Abstract
Background/Aim: Hereditary hemorrhagic telangiectasia (HHT) is a rare disease with an incidence of 1:5,000. HHT is inherited in an autosomal dominant manner and is associated with vascular malformations. It particularly affects the genes ENG (HHT1) and ACVRL1 (HHT2). Clinically, patients typically exhibit pronounced recurrent epistaxis. The aim of this study was to evaluate if ACVRL1 knockdown in HMEC-1 endothelial cells could induce a HHT2-like phenotype which could be deployed as a model for HHT2.
Materials and Methods: The human immortalized endothelial cell line HMEC-1 was used for the experiments. RNAi knockdown was performed using a pool of four siRNAs targeting the ACVRL1 gene. The gene knockdown was verified using RT-qPCR and Western blot analysis. The effects of ACVRL1 knockdown on angiogenesis were compared to a non-target (NT) small RNA control using a tube formation assay. The expression of 84 endothelia-associated genes was analyzed with RT-qPCR.
Results: Tube formation ability was significantly affected by ACVRL1 knockdown. In particular, the parameters total tube length, total segment length, total master segment length, total branching length, total mesh area and branching interval were significantly increased whereas total isolated branch length, number of master junctions, number of branches and the number of isolated segments decreased after ACVRL1 knockdown. Significant changes in the expression of angiogenesis related genes were detected by qPCR analysis and discussed.
Conclusion: Knockdown of ACVRL1 in HMEC-1 endothelial cells leads to pathological angiogenesis with enhanced tube formation capacity, which is similar to the angiogenesis in vivo in patients with HHT2. The presented HMEC-1 cell-based system therefore has the potential to be deployed as an in vitro model for HHT2 studies.
Introduction
The Rendu-Osler-Weber syndrome, also known as hereditary hemorrhagic telangiectasia (HHT), is a relatively common rare disease with an incidence of about 1:5,000 (1, 2). HHT is inherited in an autosomal dominant manner and is associated with arteriovenous malformations. Mutations mainly affect genes of the transforming growth factor (TGF)-β or bone morphogenetic protein (BMP)-9 signaling pathway. De novo mutations are rare. The main genes involved in HHT are: Endoglin (ENG, HHT1) or the activin A receptor like type 1 (ACVRL1, HHT2) (3). Other genes have also been described, including the suppressor of mothers against decapentaplegic homolog 4 (SMAD4) gene and the growth differentiation factor 2 (GDF2) gene, which encodes BMP9 (4). The original view of the underlying pathomechanism was that mutation of a single allele, typically inherited, leads to haploinsufficiency of the respective gene, resulting in disruption of angiogenesis (5-8). However, in recent years growing evidence was presented that, for development of the full-blown HHT phenotype, a second hit, e.g. an additional somatic mutation of the wild type allele (9, 10) but also UV-light, mechanical or inflammatory stress (11), is required.
Disruption of normal capillary formation leads to direct connections of arterial and venous vessels in which the capillary bed is skipped (1). As a result, arteriovenous malformations can occur in various parts of the body which can lead to life-threatening hemorrhages. The most common are recurrent spontaneous nosebleeds, which severely impair the patient’s quality of life (12). The mucous membranes of the gastrointestinal tract also frequently bleed, but also hemorrhages of vascular malformations of the central nervous system or other sites can occur (13). Typically, patients are treated by symptom-oriented approaches, but a causal therapy is not yet available (14-16). The Curaçao criteria, which were described 25 years ago, have been used clinically for HHT diagnosis (17). The criteria include epistaxis, mucocutaneous telangiectasia at typical predilection sites (e.g., lips, oral cavity, fingers and nose), visceral involvement (e.g., gastrointestinal, cerebral, pulmonary, and hepatic) and a positive family history of affected first-degree relatives. If all or at least three of the above criteria are met, HHT is considered confirmed. If two or less criteria are positive, further diagnostics like molecular genetic analysis can be sought for further validation (17, 18). Several HHT model systems have been developed for in vitro and in vivo studies, such as HHT knockout mice or the zebrafish model (19). In our study, we used RNAi to knockdown ACVRL1 in the endothelial cell line HMEC-1 aiming to create HHT2 like conditions. This could help provide an easy-to-handle cell-based in vitro system for testing therapeutic approaches aiming to achieve normalization of angiogenesis in HHT2 patients.
