Automation of 3D Liver Spheroid generation and Acetaminophen dose– response on the MO:BOT enhances assay robustness and precision

Automation of 3D Liver Spheroid generation and Acetaminophen dose– response on the MO:BOT enhances assay robustness and precision | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Automation of 3D Liver Spheroid generation and Acetaminophen dose– response on the MO:BOT enhances assay robustness and precision Dana Hellmold, Daniel S. Ziemianowicz, Frowin Ellermann, Philipp Depperschmidt, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9116861/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Toxicity is a major cause of drug failure and post-marketing withdrawals, making sensitive and reproducible toxicity readouts essential for modern drug development. FDA Modernization Act 3.0 and related strategic roadmaps explicitly encourage the use of human-relevant in vitro models as New Approach Methodologies (NAMs) as an alternative to animal models. Yet, current preclinical drug discovery and toxicity testing with human in vitro models still depend on manual three-dimensional (3D) spheroid workflows that are challenging to standardize and scale. Although 3D human models better capture tissue-like architecture, cell-cell interactions, and organ-specific functions, 3D culture protocol standardization is precluded by user-dependent protocols, variable spheroid size and morphology, and fragmented steps for seeding, maintenance, and compound dosing. This lack of standardization weakens assay robustness, complicates cross-study comparison, and slows broader adoption in decision-making pipelines. In responses to these needs, we present the MO:BOT, a modular automation platform that unifies 3D organoid and spheroid seeding, medium exchange, image-based quality control, and compound dosing, within a single enclosed system. In this article, we present one use case application focused on an automated workflow to generate HepG2 liver spheroids and screening for acetaminophen (APAP) drug-response. In direct comparison with a manual workflow, the MO:BOT reduced the well-to-well variability in liver spheroid size, stabilized spheroid area over time, and yielded spheroids with higher viability. Automated medium exchange preserved spheroid architecture, demonstrating that appropriately tuned pipetting routines can maintain the integrity of delicate 3D structures. A MO:BOT automated APAP dose–response workflow produced a clear, sigmoidal hepatotoxicity profile with an EC 50 in the clinically relevant range and concordant changes in viability, LDH release, and ALT activity are closely matching the responses. Together, these findings show that the fully automated 3D liver spheroid workflow implemented on the MO:BOT improves culture uniformity and assay readout precision, providing a standardized, scalable foundation for deploying human 3D human models as robust, decision-enabling tools in modern preclinical testing. Biological sciences/Biological techniques Biological sciences/Biotechnology Biological sciences/Drug discovery Health sciences/Medical research Figures Figure 1 Figure 2 Figure 3 Figure 4 Key Phrases Modular benchtop 3D cell culture robot Integrated imaging and liquid handling Automated liver spheroid generation Standardized APAP-dose response testing Reduced spheroid variability Increased assay robustness Introduction Human cell-based models have increasingly transformed preclinical studies by more closely recapitulating human physiology than traditional animal models 1 , 2 . However, their impact is often constrained by variability and the difficulty of achieving high-throughput with sufficient standardization 3 , 4 . Conventional preclinical testing still relies heavily on animal studies, where interspecies differences can compromise the prediction of human toxicity 5 , 6 . Recent regulatory developments, including the FDA Modernization Act 3.0 and related strategic roadmaps, explicitly encourage the use of human-relevant in vitro methods, such as organoids and microphysiological systems, as New Approach Methodologies (NAMs) and alternatives to animal testing 7 – 9 . In this context, there is a growing demand for robust, scalable human cell-based systems that can deliver high-quality, decision-enabling data in early drug development 3 , 10 . Three-dimensional (3D) human in vitro models such as organoids and spheroids address several limitations of conventional two-dimensional (2D) monolayers by more faithfully reproducing tissue-like cell-cell contacts, nutrient gradients, and microenvironments 3 , 11 . For the liver, which plays a central role in xenobiotic metabolism and detoxification, 3D cell-based models maintain liver-specific functions over extended culture periods and more accurately capture clinically-relevant hepatotoxicity than 2D cultures 3 , 5 , 6 . Hepatocarcinoma-derived cell lines, such as HepG2 assembled into 3D spheroids, are widely used in early-stage screening owing to their robustness and compatibility with high-content and high-throughput platforms 3 , 12 – 14 . Spheroids generated from primary human hepatocytes and induced pluripotent stem cell (iPSC) -derived hepatocytes are implemented as a second-tier assay, providing higher physiological relevance for drug testing 5 , 6 . The value across these 3D model types relies on the same key requirement: achieving robust standardization with consistent spheroid size, architecture, and function 5 , 6 , 13 . In practice, current 3D liver spheroid workflows remain largely manual, which hampers the realization of this potential 3 , 4 . Routine operations such as cell seeding, spheroid formation, medium exchange, quality controls, and compound dosing require extensive hands-on time. In addition, the intrinsic variability between individual spheroids makes them more difficult to handle and analyze, which reduces throughput and weakens assay reliability, and safety testing workflows 3 – 5 . Existing automation approaches only partially mitigate these issue-throughput. Liquid handlers increase scalability and speed for downstream analysis, but do not address the upstream generation of homogenous 3D cell models. 3D Bioprinters support the creation of complex, ECM-embedded constructs but typically lack integrated systems for long-term maintenance. Bioreactors can scale 3D cell cultures but often generate heterogeneous aggregates with variable size and viability. High content screening workflows still rely on manual-to-end, reproducible 3D culture process 3 , 4 , 11 . Altogether, this underscores the absence of an integrated platform that can unify spheroid generation, long-term maintenance, perturbation, and readout within a single, standardize device. To address these needs, we developed the MO:BOT, a fully automated platform specifically designed for 3D cell culture workflows. MO:BOT automates spheroid and organoid generation and maintenance, controlled medium exchange, compound dosing, and on-deck imaging and assay readouts within a single enclosed system, thereby minimizing manual plate transfers between lab devices and user-dependent steps 4 . Within the integrated workflow, MO:BOT combines plug-and-play on-deck modules (heater, cooler, imaging module, shaker) to automate different types of 3D cell culture workflows and use cases. The MO:BOT software offers an intuitive, integrated protocol library covering multiple human 3D in vitro models and applications, while enabling straightforward implementation of new, customized workflows. In this work we demonstrate that the MO:BOT enables standardized and scalable production and characterization of liver spheroids suitable for toxicity testing by combining the biological advantages of 3D liver models with the process-level benefits of automation. This approach provides a path toward high-quality, robust in vitro liver assays aligned with modern, human-relevant drug development and with emerging regulatory expectations under initiatives such as the FDA Modernization Act 2.0 and 3.0 7–9 . Results The MO:BOT establishes an integrated automated workflow for 3D liver spheroid generation and hepatotoxicity testing To address the limitations of fragmented 3D cell culture workflows, we implemented the MO:BOT as an integrated automation platform that consolidates all key phases of 3D liver spheroid culture, from cell seeding to endpoint analysis (Fig. 1 ). A typical 3D cell culture experiment comprises of four phases: (1) cell seeding and maintenance of the culture over several days or weeks, (2) iterative quality control to monitor spheroid growth and morphology, (3) treatment with a compound such as acetaminophen (APAP), again followed by intermediate quality control, and finally (4) termination of the culture by performing endpoint assays such as for viability, cytotoxicity, and functionality (Supplementary Fig. 1) . These phases form the logical backbone of the liver toxicity workflow evaluated in this study, encompassing both biochemical readouts and morphological analysis by imaging. In the MO:BOT, all four phases were executed within a single workflow. This integration eliminated repeated off-deck handling steps, which, in conventional practice, introduce time variability, potential deviations in incubation conditions, an increased risk of contamination, and precludes systematic, and time-resolved quality control across experiments. As 3D spheroid assays are particularly sensitive to subtle changes in handling, timing, and environmental conditions, this consolidated workflow is expected to reduce variability in spheroid morphology and functional outputs 15 . Automated cell seeding with