Inhibition of Bcl6b promotes gastric cancer by amplifying inflammation in mice
Chronic gastritis has been demonstrated to be a key cause of gastric cancer (GC), and control of gastric inflammation is regarded as an effective treatment for the clinical prevention of gastric carcinogenesis. However, there remains an unmet need to identify the dominant regulators of gastric oncogenesis-associated inflammation in vivo.
The mouse model for the study of inflammation-associated GC was induced by Benzo[a]pyrene (BaP) intragastric administration in Bcl6b−/− and wildtype mice on a C57BL/6 background. 5-Aza-2′-deoxycytidine (5-Aza), the demethylation drug, was intraperitoneally injected to restore Bcl6b expression. Human GC tissue array was used to analyse patient survival based on BCL6B and CD3 protein expression.
Bcl6b was gradually downregulated by its own promoter hypermethylation in parallel to an increasing inflammatory response during the progression of BaP-induced gastric carcinogenesis in mice. Moreover, knockout of Bcl6b dramatically worsened the severity of gastric cancer and aggravated the inflammatory response in the BaP-induced mice GC model. Re-activation of Bcl6b by 5-Aza impeded inflammatory amplification and BaP-induced GC development, prolonging survival time in wildtype mice, whereas no notable curative effect occurred in Bcl6b−/− mice with 5-Aza treatment. Finally, significant negative correlations were detected between the mRNA levels of BCL6B and inflammatory cytokines in human GC tissues; patients harbouring BCL6B-negetive and severe-inflammation GC tumours were found to exhibit the shortest survival time.
Epigenetic inactivation of Bcl6b promotes gastric cancer through amplification of the gastric inflammatory response in vivo and offers a new approach for GC treatment and regenerative medicine.
KeywordsGastric cancer Bcl6b Mouse gastric cancer model Inflammation 5-Aza therapy
B cell CLL/lymphoma 6
B cell CLL/lymphoma 6 member B
Bisulphite genomic sequencing
Enzyme-linked immunosorbent assay
- H&E staining
Haematoxylin and eosin staining
Interleukin 1 beta
Tumour necrosis factor alpha
As the fifth-most commonly diagnosed cancer and the third leading cause of cancer-related death, gastric cancer (GC) remains an important cancer worldwide. In particular, Asia is responsible for the largest GC incidence and mortality . Chronic gastritis, the persistent inflammation of the gastric mucosa, is a well-identified cause of GC . Currently, the effective control of gastric mucosal inflammation has highlighted the importance of GC inhibition .
H. pylori infection and carcinogenic intake are key risk factors of inflammation associated gastric oncogenesis [4, 5, 6]. During chronic inflammation, the pronounced release of cytokines and chemokines, including interleukin-1 beta (IL-1β), tumour necrosis factor alpha (TNF-α), interleukin-8 (IL-8), and interleukin-6 (IL-6) [7, 8], drive the infiltration of neutrophils, macrophages and CD3+ T cell into the gastric mucosa . Despite the well-known causal relationship between chronic gastritis and GC , the key regulators of gastric oncogenesis-associated inflammation have not been completely defined.
B-cell CLL/lymphoma 6 member B (BCL6B), also known as BAZF, is a homologue of the human proto-oncogene B-cell CLL/lymphoma 6 (BCL6) . BCL6B has a 94% identical amino acid sequence at the zinc finger motifs and a 65% identical BTB/POZ domain to BCL6, thus, they bind to similar target DNA sequences to act as transcriptional repressors. However, the tissue expression pattern and pathological function of BCL6B differ from that of BCL6 [12, 13, 14]. In recent studies, BCL6B has been identified as a novel tumour suppressor, which is silenced or downregulated by promoter hypermethylation and is associated with poor survival in colorectal carcinoma , hepatocellular carcinoma [16, 17] and GC [18, 19, 20]. In terms of a molecular mechanism, several lines of evidence have revealed that BCL6B interacts with the Notch, STAT, p53 and PI3K/AKT signalling pathways, all of which may be involved in inflammatory response regulation in cancer cells [15, 16, 21, 22]. In vivo, Bcl6b inhibits hepatocellular inflammation to obstruct liver damage and fibrogenic induction in liver-specific Bcl6b transgenic rats . However, whether BCL6B functions as a key tumour suppressor and inflammatory regulator during the progression of gastric oncogenesis in vivo requires further study.