Materials and Methods
HMEC-1 cells and cell culture. The human microvascular endothelial cell line HMEC-1 (20) was purchased from ATCC (LGC Standards GmbH, Wesel, Germany). HMEC-1 cells were cultured in uncoated 10 cm cell culture dishes under standard conditions. The culture medium was MCDB-131 (Molecular, Cellular and Developmental Biology) (Gibco Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal calf serum (FCS) (Sigma-Aldrich Life Science, St. Louis, MO, USA), 10 mmol/l glutamine (BioWest - The Serum Specialist, Nuaillé, France), 10 ng/ml epidermal growth factor (EGF) and 1 μg/ml hydrocortisone (both Sigma-Aldrich Life Science). The identity of HMEC-1 cells was validated by genotyping (Leibniz Institute DSMZ, German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany) using short tandem repeat (STR) analysis. No cross-contamination with other cell lines was detected.
Transfection of HMEC-1 cells with ACVRL1 siRNA. For transfection, 1×106 cells were seeded in 6 cm cell culture plates (Sarstedt, Nümbrecht, Germany) followed by transfection using Lipofectamine® 2000 (Thermofisher Scientific, Waltham, MA, USA) according to the manufacturer’s protocol. For transfection, 40 nmol/l siRNA directed against ACVRL1 (NM_000020.3; ON-TARGET plus SMARTpool, Dharmacon; pool of 4 target sequences: AGCCUAAAGUGAUUCAAUA, GUCAAGAUCUUCUCCUCGA, CGGGAGUGCUGGUACCCAA, GCAUCUGAGCAGGGCGACA) and 40 nmol/l control non-target (NT) RNA (ON-TARGET plus SMARTpool, Dharmacon; pool of 4 target sequences: UGGUUUACAUGUCGACUAA, UGGUUUACAUGUUGUGUGA, UGGUUUACAUGUUUUCUGA, UGGUUUACAUGUUUUCCUA) were used. During transfection, 4 ml MCDB-131 medium without additives was deployed. After 4 h, the medium was replaced with standard culture medium (see above) and cells were incubated for 24 h before further processing.
Transfection of HMEC-1 cells with an ACVRL1 encoding plasmid. The Multiporator® from Eppendorf (Hamburg, Germany) was used for electroporation. A total of 2.4×105 cells were harvested for transfection and incubated in 390 μl hypoosmolar buffer (90 mOsmol/kg, #732-6007, Eppendorf) together with 8 μg of an ACVRL1-wildtype encoding plasmid (pCMV6-AC ACVRL1, #SC321860, Origene, Rockville, FL, USA). The respective empty vector (pCMV6-AC, #PS100020, Origene) was used as a negative control. A total volume of 400 μl was transferred into cuvettes with a 2 mm electrode slit (#4307000593, Eppendorf). Transfection parameters were voltage (400 V) and pulse duration (100 μs). After transfection, the HMEC-1 cells were plated in MCDB-131 medium without additives and incubated for 24 h under standard conditions before further processing.
Tube formation assay. For the tube formation assay, Matrigel® growth factor reduced (GFR) (Corning, #354230 lot #2116002) was thawed on ice. Sixty μl of Matrigel® GFR were pipetted bubble-free in wells of a pre-cooled 96-well plate. For curing of the Matrigel®, the 96-well plates were incubated at 37°C under standard culture conditions for 30 min. 2×104 transfected HMEC-1 cells were seeded per well. These were analyzed and documented microscopically 4, 24 and 48 h after plating. Three technical replicates were prepared per experiment with a total of 3 independent experiments. Graphical analysis was performed using ImageJ (version 1.53) (21) and the Angiogenesis Analyzer Plugin Tool (22).
