{"paper_id":"ed2e9adf-455a-47fc-83ab-05c701f30bd8","body_text":"Endometriosis is considered a common public health problem, affecting 10-12 % of women at different ages, and increasing in prevalence to 20–50% in infertile women. \n 1 \n  It is characterized by endometrial glands and stroma being present outside the uterine cavity, and can result in infertility and pelvic pain, symptoms which can significantly impact patient quality of life. \n 2\nEndometriosis has been shown to significantly impact physical and psychological wellbeing, often resulting in adverse effects such as depression and anxiety. 3 , 4  As asserted by Della Corte et al. (2020), \n 4 \n  endometriosis has a detrimental effect on patients’ sex lives and social relationships.\nToday, the commonly used therapy to reduce pain is gonadotropin-releasing hormone antagonists to block the menstrual cycle, reduce estrogen and progesterone levels, minimize endometrial thickness, and progesterone production by the corpus luteum. This means putting the patient under contraception and blocking any chance of pregnancy. By investigating cfDNA quantification, we aim to explore a therapeutic approach using DNase to reduce inflammatory pathways associated with elevated cfDNA levels.\nAs an estrogen-dependent disorder, endometriosis causes a systemic immune inflammatory syndrome that is the origin of energy imbalance, mitochondrial dysfunction, metabolomics disorders and peripheral and local cell apoptosis in combination with epigenetic profile changes. 5 , 6\nIt has been reported that women with severe endometriosis suffer from elevated oxidative stress and free circulating DNA fragments in the bloodstream. \n 7\nOur previous study \n 1 \n  reported high free circulating DNA (CfDNA) levels as a consequence of apoptosis and differential methylation changes in a group of genes in patients with endometriosis compared to the control group.\nToday, endometriosis is primarily diagnosed through laparoscopy, with subsequent confirmation derived from analyzing the obtained lesions. For cases classified as mild to moderate (stages I and II), the therapeutic approach is predominantly medical. For cases classified as severe (stages III and IV), the therapeutic approach involves surgical intervention or, alternatively, initiating hormone therapy. \n 8\nRecently, many studies have focused on identifying reliable biomarkers for early endometriosis diagnosis, such as immunologic as well as genetic and biochemical markers, including specific cytokines, microRNAs, lncRNAs and circulating and mitochondrial nucleic acids, along with certain hormones, glycoproteins and signaling molecules 9 – 11\nIn 2025, we proposed novel noninvasive biomarkers for endometriosis by quantifying free circulating DNA and reanalyzing the differential methylation profile of specific genes involved in endometriosis pathophysiology. In patients with endometriosis, compared to the control group, we observed nearly 4 times as much CfDNA in the serum and a differential methylation profile of nine genes involved mainly in post translational protein regulation involved in numerous cellular cycles proliferation, invasion, cellular mobility, gene methylation and activation, chromatin modeling and DNA repair. \n 1\nAs a response to the systemic chronic inflammatory condition that is common in endometriosis with high levels of CfDNA, the natural physiological endonuclease DNase I cleaves excess DNA fragments, facilitating their elimination from the bloodstream, liver and urine to minimize the negative immunopathology impact of abnormally high free circulating DNA levels. The shelf-life of human DNase in peripheral blood is estimated to be around 30 minutes, due to its rapid degradation by proteases, renal and/or hepatic clearance. The in vitro half-life is estimated to be between three and six hours. At the physiological level, DNase I plays a pivotal role in clearing free DNA generated by excessive cell death in endometriosis.\nIn clinical medicine, the human recombinant form of deoxyribonuclease I (Pulmozyme®) is the most common mucolytic agent used for long-term daily cystic fibrosis treatment by improving easier breathing, reducing cough and sputum burden. Using DNase I for patients undergoing IVF/ICSI with repeated implantation failure and high circulating CfDNA showed that a daily injection of 2500 IU DNase I for a month reduced CfDNA by 40% with embryo implantation improvement. \n 12\nDNase I is an enzyme that catalyzes the breakdown of deoxyribonucleic acid (DNA) into nucleotides or polynucleotides and not target bacteria or inflammation directly.