In order to investigate the role of Bcl6b in gastric tumourigenesis, we assessed Bcl6b expression and the degree of inflammation during the progression of Benzo[a]pyrene (BaP)-induced gastric carcinogenesis in mice. Based on this established mouse GC model, we induced gastric tumours within wildtype and Bcl6b−/− mice, and then treated with or without the demethylation drug 5-Aza-2′-deoxycytidine (5-Aza) to induce Bcl6b reactivation. We found that eradication of Bcl6b promoted gastric carcinogenesis by amplifying inflammation in mice and that BCL6B was associated with inflammation and survival in GC patients and mice. These results highlight the role of BCL6B in the modulation of malignant GC phenotypes in vivo and provide a potential target to develop a new therapeutic strategy for gastric cancer treatment.
Animals and treatments
Bcl6b−/− mice on a C57BL/6 background were bred by Cyagen (Guangzhou, China). C57BL/6 mice were obtained from the Xiamen University Laboratory Animal Centre (Xiamen University, Xiamen, China). Mice had access to a standard chow diet and water ad libitum and were caged in a pathogen-free environment. All mice in our experiments were gender and age matched. BaP (Sigma-Aldrich, St Louis, MO, USA) was freshly prepared in sunflower oil (Sigma-Aldrich, St Louis, MO, USA) at a concentration of 5 mg/ml. At postnatal day 40–45 (P40–45), mice were treated with intragastric administration of 0.01 ml/g body weight BaP two times weekly for five weeks. The first intragastric dose was at week 0. Mice were randomized to receive either an intraperitoneal injection of 5-Aza (0.5 mg/kg body weight, twice a week; Sigma-Aldrich, St Louis, MO, USA) or saline during week 10–25 for 15 weeks. Mice were euthanized, and stomach tissues were collected for examination at specific times. All mice were used in accordance with the guidelines of the Institutional Animal Care and Use Committee of the Xiamen University Laboratory Animal Centre.
RNA extraction and quantitative PCR analysis
Total RNA from mouse tissues was extracted using TRIzol (cat. no.: 15596026, Invitrogen, Thermo Fisher Scientific, MA, USA) and was quantified via NanoDrop (Thermo Fisher Scientific, MA, USA). Equal amounts of RNA were reverse-transcribed with the HiFi-MMLV cDNA Kit from CWBIO (cat. no.: CW0744, CoWin Biotech, Beijing, China) to produce cDNA according to the manufacturer’s protocol. Quantitative PCR (qPCR) was performed using a CFX96 Touch Real-Time PCR Detection System (Bio-Rad, CA, USA) with the UltraSYBR Mixture from CWBIO (cat. no.: CW0956, Beijing CoWin Biotech, Beijing, China) according to the manufacturer’s protocol. The expression levels were quantified using the 2-ΔΔCt (where Ct is the threshold cycle) method. Eukaryotic 18S rRNA endogenous control was used as an internal standard. The following primer sequences for quantitative PCR were used:
18S: 5′- GTCTGTGATGCCCTTAGATG − 3′ (forward),
5′- AGCTTATGACCCGCACTTAC − 3′ (reverse);
hTNF-α: 5′- AGCCTGTAGCCCATGTTGTAGC − 3′ (forward),
5′- CCTTGGCCCTTGAAGAGGAC − 3′ (reverse);
hIL-8: 5′- TTGGCAGCCTTCCTGATTTCT − 3′ (forward),
5′- GGTCCACTCTCAATCACTCTCA − 3′ (reverse);
hIL-1β: 5′- GAGCTCGCCAGTGAAATGATG − 3′ (forward),
5′- TCGGAGATTCGTAGCTGGATG − 3′ (reverse);
hBCL6B: 5′- CTACGTCCGCGAGTTCACTC − 3′ (forward),
5′- CCCGGAAAATTGAATAGAAG − 3′ (reverse);
mTNF-α: 5′- TGACAAGCCTGTAGCCCACG − 3′ (forward),
5′- GGCAGCCTTGTCCCTTGAA − 3′ (reverse);
mIL-8: 5′- CAAGGCTGGTCCATGCTCC − 3′ (forward),
5′- TGCTATCACTTCCTTTCTGTTGC − 3′ (reverse);
mIL-1β: 5′- GAAATGCCACCTTTTGACAGTG − 3′ (forward),
5′- TGTTGATGTGCTGCTGCGAG − 3′ (reverse);
mBcl6b: 5′- TCCTCCGACGTGCTTAGCAAT − 3′ (forward),
5′- GGCCCCGGAAAATTGAATAGA − 3′ (reverse).