MO:BOT yields uniform liver spheroids To benchmark automated spheroid generation against the established manual workflow, we implemented an automated protocol covering cell seeding, spheroid formation, medium exchange, and endpoint characterization into the MO:BOT software, as schematized in Fig. 2 A. To determine whether automated handling affects spheroid formation and gross morphology, we first compared spheroids generated manually or with the MO:BOT during the 5-days culture period. Brightfield imaging acquired with the integrated MO:CROSCOPE on day 0, 4, and 5 demonstrated robust spheroid formation under both conditions, with compact, well-defined aggregates visible by day 4 and maintained at day 5 (Fig. 2 B). No qualitative differences in overall shape or edge definition were observed between manual and MO:BOT spheroids, indicating that the automated seeding and maintenance workflow supports efficient spheroid self-assembly and results in uniformly compact spheroids. Then, we performed a quantitative analysis of morphological parameters (area, ellipticity, and roundness) to assess how automation influences spheroid size and shape over time. From day 4 to day 5, average spheroid area decreased from 0.37 mm² ± 0.05 to 0.16 mm² ± 0.01 in the manual workflow but remained stable in the automated workflow (Fig. 2 C), suggesting that automated handling stabilizes spheroid size during liver spheroid maintenance. Ellipticity and roundness values stayed constant across all groups and time points, indicating that neither handling type nor culture duration measurably affected spheroid shape (Fig. 2 C). As medium exchange is a critical and disruptive step in 3D culture, we evaluated whether automated medium exchange affects spheroid architecture by comparing morphological metrics before and after the full medium exchange step performed on day 4 (Fig. 2 D). Quantitative analysis of area, ellipticity, and roundness before and after exchange revealed no significant changes in any metric for either manual or MO:BOT handling. These data indicate that the automated medium exchange routine is sufficiently gentle to avoid disrupting spheroid structure. To directly assess spheroid uniformity, we next compared spheroid area and variability at day 4 between both workflows. While the median spheroid area was comparable between conditions, MO:BOT cultures achieved a markedly lower coefficient of variation in spheroid area (8.1%) than manual cultures (13.3%) (Fig. 2 E), demonstrating that the automated workflow produces highly uniform spheroid populations. Such reduced variability is expected to benefit downstream toxicity testing, where assay sensitivity and statistical power depend on consistent baseline morphology. Finally, to confirm that these morphological findings translate into preserved cellular function, we assessed the viability and albumin secretion of HepG2 spheroids at day 5 of culture. Area-normalized viability expressed as x-fold change relative to manual, MO:BOT spheroids exhibited a 1.4-fold viability (Fig. 2 F), indicating that automated handling preserves the viability of the culture. Albumin secretion per spheroid was not significantly different between manual and MO:BOT cultures (Fig. 2 G). Collectively, these results demonstrate that automated spheroid generation with MO:BOT yields liver spheroids of comparable functional quality to manual culture, while reducing variability in spheroid morphology and increasing size uniformity and viability, thereby providing a robust foundation for the subsequent toxicity study. Automated acetaminophen (APAP) dosing induces concentration-dependent functional toxicity To assess liver toxicity under automated conditions, we integrated an acetaminophen (APAP) exposure routine into the MO:BOT workflow, adding the compound on day 4 and performing endpoint characterization on day 5 (Fig. 3 A). Spheroids generated by automated workflows were treated with a concentration range of 0 to 100 mM APAP. This design enabled quantitative assessment of both morphological and functional responses to APAP over a broad dose range. Brightfield imaging at day 5 revealed no gross structural disruption of spheroids across most APAP concentrations, but a concentration-dependent change in optical appearance (Fig. 3 B). At concentrations up to 30 mM, spheroids remained compact and with well-defined sharp edges in the 2D projection images, with transparency comparable to untreated controls. From ≥ 40 mM APAP, spheroids appeared progressively less transparent, and at 100 mM the spheroid borders became less well defined (i.e. “fuzzy”), indicating subtle structural alterations that are not captured by simple size metrics. The automated quantitative image analysis performed with the integrated MO:CROSCOPE of spheroid morphology at day 5 confirmed that APAP exposure only modestly affected the area, ellipticity, and roundness of the spheroids (Fig. 3 C). Fold-change analysis of spheroid area relative to untreated controls revealed a significant increase in size at 50 mM and 100 mM APAP compared with all lower concentrations, whereas spheroids treated with ≤ 40 mM remained close to baseline. Ellipticity and roundness did not exhibit any dose-dependent changes across the tested concentration range, indicating that overall spheroid shape was largely preserved despite these high-dose-induced size increases ( Supplementary Fig. 2 ). In contrast, functional readouts showed a clear concentration-dependent toxicity profile (Fig. 3 D). Fitting the ATP-based viability data with a four-parameter sigmoidal model yielded an EC₅₀ of 40.93 mM APAP (R 2 = 0.95), which we define as the toxicity study concentration for subsequent experiments. At low concentrations (≤ 20 mM), viability remained close to control levels (83.5% ± 5.31), whereas at 40 mM and 50 mM, viability dropped to 68.46% ± 3.17 and 42.96% ± 7.59, respectively, reaching 0.38% ± 0.07 at 100 mM APAP. These results show that the MO:BOT-based assay detects a clear, quantitative APAP toxicity profile with a well-defined EC₅₀, while maintaining overall spheroid morphology across the tested dose range. Automated APAP toxicity study at EC₅₀ preserves spheroid integrity while capturing hepatotoxic effects To evaluate APAP-induced toxicity under the previously established EC₅₀ condition, we exposed liver spheroids to 40.93 mM APAP on day 4 of culture and on day 5 compared spheroids generated and treated with manual or MO:BOT handling. Pre-treatment brightfield images showed compact, well-defined spheroids for both workflows and both conditions, with no visible differences between groups (Fig. 4 A). After 24 h APAP exposure, spheroids retained their overall shape but exhibited a slightly reduced transparency compared with untreated controls in both handling types. APAP treatment increased spheroid area in both workflows, with MO:BOT spheroids remaining larger than manually handled spheroids (Fig. 4 B). Ellipticity was unchanged across conditions, indicating that APAP exposure and handling type did not affect overall spheroid elongation. Roundness decreased slightly after APAP exposure in both workflows, a minor effect that did not compromise spheroid integrity or assay suitability. To examine whether automated handling affects the functional toxicity readouts at the EC₅₀ APAP concentration, we compared viability and cytotoxicity between manual and MO:BOT workflows. Functional readouts recapitulated the APAP-induced toxicity response at EC₅₀. Viability dropped to 47% in manually handled spheroids and to 45% in MO:BOT spheroids (Fig. 4 C), and cytotoxicity increased to 53% and 55%, respectively; these differences were not statistically significant. (Fig. 4 D). This indicates that automated handling produces 3D cell culture sensitive to toxicity assays. To examine whether hepatocellular injury markers were similarly captured under both handling conditions, we next quantified ALT release in response to APAP exposure. ALT release further confirmed APAP-induced hepatocellular injury (Fig. 4 E). The increase in ALT activity was significantly more pronounced in MO:BOT spheroids than in manually handled spheroids, and variability was lower in the automated condition (CV 25.0%) compared with the manual workflow (CV 43.7%). This combination of higher signal and reduced variability suggests that the more morphologically uniform spheroids produced by the automated workflow may provide a more sensitive and robust readout of hepatotoxicity. Discussion Three-dimensional liver model based on spheroid, multicellular liver microtissues, or liver microphysiological systems consistently outperform 2D hepatocyte cultures for predicting drug-induced liver injury (DILI), but their broader use is constrained by labor-intensive, user-dependent workflows that limit throughput and cross-laboratory reproducibility 5 , 6 , 16 – 18 . In this study, we demonstrate that a compact benchtop automation platform, the MO:BOT, can standardize the critical steps of a HepG2 spheroid assay, cell seeding, medium exchange, compound dosing, and on-deck brightfield imaging-based quality control, while maintaining and improving the biological performance typically reported for manually handled 3D liver systems. Automated seeding produced liver spheroids that were more uniform in size and morphology and displayed higher viability than manually generated spheroids, yet retained comparable albumin secretion, indicating that the workflow improves technical consistency without compromising basic liver-like function 3 , 17 , 19 , 20 . Using acetaminophen (APAP) as a reference hepatotoxin, the MO:BOT-generated spheroids reproduced the