\nTo date, no studies have demonstrated the effectiveness of DNase I in alleviating the pain and inflammation associated with moderate to severe endometriosis. The direct connection between DNase and endometriosis remains an active research area.\nManaging excess CfDNA and inflammation could have implications for understanding the chronic inflammatory nature of endometriosis and designing potential therapeutic approaches.\nThis study has previously been made available as a preprint titled  “Effect of Deoxyribonuclease I Therapy on Free Circulating DNA Level and Gene Methylation Profile Changes in Patients with Endometriosis: Preliminary Data.” \n 12\nIn the present preliminary prospective study, our objective was to administer a synthetic endonuclease to women diagnosed with endometriosis, as confirmed by bio-clinical ultrasound and laparoscopy, at any stage of the disease, in order to evaluate this treatment’s effects on CfDNA reduction and differential methylation of a group of genes involved in endometriosis. \n 1\n\nSixteen patients with endometriosis were enrolled for the study from a sample of women undergoing consultation for infertility from July 2024 to march 2025. Patients with primary infertility and algic symptoms along with laparoscopically proven endometriosis combined to clinical examination and radiology. Those patients were clearly classified as endometriosic patient and agreed to receive exogenous DNase I therapy for one month. The average patient age was 34 years and they all suffered from chronic abdominal–pelvic pain and excessive bleeding during their menstrual cycles.\nThe patients had at least stage II pelvic endometriosis or higher (stage III and IV); schematically, stage II is localized to the peritoneum and/or utero-sacral ligaments, utero-sacral ligaments or fallopian tubes; stage III involves the ovaries; and stage IV involves the digestive tract, particularly the recto-vaginal septum. The patients had no other pathology, had been undergoing no current medical treatment for more than three months and were not required to take any painkillers during their DNase treatment.\nPatients were tested using an electrical neural stimulation (ENS) numerical scale to evaluate pain intensity (0 to 10), where 0 indicates no pain and 10 maximal and unsupportable pain. An HRQoL (Health-Related Quality of Life) questionnaire was used to assess quality of life (1 to 10), where 1 represents the best quality of life and 10 the worst.\nThis study has been registered on Clinicaltrials. gov under  NCT05815134 .  The study was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2024.\nEach endometriosis patient was given a 2,500 IU ampoule of Pulmozyme as communal drugs sold worldwide by Roch since many years after approval by the FDA to hydrolyse the extra cellular DNA (synthetic DNase, Roche Laboratories, Switzerland) subcutaneously every 2 days for one month. Women taking part in the study were given information about the treatment and follow-up.\nPeripheral blood was sampled from each patient in an EDTA tube at two time points only: before the first DNase I administration and after completion of the one-month treatment period. No intermediate blood samples were collected during treatment. The samples were labeled and immediately centrifuged; the obtained serum was frozen at -80°C and stored for free DNA quantification and gene methylation profiling assessment.\nAll patients were informed for the study and signed the consent form.\nNucleic acids were extracted from 1 mL of frozen–thawed serum using a Qiagen QIAamp Circulating Nucleic Acid kit from Qiagen, Saint Quentin Fallavier, France, closely following the extraction protocol handbook (10/2019). Briefly, 40 µL of proteinase K (concentration 600 mAU/mL) was added to the 1 mL of thawed serum. Then, 1 mL of lysis buffer containing 1.0 µg of carrier RNA (Qiagen ACL buffer) was added to the serum and proteinase K mix and vortexed for 30 s before incubation at 56 ◦C for 10 min. Then, 840 µL of ACB buffer was added to the lysate, mixed thoroughly by pulse vortexing for 15–30 s and incubated for 5 min on ice. The lysate–ACB buffer solution was carefully applied to the QiaAmp mini-column from Qiagen, Sant Quentin Fallavier, France, to be drawn completely. After that, the column was washed and drawn successively with ACW1 and ACW2 buffers and ethanol (96–100%). Then, the QIAamp mini-column was placed in a clean 2 mL collection tube and centrifuged at full speed (14,000 rpm) for 3 min, transferred to a new collection tube and incubated for 10 min at 56◦ until dried. Finally, cell-free DNA (Cf-DNA) was eluted twice with 25 µL of TE buffer (Tris/EDTA 1 mM/0.1 mM) through 1 min of full-speed centrifugation (14,000 rpm).