Enzyme-linked immunosorbent assay
Mouse serum levels of TNF-α and IL-1β were assessed and quantified using enzyme-linked immunosorbent assay (ELISA) kits (TNF-α: cat. no.: MTA00B, R&D systems, Minneapolis, USA; IL-1β: cat. no.: 88–7013-22, Thermo Fisher Scientific, MA, USA) according to the manufacturers’ protocol. Enzyme concentration was quantified by measuring the optical density at a wavelength of 450 nm (OD450) using a spectrophotometer (Thermo Fisher Scientific, MA, USA).
Mouse tissues were lysed with SDS lysis buffer. Anti-mBcl6b (cat. no.: ab87228, Abcam, Cambridge, MA, USA) and anti-β-actin (cat. no.: A1978, Sigma-Aldrich, Darmstadt, Germany) were used for immunoblotting. β-Actin was used as an internal control.
The stomach of the mice was cut off from the greater curvature and placed on a stereomicroscope. Turn on the bottom light source, count and measure tumour nodules larger than 0.5 mm in diameter according to the lens scale. The average number of tumour nodules per tumour bearing mouse was calculated as tumour number. The average maximal diameter of tumour nodule per tumour bearing mouse was calculated as max size. Tumour incidence was the number of mice carrying at least 1 tumour nodule expressed as percentage incidence. Meanwhile, H&E staining and Pan-Cytokeratin immunostaining in the gastric tissue of mice were further used for qualitative analysis of gastric tumours.
For histologic analysis, mouse tissues were fixed with 10% neutral buffered formalin overnight, embedded in paraffin, and cut into 4.5 μm sections using a microtome (Leica RM2235; Leica Microsystems). After dewaxing and rehydration, the sections were stained in haematoxylin solution (cat. no.: D006, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) for 3 min, washed in running tap water and then differentiated in 1% acid alcohol for 30 s, followed by counterstaining with eosin (cat. no.: D006, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) for 30 s. Gastric tumours were identified and quantified on sections.
All tissues were fixed in 10% neutral buffered formalin, paraffin embedded, and sectioned. After dewaxing and rehydration, antigen retrieval was performed by boiling in citrate buffer (pH 6.0) for 20 min. Immunostaining was performed using the Maxim UltraSensitive SAP Kit (cat. no.: Kit9710, Maxim Biological Technology, Fuzhou, China). Sections were pre-treated with peroxidase blocking buffer for 10 min at room temperature. After treatment with blocking buffer (5% normal goat serum in PBS) for 10 min at room temperature, sections were incubated with anti- hBCL6B (1:500 dilution; cat. no.: ab180084, Abcam, Cambridge, MA, USA), anti- mBcl6b (1:500 dilution; cat. no.: ab87228, Abcam, Cambridge, MA, USA), anti-CD3 (1:300 dilution; cat. no.: sc-20047, Santa Cruz Biotechnology, Texas, USA), anti-Ki-67 (cat. no.: 790–4276, Ventana Medical systems, AZ, USA), or anti- Pan-Cytokeratin (cat. no.: IHC-MO67, Guangzhou ambip pharmaceutical technology, Guangzhou, China) in blocking buffer. Sections were then incubated with a secondary antibody followed by 3,3′-diaminobenzidine (DAB) staining with a DAB kit (cat. no.: DAB2031, Maxim Biological Technology, Fuzhou, China). Images were obtained using a computerized imaging system (Leica Microsystems, Imaging Solutions Ltd., Cambridge, UK). Identical settings were used for each photograph using Leica QWin Plus v3 software. To strengthen our main conclusion and further investigate the clinical significance of BCL6B and CD3 in GC, we performed tissue microarray analysis and classified all tumours that presented with more than 5% of cells over the threshold as BCL6B or CD3 positive.