Analysis of gene expression. The gene expression analysis was carried out 24 h after transfection. For this purpose, the transfected HMEC-1 cells were first detached by trypsin (Cytiva, Fisher-Scientific, Schwerte, Germany) treatment. RNA extraction was performed using the RNeasy Mini Kit (Qiagen, Venlo, the Netherlands). The RNA concentration was determined using the Implen NanoPhotometer® NP80 (Implen, Inc., Westlake Village, CA, USA). Reverse transcription was performed with the RT2 first strand kit (Qiagen) and the RT2 SYBR Green qPCR Mastermix (Qiagen) was used for quantitative PCR (QuantStudio 5, Thermo Fisher Scientific). The following primer pairs were deployed: ALK1_forward1 /ALK1_reverse1 (5′-CCCAGAGGGACCATGACCTT-3′ and 5′-CCCCTTGCAATGTGGGC-3′), ALK1_forward2/ALK1_reverse2 (5′-GGCAGCGATTACCTGGACAT-3′ and 5′-GCAATCTCCCACAGCACCA-3′), emc_forward/emc-reverse (5′-TAGCAGCGGCAGCAGTAAAAC-3′ and 5′-ACCCAGTGTTGCCGTGTTTG-3′, gpi_forward/gpi_reverse (5′-TACCAGCTCATCCACCAAGGC-3′ and 5′-AAGTTGGCCAGGAGGATCTTGTG-3′), rplp0_forward/rplp0_reverse (5′-CGAAGCCACGCTGCTGAAC-3′ and 5′-ATGCTGCCATTGTCGAACACC-3′). The RT2 Profiler™ PAHS-015Z of human endothelial cell biology (Qiagen Sciences) was used to analyze 84 genes associated with endothelia according to the manufacturer’s instructions. The data was analyzed using the online analysis tool Geneglobe from Qiagen (https://dataanalysis2.qiagen.com/pcr/analysis/139309; accession date: August 1st, 2025).
Western blot analysis. Whole cell protein was extracted by using a protein lysis buffer (20 mmol/l Tris/HCl pH 7.5, 137 mmol/l NaCl, 10% glycerol, 1% NP40 and 2 mmol/l EDTA), containing 100 μl/ml protease inhibitor cocktail 1 (Sigma-Aldrich, Saint Louis, MO, USA) and 50 μl/ml phosphatase inhibitor cocktail 2 (Sigma-Aldrich). The protein concentration was measured with the Bradford assay (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Equal amounts of protein (25 μg) were subjected to 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to nitrocellulose (NC) membranes and blocked with 3% non-fat milk in PBS for 20 min. The NC-membranes were then incubated overnight at 4°C with the following primary antibodies: ACVRL1 antibody anti-rabbit (dilution 1:500; #PA5-14921, lot #WI3392314A, Thermo Fisher), and GAPDH (dilution 1:1,000; mouse monoclonal, clone 0411; #sc-47724, lot #GO815, Santa Cruz Biotechnology, Heidelberg, Germany). The next day, the membranes were incubated for one hour in the dark with the corresponding secondary antibodies (1:2,000, goat anti-rabbit IgG HRP, #12-348; goat anti-mouse IgG HRP, #AP181P, both from Merck, Darmstadt, Germany). A chemiluminescence detection kit (WesternBright™ Sirius HRP substrate, #541021, Advansta Inc., San Jose, CA, USA) and the ChemiDoc MP imaging system (BioRad, San Francisco, CA, USA) were used to visualize the protein bands. Finally, ImageJ software (NIH, Bethesda, MD, USA) (21) was used for protein quantification.
Flow cytometry. Protein expression was additionally analyzed by flow cytometry. The cells were first detached by trypsin and then fixed in ethanol. The cell membrane was then permeabilised using 0.5% NP40/1% BSA buffer. The proteins were exposed to an ACVRL1 directed antibody (#PA5-1492, Thermo Fisher) (1:25) and the background fluorescence was determined with normal rabbit IgG (sc-2027, lot #J2909, Santa Cruz Biotechnology, Heidelberg, Germany) in NP40 0.5%/BSA 1% buffer. Incubation was performed overnight. Alexa Fluor 488 anti-Rabbit IgG (#A21206) was used as a secondary antibody at a dilution of 1:200. Incubation was carried out for at least one hour in the dark at 4°C. Flow cytometry was performed with the BD LSRFortessa™ system (Becton Dickinson, Franklin Lakes, NJ, USA). Flow cytometry parameters were determined using the forward scattered light (FSC; proportional to cell size) and side scattered light (SCC; proportional to cell granularity) settings. The FITC-A channel (FL1) was used as the measurement channel. The flow cytometry data was then analyzed with the FlowJo™ software (version 7.6.5, Tree Star Inc., Ashland, OR, USA).