\nRpp30 ( NM_006413 ) is commonly chosen to measure cfDNA levels in serum due to its role as a stable, single copy housekeeping gene. As a robust internal control, it ensures accurate and reproducible Cf-DNA measurements. Rpp30 DNA ( NM_006413 ) was quantified using qPCR genes from human serum samples, as it is a highly con-served endoribonuclease present in all living cells in the body. Triplicates of 5 µL Cf-DNA were added to 20 µL of PCR Light Cycler® 480 SYBR Green I Master (Cat. no 04707516001) along with 2.5 mM of MgCl2 and 0.5 mM of each forward and reverse RNase P primer (primer sequences: RNP30 forward—AGATTTGGACCTGCGAGCG; RNP30 reverse—GAAGCCGGGGCAACTCAC). A PCR product of 86 base pair regions spanning exon 1 and intron 2 of the Homo sapiens ribonuclease P/MRP subunit p30 gene ( NM_006413  and ENST00000371703.7) was obtained. Amplification was performed on a Light Cycler 480 II (Roche) as follows: 35 cycles of 95 ◦C for 10 s, 59 ◦C for 20 s and 72 ◦C for 15 s, followed by an elongation step of 5 min at 72 ◦C. Positive DNA controls at various concentrations and non-template controls were added to each run. Cycle threshold values were reported against a standard concentration curve, and Cf-DNA concentration was reported as the mean triplicate value.\nBisulfite DNA treatment was performed using an EZ DNA Methylation Kit (Zymo Research) following manufacturer recommendations. According to the initial concentrations, 35 µL of DNA sample was used for the reaction. At the end of the treatment, DNA was eluted with 25 µL of elution buffer and then diluted using 10 µL of H2O. Gene specific PCR reactions were performed using Taq’Ozyme HS Mix (Ozyme, Saint-Cyr-l’École, France), 1 µL of bisulfite-treated DNA and the final primers, each at a concentration of 0.4 µM, in a 20 µL final reaction volume using a C-100 thermocycler (Bio-Rad, Hercules, CA, USA). The cycling conditions were 95 ◦C/1 min; (95 ◦C/15 s–58 ◦C/15 s–72 ◦C/30 s) × 34 and 72 ◦C/5 s. All amplicon sizes were checked and validated via electrophoresis before sequencing, and sequencing was performed in the paired-end mode (2 × 150 bp) on the NextSeq Illumina Platform following the manufacturer’s protocol (BioProject record (NCBI): 1063938).\nRPP30 copy numbers were summarized descriptively using R software (v4.3.1) and compared qualitatively between paired pre-treatment and post-treatment samples. The previously reported one-way ANOVA was removed because it was not appropriate for this paired before/after design. As no additional paired statistical reanalysis was performed in this revision, no formal inferential p value is reported for RPP30 copy-number changes.”\nA post-sequencing quality check was performed with FastQC software (version 0.11.8), and sequence cleaning and paired-end read merging were performed using fastp software (version 0.21.0). A post-cleaning/merging quality check was then performed with FastQC. The targeted gene sequences were extracted and sorted using custom BLAST software. The frequency of C nucleotides relative to the total number of C and T nucleotides at each targeted position after bisulfite treatment was used to describe the DNA methylation rate. For each targeted sequence, methylated and unmethylated read counts were summarized before and after DNase I treatment to provide an exploratory overview of methylation profiles. Because the original aggregated Chi-square approach does not account for the correlation between paired measurements obtained from the same patient, no formal inferential test based on pooled C and T read counts was retained in the revised manuscript. Methylation findings are therefore presented as descriptive, hypothesis-generating observations only. All data processing was performed using R (v4.3.1) and EdgeR (v3.42).\nWe targeted the changes before and after treatment in the differential methylation target CpG sites of 9 genes selected by ENDOLIFE: CALD1, RRP1, FN1, DIP2C, RMI2, TDRD5, USP1, HDAC1 and DNMT1.\nThe reporting of this study conforms to the CONSORT statements. \n 13\n\nA total of 15 of the 16 participants reported a reduction in pain, as well as an improvement in their ability to assume professional activity and less difficulty regarding their sexuality.\nHRQoL (Health-Related Quality of Life) is a multidimensional concept, commonly used to examine the impact of health status on quality of life. It is measured by four core questions on general health status and number of unhealthy days in the Behavioral Risk Factor Surveillance System (BFRSS).