Bisulphite sequencing PCR
DNA purified from mouse tissues was subjected to bisulphite sequencing PCR using a DNA Methylation Kit (cat. no.: CW2140, Beijing CWbio Biotech, Beijing, China) following the manufacturer’s instructions. PCR primers were designed targeting the CpG sites of the mouse Bcl6b gene (starting 80-bp upstream of the mouse Bcl6b transcription start site (TSS) and ending 100-bp downstream of the TSS):
5′- GGGTTTATTATTTGGAGAGT − 3′ (forward), 5′- TTAATTCTTACCCCTATCCC − 3′ (reverse).
Human gastric cancer samples
Tissue microarrays from 186 GC patients (HStmA180Su09/HStmA180Su13) and cDNA microarrays from 45 GC patients (cDNA-HStmA030CS01/ cDNA-HStmA060CS01) were purchased from Shanghai Outdo Biotech Co., Ltd. (Shanghai, China). The studies were conducted in accordance with the International Ethical Guidelines for Biomedical Research Involving Human Subjects (CIOMS), and the research protocols were approved by the Clinical Research Ethics Committee of Zhongshan Hospital of Xiamen University.
Statistical analyses were performed with GraphPad Prism 6.0. Data are presented as the mean ± SD. The significance of the mean values between two groups was analysed by Student’s t test. Pearson correlation analysis was performed to determine the correlation between two variables. Survival analysis was performed using a log-rank test. P < 0.05 was considered statistically significant.
The expression patterns of Bcl6b and inflammatory factors correlated with the progression of BaP-induced gastric carcinogenesis
Eradication of Bcl6b aggravates gastric tumourigenesis and amplifies inflammation in BaP-treated mice
5-Aza re-activates Bcl6b and attenuates the inflammatory response in BaP-treated mice
5-Aza ameliorates BaP-induced gastric oncogenesis depended on Bcl6b in vivo
BCL6B downregulation combined with a severe inflammatory response correlates with poor survival in GC patients
To strengthen our main conclusion and further clarify the clinical significance of BCL6B and inflammation in GC, cDNA microarrays from 45 GC patients were first used to investigate the mRNA expression pattern of BCL6B and inflammatory cytokines. Notably, correlation analyses revealed strong inverse correlations between the overall expression levels of BCL6B and the inflammatory cytokines TNFα (Fig. 6a), IL-1β (Fig. 6b), and IL-8 (Fig. 6c) in the 45 GC patient cohort. Next, we performed immunohistochemical staining of BCL6B and CD3 on GC tissue microarrays containing 186 GC specimens that had long-term clinical follow-up records. As the typical images show in Fig. 6d, all tumours presenting with greater than 5% positive staining of cells over the threshold were defined as BCL6B or CD3 positive. Subsequent Kaplan-Meier survival analysis showed that the GC tissue expression levels of BCL6B protein were inversely correlated with the 5-year survival rate of GC patients. BCL6B-negative GC patients had a median survival time of 56.5 months compared to BCL6B-positive GC patients who had a median survival time of 20.0 months (Fig. 6d), which is consistent with previous reports [18, 19] and our own survival results in mice as shown in Fig. 6e. Remarkably, IHC analysis showed that among BCL6B-negative GC patients, those with CD3-positive expression exhibited the poorest survival (median survival time = 15.0 months), whereas BCL6B-negative GC patients with CD3-negative expression showed marginally different survival with a median survival time of 48.5 months (Fig. 6d). Taken together, the above results provide evidence that BCL6B is significantly negatively related to the degree of inflammation in GC and that negative BCL6B expression and a severe inflammatory response contribute jointly to the poor survival of GC patients.