expected concentration-dependent toxicity profile with concordant changes in viability, functional biomarkers, and injury markers across the tested dose range. The resulting EC₅₀ of approximately 40 mM is line with APAP potencies reported for HepG2 and other advanced liver models, and is better aligned with clinically relevant hepatotoxic plasma concentrations than typical 2D HepG2 monolayers 6 , 16 , 19 , 21 . This agreement supports the notion that introducing automation does not erode, and may even enhance, the predictive value of 3D liver spheroid assays. The observation that bulk spheroid morphology remains largely preserved across the APAP dose range, while functional and injury readouts change markedly, is consistent with reports that molecular and functional markers of hepatotoxicity often precede gross morphological breakdown in advanced 3D liver models 16 . Several groups have begun to address the scalability and standardization gap in 3D liver assays using large liquid-handling workstations, 3D bioprinters, bioreactors, and high-content screening (HCS) workflows. These platforms typically target specific bottlenecks but often focus on isolated segments of the pipeline, such as cell-based fabrication models (3D bioprinters and bioreactors) or endpoint assays (liquid-handlers and HSC) 3 , 12 , 22 , 23 . In this context, the MO:BOT unifies liver spheroid generation, on-deck imagining and downstream toxicity testing within a single enclosed system. By ensuring accurate, precisely timed execution of all workflow steps, it reduces spheroid-to-spheroid variability and increases assay robustness. Together with integrated imaging-based quality control and compatibility with standard plate-based biochemical assays, this automated 3D liver spheroid workflows is positioned as a promising candidate for inclusion in safety-testing strategies and early risk-assessment pipelines 3 , 6 , 16 . Mechanistically, the stabilization of spheroid architecture in the automated workflow—where spheroid area remained constant between day 4 and 5, in contrast to the progressive compaction observed with manual handling—likely contributes to the more pronounced and less variable ALT signal observed in MO:BOT spheroids. This is consistent with reports that more uniform 3D liver microtissues provide increased sensitivity for detecting hepatocellular injury in high-throughput platforms 3 , 6 , 24 . The lower coefficient of variation in spheroid size achieved by MO:BOT cultures, together with preserved albumin secretion and robust APAP responses, suggests that controlling physical microarchitecture is an important lever for enhancing assay sensitivity and statistical power in 3D hepatotoxicity testing Building on the current findings, several clear next steps emerge. The MO:BOT HepG2 workflow is readily transferable to primary human hepatocyte spheroids, iPSC-derived hepatocyte models, and multicellular constructs incorporating non-parenchymal cell types, which would enable to study immune-mediated, fibrotic, and cholestatic injury mechanisms alongside intrinsic hepatotoxicity 3 , 16 , 22 . The establishes APAP model establishes a clinically relevant proof-of-concept under an acute 24 h exposure and future studies can extend this to larger panels of hepatotoxic and non-hepatotoxic compounds, as well as repeated-dose and longer-term paradigms, to benchmark predictive performance against established 3D liver systems and liver-on-a-chip platforms 3 , 18 , 22 . Finally, while this work primarily addressed biological and technical performance, and systematically evaluation of economic and operational metrics represents an important next step. Quantifying labor savings, error reduction, and cross-site reproducibility using standardized performance indicators would clarify how automated spheroid workflows such as MO:BOT can support regulatory initiatives that promote human-relevant in vitro methods under frameworks inspired by FDA Modernization 2.0 and 3.0 8,18 . Regulatory initiatives that enable or encourage the use of human-relevant in vitro approaches in place of animal studies require not only biologically predictive models, but also documented assay reliability, standardization, and scalability 8 , 22 . By demonstrating that a compact automation platform can generate uniform 3D liver spheroids, perform controlled compound dosing and support multi-parametric hepatotoxicity readouts with performance equivalent to or better than manual workflows, this study addresses several of these practical prerequisites. This is further supported by the observation that bulk morphology and functional damage can dissociate in 3D spheroid models, where viability loss and molecular stress markers often precede gross structural breakdown 12 , 25 , 26 , underscoring the value of the multi-parametric readout strategy employed here. Looking forward, leveraging the advantages of the MO:BOT with more advanced analytical modalities, such as high-content phenotypic imaging, transcriptomics and metabolomics, could further increase the mechanistic depth and translational value of the generated datasets 12 . Expansion of the workflow to additional liver-relevant cell sources and to other tissues such as kidney, heart, or intestine would support construction of interconnected multi-organ platforms for systemic toxicity assessment 18 , 22 . Finally, multi-site studies comparing automated and manual workflows across laboratories will be essential to establish generalizability of the observed gains in reproducibility and to support formal qualification of automated 3D liver spheroid assays under regulatory frameworks inspired by FDA Modernization 3.0 and related initiatives 8 , 22 . Methods MO:BOT, the 3D cell culture automated device The MO:BOT, a modular benchtop automation system developed by mo:re GmbH, was employed for the automated generation of human three-dimensional (3D) liver spheroids and subsequent drug exposure studies in high-throughput 96-well plates. This platform integrates multiple interchangeable plug‑and‑play modules, without the need for cables or screws, across ten configurable work fields to support specific cell culture workflows. The system includes the following functional modules: 1) MO:CROSCOPE, a brightfield imaging unit used to collect and analyze morphological parameters of individual 3D cell aggregates, 2) MO:HEAT, a heating module capable of reaching up to 40°C for applications such as sample incubation or enzymatic reactions, 3) MO:COOL, a cooling module capable of maintaining temperatures down to − 5°C to ensure thermal stability for temperature sensitive experiments or short‑term sample storage, 4) MO:TILT, a tilting module enabling gentle mixing to enhance medium exchange and homogeneous distribution of cells and reagents, and 5) MO:SHAKE, a shaker operating up to 1,000 rpm for vigorous mixing and ensuring reaction uniformity. Additionally, the MO:BOT features several passive modules: 1) MULTIRAK, a reagent rack compatible with various tubes sizes ranging from 1.5 mL to 50 mL, 2) TIP BOX MODULE, used to register the placement of pipette tips of different volumes such as 250 µL and 1000 µL, and 3) PLATE HOLDER, a module designed to accommodate various plates formats (U-Bottom, V-Bottom, F-Bottom plates) and sizes from 6-well to 384-well plates, with optional adapters for organ-on-a-chip devices. The MO:BOT can be equipped with single and multi‑channel positive‑displacement pipette capable of handling both viscous and non‑viscous liquids, supporting precise cell seeding, medium exchange, and compound dosing to promote consistent culture conditions. In addition, a robotic gripper allows automated transfer of plates between modules within the workspace. For the present study, the integrated single‑channel pipette, MO:CROSCOPE (QM, mo:re GmbH), Plate holder (HM, mo:re GmbH), Tip box module (PM, mo:re GmbH), and Reagent rack (MM, mo:re GmbH GmbH) were utilized. The “Liver spheroids generation and hepatotoxicity testing” protocol, available in the MO:BOT software (moreOS), was selected for both spheroid formation and subsequent drug stimulation steps. The protocol comprises three routines executed at different timepoints: day 0, cell seeding, day 4, medium exchange with or without the hepatotoxic compound, and day 5, sample collection or endpoint analysis. As part of the day 4 and 5 routines, brightfield images of each well were acquired and analyzed by the MO:CROSCOPE immediately before and after medium exchange. Additionally, the protocol “Liver spheroids APAP dose titration” comprising two routines (cell seeding and medium exchange) and it was performed to determine the acetaminophen (APAP) EC 50 . Upon protocol initiation, the system provides guidance for correct placement of modules and consumables across designated workfields. Each module is automatically recognized and activated at predefined steps of the workflow. The automated pipetting actions were executed with high precision and accuracy to ensure uniform cell distribution and reagent handling. Data generated by the MO:CROSCOPE module were used to monitor and analyze spheroid morphology and culture evolution throughout the experimental process. Liver cell culture HepG2 cells (ATCC, HB-8065) were maintained in Dulbecco's Modified Eagle Medium (DMEM) (DMEM, high glucose, Sigma-Aldrich, D0819-500ML) supplemented with 10% Fetal Bovine Serum (FBS) (ThermoFisher Scientific, A5670701) and 1% Penicillin–Streptomycin (P/S) (ThermoFisher Scientific, 15140-148), the DMEM complete medium. Cells were incubated at 37°C and passaged at 70% confluency using Trypsin-EDTA Solution (Himeda, TCL033) to