Statistical analysis. The Graph PadPrism software (version 10.2.2 for MAC; GraphPad Software, Inc., San Diego, CA, USA) was used for the statistical analysis except for the RT-2 Profiler Assays, which were analyzed by using the online RT2 Profiler PCR Array Data Analysis Tool (https://geneglobe.qiagen.com/us/analyze). A one-tailed, unpaired Student’s t-test was used to calculate ACVRL1 mRNA (RT-PCR) and protein (FACS) expression of transfected HMEC-1 cells. For analysis of data of the tube formation assay, a 2-tailed Student’s t-test was used. If the variances of the two tested groups were significantly different, a Welch correction was performed in each case. In case of the RT2 Profiler PCR Array Data, the threshold cycle (Ct) values of each sample (n =3 for control, n=3 ACVRL1) were uploaded onto the SABioscience website (August, 1st, 2025) and relative gene expression was analyzed using 2−ΔΔCt. For statistical calculations (unpaired, 2-tailed Student’s t-test), the online software does not use the ΔΔCT values, but rather the 2−ΔCT values of the biological replicates, so that the significance test is performed on the linear expression values. Data represents the mean ± standard deviation (SD), with p<0.05 considered statistically significant. Statistical differences were indicated as *p<0.05, **p<0.01, and *** p<0.001.
Results
Tube formation features are affected after ACVRL1 knockdown in HMEC-1 cells. Knockdown of ACVRL1 transcripts was achieved by transfection of HMEC-1 cells with ACVRL1 specific siRNAs. Significant reduction in ACVRL1 mRNA could be demonstrated by qRT-PCR (Figure 1A). Although not reaching significance, a clear reduction of ACVRL1 protein levels could be observed by flow cytometry (Figure 1B) and Western blot analysis (Figure 1C). The angiogenic potential of ACVRL1 knockdown and control cells, the latter being transfected with non-target small RNA (siNT), was evaluated with a Matrigel®-based tube formation assay and quantitated with the image processing software ImageJ (21) and the plugin tool Angiogenesis-Analyzer (22). Microscopic images were taken at 4, 24 and 48 h after seeding (Figure 2A). A distinct tube formation occurred in ACVRL1 knockdown as well as in siNT control cells. In comparison to the siNT control cells, ACVRL1 knockdown cells exhibited an increase in total tube length (Figure 2B, significant at 4 h), total segment length (Figure 2C, significant at 24 h), total master segment length (Figure 2D, significant at 24 h), total branching length (Figure 2E, significant at 24 h), total mesh area (Figure 2G, significant at 4 h) and branching intervals (Figure 2K, significant at 4 h). Compared to controls, ACVRL1 knockdown cells exhibited a reduction in total isolated branch length (Figure 2F, significant at 4 h), number of master junctions (Figure 2H, significant at 24 h), number of branches (Figure 2I, significant at 4 h) and number of isolated segments (Figure 2J, significant at 4 and 24 h).
RNAi knockdown of ACVRL1 in HMEC-1 cells. (A) The graph illustrates the alteration of ACVRL1 mRNA levels after RNAi using qRT-PCR analysis. Two different primer pairs were used for the analysis of the ACVRL1 gene (see materials and methods). A significant reduction in ACVRL1 mRNA expression was seen after RNAi knockdown. Although significance was not reached, on average ACVRL1 reduction at the protein level was also detected by flow cytometry (B) as well as Western blot analysis (C) following siACVRL1 RNAi. A one-tailed, unpaired Student’s t-test was used. Significant results (p≤0.05) are indicated with an asterisk (*). Experiments were conducted as triplicate (n=3). GAPDH was used as a loading control. ImageJ was used for protein quantification of Western blot results. The image in (A) was created with Affinity Designer 2 (version 2.6.4).
Following siACVRL1 and siNT treatment, alterations in tubule formation were documented over 2 days. (A) Representative images of the vascular network at 4, 24 and 48 h after seeding on Matrigel® are provided as an example. Upper images were taken after the siNT treatment whereas the lower images were taken after siACVRL1 knockdown. All images were captured by phase-contrast microscopy and processed using ImageJ (Fiji) and the Angiogenesis Analyzer. Shown in (B-K) are angiogenesis-related features exhibiting statistical differences after ACVRL1 knockdown. Statistical analysis was conducted using an unpaired t-test, and significant results are indicated by asterisks: *p<0.05, **p<0.01, ***p<0.001. The number of experiments was n=3, with three technical replicates per experimental run.