In this study, we used Bcl6b-deficient mice and 5-Aza-treated mice to evaluate the role of Bcl6b in BaP induced gastric inflammation and GC development. This study provided compelling evidence that Bcl6b functions as a tumour suppressor to inhibit GC and suggested a novel inflammatory strategy to control GC. Previous studies have showed that inflammation is a hallmark of cancer and a pro-tumourigenic factor [25, 26]. Furthermore, gastritis stage is used to quantify GC risk  and GC patients with a severe inflammatory response exhibit poor survival . Thus, effective inflammation control has been recognized as both a preventative and treatment strategy for GC [29, 30, 31]. In the present study, we found that Bcl6b was gradually downregulated by its own promoter hypermethylation in parallel to an increased inflammatory response during the progression of BaP-induced gastric carcinogenesis in mice. Moreover, knockout of Bcl6b amplified the inflammatory response and aggravated gastric tumourigenesis in mice upon BaP treatment. Re-activation of Bcl6b by the demethylation drug 5-Aza could impede inflammatory amplification and Bap-induced GC development, further prolonging the survival time in wildtype mice, whereas no notable curative effect was observed in Bcl6b−/− mice with 5-Aza treatment. Analyses of clinical GC specimens showed that negative BCL6B expression together with a severe inflammatory response resulted in the poorest survival in GC patients. We therefore conclude that Bcl6b plays a crucial role in regulating inflammation-associated GC initiation and development in vivo and that 5-Aza can be used as an anti-tumour compound through re-activation of Bcl6b.
Given that the inflammatory response has been reported to modulate GC development [32, 33, 34] and that BaP is a carcinogen that is most likely ingested during daily human life [35, 36], BaP can induce cancerous lesions in the human stomach and result in gastric carcinogenesis by promoting a pro-inflammatory phenotype [5, 37, 38, 39]. In addition, feeding of mice with BaP has been an extensively applied method for the establishment of mouse GC models [40, 41]. Thus, BaP-induced mouse GC is therefore an ideal model for the study of inflammation-associated GC. Consistent with our observation in our mouse GC model induced by BaP gavage, inflammation gradually increased during gastric carcinogenesis with a corresponding decreasing trend for Bcl6b. In human GC samples, Bcl6b expression exhibited a negative correlation with inflammatory cytokines. Based on the previous report that BCL6B is an independent predictor of poor outcome in patients with GC . Our study further indicates that combination of BCL6B expression and inflammatory response may improve the prognosis accuracy in GC patient survival. Moreover, our results showed that Bcl6b inhibition remarkably aggravated gastric inflammation in BaP-induced mice. These findings indicate that BCL6B functions as a key inflammatory regulator in the progression of gastric oncogenesis in vivo. However, the precise regulatory mechanisms underlying the relationship between BCL6B and inflammatory cytokines has not yet been elucidated. Several previous studies revealed that BCL6B interacts with the Notch, STAT, p53 and PI3K/AKT signalling pathways [15, 16, 21, 22], which may be involved in the regulation of the inflammatory response in cancer cells. Future studies should verify which core signalling pathway BCL6B depends on for the regulation of the inflammatory response in inflammation-associated GC development.
From a clinical point of view, both BCL6B and gastric inflammation should be precisely regulated within a suitable range to control normal development and homeostasis. However, this balance may be disrupted during carcinogenesis. It has been reported that BCL6B is silenced via its own promoter hypermethylation in GC , which is consistent with our observation that the methylated levels of the Bcl6b promoter were gradually up-regulated during the process of gastric tumourigenesis by BaP induction. In our study, which aimed to investigate an existing clinical drug to target Bcl6b in inflammation-associated GC in vivo. We used the DNA demethylation drug 2′-deoxy-5-azacytidine (5-Aza), which has been applied for the clinical treatment of multiple haematologic malignancies and solid tumours [42, 43]. Our results revealed that 5-Aza treatment effectively restored Bcl6b expression and dramatically blocked gastric inflammation and GC development. In contrast, 5-Aza treatment had a weak therapeutic effect on BaP-induced GC in Bcl6b−/− mice. This study, along with another complementary reports [44, 45] confirmed that a specific panel of candidate genes (BCL6B, GDF1, FBP1, BNIP3, CDX1, CHFR, MGMT, MLH1, etc.) are aberrantly activated or silenced by methylation in stomach tumours. Our results in Bcl6b−/− mice suggest that the therapeutic effect of 5-Aza on BaP-induced GC depends on the activation of Bcl6b, which highlights the key role of BCL6B methylation in the occurrence and development of GC. In clinical trials, the demethylation drug 5-Aza, which broadly targets DNA, showed high toxicity [46, 47, 48]. Thus, a novel anti-GC drug that can selectively target BCL6B with low toxicity is urgently needed.
