Abstract
Background:
The clinical efficacy of anticancer drugs is often limited by dose-dependent toxicity to normal cells, particularly to rapidly dividing stem and progenitor cell populations. Traditional preclinical toxicity screening relies on animal models or primary human cells/immortalized lines, both of which have significant limitations regarding scalability, reproducibility, genetic diversity, and translational relevance.
Objective:
This study aims to establish and validate a human pluripotent stem cell -based platform for profiling the toxicity of cytostatic drugs on healthy, dividing cell populations, and to compare their sensitivity to that of cancer cell lines.
Methods:
Two hiPSC lines (male and female) were reprogrammed from umbilical cord blood cells, fully characterized, and differentiated into neural progenitors and mesenchymal derivatives. Dose–response curves were generated using a dilution series of four anticancer drugs and four other pharmacologically active compounds based on cell viability. Toxic half-maximal values (IC50) were determined for each cell type and compared to those obtained from conventional cancer cell lines.
Results:
The highest sensitivity to DNA-damaging agents was exhibited by pluripotent stem cells, followed by neural progenitors, with mesenchymal derivatives being the least sensitive. Notably, except for mesenchymal cells, both pluripotent stem cells and their differentiated derivatives were more sensitive to cytostatics than cancer cell lines.
Conclusion:
Human iPSC-derived stem/progenitor cells recapitulate the in vivo sensitivity of healthy dividing cell populations to chemotherapy. This platform offers unlimited, genetically consistent, and reproducible access to human cells, rendering it as a powerful complementary approach or even an alternative to animal models and primary human cells for early-phase drug safety profiling.
Introduction
A major limitation of conventional anticancer chemotherapy is its lack of selectivity for malignant tissues, leading to severe damage to rapidly renewing healthy tissues. Consequently, clinical dose-limiting toxicities—such as myelosuppression, mucositis, hair loss, infertility, and gastrointestinal damage—result from the destruction of healthy stem and progenitor cell compartments, including hematopoietic, intestinal crypt, and germ cell populations [–]. The inability to distinguish between malignant and healthy proliferating cells directly narrows the therapeutic window and treatment options. Rapid and accurate preclinical prediction of toxicity in normal human stem and progenitor populations is an urgent, unmet need [].
This problem is exacerbated by the limited predictive value of traditional preclinical models, as more than 90% of compounds that demonstrate safety and efficacy in animal studies ultimately fail in human clinical trials due to unexpected toxicity or lack of efficacy []. In order to improve human relevance and adhere to the 3Rs (replace, reduce, refine animal models), global regulatory frameworks have undergone a significant paradigm shift towards new approach methodologies (NAMs). Landmark initiatives—including the US FDA (Food and Drug Administration) Modernization Act 2.0 [], the subsequent FDA Strategic Roadmap 2025 (FDA, 2025), the revised guidelines of the European Medicines Agency (EMA) and the 3R Working Group (EMA, 2023; EMA 3R Working Group Biennial Report, 2023–2024), and international harmonization efforts (IMRWG3R, 2024)—now formally recognize and encourage human cell-based studies to establish drug safety and efficacy.
Despite the above regulatory directions, the preclinical application of healthy human stem and progenitor controls remains rare, and their lack limits more accurate
in vitrotoxicity assessments. When using healthy human controls (primary cells or cell lines) to calculate the selectivity index (SI) of a compound [
,
], current experimental setups have significant limitations:
Inadequately representative models: Many studies rely on a single normal human control line, such as fibroblasts or epithelial cells. This does not reflect clinical reality, as the primary off-target effects of chemotherapy are related to actively dividing stem/progenitor cells, not quiescent fibroblasts.
Diverse genetic background: Although the inclusion of multiple normal human cell types may improve the estimation of compound selectivity [], these lines typically originate from different donors. This genetic heterogeneity makes it difficult to determine whether changes in drug sensitivity arise from true tissue-specific vulnerabilities or from underlying donor variability.
In vitro culture limitations: The procurement of primary human cells is limited by finite proliferative capacity and rapid functional decline ex vivo [–]. Immortalized cell lines offer better reproducibility, but often harbor genetic mutations and altered cell cycle regulation, or are themselves of tumor origin, making their toxicological data difficult to translate into physiological human responses [, ].
Human induced pluripotent stem cells (hiPSCs) offer an ethical, non-malignant alternative to embryonic stem cells and tumor lines, providing a virtually inexhaustible source of human cells for disease modeling and drug screening [–]. Because undifferentiated hiPSCs exhibit a rapid cell cycle, stringent DNA damage checkpoints, and a robust p53-mediated apoptotic response [, ], their heightened sensitivity to genotoxic stress effectively models the vulnerability of rapidly dividing healthy tissue []. Crucially, a single hiPSC line can be differentiated into multiple lineage-specific cell types. This allows for direct comparison of intrinsic, lineage-dependent toxicity profiles while minimizing the confounding effects of inter-donor genetic heterogeneity [, ].
The choice of somatic donor tissue directly impacts the stability and quality of the resulting hiPSC lines. Umbilical cord blood-derived cells are biologically young, exhibiting high telomerase activity [22], fewer somatic mutations [, 23], and a “naïve” epigenetic state that facilitates high reprogramming efficiency [24]. Collected via non-invasive, routine biobanking methods [25], these newborn cells possess minimal immunological memory and lack the age-related cellular damage, chronic inflammation, and epigenetic drift typical of adult donor cells—alterations that can persist post-reprogramming [26–28]. While their early developmental origin may result in altered metabolic profiles or immature drug-metabolizing enzyme activity relative to adult cells [29], their high baseline fidelity makes them an ideal starting material for predictive toxicological platforms.
Despite clinical evidence showing that women face a higher risk of adverse drug reactions (ADRs) [30], biological sex remains severely underrepresented in preclinical research [31, 32]. This omission is critical because sex-dependent differences in drug sensitivity and mitochondrial stress responses manifest at the cellular level [33]. To align with evolving regulatory mandates emphasizing sex as a biological variable, this study utilizes two established cord blood-derived hiPSC lines: KUT-2 (XX) and KUT-6 (XY), enabling the systematic evaluation of sex-specific cytotoxic liabilities.
One of the key advantages of hiPSC-based systems is the ability to generate and study multiple cell types derived from the same genetic background. This enables the investigation of lineage-specific toxicological responses, as well as the underlying mechanism-specific sensitivities []. In the present study, we hypothesized that different cell types may exhibit distinct sensitivities to cytotoxic compounds. To evaluate cell type-specific cytotoxicity across a biological gradient [], we examined undifferentiated hiPSCs alongside their differentiated derivatives representing distinct embryonic germ layers. hiPSCs are already routinely and successfully used in embryotoxicity and teratogenicity assays [34, 35]. Their differentiated stem/progenitor derivatives, although often retaining a fetal-like phenotype, provide valuable in vitro models for tissue-specific stem cell populations and enable the investigation of lineage-dependent toxic responses. The nervous system is particularly sensitive to toxicants, so neural models are essential for studying neurotoxicity [36]. Neurogenesis produces Neural Stem Cells (NSCs), which can self-renew and generate other types of neural tissue cells including neurons, astrocytes and oligodendrocytes [37]. hiPSC-derived NSCs represent an ectodermal model that is highly sensitive to metabolic stress, ion channel disruption and proteostasis impairment [36] and can be used as a model of the developing brain. Our other neural model is based on a specific differentiation protocol to generate hippocampal Neural Progenitor Cells (hNPCs) from hiPSCs. Since the adult dentate gyrus of the hippocampus contains neuronal stem cells that give rise to immature and subsequently mature functional granule neurons [38], hNPCs can be used as a model not only of the developing hippocampus, but also of adult neurogenesis [39]. In contrast, Mesenchymal Stem Cell-like cells (MSCs), differentiated from hiPSCs, are mesodermal stromal models. MSCs play a key role in the stromal microenvironment, contribute to tissue regeneration, and are characterized by lower proliferative activity, a different metabolism, and a more resistant stress response profile [40–42]. In addition to modeling stromal cells involved in regenerative processes, they may also contribute to the understanding of tumor-associated mesoderm cells that appear in the tumor environment [43]. To characterize cytotoxic responses across mechanistically distinct cellular stress pathways, we assembled a panel of eight pharmacologically active compounds with diverse and well-defined modes of action (Supplementary Table 1). This approach enables the identification of pathway-specific vulnerabilities while allowing comparison of cell type–specific sensitivities across hiPSCs, MSCs, and NPCs [, ].