\nTo quantify the absolute Hs_Rpp30 gene amount in a 1 mL serum sample, we calculated a standard curve using a plasmid containing a portion of the Rpp30 gene, a single copy gene present in the human genome. The Hs_Rpp30-positive control (IDT, ref. 10006626) was calibrated at a concentration of 200,000 copies/µL in Tris/EDTA at a pH of 8.0. The absolute quantification and mean absolute quantification from the 16 endometriosis patients before and after DNase I therapy are shown in  Figure 1 . The median RPP30 copy number decreased descriptively from 2,767 copies before treatment to 2,234 copies after treatment. No formal inferential p value is reported because no paired statistical reanalysis was performed.” Figure 1. Median absolute quantification of RPP30 gene for follow-up cohort, expressed as copy number. Green color indicates patients “Before the treatment” and blue “After the treatment”.\nMedian absolute quantification of RPP30 gene for follow-up cohort, expressed as copy number. Green color indicates patients “Before the treatment” and blue “After the treatment”.\nAfter gene depletion and selection, we described methylation patterns for nine targeted genes (CALD1, RRP1, FN1, DIP2C, RMI2, TDRD5, USP1, HDAC1 and DNMT1) before and after treatment (see the details in  Table 2 ).\nThe aggregated Chi-square p values initially generated from C and T nucleotide distributions are reported only as exploratory descriptive indicators of sequencing read composition. They were not used to draw formal paired statistical conclusions, because pooled read-count comparisons do not account for intra-subject dependence in this before/after design.\nThe green and blue colors indicate the descriptive direction of the methylation pattern after treatment compared with before treatment and should not be interpreted as formal paired statistical significance.\nIn this exploratory read-count analysis, CALD1, DNMT1, HDAC1 and RRP1 showed a hypermethylation pattern, while DIP2C and USP1 showed a hypomethylation pattern after treatment compared with before treatment.\nHistograms display the distribution of methylated (C) and unmethylated (T) read counts for all nine targeted CpG loci (CALD1, RRP1, FN1, DIP2C, RMI2, TDRD5, USP1, HDAC1 and DNMT1) across the analyzed samples. Each panel illustrates the relative C and T nucleotide proportions obtained after bisulfite sequencing, providing a visual overview of methylation patterns for each gene target. The histograms summarize the variability in read composition among participants and allow qualitative methylation profile comparisons between the examined loci (see  Figure 2  for the different histo-grams). Figure 2. Histograms for each sequence in the methylation experiment. Nb_C: number of C reads; Nb_T: number of T reads. Blue and orange bars correspond to before and after treatment, respectively.\nHistograms for each sequence in the methylation experiment. Nb_C: number of C reads; Nb_T: number of T reads. Blue and orange bars correspond to before and after treatment, respectively.\nThe two genes DIP2C and USP1 showed a hypomethylation pattern in this exploratory analysis. DIP2C (Disco Interacting Protein 2 Homolog C) is a gene encoding a protein that interacts with the disco transcription factor and is expressed in the nervous system. USP1 is a negative regulator of DNA damage repair that is also involved in PCNA-mediated translational synthesis (TLS) by deubiquitinating monoubiquitinated PCNA. This pattern could be compatible with increased transcriptional permissiveness; however, no gene expression analysis was performed, and this interpretation remains hypothesis-generating.\nOn the contrary, CALD1, DNMT1, HDAC1 and RRP1 are hypermethylated, blocking any possibility of transcription.\n\nThis preliminary study reported that 15 of the 16 participants reported reduced pain, improved ability to assume professional activity and less difficult sexuality ( Table 1 ). Because this was an open-label study without placebo control, these subjective outcomes cannot be causally attributed to DNase I therapy and may include expectation or placebo effects. Table 1. This table report the changes of Pain and Quality of life evaluation in 16 endometriosis patients before and after treatment. Patient number Age Endometriosis stage ENS (1 to 10) HRQoL (1 to 10) Before tt After tt Before tt After tt 1 31 3 5 5 7.5 4 2 28 3–4 3 2 7.5 6 3 39 4 6 4 9 5 4 29 3 1 1 8 4 5 36 4 7 ​ 6 3 6 36 3 5 3 9 5 7 39 3 4 4 8 3 8 30 2 5 1 5 2 9 37 2 7 3 4 3 10 41 4 9.5 1 2 1 11 35 4 9 3 7 3 12 41 2 8 ​ 1 1 13 29 3 4 1 10 6 14 28 2 9.5 8 3 3 15 34 4 9.7 3 3 1 16 31 2 6 2 9.5 6 ENS = numerical evaluation scale. HRQoL = Health-Related Quality of Life.\nThis table report the changes of Pain and Quality of life evaluation in 16 endometriosis patients before and after treatment.\nENS = numerical evaluation scale. HRQoL = Health-Related Quality of Life.