maintain exponential growth. Cells were routinely screened for mycoplasma contamination using a PCR-based detection assay (ThermoFisher Scientific, M7006) and all cell cultures tested negative throughout the study. Automated liver spheroid generation For all experiments, HepG2 cells were used within a defined low passage range (passage 18–25), and the same passage number was applied across biological replicates and assay runs to ensure standardized handling. Before seeding, cells were harvested, counted using a hemocytometer, and resuspended in complete DMEM medium at the desired concentration. Cell seeding was performed with the MO:BOT and an additional plate was seeded manually to evaluate the impact of automated handling on spheroid quality and uniformity. For automated spheroid generation, the MO:BOT protocol “Liver spheroids generation and hepatotoxicity testing” routine 1, was selected from the integrated protocol library. Required consumables were positioned on the corresponding mo:re modules (described in section 4.1), including a 96-well U-bottom ULA plate (Facellitate, F202003), a 15 mL tube containing the prepared HepG2 cell suspension, and a box of 1000 µL positive-displacement tips (mo:re GmbH, MO:RE 1000 µL piston tips), and placed on the MO:BOT workfield. Upon protocol initiation, the MO:BOT’s single-channel pipetting tool automatically picked up a tip, aspirated the defined volume of cell suspension, and dispensed 2,000 cells in 150 µL of medium per well to support uniform spheroid formation. After seeding under both handling conditions, plates were incubated at 37°C, 5% CO 2 for 4 days to allow spheroid aggregation. Further details about the MO:BOT system are provided in Supplementary Fig. 1 and at https://more.science . MO:CROSCOPE, automated bright-field imaging module Spheroid morphology was assessed using the MO:CROSCOPE imaging system integrated into the MO:BOT. Cell culture plates were loaded into the MO:CROSCOPE carrier, and brightfield images were acquired under standardized illumination and focus settings using a 5X ZEISS A-Plan objective, IMX477 sensor (Arducam B0279) with an image resolution 2160 × 2160 pixels (center-cropped from full sensor) and pixel size of 0.79 µm/px. The MO:CROSCOPE automatically captured brightfield images of each individual spheroid across a full 96-well plate and analyzed in real-time key morphological parameters, such as size, roundness, and ellipticity. Imaging of a full 96-sample plate occurred within 10 minutes, limiting the time of the plate outside of the incubator. The MO:CROSCOPE served as a quality control check for spheroid growth and was used before and after medium exchange and drug stimulation to monitor potential morphological changes. Viability assay To assess spheroid quality, cellular ATP content was measured using the CellTiter-Glo 3D Cell Viability Assay (Promega, G9681). The assay was performed following manufacturer’s instruction. For each handling condition, 50 µl of culture supernatant containing the spheroid was transferred from the culture plate to an opaque 96-well assay plate (Corning, 3610) prior to assay reagent addition. Spheroid plates were equilibrated at room temperature (RT) for approximately 30 min prior to reagent addition to optimize cell lysis and signal stability. CellTiter-Glo 3D reagent was added at a 1:1 ratio to the transferred 50 µl spheroid-containing supernatant in each well, and the assay plate was shaken for 5 min at RT at 1,000 rpm to ensure efficient lysis of the 3D structures and thorough mixing of reagent and sample. Plates were incubated for an additional 25 min at RT to allow complete ATP extraction and development of a stable glow-type luminescent signal. Luminescence was measured on a ClarioStar plate reader (BMG Labtech, 1 s integration), and for APAP stimulation values were normalized to DMSO vehicle controls and expressed as percentage viability. Functional assay: Albumin secretion Albumin secretion, a key marker of liver-specific functionality, was quantified in spheroids generated manually and with the MO:BOT using the Human Albumin ELISA Kit (Abcam, ab108788). Kit components and samples were brought to RT, and albumin standards were prepared by serial dilution according to the manufacturer's instructions. After thawing and centrifugation (1,500 rpm, 10 min, 4°C), 50 µL of each supernatant or standard was added to the pre-coated wells and incubated for 1 h at RT. Wells were washed five times with 1X wash buffer, followed by sequential incubations with 50 µl biotinylated anti-albumin antibody and 50 µl streptavidin–HRP conjugate (SP conjugate) for 30 min each, with wash steps between incubations. After a final wash, TMB substrate was added and incubated for 30 min before adding 50 µL of stop solution to each well, and absorbance at 450 nm was recorded using a ClarioStar plate reader. Albumin concentrations were interpolated from the standard curve. Automated Acetaminophen (APAP) dose titration To determine the half-maximal effective concentration (EC₅₀) of acetaminophen (APAP), a dose-response titration experiment was conducted. A 5 M APAP stock solution was prepared in DMSO (Sigma Aldrich, D2438-5X10ML) and subsequently diluted in fresh complete DMEM to the desired working concentrations ranging from 1 mM to 100 mM. Vehicle control wells received complete DMEM containing the corresponding DMSO concentrations without APAP. The protocol “Liver spheroids APAP dose titration” was selected from the MO:BOT software library. Cell culture consumables were arranged in the designated MO:BOT work fields, including the liver spheroid culture plate, 1000‑µL tips, seven tubes containing culture medium with different APAP concentrations, one tube containing the vehicle control medium, and the liquid and tip waste modules. Upon protocol initiation, the MO:BOT automatically acquired a fresh tip, aspirated 80% of the culture medium without disrupting spheroid integrity or 3D morphology, and dispensed the respective APAP treatment solutions according to the predefined plate layout, using new tips between conditions to avoid cross‑contamination. Following 24 h incubation under standard conditions, spheroid responses were evaluated based on morphological parameters and cell viability. Dose-response curves were generated from normalized viability data, and EC₅₀ values were calculated by nonlinear regression (log[concentration] vs. response, variable-slope model), yielding an EC₅₀ of 40 mM that was subsequently used in downstream experiments. Automated Medium Exchange for EC 50 APAP stimulation After establishing 40 mM APAP as the EC₅₀ in liver spheroids, spheroid responses were characterized in both MO:BOT-generated and manually generated spheroids. On day 4 post-seeding, the MO:BOT performed routine 2 of the “Liver spheroids generation and hepatotoxicity testing” protocol. The MO:BOT aspirated the culture medium from each well and replaced it with either complete DMEM containing the corresponding DMSO concentration (vehicle control) or complete DMEM containing 40 mM APAP (treatment), following the workflow described in section 4.7. Plates were then incubated for 24 h before endpoint analyses were performed. Morphological, functional, and viability readouts were assessed in spheroids generated with the MO:BOT and compared with those generated manually. Cytotoxicity assay in APAP-stimulated liver spheroids The cytotoxicity of spheroids stimulated with APAP from both handling conditions was analyzed using the LDH-Glo Cytotoxicity Assay (Promega, J2380) following the manufacturer’s instructions. The cell culture plates were equilibrated to RT for approximately 30 min before reagent addition to stabilize LDH activity and luminescent output. For each condition, 50 µL of culture supernatant and 50 µL of LDH-Glo detection reagent were combined in an opaque-walled 96-well plate (Corning, 3610), mixed briefly (500 rpm) on a plate shaker, and incubated for 60 min at RT protected from light. Luminescence was recorded using a ClarioStar plate reader, and signal intensity was proportional to extracellular LDH levels. Metabolic function: Alanine Aminotransferase (ALT) activity Drug-induced hepatocellular toxicity was further evaluated by measuring Alanine Aminotransferase (ALT) activity in culture supernatants using the Human ALT SimpleStep ELISA Kit (Abcam, #ab234578). Reagents, standards, and samples were equilibrated at RT, and the ALT standard curve was prepared as described in the kit protocol. For each well, 50 µL of sample or standard and 50 µL of the antibody cocktail were added, mixed gently, and incubated for 1 h at RT to allow in-well immunocapture. Wells were then washed three times with 1× wash buffer, followed by addition of 100 µL TMB substrate and incubation for approximately 10 min at RT protected from the light. The reaction was stopped by adding 100 µL stop solution, and absorbance was measured at 450 nm using a ClarioStar plate reader. ALT concentrations were calculated from the standard curve and expressed as fold change relative to DMSO vehicle controls, providing a metabolic toxicity readout complementary to viability and LDH release. MO:CROSCOPE spheroid image segmentation Spheroid boundaries were delineated using the MIT-B0 SegFormer semantic segmentation model fine-tuned on organoid bright-field images deployed as an ONNX model (ONNX Runtime v1.23.0). Morphological features were computed from the segmented binary masks using OpenCV (v4.11.0) and an in-house Python (Python v3.10 ) library. Spheroid area was computed as the number of pixels enclosed by the contour boundary and converted to mm² using the calibrated pixel