The selected compounds in this study can be broadly grouped into two functional categories. The first comprises agents that primarily target DNA integrity and replication. Doxorubicin exerts its cytotoxic effects through multiple mechanisms, including DNA intercalation, topoisomerase II inhibition, and free radical-mediated DNA-damage, ultimately inducing, apoptosis, autophagy, senescence, necrosis, and metabolic alterations. Due to its high cytotoxic efficacy, it is widely used in the treatment of both adult and pediatric hematological malignancies and solid tumors, (breast, liver, bile ducts, endometrial tissue, osteosarcomas, esophagus) [44, 45]. Unfortunately, normal cells are also sensitive to doxorubicin, which primarily affects the heart, brain, liver, and kidneys, often long after the treatment. This cytotoxic profile underlines the importance of a precise application dose [45]. Mitoxantrone acts as DNA intercalators and topoisomerase II inhibitors, promoting strand breaks and replication arrest [46]; SN-38 stabilizes the topoisomerase I–DNA cleavage complex, inducing replication-associated single- and double-strand breaks [47]; and Laromustine functions as a sulfonylhydrazine alkylating agent that induces O6-guanine DNA alkylation, leading to the formation of DNA crosslinks, while simultaneously inhibiting the repair enzyme O6-alkylguanine DNA transferase [48]. The second category encompasses compounds that modulate non-genotoxic stress pathways: Deferiprone exerts its effects by chelating intracellular iron, thereby disrupting iron-dependent mitochondrial processes and promoting the accumulation of reactive oxygen species (ROS), including mitochondrial superoxide [49]; Celecoxib, a selective COX-2 inhibitor, exerts pro-apoptotic and anti-proliferative effects through inhibition of COX-2–mediated Prostaglandin E2 (PGE2) signaling and its downstream pathways, including NF-κB and Akt [50]; Phenothiazine interferes with multiple tumor-promoting pathways, including PDK1/Akt, MAPK/ERK1/2, and Akt/mTOR, and can promote lysosome-mediated cell death while inhibiting drug efflux pumps [51], while Quinacrine is primarily an antimalarial agent, but it is also known to have antitumor effects; it is capable of inhibiting phospholipase A and intercalating into DNA [52, 53]. Together, this panel enables a systematic evaluation of how hiPSC-derived cell types of distinct lineages respond to genotoxic, oxidative, autophagic, and signaling-mediated cytotoxic stimuli, thereby providing mechanistic insight into lineage-dependent differences in cellular stress tolerance.
Taken together, the present study employs a multidimensional experimental framework that integrates several key design principles of modern preclinical research. By utilizing hiPSC lines derived from umbilical cord blood of both a male and a female donor, this study combines the biological advantages of neonatal-derived cells with the ability to explore sex as a biological variable within a controlled genetic context. Differentiation of these lines into mesenchymal stem cells and neural progenitor cells, representing distinct germ layer derivatives, enables the investigation of lineage-specific cytotoxic responses. The application of a mechanistically diverse compound panel further allows the systematic interrogation of multiple cytotoxic pathways across cell types. This approach is designed to generate translatable, human-relevant data that addresses recognized limitations of conventional preclinical models, while contributing to the emerging evidence base for hiPSC-based platforms as physiologically meaningful tools for drug safety assessment.
Materials and methods
Detailed information about the materials used can be found in Supplementary Table 2.
Generation of hiPSC lines from umbilical cord blood
According to the permission for reprogramming and the study by the Human Reproduction Committee of the Hungarian Health Science Council (ETT HRB- (Approval n. 3883-2/2018/EÜIG), umbilical cord blood samples were obtained after written informed consent. Umbilical cord blood mononuclear cells (UCB-MNCs) were isolated via BD Vacutainer CPT tubes (sodium citrate) per manufacturer protocols. UCB-MNCs were seeded at 105 cells/well in 24-well plates using supplemented StemPro-34 medium containing 2 mM GlutaMax, 10 µM β-mercaptoethanol, 1% Antibiotic-Antimycotic and cytokines: 20 ng/mL IL-3, 100–100 ng/mL IL-6, FLT-3L and SCF. Following 2 days of expansion with daily half-medium changes, cells were transduced using the CytoTune-iPS 2.0 Sendai Reprogramming Kit (Oct3/4, Sox2, Klf4, c-Myc). Vectors were removed after 24 h, and cells were maintained in UCB-MNC medium for an additional 48 h. Transduced cells were subsequently transitioned onto mitomycin C-treated CF-1 feeder cells in cytokine-free StemPro-34 medium. After 2 days, the culture was transitioned to hiPSC medium (KO-DMEM, 10% KO-Serum Replacement, 100 mM NEAA, 2 mM GlutaMax, 10 mM β-mercaptoethanol, 4 ng/mL bFGF, 1% Antibiotic-Antimycotic) via gradual replacement. Emerging colonies were manually picked and expanded on feeder before being transitioned to feeder-free conditions on hESC-qualified Matrigel-coated plates in mTeSR1 medium. Two hiPSC lines were established, characterized, and registered at hpscreg.eu under the following names: RCNSi009-A for KUT-2 and RCNSi0010-A for KUT-6.
iPSC cell culture
KUT-2 and KUT-6 hiPSC lines were cultured on plates coated with Matrigel and maintained in mTeSR1 medium. Culture medium was refreshed daily. Cells were passaged every 3–4 days at an approximate ratio of 1:10 using ReLeSR, and subsequently replated onto freshly coated Matrigel surfaces in mTeSR1 medium.
Spontaneous differentiation of hiPSC lines
To demonstrate pluripotency, we examined the differentiation of the three germ layer directions during spontaneous differentiation. The spontaneous in vitro differentiation potential was examined by the formation of embryoid bodies (EBs), as described previously [54]. The EBs were cultured for 6 days in a free-floating culture and then maintained for 6 and 12 days on gelatin-coated Nunc Lab-Tek confocal chambers in DMEM medium supplemented with 10% ES tested FBS and 2 mM GlutaMAX™.
Directed differentiation of hiPSCs
The generation of hippocampal neuronal progenitor cells has been described previously [55]. The hNPCs were maintained on Matrigel coated plates in DMEM/F-12, 2 mM GlutaMAX™/N2/B27 medium containing 1% FGF2, 2% laminin and 1% Antibiotic-Antimycotic. For neural stem cell (NSC) induction, PSC Neural Induction was used according to the manufacturer’s instructions. The NSCs were cultured in Neuronal Expansion medium (AdvancedTM DMEM⁄F-12 and Neurobasal® Medium in 1:1 ratio, 2% Neuronal Induction Supplement, 1% Antibiotic-Antimycotic on Matrigel coated plates. We performed immunocytochemical staining to identify neural progenitor markers.
Mesenchymal-like cells (MSC) were generated from hiPSC lines as described previously [56]. MSCs were cultured in DMEM medium supplemented with 10% ES tested FBS and 2 mM GlutaMAX™. For adipogenic and osteogenic differentiation, Human Mesenchymal Stem Cell Functional Identification Kit was used according to the manufacturers’ instructions. To determine MSC markers, flow cytometry was used.
Human tumor cell line culturing
The human uterine sarcoma cell line Mes-Sa was obtained from American Type Culture Collection (ATCC). The human ovarian serous adenocarcinoma cell line OVCAR-8 (OVC-8) was provided by the National Cancer Institute Developmental Therapeutics Program of the National Institute of Health (NIH NCI DTP) as part of the NCI-60 human tumor cell line panel. Mes-Sa cells were cultured in DMEM, while OVC-8 cells were maintained in RPMI, both supplemented with 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin, and 2 mM L-glutamine.