In summary, our study showed that BCL6B acts as a dominant tumour suppressor in GC and plays a crucial role in inflammatory response control in vivo. Manipulating the expression of BCL6B may provide a new entry point for GC treatment and regenerative medicine.
Study concept and design: WYC, LW & QCL; acquisition and analysis of data: WYC, LYL, LY, LW & JC; drafting of the manuscript: WYC, LYL & QCL; statistical analysis: QZ, YYX & MLC; obtained funding: WYC, QCL & GDY; technical or material support: GDY, MLC & LW. All authors read and approved the final manuscript.
This work was supported by grants from the National Natural Science Foundation of China (grant number 81602148 to Wang-Yu Cai; grant number 81872422, 81672871 to Qi-Cong Luo; grant number 81602563, 81871977 to Guo-Dong Ye); the Natural Science Foundation of Fujian Province (grant number 2015 J01523 to Wang-Yu Cai); and the Science and Technology Projects of Xiamen City (grant number 3502Z20154025 to Wang-Yu Cai).
Ethics approval and consent to participate
The studies were conducted in accordance with the international ethical guidelines for biomedical research involving human subjects (CIOMS), and the research protocols were approved by the clinical research ethics Committee of Zhongshan Hospital of Xiamen University. Written informed consents were obtained from the participants before sampling.
Consent for publication
The authors declare that they have no competing interests.
- 2.Dinis-Ribeiro M, Areia M, de Vries AC, Marcos-Pinto R, Monteiro-Soares M, O'Connor A, Pereira C, Pimentel-Nunes P, Correia R, Ensari A, et al. Management of precancerous conditions and lesions in the stomach (MAPS): guideline from the European Society of Gastrointestinal Endoscopy (ESGE), European helicobacter study group (EHSG), European Society of Pathology (ESP), and the Sociedade Portuguesa de Endoscopia Digestiva (SPED). Endoscopy. 2012;44:74–94.CrossRefGoogle Scholar
- 7.Tafreshi M, Guan J, Gorrell RJ, Chew N, Xin Y, Deswaerte V, Rohde M, Daly RJ, Peek RM Jr, Jenkins BJ, et al. Helicobacter pylori type IV secretion system and its Adhesin subunit, CagL, mediate potent inflammatory responses in primary human endothelial cells. Front Cell Infect Microbiol. 2018;8:22.CrossRefGoogle Scholar
- 27.Rugge M, Genta RM, Fassan M, Valentini E, Coati I, Guzzinati S, Savarino E, Zorzi M, Farinati F, Malfertheiner P. OLGA gastritis staging for the prediction of gastric Cancer risk: a long-term follow-up study of 7436 patients. Am J Gastroenterol. 2018.Google Scholar
- 38.Fueldner C, Kohlschmidt J, Riemschneider S, Schulze F, Zoldan K, Esser C, Hauschildt S, Lehmann J. Benzo(a)pyrene attenuates the pattern-recognition-receptor induced proinflammatory phenotype of murine macrophages by inducing IL-10 expression in an aryl hydrocarbon receptor-dependent manner. Toxicology. 2018;409:80–90.CrossRefGoogle Scholar
- 47.Lubbert M, Suciu S, Baila L, Ruter BH, Platzbecker U, Giagounidis A, Selleslag D, Labar B, Germing U, Salih HR, et al. Low-dose decitabine versus best supportive care in elderly patients with intermediate- or high-risk myelodysplastic syndrome (MDS) ineligible for intensive chemotherapy: final results of the randomized phase III study of the European Organisation for Research and Treatment of Cancer leukemia group and the German MDS study group. J Clin Oncol. 2011;29:1987–96.CrossRefGoogle Scholar
- 48.Kantarjian H, Oki Y, Garcia-Manero G, Huang X, O'Brien S, Cortes J, Faderl S, Bueso-Ramos C, Ravandi F, Estrov Z, et al. Results of a randomized study of 3 schedules of low-dose decitabine in higher-risk myelodysplastic syndrome and chronic myelomonocytic leukemia. Blood. 2007;109:52–7.CrossRefGoogle Scholar
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.