size. Roundness was calculated as the isoperimetric quotient R = 4πA / P², where P is the contour perimeter, yielding R = 1 for a perfect circle. Ellipticity was defined as the minor-to-major axis ratio of a least-squares fitted ellipse (E = a_minor / a_major), with E = 1 indicating a circle and lower values indicating elongation. Data analysis Experimental data were collected from plate reader output files and MO:CROSCOPE raw data exports and processed using custom R scripts (R v4.4.2). Prior to quantitative image analysis, all MO:CROSCOPE images were screened for artefacts such as out of focus wells, debris, or edge located spheroids, and wells failing these basic quality checks were excluded from further analysis. Raw luminescence and absorbance values were first background corrected using medium or blank wells, and technical replicates were averaged to yield a single value per biological replicate for each condition. Imaging-derived morphological parameters (area, roundness, and ellipticity) were exported from the analysis pipeline described in section 4.4 and linked to treatment metadata for downstream statistical comparisons. Where applicable, area measurements were normalized within each plate by dividing by the median of vehicle control wells (0 mM APAP), yielding fold-change values relative to untreated controls. Dose-response curves were fitted with a four-parameter logistic (4PL) model using log10[concentration] as the independent variable to derive EC₅₀ values and associated 95% confidence intervals. Statistical analysis and Manuscript preparation Group comparisons for functional readouts (viability, cytotoxicity, albumin, ALT) were performed using one-way ANOVA when more than two conditions were compared, followed by Tukey's HSD post-hoc test to correct for multiple pairwise comparisons. For two group comparisons (e.g. MO:BOT vs. manual at the same treatment condition), unpaired two tailed Welche’s t test (unequal variances) were applied after confirming approximate normality and homoscedasticity of residuals. Normality was assessed using the Shapiro-Wilk test and homogeneity of variance using Levene's test; when assumptions were violated, non-parametric alternatives (Kruskal-Wallis test with Dunn's post-hoc) were used. Dose-response relationships for APAP titration were modelled by nonlinear regression (log[concentration] vs. response, variable slope) to estimate EC₅₀ values and 95% confidence intervals. Coefficient of variation (CV) values were calculated for each condition as CV = (SD / mean) × 100 across biological replicates to assess assay variability and robustness. Statistical significance was evaluated with a p value < 0.05 considered statistically significant. Significance levels: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. A large language model-based assistant (Perplexity, powered by GPT-5.1) was used to support manuscript text editing and phrasing. All scientific content, data interpretation and final text were reviewed and approved by the authors. Declarations Author contributions statement D.He. designed and conducted the experiments, analysed the data, and wrote the manuscript. D.S.Z. developed hardware and software for morphological parameter analysis and reviewed the manuscript. F.E. and M.A. developed the moreOS software. P.D., A.S.O., J.K., T.P.H. and D.Ha. designed and developed the MO:BOT hardware. F.K.S. was involved in project communication and dissemination. L.G. supervised the budget and reviewed the manuscript. J.V. led the project and reviewed the manuscript. Competing Interests All authors are employees of mo:re GmbH, a for-profit organization that develops an automated 3D cell culture and organoid platform, namely the MO:BOT, related to the subject of this research. The research reported here was conducted as part of the authors’ employment at mo:re GmbH, which provided salary support and research funding. These relationships may be considered competing financial interests. The authors declare that they have no additional financial or non financial competing interests. Funding Declaration No funding was received for conducting this study. Author Contribution D.He. designed and conducted the experiments, analysed the data, and wrote the manuscript. D.S.Z. developed hardware and software for morphological parameter analysis and reviewed the manuscript. F.E. and M.A. developed the moreOS software. P.D., A.S.O., J.K., T.P.H. and D.Ha. designed and developed the MO:BOT hardware. F.K.S. was involved in project communication and dissemination. L.G. supervised the budget and reviewed the manuscript. J.V. led the project and reviewed the manuscript. Acknowledgement We thank Prof. Dr. Søren Gersting (University Medical Center Hamburg-Eppendorf, UKE) for valuable scientific input and technical expertise related to cell culture methodologies. A BioRender account was used to create schematic illustrations for this work. Data Availability The datasets generated and analyzed during the current study are available from the corresponding authors on reasonable request. References Loewa, A., Feng, J. J. & Hedtrich, S. Human disease models in drug development. Nat. Rev. Bioeng. 1 , 545–559 (2023). Leist, M. et al. The biological and ethical basis of the use of human embryonic stem cells for in vitro test systems or cell therapy. ALTEX 25 , 163–190 (2008). Yang, S., Ooka, M., Margolis, R. J. & Xia, M. Liver three-dimensional cellular models for high-throughput chemical testing. Cell. Rep. Methods . 3 , 100432 (2023). Mysior, M. M. & Simpson, J. C. An automated high-content screening and assay platform for the analysis of spheroids at subcellular resolution. PLOS ONE . 19 , e0311963 (2024). Vorrink, S. U., Zhou, Y., Ingelman-Sundberg, M. & Lauschke, V. M. Prediction of Drug-Induced Hepatotoxicity Using Long-Term Stable Primary Hepatic 3D Spheroid Cultures in Chemically Defined Conditions. Toxicol. Sci. Off J. Soc. Toxicol. 163 , 655–665 (2018). Proctor, W. R. et al. Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury. Arch. Toxicol. 91 , 2849–2863 (2017). Zhou, L. et al. Organoids and organs-on-chips: Recent advances, applications in drug development, and regulatory challenges. Med 6 , 100667 (2025). Zushin, P. J. H., Mukherjee, S. & Wu, J. C. FDA Modernization Act 2.0: transitioning beyond animal models with human cells, organoids, and AI/ML-based approaches. J. Clin. Invest. 133 , e175824 (2023). Han, J. J. F. D. A. & Modernization Act 2.0 allows for alternatives to animal testing. Artif. Organs . 47 , 449–450 (2023). Underhill, G. H. & Khetani, S. R. Advances in Engineered Human Liver Platforms for Drug Metabolism Studies. Drug Metab. Dispos. Biol. Fate Chem. 46 , 1626–1637 (2018). Bircsak, K. M. et al. A 3D microfluidic liver model for high throughput compound toxicity screening in the OrganoPlate®. Toxicology 450 , 152667 (2021). Hong, S. & Song, J. M. A 3D cell printing-fabricated HepG2 liver spheroid model for high-content in situ quantification of drug-induced liver toxicity. Biomater. Sci. 9 , 5939–5950 (2021). Nguyen, D. G. et al. Bioprinted 3D Primary Liver Tissues Allow Assessment of Organ-Level Response to Clinical Drug Induced Toxicity In Vitro. PloS One . 11 , e0158674 (2016). Schofield, C. A. et al. Evaluation of a Three-Dimensional Primary Human Hepatocyte Spheroid Model: Adoption and Industrialization for the Enhanced Detection of Drug-Induced Liver Injury. Chem. Res. Toxicol. 34 , 2485–2499 (2021). Haycock, J. W. 3D Cell Culture: A Review of Current Approaches and Techniques. in 3D Cell Culture (ed Haycock, J. W.) vol. 695 1–15 (Humana, Totowa, NJ, (2011). Bell, C. C. et al. Characterization of primary human hepatocyte spheroids as a model system for drug-induced liver injury, liver function and disease. Sci. Rep. 6 , 25187 (2016). Zhou, Y., Shen, J. X. & Lauschke, V. M. Comprehensive Evaluation of Organotypic and Microphysiological Liver Models for Prediction of Drug-Induced Liver Injury. Front. Pharmacol. 10 , 1093 (2019). Tutty, M. A., Movia, D. & Prina-Mello, A. Three-dimensional (3D) liver cell models - a tool for bridging the gap between animal studies and clinical trials when screening liver accumulation and toxicity of nanobiomaterials. Drug Deliv Transl Res. 12 , 2048–2074 (2022). Fey, S. J. & Wrzesinski, K. Determination of drug toxicity using 3D spheroids constructed from an immortal human hepatocyte cell line. Toxicol. Sci. Off J. Soc. Toxicol. 127 , 403–411 (2012). Basharat, A., Rollison, H. E., Williams, D. P. & Ivanov, D. P. HepG2 (C3A) spheroids show higher sensitivity compared to HepaRG spheroids for drug-induced liver injury (DILI). Toxicol. Appl. Pharmacol. 408 , 115279 (2020). Wang, Z. et al. Generation of hepatic spheroids using human hepatocyte-derived liver progenitor-like cells for hepatotoxicity screening. Theranostics 9 , 6690–6705 (2019). Kulsharova, G. & Kurmangaliyeva, A. Liver microphysiological platforms for drug metabolism applications. Cell. Prolif. 54 , e13099 (2021). Jeon, H., Kim, G., Carpenter, J., Colón, Y. J. & Wang, Y. Automated High-Content, High-Throughput Spatial Analysis Pipeline for Drug Screening in 3D Tumor Spheroid Inverted Colloidal Crystal Arrays. ACS Appl. Mater. Interfaces . 