Karyotype and STR analysis
The genetic identity and chromosomal integrity of the hiPSC lines were verified by short tandem repeat (STR-data can be presented upon request) profiling and G-banding karyotype analysis. These analyses were performed by UD-GenoMed Medical Genomic Technologies Ltd.
Flow cytometry measurements
Flow cytometric analysis of SSEA-4 and MSC marker expression was carried out as described previously [54], with minor modifications. Briefly, single-cell suspensions were generated using StemPro Accutase. Cells were then incubated with PE-conjugated anti-human SSEA-4 antibody (1:100) in PBS containing 0.5% bovine serum albumin at 37 °C for 30 min. To exclude non-viable cells, propidium iodide staining was applied during analysis. Appropriate isotype controls were included in all experiments (1:100). MSC identity was assessed by flow cytometry using antibodies against CD73, CD44, CD45, CD31, CD34, and CD90. Appropriate isotype controls were included in all experiments.
Immunocytochemistry
For immunocytochemical analysis, cells were seeded onto eight-well Nunc Lab-Tek II chambered cover glass slides. Cells were fixed and permeabilized using 4% paraformaldehyde (PFA) in DPBS for 15 min at room temperature. Following fixation, samples were blocked for 60 min at room temperature in a blocking buffer consisting of DPBS supplemented with 2 mg/mL BSA, 0.1% Triton X-100, and 1% fish gelatin, with or without 5% goat serum. Cells were subsequently incubated for 60 min at room temperature with primary antibodies diluted in blocking buffer. Pluripotency markers included OCT3/4 (1:50) and NANOG (1:100), while lineage-specific markers included AFP (α-fetoprotein) (1:500), SMA (α-smooth muscle actin) (1:500), and β-III tubulin (1:2000). Neuronal progenitor markers included Sox1, (1:100) Sox2 (1:20), Pax6 (1:100), and Nestin (1:250). After washing with DPBS, cells were incubated for 60 min at room temperature with Alexa Fluor 647-conjugated goat anti-mouse and Alexa Fluor 594-conjugated donkey anti-goat secondary antibodies (1:250). Nuclear staining was performed using DAPI. Samples were imaged using a Zeiss LSM 710 confocal laser scanning microscope.
RNA isolation and gene expression analysis
Total RNA was isolated from KUT-2 and KUT-6 cells and differentiated derivatives using TRIzol reagent. RNA integrity was assessed by agarose gel electrophoresis, while concentration and purity were determined using a Nanodrop spectrophotometer. For mRNA expression analysis, cDNA was synthesized from 400 ng total RNA using the Promega Reverse Transcription system according to the manufacturer’s protocol. Expression levels of SEV, BRY, OCT4, GATA4, NANOG, AFP, and Nestin were quantified using TaqMan® gene expression assays. Quantitative PCR reactions were performed on a StepOne™ Real-Time PCR System (Applied Biosystems) following the manufacturer’s recommendations. Gene expression levels were normalized to endogenous control RPLP0.
Cytotoxicity measurements
After trypsinization, both cancer cell lines and the MSCs were seeded on 384-well microplates in a density of 2500 cells/well in 40 μL complete medium. Prior to seeding hiPSC and NSC lines, 20 μL of cold Matrigel was dispensed into pre-chilled 384-well plates; for hNPC cultures, cold Matrigel was used instead. Plates were incubated at 37 °C for 1 h, after which excess coating solution was removed manually using a vacuum hand operator (Vacuboy, Integra Biosciences), and 2500 cells/well in 40 μL cell seeding was performed immediately, as described for the other lines. The following day, when cells were attached, serial dilution of the test drugs were aspirated in an additional 20 μL complete medium, and plates were incubated for 72 h.
Cytotoxicity was assessed by adding resazurin-based PrestoBlue viability reagent to a final concentration of 10%, followed by 1 h incubation, after which fluorescence of the reduced dye (resorufin) was measured at 555/585 nm ex./em. wavelengths with an EnSpire (Perkin Elmer) plate reader. Cytotoxicity testing-related liquid aspirations were performed by a Hamilton StarLet robot.
Raw data files were uploaded into our custom program, which performed automated data normalization, sigmoidal curve fitting, and pIC50 calculation [57]. The mean and standard deviation of pIC50 values were determined from at least three independent experiments. IC50 values were then obtained by logarithmic transformation, with corresponding ± SD reported, noting that symmetric SD values in pIC50 become asymmetric after transformation.
Statistical analysis
Statistical analyses were performed using GraphPad Prism. Differences in cytotoxicity among cell types and compounds were evaluated using ordinary two-way ANOVA with Tukey’s multiple comparisons post hoc test. The significance threshold (α) was set at 0.05. Cell identity was used as the row factor, while compound identity was used as the column factor. Type III sum-of-squares analysis was applied to determine the contribution of each factor and their interaction to the total variance (Supplementary Tables 3A,B). Adjusted p values from Tukey’s multiple comparisons test were used to compare estimated marginal means between cell types. Pairwise comparisons were performed among all cell derivatives across the KUT-2 and KUT-6 backgrounds to identify statistically significant differences in cytotoxicity. Tukey’s correction was applied to reduce the probability of type I error arising from multiple comparisons, and adjusted p values < 0.05 were considered statistically significant.
Results
Our aim was to determine whether pluripotent and lineage-committed derivatives of human umbilical cord blood-derived induced pluripotent stem cells (hiPSCs) exhibit differential sensitivity to cytotoxic compounds. We hypothesized that pluripotent hiPSCs, due to their rapid proliferation rate and increased susceptibility to DNA damage, would display higher sensitivity, whereas differentiated mesenchymal stem cells (MSCs) and neural progenitor cells (NPCs) would show distinct, lineage-specific stress responses resulting in different patterns of drug sensitivity. Two independent hiPSC lines derived from different donors with sex differences were used in this study, allowing limited assessment of donor-dependent variability in drug responses at the pluripotent stage. Since our goal here was to provide detailed examples for NAMs, potentially replacing some of the animal studies, we had to comply with the regulatory requirements (see ref 6). These include defined test methodologies; establishing the relevance within the context of use; and demonstration of reliability and robustness. Thus, a detailed characterization of the applied iPS cells and their differentiated derivatives, as well as their response to drug treatment had to be performed.
Generation and validation of hiPSC lines derived from umbilical cord blood
Two distinct hiPSC lines (KUT-2- female and KUT-6 male) were generated from umbilical cord blood–derived mononuclear cells using Sendai virus–mediated reprogramming. Sendai virus reprogramming is widespread because the reprogramming factors (OCT4, SOX2, cMYC and KLF4) they are expressed only transiently and in the cytoplasm, do not integrate into the nucleus, and are diluted with cell division. Thus, reprogramming occurs without a genetic “trace”. The clearance of the virus should be checked after 10 passages. Clonal lines exhibiting normal diploid karyotypes were selected for further analysis, confirming genomic integrity (Figure 1A).
FIGURE 1
In order for hiPSC-based models to produce reproducible results, the starting hiPSCs must be thoroughly characterized according to the strict standards of the field. The most commonly used characterization methods include assessment of morphology, immunostaining for surface (SSEA4) and nuclear pluripotency markers (OCT4, NANOG) and verification of the pluripotent state by spontaneous differentiation. During spontaneous differentiation induced by serum-containing three dimensional culture conditions, pluripotent cells give rise to cell types representing all three germ layers. Ectodermal, mesodermal and endodermal cell populations appearing after 12 days of differentiation must be detected using at least two independent methods (e.g., immunostaining and RT-PCR) and the loss of pluripotency markers must also be confirmed.
Both hiPSC lines displayed typical human pluripotent stem cell morphology, characterized by compact colonies with defined borders and a high nuclear-to-cytoplasmic ratio. Pluripotency was further confirmed by immunofluorescence staining, demonstrating robust expression of the core transcription factors OCT4 and NANOG (Figure 1C).
Flow cytometric analysis revealed high surface expression of SSEA4 (approximately 96%–98%), indicating a homogeneous pluripotent population (Figure 1B).