Overexpression of ACVRL1 in HMEC-1 cells. To observe a possible effect of ACVRL1 overexpression on angiogenesis, an ACVRL1 encoding vector was transfected into HMEC-1 cells and compared to HMEC-1 control cells that received the empty vector only (see Materials and Methods). A moderate ACVRL1 overexpression could be achieved in HMEC-1 cells (significant on the mRNA level and during flow cytometry). The subsequent tube formation analysis demonstrated a rather opposite effect on angiogenesis features as seen after ACVRL1 RNAi knockdown. However, none of the observed changes reached significance (see Supplementary Material).
Impact of ACVRL1 knockdown on the expression of endothelia-related genes. The expression of 84 endothelia-related genes was evaluated by qRT-PCR in HMEC-1 control and ACVRL1 knockdown cells. The respective genes could be assigned to the following 8 groups: “proteases”, “receptors and ligands”, “cell adhesion molecules”, “blood clotting”, “enzymes”, “apoptosis”, “cytokines” and “others” (Figure 3). The “proteases” MMP1 and ADAM17 were significantly elevated in ACVRL1 knockdown cells whereas in the “receptors and ligands group”, TEK, FLT1, FGF2, F2R and ENG showed significant induction. VCAM1 was significantly downregulated and ITGA5 and CDH5 significantly upregulated in the “cell adhesion molecule group”. The “blood clotting related genes” TFP1 and PLAU were significantly upregulated, while PLG was significantly downregulated. Furthermore, TGFβ1 was highly upregulated and IL6, IL1B, CXCL1, CCL5 downregulated in the “cytokines” group. In addition, the “apoptosis” related genes CFLAR, CASP3 and BAX exhibited significant upregulation in HMEC-1 knockdown cells. In the “enzymes” group, SPHK1, SOD1, SERPINE1, and PTK2 were upregulated, while ALOX5 and AGT were downregulated. Other significantly upregulated genes included CAV1 and ANAX5. Interestingly, as observed in HHT patients (23), VEGFA was also upregulated in ACVRL1 knockdown cells compared to control cells, however without reaching significance.
Effect of ACVRL1 knockdown on the expression of endothelia-related genes. The differences between various endothelia-related genes between ACVRL1 knockdown and siNT-transfected cells are shown. The following groups are presented: proteases, receptors and ligands, cell adhesion molecules, blood clotting, enzymes, apoptosis, cytokines and others. A total of n=3 samples were tested for all groups. The analysis of the RT2-Profiler arrays was conducted by using the RT2 Profiler PCR Array Data Analysis Tool (https://geneglobe.qiagen.com/us/analyze). For statistical calculations (t-test), the online software is using the 2−ΔCT values of the biological replicates, so that the significance test is performed on the linear expression values. Significant results are indicated with asterisks: *p<0.05, **p<0.01. For gene names please visit https://www.genenames.org/.
Discussion
Since HHT is a rare disease, there is only limited availability of biomaterials from patients that could be used in HHT research. Furthermore, biomaterials are difficult to standardize. The aim of the present study was to evaluate if downregulation of ACVRL1 in the human microvascular endothelial cell line HMEC-1 could mimic the pathological vascular phenotype as observed in HHT2 patients.
Similar approaches were explored by other groups. Lamouille and colleagues expressed a constitutively active ALK-1 mutant in HMVEC, HMEC-1 and HUVEC endothelial cells and observed inhibition of migration and a cell cycle arrest in the G1 phase consistent with maturation of the endothelial cells (24). In the present study, we observed a similar tendency to inhibition of angiogenesis after ACVRL1 overexpression, however, the data didn’t reach significance. Overall, the findings are consistent with observations in HHT2 patients, which exhibit a pathological hyperproliferation of endothelial cells, when one of the ACVRL1 alleles is inactivated by mutation (25). The protease MMP1, which is involved in vascular remodeling and angiogenesis (26), and ADAM17, a sheddase promoting angiogenesis (27), were found significantly upregulated in HMEC-1 cells after ACVRL1 downregulation. The TEK tyrosine kinase (TIE2), FLT1 (VEGFR-1), FGF2, F2R and ENG were all significantly upregulated after ACVRL1 knockdown. TEK (TIE2) was shown to be involved in the pathogenesis of venous and lymphatic malformations (28, 29). ANGPT2 can destabilize TEK (TIE2), thereby promoting vascular remodeling (30). Although some reports hypothesize that FLT1 (VEGFR-1) does not exhibit high activation levels after ligand binding and could act rather as an antiangiogenic decoy