\nDescriptively, for the 16 patients, the median copy number of RPP30 changed from 2767 copies before treatment to 2234 copies after treatment ( Figure 1 ). No formal inferential p value is reported in the revised manuscript because the previous one-way ANOVA was not appropriate for this paired design and no paired reanalysis was performed.\nConcerning methylation, the read-count profiles suggested changes in CALD1, DIP2C, DNMT1, HDAC1, RRP1 and USP1 ( Table 2 ). These findings are presented as exploratory and descriptive, because the aggregated Chi-square approach does not account for intra-subject dependence in the before/after design. CALD1, DNMT1, HDAC1 and RRP1 showed a hypermethylation pattern, while DIP2C and USP1 showed a hypomethylation pattern. Table 2. Exploratory aggregated read-count p values from the original Chi-square test of homogeneity, presented descriptively only and not used for paired inference. Target Khi2.pval CALD1 4.76 × 10 −7 DIP2C 0.02707 DNMT1 0.047 FN1 0.9722 HDAC1 0.01928 RMI2 1 RRP1 0.005371 TDRD5 0.8327 USP1 0.0183\nExploratory aggregated read-count p values from the original Chi-square test of homogeneity, presented descriptively only and not used for paired inference.\nTwo genes,  DIP2C and USP1 , are hypo methylated.  DIP2C  (Disco Interacting Protein 2 Homolog C) is a gene that encodes a protein. The protein interacts with the disco transcription factor and is expressed in the nervous system.  USP1  is a negative regulator of DNA damage repair. It is also involved in PCNA-mediated translational synthesis (TLS) by deubiquitinating mon oubiquitinated PCNA. The hypo methylation status of these two genes gives them a major expression potential. On the contrary,  CALD1, DNMT1, HDAC1 and RRP1  are hyper methylated, blocking any possibility of transcription ( Figure 2 ).\nAs a systemic immune inflammatory syndrome, endometriosis mainly causes pelvic pain, oxidative stress, mitochondrial dysfunction, adhesions with a high risk of reduced infertility potential and epigenetic status changes in some genes.\nFor endometriosis symptom therapy, there is an increase in clinical management options and the drug range available to reduce painful systemic immune inflammatory symptoms, improve social life conditions and increase fertility potential reservation. 13 – 15\nA correlation has been demonstrated between endometriosis and elevated levels of free DNA in the blood. 1 , 16  CfDNA is a complex biomarker that can be traced back to multiple sources, including cellular apoptosis, necrosis and active secretion, making it a valuable tool in noninvasive diagnostics such as oxidative stress.\nTo the best of our knowledge, this preliminary study is the first to combine diagnosis and treatment of women suffering from endometriosis.\nIn a normal situation, an endonuclease ensures excess free DNA is eliminated. In endometriosis, this mechanism is unfortunately insufficient to ensure all cfDNA is removed, which means it is significantly increased. Logically, an exogenous supply of synthetic DNase I has become necessary to help the body promote faster necrotic and apoptotic DNA fragment excretion.\nHowever, when we compared our clinical observations with the free DNA levels of all patients before and after treatment, we observed a descriptive decrease; this observation should be interpreted with care, and no formal statistical conclusion is drawn in the revised manuscript. We can suggest several possible explanations: The first is a dose effect; given the product’s very short half-life, it is likely that 2,500 IU administered every other day is insufficient. A daily, or even twice daily, dose would be more appropriate. For treatment compliance reasons, designing a long-acting DNase should be considered.\nWhen administered nasally, the Pulmozyme plasma level was actually 10 to 15% of the maximum, an absorption rate that was insufficient to significantly reduce free circulating DNA (Ferring patent 2013). Only intravenous (and/or IM and subcutaneous) administration, used for toxicological testing, provided sufficient concentrations without harmful effects (Pulmozyme Roche file) when tested on 20 volunteers. The short half-life of the product required higher concentrations, but the 2013 experiment taught us that free DNA stock replenishment was delayed. Because of this, as well as for practical reasons, we decided to inject only one ampoule every two days. In addition, treatment compliance and product storage appeared to be better with this treatment schedule. Given the opportunistic events that can occur during a month (colds and infectious or inflammatory problems), it seems more appropriate to treat patients daily with one ampoule to achieve a significant difference in free DNA concentration before and after treatment. The one-month treatment period makes it possible for such phenomena to occur, which could interfere with the result interpretation.