The differentiation capacity of the hiPSC lines was assessed through spontaneous embryoid body (EB) formation. Following differentiation, cells lost pluripotency markers (OCT4 and NANOG see Supplementary Figure 1) and expressed lineage-specific markers representing all three germ layers, including AFP (endoderm), SMA (mesoderm), and β-III tubulin (ectoderm), as confirmed by immunofluorescence staining (Figure 1D). The differentiation capacity to the three germ layers was also validated by qPCR (Supplementary Figure 1).
Quality control and genomic stability of hiPSC lines
Comprehensive quality control analyses confirmed the stability and suitability of the generated hiPSC lines for downstream applications. Clearance of the Sendai viral vectors was verified by RT-PCR, demonstrating the absence of residual reprogramming vectors after hiPSC establishment (Supplementary Figure 1A). Karyotype analysis confirmed normal chromosomal composition, while STR profiling verified cell line identity. In addition, all samples tested negative for mycoplasma contamination (Supplementary Figure 1B).
Validation of neural and mesenchymal lineage differentiation
To evaluate lineage-specific differentiation potential, hiPSC lines were directed toward both neural and mesenchymal fates. Neural stem cells (NSCs) and hippocampal neural progenitor cells (hNPCs) were successfully generated from hiPSCs derived from both donors using two independent differentiation protocols (Figures 2A,B). Neural differentiation was confirmed by immunofluorescence staining for canonical neural progenitor markers. Cells derived from both hiPSC lines exhibited robust expression of Nestin and SOX2, indicating successful acquisition of neural progenitor identity (Figures 2A,B). The observed staining patterns and cellular morphology were consistent with neural stem/progenitor-like phenotypes, including elongated cellular structures and dense, progenitor-like populations. Both KUT-2 and KUT-6 lines demonstrated comparable marker expression profiles, suggesting reproducible and efficient neural induction across independent genetic backgrounds.
FIGURE 2
To validate the quality of the NSCs, we have further differentiated the resulting NSCs into neurons (Supplementary Figures 2A,C). We have confirmed that the neuronal cultures are functional, capable of both spontaneous electrical activity, which was validated by multielectrode array measurements (Supplementary Figures 2B,D,E,F), and ligand-induced responses, measured by calcium imaging (Supplementary Figures 2G,H). The resulting neurons are glutamatergic, therefore after the addition of 50 μM glutamate and in both cell lines we obtained a Ca-signal and most of the cells responded to the addition of glutamate.
These findings confirm that both hiPSC lines can generate functionally active neuronal populations, supporting their suitability for downstream applications involving neurobiological and pharmacological studies.
In parallel, mesenchymal differentiation capacity was assessed by directing hiPSCs toward an MSC-like lineage. Mesenchymal differentiation resulted in adherent, fibroblast-like cells displaying characteristic spindle-shaped morphology and plastic adherence, consistent with mesenchymal lineage commitment. Flow cytometric analysis demonstrated expression of canonical mesenchymal surface markers, including CD44 and CD90, with partial expression of CD73 (Figure 2C; Supplementary Figures 4A–C) and absence or low expression of hematopoietic and endothelial marker (CD45, CD34 and CD31) (Supplementary Figures 3A–C). Differentiation assays further supported mesenchymal identity: KUT-2-derived MSCs showed successful adipogenic and osteogenic differentiation (Supplementary Figures 3D–G). In contrast, KUT-6-derived cells showed limited differentiation capacity under the applied conditions, suggesting donor-dependent variability and incomplete mesenchymal maturation (Supplementary Figures 4D–G). Furthermore, they quickly entered a senescent state, so we could not perform cytotoxicity measurements on them. Together, these results establish a comparative platform for investigating lineage-dependent differences in cellular stress responses and drug sensitivity.
Cytotoxicity patterns of KUT-2- and KUT-6-derived derivatives against DNA-damaging chemotherapeutics
We evaluated the cytotoxicity effects of a panel of DNA-damaging anticancer agents across KUT-2 and KUT-6 hiPSCs as well as in their differentiated derivatives, including MSCs, NSCs, and hNPCs (Figure 3). Two-way ANOVA followed by Tukey’s post hoc test using the respective log IC50 values revealed that both KUT-2 and KUT-6 originated hiPSCs were the most sensitive, exhibiting the lowest IC50 values. hNPC and NSC derivatives displayed slightly lower sensitivity, with small but statistically detectable differences between them. MSC was consistently the most resistant derivative with a 1-2 orders of magnitude higher IC50 values compared to the other lines. These patterns were largely preserved across all tested compounds, suggesting that sensitivity is primarily dictated by cell type rather than compound-specific effects (Supplementary Table 3A). Differences in IC50 values between non-isogenic KUT-2- and KUT-6-derived cell types were minimal for the tested oncology drugs, further reinforcing that cell type is the main determinant of sensitivity among our cell lines.
FIGURE 3
When these four compounds were tested against the cancer cell lines OVC-8 and Mes-Sa, the IC50 values were similar to those measured in the MSC cell line: the MSC line was approximately only 2–3-fold more sensitive to doxorubicin, mitoxantrone, and laromustine, while it was several-fold more resistant to SN-38, even compared with the cancer cell lines. As a secondary observation, laromustine generally exhibited lower cytotoxicity against all cell lines, while doxorubicin, mitoxantrone and SN-38 were broadly comparable.
Cytotoxicity patterns of KUT-2- and KUT-6-derived derivatives against non-genotoxic medicines
Cytotoxicity of four additional clinically used non-genotoxic drugs being repurposed for oncology applications in preclinical research (quinacrine, deferiprone, phenothiazine, and celecoxib), was also evaluated (Figure 3). The statistical analysis demonstrated that the primary determinant of drug sensitivity is the biological (differentiation) state of the cells, as most of the variation in IC50 values is explained by cell type rather than compound identity (Supplementary Table 3B). Post hoc comparisons confirmed this clear cell-type–dependent pattern. MSC, hiPSC, and NSC cells exhibited comparable IC50 values, with no significant differences among them, indicating similar levels of drug sensitivity within this group. In contrast, hNPC cells showed consistently and significantly lower IC50 values compared to all other cell types, reflecting a markedly higher sensitivity. This pattern was observed across both KUT-2 and KUT-6 origins, only minor numerical shifts were observed, most notably in hNPC cells (∼0.04 log units, that is roughly a 10% difference), which contributed slightly to overall variability but did not alter the general pattern.
When compared to the cancer cell lines Mes-Sa and OVC-8, the IC50 values were generally similar to those observed in MSC, hiPSC, and NSC populations, indicating comparable levels of drug sensitivity. Notably, OVC-8 cells exhibited the highest resistance, most prominently in response to phenothiazine, while responses to other compounds remained within the general range observed across cell types. Overall, these findings suggest that the cytotoxic effects of these non-genotoxic drugs are not cancer cell specific.
Compound identity had a secondary but detectable effect on cytotoxicity. Across all cell types, quinacrine exhibited the highest cytotoxicity, followed by celecoxib and deferiprone, while phenothiazine showed the lowest overall activity.
Discussion
Several data-sets are available on the in vitro cytotoxicity of the active compounds examined in the present study; these reports have examined the effects of the substances using a wide range of cell lines, different methods, and exposure conditions, primarily in tumor cell lines (Supplementary Table 4). The advantage of the model system we have established is that it allows for the examination of different cell types with the same genetic background, thereby creating a common platform where the effects of individual substances can be determined relative to one another and which also provides the opportunity for combination of drug testing.
Response to genotoxic stress: cell type is a more significant factor than biological sex or genetic background
Cellular drug sensitivity depends heavily on cell type and tissue of origin, driven by lineage-specific variations in metabolic state, cell cycle dynamics, DNA repair, and stress responses [58, 59]. Consequently, a single normal cell type cannot serve as a uniform control for lineage-specific cytotoxicity or predict clinical tissue toxicities, such as doxorubicin cardiotoxicity, cisplatin nephrotoxicity, or taxane neurotoxicity [60–62], whose underlying cellular mechanisms remain poorly defined.