17 , 49210–49226 (2025). Vilas-Boas, V. et al. Primary Human Hepatocyte Spheroids as Tools to Study the Hepatotoxic Potential of Non-Pharmaceutical Chemicals. Int. J. Mol. Sci. 22 , 11005 (2021). Shin, D. S. et al. Hepatotoxicity evaluation method through multiple-factor analysis using human pluripotent stem cell derived hepatic organoids. Sci. Rep. 15 , 10804 (2025). Elje, E. et al. Hepato(Geno)Toxicity Assessment of Nanoparticles in a HepG2 Liver Spheroid Model. Nanomaterials 10 , 545 (2020). Additional Declarations Competing interest reported. All authors are employees of mo:re GmbH, a for-profit organization that develops an automated 3D cell culture and organoid platform, namely the MO:BOT, related to the subject of this research. The research reported here was conducted as part of the authors’ employment at mo:re GmbH, which provided salary support and research funding. These relationships may be considered competing financial interests. The authors declare that they have no additional financial or non financial competing interests. Supplementary Files HellmoldD.etalSupplementaryFigure1and2.pdf HellmoldD.etal.Supplementaryvideo.mp4 Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 27 Apr, 2026 Reviews received at journal 25 Apr, 2026 Reviewers agreed at journal 16 Apr, 2026 Reviews received at journal 01 Apr, 2026 Reviewers agreed at journal 23 Mar, 2026 Reviewers invited by journal 22 Mar, 2026 Editor invited by journal 20 Mar, 2026 Editor assigned by journal 16 Mar, 2026 Submission checks completed at journal 16 Mar, 2026 First submitted to journal 13 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9116861","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":609966395,"identity":"84f46350-d8a6-421a-9d63-b8aa940ae414","order_by":0,"name":"Dana Hellmold","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Dana","middleName":"","lastName":"Hellmold","suffix":""},{"id":609966396,"identity":"e72dcf77-e9b7-470e-813c-46317d751aed","order_by":1,"name":"Daniel S. Ziemianowicz","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"S.","lastName":"Ziemianowicz","suffix":""},{"id":609966397,"identity":"54337a4f-1c11-478d-bb05-25cdc59ce59d","order_by":2,"name":"Frowin Ellermann","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Frowin","middleName":"","lastName":"Ellermann","suffix":""},{"id":609966398,"identity":"d31f3396-1b89-4d54-b040-e3dde4808d57","order_by":3,"name":"Philipp Depperschmidt","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Philipp","middleName":"","lastName":"Depperschmidt","suffix":""},{"id":609966399,"identity":"78d05ab6-93a3-4159-8dff-06604e56658d","order_by":4,"name":"Max Appold","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Max","middleName":"","lastName":"Appold","suffix":""},{"id":609966400,"identity":"a88cc599-d3a9-4e91-8de9-bae04ea66522","order_by":5,"name":"Ahmed S. Omar","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Ahmed","middleName":"S.","lastName":"Omar","suffix":""},{"id":609966401,"identity":"cccefc30-adc5-4b0c-84cc-b66085fa56ac","order_by":6,"name":"Jonathan Kurz","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"","lastName":"Kurz","suffix":""},{"id":609966402,"identity":"dcb50dad-9d9c-4cfb-91fb-5dcabfe8f2a8","order_by":7,"name":"Frank Krieg-Schneider","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Krieg-Schneider","suffix":""},{"id":609966403,"identity":"05e069f7-7c38-4313-b1c9-7100b42056d0","order_by":8,"name":"Thorben Pascal Hoppe","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Thorben","middleName":"Pascal","lastName":"Hoppe","suffix":""},{"id":609966404,"identity":"482fc249-f3b3-4f41-80e5-fac7098e5a45","order_by":9,"name":"David Hackenberger","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Hackenberger","suffix":""},{"id":609966405,"identity":"676996b7-2fce-4762-aadd-9e8372d2a782","order_by":10,"name":"Lukas Gaats","email":"","orcid":"","institution":"mo:re GmbH","correspondingAuthor":false,"prefix":"","firstName":"Lukas","middleName":"","lastName":"Gaats","suffix":""},{"id":609966406,"identity":"6f6f94f0-64d3-40a2-8476-80196a937d7a","order_by":11,"name":"Julia Vallverdú","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYBACxgbmhgMMFTCuAVFaGIFazpCiBaSJgbGNFIcxz0hsPPBz3jY58/YG5g8/Cmzy+NsbCNgxI7HhYO+228YyZw6wSfYYpBVLnDlAWMsB3m23E2dIJLAxMxgcTtwgkUCELX/nALXIP2D+zGDwP3GD/APCWg7zNoBsYWCQZjA4ALQFvw4Gxp6HDYdljt02luBJbAP6JTlxxhkCDjNsTz788U3NbTkJ9sOHP/z4Y5fY336AgJYGhIUNOFWhAHnilI2CUTAKRsGIBgB0I0nccw8U+AAAAABJRU5ErkJggg==","orcid":"","institution":"mo:re GmbH","correspondingAuthor":true,"prefix":"","firstName":"Julia","middleName":"","lastName":"Vallverdú","suffix":""}],"badges":[],"createdAt":"2026-03-13 16:53:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9116861/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9116861/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105297693,"identity":"a3c6e631-7ff6-4408-b3a6-16446826feed","added_by":"auto","created_at":"2026-03-24 13:21:12","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":115826,"visible":true,"origin":"","legend":"\u003cp\u003eThe MO:BOT, an integrated automation platform that consolidates the traditional cell culture workflow. (\u003cstrong\u003eA\u003c/strong\u003e) A typical 3D cell culture workflow occurs in four phases: (1) cell seeding and maintenance of the culture over days or weeks, wherein (2) quality control is performed throughout to track growth and development, followed by (3) treatment by e.g. stimulation with a small molecule and additional (2) quality control and terminated by (4) an endpoint assay. (\u003cstrong\u003eB\u003c/strong\u003e) Cell culture workflow integration in the MO:BOT. The MO:BOT consolidates all steps with interconnected on-deck modules. This enables e.g. temperature control during pipetting steps and seamless data collection via brightfield imaging and plate reading without interrupting sterility or increasing handling overhead. (\u003cstrong\u003eC\u003c/strong\u003e) Traditional workflows require transfers between separate instruments (e.g. incubator, biosafety cabinet, liquid handler, shaker, plate reader, microscope etc.), introducing sterility risks, time variability at each transfer step, increasing overhead and does not allow for systematic progress tracking and data collection.\u003c/p\u003e","description":"","filename":"image0.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/bd763659bff7015a2fbc0be3.jpg"},{"id":105297698,"identity":"e87f9450-0be4-477d-bd25-63562306f57c","added_by":"auto","created_at":"2026-03-24 13:21:13","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":95548,"visible":true,"origin":"","legend":"\u003cp\u003eThe MO:BOT automates the manual cell culture workflow without protocol modification, reducing spheroid variability and enhancing viability. (\u003cstrong\u003eA\u003c/strong\u003e) Experimental design for HepG2 liver spheroid formation and characterization performed with both handling methods, Manual and MO:BOT. (\u003cstrong\u003eB\u003c/strong\u003e) Representative sample of brightfield images collected with the MO:CROSCOPE at day 4 and 5 of culture (500 μm scale bars). (\u003cstrong\u003eC\u003c/strong\u003e) Time-course analysis reveals significant differences in area between handling methods, prior to medium exchange, at day 4 and 5, with comparable shape metrics (ellipticity, roundness). (\u003cstrong\u003eD\u003c/strong\u003e) Medium exchange timing on day 4 shows little or no effect on spheroid morphology, regardless of handling type. (\u003cstrong\u003eE\u003c/strong\u003e) Automation with the MO:BOT achieves lower coefficient of variation in spheroid area compared to manual pipetting on day 4 (p\u0026lt;0.001). (\u003cstrong\u003eF) \u003c/strong\u003eLiver spheroids on day 5 has a higher viability than ones generated manually). \u003cstrong\u003e(G\u003c/strong\u003e) Liver spheroids generated with both handling conditions present an equivalent albumin secretion.\u003c/p\u003e","description":"","filename":"image1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/85dd9dbd735714fc9ad65956.jpg"},{"id":105297692,"identity":"5d530da9-fa5b-4b83-b3a0-37c4427f4252","added_by":"auto","created_at":"2026-03-24 13:21:12","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":127202,"visible":true,"origin":"","legend":"\u003cp\u003eThe MO:BOT enables hepatotoxicity screening. (\u003cstrong\u003eA\u003c/strong\u003e) Experimental design for dose-response assessment; APAP exposure 4 days after seeding. (\u003cstrong\u003eB\u003c/strong\u003e) Representative brightfield images, collected with the MO:CROSCOPE, of spheroids exposed to a range of APAP concentrations show no obvious morphological disruption (500 μm scale bars). (\u003cstrong\u003eC\u003c/strong\u003e) Morphological analysis reveals a dose-dependent reduction in organoid cross-sectional area, while shape metrics show no significant differences across concentration range. (\u003cstrong\u003eD\u003c/strong\u003e) The viability dose-response curve yields an EC₅₀ = 40.93 mM.\u003c/p\u003e","description":"","filename":"image2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/e56162e9062a808a305a36c5.jpg"},{"id":105297695,"identity":"097389b7-9f4f-44e1-8748-642fed327753","added_by":"auto","created_at":"2026-03-24 13:21:12","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":121950,"visible":true,"origin":"","legend":"\u003cp\u003eAutomated handling with the MO:BOT matches or improves on manual handling for detection of hepatocytotoxic effects. (\u003cstrong\u003eA\u003c/strong\u003e) Representative brightfield images showing spheroid morphology before and 24 h after 40.93 mM APAP treatment for both handling methods (500 μm scale bars). (\u003cstrong\u003eB\u003c/strong\u003e) Morphological analysis reveals APAP-induced increase in area (p\u0026lt;0.001) with preserved shape metrics, consistently across handling methods. (\u003cstrong\u003eC,D\u003c/strong\u003e) Viability decreases and cytotoxicity increases (p\u0026lt;0.0001) independently of handling type. (\u003cstrong\u003eE\u003c/strong\u003e) The automated workflow shows a greater effect of hepatocyte injury as measured by ALT (p\u0026lt;0.01).