Ackermann S, Langbein M, Swiatek J, Schubert SW, Hashemolhosseini S, et al. Proliferation and cell-cell fusion of endometrial carcinoma are induced by the human endogenous retroviral syncytin-1 and regulated by TGF-beta. J Mol Med. 2007;85(1):23–38.View ArticlePubMedGoogle Scholar
- Moch H, Cubilla AL, Humphrey PA, Reuter VE, Ulbright TM. The 2016 WHO classification of tumours of the urinary system and male genital organs-part A: renal, penile, and testicular tumours. Eur Urol. 2016;70(1):93–105.View ArticlePubMedGoogle Scholar
- Almstrup K, Nielsen JE, Mlynarska O, Jansen MT, Jorgensen A, Skakkebaek NE, et al. Carcinoma in situ testis displays permissive chromatin modifications similar to immature foetal germ cells. Br J Cancer. 2010;103(8):1269–76.View ArticlePubMedPubMed CentralGoogle Scholar
- Kristensen DG, Nielsen JE, Jorgensen A, Skakkebaek NE, Rajpert-De Meyts E, Almstrup K. Evidence that active demethylation mechanisms maintain the genome of carcinoma in situ cells hypomethylated in the adult testis. Br J Cancer. 2014;110(3):668–78.View ArticlePubMedGoogle Scholar
- Looijenga LH, Gillis AJ, van Gurp RJ, Verkerk AJ, Oosterhuis JW. X inactivation in human testicular tumors. XIST expression and androgen receptor methylation status. Am J Pathol. 1997;151(2):581–90.PubMedPubMed CentralGoogle Scholar
- Netto GJ, Nakai Y, Nakayama M, Jadallah S, Toubaji A, Nonomura N, et al. Global DNA hypomethylation in intratubular germ cell neoplasia and seminoma, but not in nonseminomatous male germ cell tumors. Mod Pathol. 2008;21(11):1337–44.View ArticlePubMedPubMed CentralGoogle Scholar
- Peltomaki P. DNA methylation changes in human testicular cancer. Biochim Biophys Acta. 1991;1096(3):187–96.View ArticlePubMedGoogle Scholar
- Smiraglia DJ, Szymanska J, Kraggerud SM, Lothe RA, Peltomaki P, Plass C. Distinct epigenetic phenotypes in seminomatous and nonseminomatous testicular germ cell tumors. Oncogene. 2002;21(24):3909–16.View ArticlePubMedGoogle Scholar
- Wermann H, Stoop H, Gillis AJ, Honecker F, van Gurp RJ, Ammerpohl O, et al. Global DNA methylation in fetal human germ cells and germ cell tumours: association with differentiation and cisplatin resistance. J Pathol. 2010;221(4):433–42.PubMedGoogle Scholar
- Gkountela S, Zhang KX, Shafiq TA, Liao WW, Hargan-Calvopina J, Chen PY, et al. DNA demethylation dynamics in the human prenatal germline. Cell. 2015;161(6):1425–36.View ArticlePubMedPubMed CentralGoogle Scholar
- Guo F, Yan L, Guo H, Li L, Hu B, Zhao Y, et al. The transcriptome and DNA methylome landscapes of human primordial germ cells. Cell. 2015;161(6):1437–52.View ArticlePubMedGoogle Scholar
- Tang WW, Dietmann S, Irie N, Leitch HG, Floros VI, Bradshaw CR, et al. A unique gene regulatory network resets the human germline epigenome for development. Cell. 2015;161(6):1453–67.View ArticlePubMedPubMed CentralGoogle Scholar
- Benesova M, Trejbalova K, Kucerova D, Vernerova Z, Hron T, Szabo A, et al. Overexpression of TET dioxygenases in seminomas associates with low levels of DNA methylation and hydroxymethylation. Mol Carcinog. 2017. doi:10.1002/mc.22638.PubMedGoogle Scholar
- Ito S, D’Alessio AC, Taranova OV, Hong K, Sowers LC, Zhang Y. Role of tet proteins in 5mC to 5hmC conversion, ES-cell self-renewal and inner cell mass specification. Nature. 2010;466(7310):1129–33.View ArticlePubMedPubMed CentralGoogle Scholar
- He YF, Li BZ, Li Z, Liu P, Wang Y, Tang Q, et al. Tet-mediated formation of 5-carboxylcytosine and its excision by TDG in mammalian DNA. Science. 2011;333(6047):1303–7.View ArticlePubMedPubMed CentralGoogle Scholar
- 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.View ArticlePubMedPubMed CentralGoogle Scholar