To address this in a human-relevant NAM framework, we evaluated isogenic hiPSC, MSC, NSC, and hNPC lines from two donors (one female, one male) across compounds triggering diverse cytotoxic mechanisms. Distinct sensitivity patterns emerged across cell types, primarily driven by developmental state, proliferative activity, and lineage identity rather than donor sex, which yielded no pronounced differences. These findings align with prior work demonstrating differential toxicity across isogenic neural lineages []. Moreover, the demonstrated sensitivity of NPCs to cytotoxic agents could serve as a model of chemotherapy-related cognitive impairment („chemobrain”) mechanisms in future studies.
DNA-damaging agents and iPSC hypersensitivity: a mechanism-specific pattern reflecting the genomic integrity strategy of pluripotent cells
One of the most consistent and mechanistically informative findings of the present study was the markedly elevated sensitivity of hiPSCs to DNA-damaging agents. Across all four compounds in this category—doxorubicin, mitoxantrone, SN-38, and laromustine—hiPSCs consistently showed the lowest IC50 values, while MSCs were the most resistant. Critically, this pattern was reproducible across both donor lines and was not compound-specific, but rather reflected a broader response to genotoxic stress.
This observation is consistent with a well-established body of literature demonstrating that human pluripotent stem cells (hPSCs) exhibit markedly higher sensitivity to DNA-damaging agents than their differentiated derivatives. Multiple studies have shown that hPSCs respond to genotoxic stress with rapid apoptosis, whereas differentiated progeny display greater resistance and preferentially activate cell cycle arrest and DNA repair pathways. This difference arises from the fundamentally distinct DNA damage response of pluripotent cells compared to differentiated cells.
Liu et al. demonstrated that human embryonic stem cells undergo significantly more rapid p53-dependent apoptosis following DNA damage than isogenic differentiated cells. This difference is not due to altered p53 function itself, but rather to the fact that pluripotent cells exist in a state of higher mitochondrial priming—meaning that even minor damage is sufficient to trigger cell death. The underlying basis is a shifted balance of pro- and anti-apoptotic proteins of the Bcl-2 family, which places these cells closer to the apoptotic threshold [63]. In a complementary study, Luo et al. showed that even at low doses of UV radiation, human pluripotent stem cells preferentially undergo apoptosis, while differentiated cells fail to mount a comparable apoptotic response, despite the pluripotent cells actually sustaining less UV-induced DNA damage than their differentiated counterparts [64]. Together, these findings indicate that the hypersensitivity of pluripotent cells to genotoxic stress is not simply a consequence of impaired DNA repair, but rather reflects a fundamentally different cellular decision-making strategy in response to genomic damage [65].
The differentiation-dependent loss of this hypersensitive apoptotic priming provides a mechanistic explanation for the gradient of drug sensitivity observed in our study across the hiPSC → NPC/MSC differentiation axis: as cells transition from an apoptosis-prone pluripotent phenotype toward a repair-tolerant differentiated state, a consistent biological gradient emerges that was well reflected in the IC50 patterns obtained with all four DNA-damaging agents. Moreover, this response was consistent across four distinct genotoxic classes—spanning topoisomerase I/II inhibitors, intercalators, and alkylating agents—these differences represent a fundamental, differentiation-dependent lowering of the apoptotic threshold rather than compound-specific toxicity.
MSC resistance to cytotoxic agents: a consistent but mechanistically complex pattern
Among the cell types examined in the present study, MSCs derived from the KUT-2 donor line consistently displayed the highest resistance to cytotoxic compounds, with IC50 values one to two orders of magnitude higher than those observed in hiPSCs and NPC lineages. Notably, this resistance extended to comparisons with cancer cell lines: when the four DNA-damaging agents were tested against the OVC-8 and Mes-Sa tumor lines, the MSC line was only slightly more sensitive to doxorubicin, mitoxantrone, and laromustine, while it was even more resistant to SN-38. This pronounced tolerance aligns with previous studies showing naive MSCs frequently exhibit drug resistance comparable to or higher than fibroblasts [66, 67] and cancer cells [68]. Mechanistically, MSC resistance is non-universal and heavily context-dependent, governed by tissue source, compound mechanism of action [67, 68], and complex stress-response networks [69, 70]. For instance, Nicolay et al. demonstrated that bone marrow-derived MSCs resist cisplatin via high constitutive heat shock protein expression and suppressed p73/Bax/p21 pro-apoptotic signaling [66], yet remain uniquely vulnerable to bleomycin [67]. Taken together, our findings demonstrate that MSC resistance stems from overlapping survival strategies, including slow proliferation and attenuated apoptotic priming, and underscore that selecting an appropriate “normal cell” comparator is critical when evaluating therapeutic windows in drug discovery platforms.
Lineage-specific vulnerability of neural progenitors
Our finding that hiPSC-derived neural progenitor cells (NPCs) exhibit greater sensitivity to non-genotoxic, metabolically active compounds than parental hiPSCs aligns with NPC dependency on lysosomal homeostasis and autophagic flux [71]. Compounds disrupting these pathways—such as quinacrine (lysosomal pH/autophagy inhibition) and deferiprone (iron homeostasis/mitochondrial stress)—uncovered a lineage-specific vulnerability that warrants further mechanistic study. Crucially, as NPCs represent a temporal window vulnerable to developmental and adult neurotoxicity [72], hippocampal NPCs (hNPCs) serve as sensitive models beyond conventional cytotoxicity endpoints. While CNS-penetrant compounds like phenothiazine directly exploit this vulnerability [73], our findings imply that non-permeable DNA-damaging agents could similarly trigger severe neural damage if the blood-brain barrier is compromised.
A conceptually similar protective role is fulfilled by the placenta during embryonic development. Like the BBB, the placental barrier limits fetal exposure to xenobiotics while maintaining nutrient and oxygen exchange [74]. However, placental transfer protection is incomplete, and numerous small molecules—particularly lipophilic, weakly basic, or transporter-favored compounds—can cross the placenta and reach the developing embryo [75]. This is particularly important because pluripotent cells of the early embryo are characterized by extremely high sensitivity to genotoxic stress, as demonstrated in our hiPSC experiments. Their rapid apoptotic response to DNA damage reflects the well-established “genomic guardian” strategy of pluripotent cells, in which preservation of genomic integrity is prioritized over survival. Consequently, hiPSCs provide a highly relevant in vitro model for embryotoxicity and teratogenicity assessment, particularly for compounds capable of crossing the placental barrier and interfering with early developmental processes.
The combination of hiPSC and hNPC models therefore provides complementary toxicological information across two critical developmental windows: early embryogenesis and early neurodevelopment. While hiPSCs serve as highly sensitive indicators of embryotoxic and genotoxic risk, hNPCs are particularly informative for detecting potential developmental neurotoxicity. This dual sensitivity strengthens the translational relevance of our model system and suggests that cytotoxicity testing based exclusively on rapidly proliferating tumor models may underestimate developmental and lineage-specific toxicity, whereas hiPSC-derived developmental models may provide a more physiologically relevant and predictive approach for human toxicity assessment.
Limitations of the study
The core cytotoxic patterns observed — hiPSC hypersensitivity to DNA-damaging agents, hNPC sensitivity to metabolic stressors, and MSC resistance — were consistent across both donor lines, supporting the robustness of the main findings; nevertheless, replication in additional donor lines would be required to fully exclude donor-specific effects.
The experimental readout was based exclusively on cell viability and IC50 determination. While this approach allows for systematic quantitative comparison across cell types and compounds, it does not resolve the mechanistic fate of drug-exposed cells. Identical IC50 values may reflect fundamentally different cellular outcomes — including apoptosis, necrosis, quiescence, or senescence — which cannot be distinguished by viability assays alone. This is particularly relevant when interpreting the MSC data: as discussed above, the high IC50 values in MSCs may reflect a shift toward drug-induced quiescence rather than true resistance in the conventional pharmacological sense [69]. Complementary mechanistic endpoints — including caspase activation, cell cycle analysis, senescence markers, ROS quantification, and lysosomal or mitochondrial function assays — would be needed to fully characterize the cell fate consequences of drug exposure in each lineage.