\nA second possible explanation is that some patients may have had a viral or other such infection during treatment that could have increased their free DNA levels. Therefore, in a future controlled study, the treatment time should be shortened to 15 days to confirm a significant and objective effect on cfDNA levels.\nA final possible explanation is that some patients with very high levels of cfDNA after treatment may have forgotten to take certain injections, an issue which could be avoided by formulating DNase differently, for example, as a delayed-release formulation.\nRegardless of the free cfDNA level, it is interesting to note that a minimal concentration of DNase substantially reduced pain in all but one patient. We have no explanation for these two contrasting effects; however, DNase may have a dose-dependent effect, resulting in apparent resistance to treatment due to a particular genomic profile.\nThe difficulty of this work lies in the treatment time, during which even the slightest viral or bacterial infection can cause an excessive increase in free DNA levels. This is why we believe that treatment could be more intense and shorter, or delivered in a long-acting form.\nThe rise in cfDNA levels should enable endometriosis progression to be monitored and, if necessary, indicate when treatment should be resumed.\nThe second difficulty will be to introduce a placebo arm, as patients suffering from this condition are keen to receive symptomatic treatment. This will need to be performed in our next study.\nAt the genetic level, a specific necroptosis-related gene signature was reported by Wang et al. 2025  16 , and a specific model analysis reported seven specific genes \n 17 \n  as diagnostic markers of endometriosis. Endometrial single cell ribonucleic acid sequencing (scRNA-seq) \n 18 \n  and piwi RNA saliva-reverse transcription and sequencing \n 19 \n  were proposed as diagnostic signatures of endometriosis.\nIn our preliminary study, the two hypomethylated genes in the endometriosis group were  DIP2C and USP1 .\nBoth genes can influence the ubiquitination state of histones, such as H2A or H2B. This is critical for chromatin remodeling, which can either facilitate or inhibit the access of DNA methyl transferases (DNMTs) or other methyl transferases. In endometriosis, inflammatory processes may be aberrantly activated due to the overactivity of hypomethylated genes. The four genes  CALD1, DNMT1, HDAC1 and RRP1  are hypermethylated. This study need caution of interpretation before extrapolating from molecular data to clinical implication. Large double blind study with probably daily injection of DNase is recommended before final conclusions.\nThese findings pave the way for precision-based approaches to endometriosis through gene-targeted interventions. However, these results should be interpreted with caution; future studies should validate the effective dosage and long-term safety of DNase therapy.\nThe ideal option for this study would be a placebo-controlled design and daily injection of DNase. Owing to the complexity and cost of the enzyme, it was difficult to include a placebo group or daily DNase injections in this preliminary study. This study is preliminary, includes a limited dataset, and has no placebo or control group, which precludes definitive therapeutic conclusions. Because all women were informed that they received DNase I, changes in ENS and HRQoL may include expectation or placebo effects and cannot be interpreted as proof of treatment efficacy. Also the sample size calculation was not performed. In addition, the original aggregated methylation read-count analysis does not account for intra-subject dependence; therefore, methylation results should be considered exploratory and descriptive. Future research with larger samples, randomized double-blind placebo-controlled designs and longer follow-up is needed to validate and expand upon these findings.\n\nTo our knowledge, this is a preliminary study evaluating DNase treatment in relation to free circulating DNA levels, methylation profiles and patient-reported symptoms before and after treatment. These data should be interpreted with care because of the limited sample size, the absence of a placebo group and the exploratory descriptive nature of the molecular analyses. The observed patient-reported improvements are encouraging, but they cannot be causally attributed to DNase I therapy in this open-label uncontrolled design. Further research including the modifications suggested in the Discussion is warranted.","source_license":"public-domain-us","license_restricted":false}