\u003c/p\u003e","description":"","filename":"image3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/03ba3459e6db518e50d2286a.jpg"},{"id":105565681,"identity":"9f15e508-2367-47ef-b358-d803ee575217","added_by":"auto","created_at":"2026-03-27 12:54:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1481501,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/e3d9da4b-e49e-4327-8ca1-8f5eeab1f804.pdf"},{"id":105297694,"identity":"a8e076fc-6e79-4fc1-b149-1b5f5e66d71b","added_by":"auto","created_at":"2026-03-24 13:21:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":252915,"visible":true,"origin":"","legend":"","description":"","filename":"HellmoldD.etalSupplementaryFigure1and2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/5898d015d2a3e3d655cc5010.pdf"},{"id":105297697,"identity":"97650213-70d1-4f7f-8f68-da35f2505b2c","added_by":"auto","created_at":"2026-03-24 13:21:13","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":47454749,"visible":true,"origin":"","legend":"","description":"","filename":"HellmoldD.etal.Supplementaryvideo.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9116861/v1/bd8dcbc8cfa3b29208f21d2a.mp4"}],"financialInterests":"Competing interest reported. All authors are employees of mo:re GmbH, a for-profit organization that develops an automated 3D cell culture and organoid platform, namely the MO:BOT, related to the subject of this research. The research reported here was conducted as part of the authors’ employment at mo:re GmbH, which provided salary support and research funding. These relationships may be considered competing financial interests. The authors declare that they have no additional financial or non financial competing interests.","formattedTitle":"Automation of 3D Liver Spheroid generation and Acetaminophen dose– response on the MO:BOT enhances assay robustness and precision","fulltext":[{"header":"Key Phrases","content":"\u003cul type=\"disc\"\u003e\n \u003cli\u003eModular benchtop 3D cell culture robot\u003c/li\u003e\n \u003cli\u003eIntegrated imaging and liquid handling\u003c/li\u003e\n \u003cli\u003eAutomated liver spheroid generation\u003c/li\u003e\n \u003cli\u003eStandardized APAP-dose response testing\u003c/li\u003e\n \u003cli\u003eReduced spheroid variability\u003c/li\u003e\u003cli\u003eIncreased assay robustness\u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e\n"},{"header":"Introduction","content":"\u003cp\u003eHuman cell-based models have increasingly transformed preclinical studies by more closely recapitulating human physiology than traditional animal models\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. However, their impact is often constrained by variability and the difficulty of achieving high-throughput with sufficient standardization\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Conventional preclinical testing still relies heavily on animal studies, where interspecies differences can compromise the prediction of human toxicity\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Recent regulatory developments, including the FDA Modernization Act 3.0 and related strategic roadmaps, explicitly encourage the use of human-relevant \u003cem\u003ein vitro\u003c/em\u003e methods, such as organoids and microphysiological systems, as New Approach Methodologies (NAMs) and alternatives to animal testing\u003csup\u003e\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In this context, there is a growing demand for robust, scalable human cell-based systems that can deliver high-quality, decision-enabling data in early drug development\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThree-dimensional (3D) human \u003cem\u003ein vitro\u003c/em\u003e models such as organoids and spheroids address several limitations of conventional two-dimensional (2D) monolayers by more faithfully reproducing tissue-like cell-cell contacts, nutrient gradients, and microenvironments\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. For the liver, which plays a central role in xenobiotic metabolism and detoxification, 3D cell-based models maintain liver-specific functions over extended culture periods and more accurately capture clinically-relevant hepatotoxicity than 2D cultures\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Hepatocarcinoma-derived cell lines, such as HepG2 assembled into 3D spheroids, are widely used in early-stage screening owing to their robustness and compatibility with high-content and high-throughput platforms\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Spheroids generated from primary human hepatocytes and induced pluripotent stem cell (iPSC) -derived hepatocytes are implemented as a second-tier assay, providing higher physiological relevance for drug testing\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. The value across these 3D model types relies on the same key requirement: achieving robust standardization with consistent spheroid size, architecture, and function\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn practice, current 3D liver spheroid workflows remain largely manual, which hampers the realization of this potential\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Routine operations such as cell seeding, spheroid formation, medium exchange, quality controls, and compound dosing require extensive hands-on time. In addition, the intrinsic variability between individual spheroids makes them more difficult to handle and analyze, which reduces throughput and weakens assay reliability, and safety testing workflows\u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Existing automation approaches only partially mitigate these issue-throughput. Liquid handlers increase scalability and speed for downstream analysis, but do not address the upstream generation of homogenous 3D cell models. 3D Bioprinters support the creation of complex, ECM-embedded constructs but typically lack integrated systems for long-term maintenance. Bioreactors can scale 3D cell cultures but often generate heterogeneous aggregates with variable size and viability. High content screening workflows still rely on manual-to-end, reproducible 3D culture process\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Altogether, this underscores the absence of an integrated platform that can unify spheroid generation, long-term maintenance, perturbation, and readout within a single, standardize device.\u003c/p\u003e \u003cp\u003eTo address these needs, we developed the MO:BOT, a fully automated platform specifically designed for 3D cell culture workflows. MO:BOT automates spheroid and organoid generation and maintenance, controlled medium exchange, compound dosing, and on-deck imaging and assay readouts within a single enclosed system, thereby minimizing manual plate transfers between lab devices and user-dependent steps\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Within the integrated workflow, MO:BOT combines plug-and-play on-deck modules (heater, cooler, imaging module, shaker) to automate different types of 3D cell culture workflows and use cases. The MO:BOT software offers an intuitive, integrated protocol library covering multiple human 3D \u003cem\u003ein vitro\u003c/em\u003e models and applications, while enabling straightforward implementation of new, customized workflows.\u003c/p\u003e \u003cp\u003eIn this work we demonstrate that the MO:BOT enables standardized and scalable production and characterization of liver spheroids suitable for toxicity testing by combining the biological advantages of 3D liver models with the process-level benefits of automation. This approach provides a path toward high-quality, robust \u003cem\u003ein vitro\u003c/em\u003e liver assays aligned with modern, human-relevant drug development and with emerging regulatory expectations under initiatives such as the FDA Modernization Act 2.0 and 3.0\u003csup\u003e7\u0026ndash;9\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eThe MO:BOT establishes an integrated automated workflow for 3D liver spheroid generation and hepatotoxicity testing\u003c/h2\u003e \u003cp\u003eTo address the limitations of fragmented 3D cell culture workflows, we implemented the MO:BOT as an integrated automation platform that consolidates all key phases of 3D liver spheroid culture, from cell seeding to endpoint analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). A typical 3D cell culture experiment comprises of four phases: (1) cell seeding and maintenance of the culture over several days or weeks, (2) iterative quality control to monitor spheroid growth and morphology, (3) treatment with a compound such as acetaminophen (APAP), again followed by intermediate quality control, and finally (4) termination of the culture by performing endpoint assays such as for viability, cytotoxicity, and functionality \u003cb\u003e(Supplementary Fig.\u0026nbsp;1)\u003c/b\u003e. These phases form the logical backbone of the liver toxicity workflow evaluated in this study, encompassing both biochemical readouts and morphological analysis by imaging.