The cell type panel was intentionally restricted to hiPSCs, MSCs, and NPC lineages, representing two of three embryonic germ layers. The inclusion of a third germ layer derivative — most relevantly hiPSC-derived cardiomyocytes — would have further extended the comparative framework. At the same time, cardiotoxicity assessment in hiPSC-derived cardiomyocytes is a well-established and extensively studied area [76, 77], that certain compound-specific toxicities — in particular, the cardiotoxic potential of doxorubicin — remain unaddressed in the present dataset.
We still suggest that the presented screening methods are superior in several aspects as compared to studies performed in fibroblast-cancer cell duets. This is especially true regarding the less differentiated, stem cell-like cells, potentially present in inhomogeneous tissue samples with greatly different drug sensitivities. In vivo animal studies could reinforce our conclusions, when undifferentiated, stem cell like cells can be separately examined.
Taken together, while these limitations should be considered in interpreting the findings, the main cytotoxic patterns identified in this study — genotoxic hypersensitivity in hiPSCs, metabolic vulnerability in hippocampal NPCs, and broad cytotoxic resistance in MSCs — were internally consistent, reproducible across both donor lines, and mechanistically coherent with the existing literature. These features support the biological relevance of the model and the robustness of the principal conclusions.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by According to the permission for reprogramming and the study by the Human Reproduction Committee of the Hungarian Health Science Council (ETT HRB- (Approval number: 3883-2/2018/EÜIG)), umbilical cord blood samples were obtained after written informed consent. The iPSC lines used in this study were obtained from anonymized donor. The investigation conformed to the principles outlined in the Declaration of Helsinki. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.
Author contributions
NV and KV: visualization, conceptualization, methodology, investigation, writing – original draft; ST and AK, Formal Analysis, Methodology, Investigation; ZE and JR: methodology, resources and Review and Editing; AÁ and BS: conceptualization, funding acquisition, methodology, visualization, supervision, and writing Original Draft and Review and Editing. All authors contributed to the article and approved the submitted version.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the National Research, Development and Innovation Office (Hungarian Scientific Research Fund (OTKA) K-128369 to ÁA, 2024-1.1.1-KKV_FÓKUSZ-2024-00022.to ZE and GINOP-2.1.1-15-2015-00369 to AÁ); the National Research, Development and Innovation Office had no further role in study design; in the col-lection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Acknowledgments
The authors would like to thank Bea Haraszti and Anna Szigeti for their excellent technical assistance and György Várady for his help with cell sorting.
Conflict of interest
KV, RB, and ZE are full-time, and SB is a part time employee of Salus Ltd. and Innocell Ltd. respectively. There is a research partnership between Salus Ltd., Innocell Ltd. and the Institute for Molecular Life Sciences to explore the potential application of hiPSC models in toxicological studies. Salus Ltd. and Innocell Ltd. were not involved in the actual planning, evaluation or interpretation of the results thus could not inappropriately influence or bias this work.
The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.por-journal.com/articles/10.3389/pore.2026.1612490/full#supplementary-material
Glossary
- 3R
replace, reduce, and refine animal models
- ADR
Adverse drug reactions
- AFP
α-fetoprotein
- Akt
Protein kinase B
- ATCC
American Type Culture Collection
- BBB
Blood-brain-barrier
- BSA
Body surface area
- CNS
Central nervous system
- EB
Embryoid body
- EMA
European Medicines Agency
- ERK1/2
extracellular signal-regulated kinase
- FDA
Food and Drug Administration
- hiPSC
Human induced pluripotent stem cell
- hNPC
Hippocampal neural progenitor cell
- hPSC
Human pluripotent stem cell
- IMRWG3R
International Medicines Regulators’ Working Group on 3Rs
- MAPK
Mitogen-activated protein kinase
- MSC
Mesenchymal stem cell-like cell
- mTOR
mechanistic Target of Rapamycin
- MTT
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide
- NAM
New approach methodologies
- NF-κB
Nuclear factor-κB
- NIH NCI DTP
National Cancer Institute Developmental Therapeutics Program of the National Institute of Health
- NSC
Neural stem cell
- Pax6
paired box 6
- PDK1
phosphoinositide-dependent kinase 1
- PGE2
Prostaglandin E2
- SI
Selectivity index
- SMA
α-smooth muscle actin
- Sox1
Sry-box transcription factor 1
- Sox2
Sry-box transcription factor 2
- SSEA4
Stage-specific embryonic antigen-4
- STR
Short tandem repeat
- TUB
β-III tubulin
References
1.
YuJ. Intestinal stem cell injury and protection during cancer therapy. Translational Cancer Research (2013) 2(5):384–96.
2.
PoorvuPDFrazierALFeracoAMManleyPEGinsburgESLauferMRet alCancer treatment-related infertility: a critical review of the evidence. JNCI Cancer Spectrum (2019) 3(1):pkz008. 10.1093/jncics/pkz008
3.
MaxwellMBMaherKE. Chemotherapy-induced myelosuppression. Semin Oncology Nursing (1992) 8(2):113–23. 10.1016/0749-2081(92)90027-z
4.
PeiYPengJBehlMSipesNSShockleyKRRaoMSet alComparative neurotoxicity screening in human iPSC-derived neural stem cells, neurons and astrocytes. Brain Research (2016) 1638(Pt A):57–73. 10.1016/j.brainres.2015.07.048
5.
HayMThomasDWCraigheadJLEconomidesCRosenthalJ. Clinical development success rates for investigational drugs. Nat Biotechnology (2014) 32(1):40–51. 10.1038/nbt.2786
6.
HanJJ. FDA modernization act 2.0 allows for alternatives to animal testing. Artif Organs (2023) 47(3):449–50. 10.1111/aor.14503
7.
TaylorPColmanLBajoonJ. The search for plants with anticancer activity: pitfalls at the early stages. J Ethnopharmacology (2014) 158(Pt A):246–54. 10.1016/j.jep.2014.10.034
8.
JaegerSDuran-FrigolaMAloyP. Drug sensitivity in cancer cell lines is not tissue-specific. Mol Cancer (2015) 14:40. 10.1186/s12943-015-0312-6
9.
NogueiraDRMitjansMInfanteMRVinardellMP. Comparative sensitivity of tumor and non-tumor cell lines as a reliable approach for in vitro cytotoxicity screening of lysine-based surfactants with potential pharmaceutical applications. Int Journal Pharmaceutics (2011) 420(1):51–8. 10.1016/j.ijpharm.2011.08.020
10.
ElautGHenkensTPapeleuPSnykersSVinkenMVanhaeckeTet alMolecular mechanisms underlying the dedifferentiation process of isolated hepatocytes and their cultures. Curr Drug Metabolism (2006) 7(6):629–60. 10.2174/138920006778017759
11.
HeslopJARoweCWalshJSison-YoungRJenkinsRKamalianLet alMechanistic evaluation of primary human hepatocyte culture using global proteomic analysis reveals a selective dedifferentiation profile. Arch Toxicology (2017) 91(1):439–52. 10.1007/s00204-016-1694-y
12.
WagnerWHornPCastoldiMDiehlmannABorkSSaffrichRet alReplicative senescence of mesenchymal stem cells: a continuous and organized process. PloS One (2008) 3(5):e2213. 10.1371/journal.pone.0002213
13.
GilletJPVarmaSGottesmanMM. The clinical relevance of cancer cell lines. J Natl Cancer Inst (2013) 105(7):452–8. 10.1093/jnci/djt007
14.
AllenDDCaviedesRCardenasAMShimaharaTSegura-AguilarJCaviedesPA. Cell lines as in vitro models for drug screening and toxicity studies. Drug Development Industrial Pharmacy (2005) 31(8):757–68. 10.1080/03639040500216246
15.
EbertADLiangPWuJC. Induced pluripotent stem cells as a disease modeling and drug screening platform. J Cardiovascular Pharmacology (2012) 60(4):408–16. 10.1097/FJC.0b013e318247f642
16.