receptor (31), there are also reports about a proangiogenic role of this receptor if acting as a heterodimer bound to VEGFR-2 (32). Furthermore, reports found mutant FLT3, a related receptor tyrosine kinase that is involved in the pathogenesis of acute myeloid leukemia (AML), to be regulated by circRNA MYBL2 (33). However, a possible involvement of circRNAs in the dysregulation of endothelia related genes such as FLT1 still needs to be elucidated. F2R is involved in the maintenance of the endothelial barrier integrity and endoglin is part of the BMP9/TGFβ signaling pathway and found to be mutated in HHT1 patients. It appears that ACVRL1 knockdown could induce a reactive upregulation of ENG in this experimental setting. Reportedly, FGF2 is involved in retinal neovascularization (34) and similarly could contribute to the pathologic increase in tube formation as observed after ACVRL1 knockdown. Interestingly, VCAM-1, which binds to α4β1-integrins on the surface of leukocytes, mediates attachment of these immune cells to the endothelial surface (35) and was found significantly reduced after ACVRL1 RNAi knockdown. HIF-1α, promoting angiogenesis during hypoxia, also plays a major role in tumor angiogenesis (36). After ACVRL1 knockdown in HMEC-1 cells, HIF-1α appeared elevated but did not reach significance. ITGA5 and CDH5 (VE-Cadherin) are essential for proper endothelial cell contact and found overexpressed in ACVRL1low HMEC-1 cells, which appears consistent with the observed enhanced level of tube formation ability of these cells. The enzymes SPHK1, SOD1, SERPINE1 and PTK2 were upregulated and the enzymes ALOX5, AGT downregulated in ACVRL1-knockdown cells. The anti-apoptotic protein CFLAR and the pro-apoptotic proteins CASP3 and BAX were found upregulated in the respective knockdown cells. TGFβ1 appeared as one of the most up-regulated genes after RNAi knockdown of ACVRL1. This is in accordance with own observations on iPS cells in which one allele of the ACVRL1 gene was mutant. In these cells, we also observed an upregulation of TGFβ1, presumably the precursor form (3). Similarly, as observed for ENG, TGFβ1 upregulation could be a mean of compensation for the BMP9/TGFβ signaling pathway to account for the lack of ALK1 (ACVRL1) contribution to signaling.
Conclusion
Here, we could demonstrate that RNAi knockdown of ACVRL1 in HMEC-1 endothelial cells affects angiogenesis in a similar manner as expected to occur in HHT2 patients. Furthermore, gene expression analysis revealed significant changes in mRNA levels of endothelia and angiogenesis related genes after ACVRL1 knockdown. The system furthermore has the potential to mimic the endothelial phenotype after inactivation of one as well as two ACVRL1 alleles, as proposed by the second hit hypothesis (9). This could be achieved by the degree of ACVRL1 knockdown, e.g., ~50% or ~100%. Future studies need to expand and standardize this model. Such a simplified in vitro model of HHT2 could therefore allow to more easily perform preclinical tests such as high throughput screening of candidate therapeutic compounds prior to testing these in more complex biosystems such as embryoid bodies, organoids or animals.
Acknowledgements
Technical assistance by Ms. Maria Sadowski and Ms. Roswitha Peldszus (Department of Otorhinolaryngology, Head and Neck Surgery, University Hospital Marburg and Philipps-Universität Marburg, Marburg, Germany) was greatly appreciated. The authors thank Christian Möbs (Clinical & Experimental Allergology, Department of Dermatology and Allergology, Philipps-Universität Marburg, Marburg, Germany) for providing the Multiporator® for the electroporation of HMEC-1 cells. Data of this manuscript is part of the doctoral thesis of JRR.
Footnotes
Authors’ Contributions
JRR: Experimental work, data curation, analyzed and interpreted the results, contributed to original draft. MB: Experimental work with the RT2-Profiler Assay, data curation, analyzed and interpreted the results, statistics, contributed to original draft. BAS: Review and editing, funding support. UB: Review and editing. UWG: Review and editing. RM: Conceived and designed the study, supervision of research, data curation, analyzed and interpreted the results, contributed to original draft.
Supplementary Material
Supplementary Figures 1 and 2 can be accessed via the following link: https://doi.org/10.5281/zenodo.17632452
Conflicts of Interest
The Authors declare no conflicts of interest.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received November 4, 2025.
- Revision received November 18, 2025.
- Accepted November 21, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.