- Yamaguchi S, Shen L, Liu Y, Sendler D, Zhang Y. Role of Tet1 in erasure of genomic imprinting. Nature. 2013;504(7480):460–4.View ArticlePubMedPubMed CentralGoogle Scholar
- Kagiwada S, Kurimoto K, Hirota T, Yamaji M, Saitou M. Replication-coupled passive DNA demethylation for the erasure of genome imprints in mice. EMBO J. 2013;32(3):340–53.View ArticlePubMedGoogle Scholar
- Rengstl B, Newrzela S, Heinrich T, Weiser C, Thalheimer FB, Schmid F, et al. Incomplete cytokinesis and re-fusion of small mononucleated Hodgkin cells lead to giant multinucleated Reed-Sternberg cells. Proc Natl Acad Sci USA. 2013;110(51):20729–34.View ArticlePubMedPubMed CentralGoogle Scholar
- Strissel PL, Ellmann S, Loprich E, Thiel F, Fasching PA, Stiegler E, et al. Early aberrant insulin-like growth factor signaling in the progression to endometrial carcinoma is augmented by tamoxifen. Int J Cancer. 2008;123(12):2871–9.View ArticlePubMedGoogle Scholar
- Kassiotis G. Endogenous retroviruses and the development of cancer. J Immunol. 2014;192(4):1343–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Muroi Y, Sakurai T, Hanashi A, Kubota K, Nagaoka K, Imakawa K. CD9 regulates transcription factor GCM1 and ERVWE1 expression through the cAMP/protein kinase A signaling pathway. Reproduction. 2009;138(6):945–51.View ArticlePubMedGoogle Scholar
- von Hochstetter AR, Sigg C, Saremaslani P, Hedinger C. The significance of giant cells in human testicular seminomas. Virchows Arch A. 1985;407(3):309–22.View ArticleGoogle Scholar
- Butcher DN, Gregory WM, Gunter PA, Masters JR, Parkinson MC. The biological and clinical significance of HCG-containing cells in seminoma. Br J Cancer. 1985;51(4):473–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Trejbalova K, Kovarova D, Blazkova J, Machala L, Jilich D, Weber J, et al. Development of 5 LTR DNA methylation of latent HIV-1 provirus in cell line models and in long-term-infected individuals. Clin Epigenetics. 2016;8:19. doi:10.1186/s13148-016-0185-6.View ArticlePubMedPubMed CentralGoogle Scholar
- Bestor TH, Tycko B. Creation of genomic methylation patterns. Nat Genet. 1996;12(4):363–7.View ArticlePubMedGoogle Scholar
- Roulois D, Yau HL, Singhania R, Wang YD, Danesh A, Shen SY, et al. DNA-demethylating agents target colorectal cancer cells by inducing viral mimicry by endogenous transcripts. Cell. 2015;162(5):961–73.View ArticlePubMedPubMed CentralGoogle Scholar
- Chiappinelli KB, Strissel PL, Desrichard A, Li HL, Henke C, Akman B, et al. Inhibiting DNA methylation causes an interferon response in cancer via dsRNA including endogenous retroviruses. Cell. 2015;162(5):974–86.View ArticlePubMedPubMed CentralGoogle Scholar
- Stengel S, Fiebig U, Kurth R, Denner J. Regulation of human endogenous retrovirus-K expression in melanomas by CpG methylation. Gene Chromosome Cancer. 2010;49(5):401–11.View ArticleGoogle Scholar
- Santoni FA, Guerra J, Luban J. HERV-H RNA is abundant in human embryonic stem cells and a precise marker for pluripotency. Retrovirology. 2012;9:111.View ArticlePubMedPubMed CentralGoogle Scholar
- Friedli M, Turelli P, Kapopoulou A, Rauwel B, Castro-Diaz N, Rowe HM, et al. Loss of transcriptional control over endogenous retroelements during reprogramming to pluripotency. Genome Res. 2014;24(8):1251–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Fuchs NV, Loewer S, Daley GQ, Izsvak Z, Lower J, Lower R. Human endogenous retrovirus K (HML-2) RNA and protein expression is a marker for human embryonic and induced pluripotent stem cells. Retrovirology. 2013;10:115.View ArticlePubMedPubMed CentralGoogle Scholar