RaabSKlingensteinMLiebauSLintaL. A comparative view on human somatic cell sources for iPSC generation. Stem Cells International (2014) 2014:768391. 10.1155/2014/768391
17.
YoshidaYYamanakaS. Recent stem cell advances: induced pluripotent stem cells for disease modeling and stem cell-based regeneration. Circulation (2010) 122(1):80–7. 10.1161/CIRCULATIONAHA.109.881433
18.
MarionRMStratiKLiHMurgaMBlancoROrtegaSet alA p53-mediated DNA damage response limits reprogramming to ensure iPS cell genomic integrity. Nature (2009) 460(7259):1149–53. 10.1038/nature08287
19.
ChenACHPengQFongSWLeeKCYeungWSBLeeYL. DNA damage response and cell cycle regulation in pluripotent stem cells. Genes (2021) 12(10):1548. 10.3390/genes12101548
20.
Abu-DawudRGraffmannNFerberSWruckWAdjayeJ. Pluripotent stem cells: induction and self-renewal. Philosophical Transactions R Soc Lond Ser B, Biol Sciences (2018) 373(1750):20170213. 10.1098/rstb.2017.0213
21.
KamataSHashiyamaRHana-IkaHOhkuboISaitoRHondaAet alCytotoxicity comparison of 35 developmental neurotoxicants in human induced pluripotent stem cells (iPSC), iPSC-derived neural progenitor cells, and transformed cell lines. Toxicol Vitro : An International Journal Published Association BIBRA (2020) 69:104999. 10.1016/j.tiv.2020.104999
22.
IngramDAMeadLETanakaHMeadeVFenoglioAMortellKet alIdentification of a novel hierarchy of endothelial progenitor cells using human peripheral and umbilical cord blood. Blood (2004) 104(9):2752–60. 10.1182/blood-2004-04-1396
23.
KohHZhenXKimJHaHYLeeJH. Generation and characterization of human umbilical cord blood-derived induced pluripotent stem cells (KRIBBi005-A). Stem Cell Research (2022) 60:102674. 10.1016/j.scr.2022.102674
24.
GaoXYourickJJSprandoRL. Comparative transcriptomic analysis of endothelial progenitor cells derived from umbilical cord blood and adult peripheral blood: implications for the generation of induced pluripotent stem cells. Stem Cell Research (2017) 25:202–12. 10.1016/j.scr.2017.11.004
25.
BadowskiMSHarrisDT. Collection, processing, and banking of umbilical cord blood stem cells for transplantation and regenerative medicine. Methods Molecular Biology (2012) 879:279–90. 10.1007/978-1-61779-815-3_16
26.
IssaJP. Aging and epigenetic drift: a vicious cycle. The J Clinical Investigation (2014) 124(1):24–9. 10.1172/JCI69735
27.
LoSVFergusonWEriksonGATopolEJBaldwinKKTorkamaniA. Influence of donor age on induced pluripotent stem cells. Nat Biotechnology (2017) 35(1):69–74. 10.1038/nbt.3749
28.
KikutaKLeeEMenezesTFungHAmorinAAgrawalAet alImmune composition of the mononuclear cell fraction of human umbilical cord blood. Front Immunology (2025) 16:1614230. 10.3389/fimmu.2025.1614230
29.
BlakeMJCastroLLeederJSKearnsGL. Ontogeny of drug metabolizing enzymes in the neonate. Semin Fetal and Neonatal Medicine (2005) 10(2):123–38. 10.1016/j.siny.2004.11.001
30.
GandhiMAweekaFGreenblattRMBlaschkeTF. Sex differences in pharmacokinetics and pharmacodynamics. Annu Review Pharmacology Toxicology (2004) 44:499–523. 10.1146/annurev.pharmtox.44.101802.121453
31.
TaylorKEVallejo-GiraldoCSchaibleNSZakeriRMillerVM. Reporting of sex as a variable in cardiovascular studies using cultured cells. Biol Sex Differences (2011) 2:11. 10.1186/2042-6410-2-11
32.
BeeryAKZuckerI. Sex bias in neuroscience and biomedical research. Neurosci Biobehavioral Reviews (2011) 35(3):565–72. 10.1016/j.neubiorev.2010.07.002
33.
MennecozziMLandesmannBPalosaariTHarrisGWhelanM. Sex differences in liver toxicity-do female and Male human primary hepatocytes react differently to toxicants in vitro?PloS One (2015) 10(4):e0122786. 10.1371/journal.pone.0122786
34.
AikawaN. A novel screening test to predict the developmental toxicity of drugs using human induced pluripotent stem cells. The J Toxicological Sciences (2020) 45(4):187–99. 10.2131/jts.45.187
35.
GalanjukSZuhrEDonmezABartschDKurianLTiggesJet alThe human induced pluripotent stem cell test as an alternative method for embryotoxicity testing. Int Journal Molecular Sciences (2022) 23(6):3295. 10.3390/ijms23063295
36.
OrfaliRAlwatbanAZOrfaliRSLauLCheaNAlotaibiAMet alOxidative stress and ion channels in neurodegenerative diseases. Front Physiology (2024) 15:1320086. 10.3389/fphys.2024.1320086
37.
Maldonado-SotoAROakleyDHWichterleHSteinJDoetschFKHendersonCE. Stem cells in the nervous system. Am Journal Physical Medicine and Rehabilitation (2014) 93(11 Suppl. 3):S132–44. 10.1097/PHM.0000000000000111
38.
GageFH. Adult neurogenesis in the human dentate gyrus. Hippocampus (2025) 35(1):e23655. 10.1002/hipo.23655
39.
YuDXDi GiorgioFPYaoJMarchettoMCBrennandKWrightRet alModeling hippocampal neurogenesis using human pluripotent stem cells. Stem Cell Reports (2014) 2(3):295–310. 10.1016/j.stemcr.2014.01.009
40.
CiavarellaCValenteSPasquinelliG. The characteristics and survival potential under sub-lethal stress of mesenchymal stromal/stem cells isolated from the human vascular wall. Stem Cells (2022) 40(12):1071–7. 10.1093/stmcls/sxac066
41.
LiHDaiHLiJ. Immunomodulatory properties of mesenchymal stromal/stem cells: the link with metabolism. J Advanced Research (2023) 45:15–29. 10.1016/j.jare.2022.05.012
42.
SagaradzeGDBasalovaNAEfimenkoAYTkachukVA. Mesenchymal stromal cells as critical contributors to tissue regeneration. Front Cell Developmental Biology (2020) 8:576176. 10.3389/fcell.2020.576176
43.
DutourAPaselloMFarrowLAmerMHEntz-WerleNNathrathMet alMicroenvironment matters: insights from the FOSTER consortium on microenvironment-driven approaches to osteosarcoma therapy. Cancer Metastasis Reviews (2025) 44(2):44. 10.1007/s10555-025-10257-3
44.
GewirtzDA. A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacology (1999) 57(7):727–41. 10.1016/s0006-2952(98)00307-4
45.
MeredithAMDassCR. Increasing role of the cancer chemotherapeutic doxorubicin in cellular metabolism. The J Pharmacy Pharmacology (2016) 68(6):729–41. 10.1111/jphp.12539
46.
EvisonBJSleebsBEWatsonKGPhillipsDRCuttsSM. Mitoxantrone, more than just another topoisomerase II poison. Med Research Reviews (2016) 36(2):248–99. 10.1002/med.21364
47.
KumarSShermanMY. Resistance to TOP-1 inhibitors: good old drugs still can Surprise us. Int Journal Molecular Sciences (2023) 24(8):7233. 10.3390/ijms24087233
48.
PigneuxA. Laromustine, a sulfonyl hydrolyzing alkylating prodrug for cancer therapy. IDrugs : The Investigational Drugs Journal (2009) 12(1):39–53.
49.
FiorilloMTothFBrindisiMSotgiaFLisantiMP. Deferiprone (DFP) targets cancer stem cell (CSC) propagation by inhibiting mitochondrial metabolism and inducing ROS production. Cells (2020) 9(6):1529. 10.3390/cells9061529
50.
WenBWeiYTMuLLWenGRZhaoK. The molecular mechanisms of celecoxib in tumor development. Medicine (2020) 99(40):e22544. 10.1097/MD.0000000000022544
51.
ZouXXieB. Therapeutic mechanisms of phenothiazine drugs: a mini-review of advances in cancer treatment and antibiotic resistance. Iranian Journal Pharmaceutical Research : IJPR (2025) 24(1):e157923. 10.5812/ijpr-157923
52.
ErikssonAOsterroosAHassanSGullboJRickardsonLJarviusMet alDrug screen in patient cells suggests quinacrine to be repositioned for treatment of acute myeloid leukemia. Blood Cancer Journal (2015) 5(4):e307. 10.1038/bcj.2015.31
53.
ErikssonAChantziEFryknasMGullboJNygrenPGustafssonMet alTowards repositioning of quinacrine for treatment of acute myeloid leukemiaPromising synergies and in vivo effects. Leuk Research (2017) 63:41–6. 10.1016/j.leukres.2017.10.012
54.
ErdeiZLorinczRSzebenyiKPentekAVargaNLikoIet alExpression pattern of the human ABC transporters in pluripotent embryonic stem cells and in their derivatives. Cytometry part B. Clin Cytometry (2014) 86(5):299–310. 10.1002/cyto.b.21168
55.
HathyESzaboEVinczeKHaltrichIKissEVargaNet alGeneration of multiple iPSC clones from a male schizophrenia patient carrying de novo mutations in genes KHSRP, LRRC7, and KIR2DL1, and his parents. Stem Cell Research (2021) 51:102140. 10.1016/j.scr.2020.102140
56.
VargaNVerebZRajnavolgyiENemetKUherFSarkadiBet alMesenchymal stem cell like (MSCl) cells generated from human embryonic stem cells support pluripotent cell growth. Biochem Biophysical Research Communications (2011) 414(3):474–80. 10.1016/j.bbrc.2011.09.089
57.
WindtTTothSPatikISesslerJKucsmaNSzepesiAet alIdentification of anticancer OATP2B1 substrates by an in vitro triple-fluorescence-based cytotoxicity screen. Arch Toxicology (2019) 93(4):953–64. 10.1007/s00204-019-02417-6
58.
WoodKC. Drivers of intrinsic resistance. Nat Chemical Biology (2022) 18(6):579–81. 10.1038/s41589-022-00980-1
59.
YaoFMadani TonekaboniSASafikhaniZSmirnovPEl-HachemNFreemanMet alTissue specificity of in vitro drug sensitivity. J Am Med Inform Assoc : JAMIA (2018) 25(2):158–66. 10.1093/jamia/ocx062
60.
SheibaniMAziziYShayanMNezamoleslamiSEslamiFFarjooMHet alDoxorubicin-induced cardiotoxicity: an overview on pre-clinical therapeutic approaches. Cardiovasc Toxicology (2022) 22(4):292–310. 10.1007/s12012-022-09721-1
61.
VelascoRBrunaJ. Taxane-induced peripheral neurotoxicity. Toxics (2015) 3(2):152–69. 10.3390/toxics3020152
62.
YaoXPanichpisalKKurtzmanNNugentK. Cisplatin nephrotoxicity: a review. The Am Journal Medical Sciences (2007) 334(2):115–24. 10.1097/MAJ.0b013e31812dfe1e
63.
LiuJCGuanXRyanJARiveraAGMockCAgrawalVet alHigh mitochondrial priming sensitizes hESCs to DNA-damage-induced apoptosis. Cell Stem Cell (2013) 13(4):483–91. 10.1016/j.stem.2013.07.018
64.
LuoLZGopalakrishna-PillaiSNaySLParkSWBatesSEZengXet alDNA repair in human pluripotent stem cells is distinct from that in non-pluripotent human cells. PloS One (2012) 7(3):e30541. 10.1371/journal.pone.0030541
65.
LiuJCLerouPHLahavG. Stem cells: balancing resistance and sensitivity to DNA damage. Trends Cell Biology (2014) 24(5):268–74. 10.1016/j.tcb.2014.03.002
66.
NicolayNHLopez PerezRRuhleATrinhTSisombathSWeberKJet alMesenchymal stem cells maintain their defining stem cell characteristics after treatment with cisplatin. Scientific Reports (2016) 6:20035. 10.1038/srep20035
67.
NicolayNHRuhleAPerezRLTrinhTSisombathSWeberKJet alMesenchymal stem cells are sensitive to bleomycin treatment. Scientific Reports (2016) 6:26645. 10.1038/srep26645
68.
VajdaFSzepesiAVaradyGSesslerJKissDErdeiZet alComparison of different clinical chemotherapeutical agents' toxicity and cell response on mesenchymal stem cells and cancer cells. Cells (2022) 11(19):2942. 10.3390/cells11192942
69.
BoscoDBKenworthyRZorioDASangQX. Human mesenchymal stem cells are resistant to paclitaxel by adopting a non-proliferative fibroblastic state. PloS One (2015) 10(6):e0128511. 10.1371/journal.pone.0128511
70.
QiZZhangYLiuLGuoXQinJCuiG. Mesenchymal stem cells derived from different origins have unique sensitivities to different chemotherapeutic agents. Cell Biology International (2012) 36(9):857–62. 10.1042/CBI20110637
71.
LeemanDSHebestreitKRuetzTWebbAEMcKayAPollinaEAet alLysosome activation clears aggregates and enhances quiescent neural stem cell activation during aging. Science (2018) 359(6381):1277–83. 10.1126/science.aag3048
72.
BreierJMGassmannKKayserRStegemanHDe GrootDFritscheEet alNeural progenitor cells as models for high-throughput screens of developmental neurotoxicity: state of the science. Neurotoxicology and Teratology (2010) 32(1):4–15. 10.1016/j.ntt.2009.06.005
73.
OhlowMJMoosmannB. Phenothiazine: the seven lives of pharmacology's first lead structure. Drug Discovery Today (2011) 16(3-4):119–31. 10.1016/j.drudis.2011.01.001
74.
Al-EnazySAliSAlbekairiNEl-TawilMRyttingE. Placental control of drug delivery. Adv Drug Delivery Reviews (2017) 116:63–72. 10.1016/j.addr.2016.08.002
75.
LewisRMUmapathyAClealJKSengersBG. Physiological and anatomical determinants of placental drug transfer. The J Physiology (2026). 10.1113/JP289604
76.
BereczTYiuAVittayOOrsolitsBMioulaneMDos RemediosCGet alTranscriptional co-activators YAP1-TAZ of hippo signalling in doxorubicin-induced cardiomyopathy. ESC Heart Failure (2022) 9(1):224–35. 10.1002/ehf2.13756
77.
DohertyKRTalbertDRTruskPBMoranDMShellSABacusS. Structural and functional screening in human induced-pluripotent stem cell-derived cardiomyocytes accurately identifies cardiotoxicity of multiple drug types. Toxicol Applied Pharmacology (2015) 285(1):51–60. 10.1016/j.taap.2015.03.008
Summary
Keywords
cytotoxicity, induced pluripotent stem cells (iPSCs), isogenic in vitro model, lineage-specific toxicity, mesenchymal stem cells
Citation
Vincze K, Varga N, Tóth S, Kis AK, Béres R, Réthelyi JM, Veréb ZJ, Erdei Z, Sarkadi B and Apáti Á (2026) Beyond animal models: using human pluripotent stem cell-derived stem and progenitor cells to predict cytotoxicity in normal tissues. Pathol. Oncol. Res. 32:1612490. doi: 10.3389/pore.2026.1612490
Received
26 May 2026
Revised
03 August 2026
Accepted
11 August 2026
Published
20 August 2026
Volume
32 - 2026
Edited by
Szabó Györgyi, Semmelweis University, Hungary
Updates
Copyright
© 2026 Vincze, Varga, Tóth, Kis, Béres, Réthelyi, Veréb, Erdei, Sarkadi and Apáti.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Ágota Apáti, apati.agota@ttk.hu
† These authors have contributed equally to this work
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.