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn the MO:BOT, all four phases were executed within a single workflow. This integration eliminated repeated off-deck handling steps, which, in conventional practice, introduce time variability, potential deviations in incubation conditions, an increased risk of contamination, and precludes systematic, and time-resolved quality control across experiments. As 3D spheroid assays are particularly sensitive to subtle changes in handling, timing, and environmental conditions, this consolidated workflow is expected to reduce variability in spheroid morphology and functional outputs\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eAutomated cell seeding with MO:BOT yields uniform liver spheroids\u003c/h3\u003e\n\u003cp\u003eTo benchmark automated spheroid generation against the established manual workflow, we implemented an automated protocol covering cell seeding, spheroid formation, medium exchange, and endpoint characterization into the MO:BOT software, as schematized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA. To determine whether automated handling affects spheroid formation and gross morphology, we first compared spheroids generated manually or with the MO:BOT during the 5-days culture period. Brightfield imaging acquired with the integrated MO:CROSCOPE on day 0, 4, and 5 demonstrated robust spheroid formation under both conditions, with compact, well-defined aggregates visible by day 4 and maintained at day 5 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). No qualitative differences in overall shape or edge definition were observed between manual and MO:BOT spheroids, indicating that the automated seeding and maintenance workflow supports efficient spheroid self-assembly and results in uniformly compact spheroids. Then, we performed a quantitative analysis of morphological parameters (area, ellipticity, and roundness) to assess how automation influences spheroid size and shape over time. From day 4 to day 5, average spheroid area decreased from 0.37 mm\u0026sup2; \u0026plusmn; 0.05 to 0.16 mm\u0026sup2; \u0026plusmn; 0.01 in the manual workflow but remained stable in the automated workflow (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC), suggesting that automated handling stabilizes spheroid size during liver spheroid maintenance. Ellipticity and roundness values stayed constant across all groups and time points, indicating that neither handling type nor culture duration measurably affected spheroid shape (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAs medium exchange is a critical and disruptive step in 3D culture, we evaluated whether automated medium exchange affects spheroid architecture by comparing morphological metrics before and after the full medium exchange step performed on day 4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). Quantitative analysis of area, ellipticity, and roundness before and after exchange revealed no significant changes in any metric for either manual or MO:BOT handling. These data indicate that the automated medium exchange routine is sufficiently gentle to avoid disrupting spheroid structure. To directly assess spheroid uniformity, we next compared spheroid area and variability at day 4 between both workflows. While the median spheroid area was comparable between conditions, MO:BOT cultures achieved a markedly lower coefficient of variation in spheroid area (8.1%) than manual cultures (13.3%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE), demonstrating that the automated workflow produces highly uniform spheroid populations. Such reduced variability is expected to benefit downstream toxicity testing, where assay sensitivity and statistical power depend on consistent baseline morphology.\u003c/p\u003e \u003cp\u003eFinally, to confirm that these morphological findings translate into preserved cellular function, we assessed the viability and albumin secretion of HepG2 spheroids at day 5 of culture. Area-normalized viability expressed as x-fold change relative to manual, MO:BOT spheroids exhibited a 1.4-fold viability (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF), indicating that automated handling preserves the viability of the culture. Albumin secretion per spheroid was not significantly different between manual and MO:BOT cultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG). Collectively, these results demonstrate that automated spheroid generation with MO:BOT yields liver spheroids of comparable functional quality to manual culture, while reducing variability in spheroid morphology and increasing size uniformity and viability, thereby providing a robust foundation for the subsequent toxicity study.\u003c/p\u003e\n\u003ch3\u003eAutomated acetaminophen (APAP) dosing induces concentration-dependent functional toxicity\u003c/h3\u003e\n\u003cp\u003eTo assess liver toxicity under automated conditions, we integrated an acetaminophen (APAP) exposure routine into the MO:BOT workflow, adding the compound on day 4 and performing endpoint characterization on day 5 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Spheroids generated by automated workflows were treated with a concentration range of 0 to 100 mM APAP. This design enabled quantitative assessment of both morphological and functional responses to APAP over a broad dose range.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBrightfield imaging at day 5 revealed no gross structural disruption of spheroids across most APAP concentrations, but a concentration-dependent change in optical appearance (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). At concentrations up to 30 mM, spheroids remained compact and with well-defined sharp edges in the 2D projection images, with transparency comparable to untreated controls. From \u0026ge;\u0026thinsp;40 mM APAP, spheroids appeared progressively less transparent, and at 100 mM the spheroid borders became less well defined (i.e. \u0026ldquo;fuzzy\u0026rdquo;), indicating subtle structural alterations that are not captured by simple size metrics.\u003c/p\u003e \u003cp\u003eThe automated quantitative image analysis performed with the integrated MO:CROSCOPE of spheroid morphology at day 5 confirmed that APAP exposure only modestly affected the area, ellipticity, and roundness of the spheroids (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Fold-change analysis of spheroid area relative to untreated controls revealed a significant increase in size at 50 mM and 100 mM APAP compared with all lower concentrations, whereas spheroids treated with \u0026le;\u0026thinsp;40 mM remained close to baseline. Ellipticity and roundness did not exhibit any dose-dependent changes across the tested concentration range, indicating that overall spheroid shape was largely preserved despite these high-dose-induced size increases (\u003cb\u003eSupplementary Fig.\u0026nbsp;2\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eIn contrast, functional readouts showed a clear concentration-dependent toxicity profile (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Fitting the ATP-based viability data with a four-parameter sigmoidal model yielded an EC₅₀ of 40.93 mM APAP (R\u003csup\u003e2\u003c/sup\u003e= 0.95), which we define as the toxicity study concentration for subsequent experiments. At low concentrations (\u0026le;\u0026thinsp;20 mM), viability remained close to control levels (83.5% \u0026plusmn; 5.31), whereas at 40 mM and 50 mM, viability dropped to 68.46% \u0026plusmn; 3.17 and 42.96% \u0026plusmn; 7.59, respectively, reaching 0.38% \u0026plusmn; 0.07 at 100 mM APAP. These results show that the MO:BOT-based assay detects a clear, quantitative APAP toxicity profile with a well-defined EC₅₀, while maintaining overall spheroid morphology across the tested dose range.\u003c/p\u003e\n\u003ch3\u003eAutomated APAP toxicity study at EC₅₀ preserves spheroid integrity while capturing hepatotoxic effects\u003c/h3\u003e\n\u003cp\u003eTo evaluate APAP-induced toxicity under the previously established EC₅₀ condition, we exposed liver spheroids to 40.93 mM APAP on day 4 of culture and on day 5 compared spheroids generated and treated with manual or MO:BOT handling. Pre-treatment brightfield images showed compact, well-defined spheroids for both workflows and both conditions, with no visible differences between groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). After 24 h APAP exposure, spheroids retained their overall shape but exhibited a slightly reduced transparency compared with untreated controls in both handling types.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAPAP treatment increased spheroid area in both workflows, with MO:BOT spheroids remaining larger than manually handled spheroids (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Ellipticity was unchanged across conditions, indicating that APAP exposure and handling type did not affect overall spheroid elongation. Roundness decreased slightly after APAP exposure in both workflows, a minor effect that did not compromise spheroid integrity or assay suitability.\u003c/p\u003e \u003cp\u003eTo examine whether automated handling affects the functional toxicity readouts at the EC₅₀ APAP concentration, we compared viability and cytotoxicity between manual and MO:BOT workflows. Functional readouts recapitulated the APAP-induced toxicity response at EC₅₀. Viability dropped to 47% in manually handled spheroids and to 45% in MO:BOT spheroids (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), and cytotoxicity increased to 53% and 55%, respectively; these differences were not statistically significant. (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). This indicates that automated handling produces 3D cell culture sensitive to toxicity assays.\u003c/p\u003e \u003cp\u003eTo examine whether hepatocellular injury markers were similarly captured under both handling conditions, we next quantified ALT release in response to APAP exposure. ALT release further confirmed APAP-induced hepatocellular injury (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). The increase in ALT activity was significantly more pronounced in MO:BOT spheroids than in manually handled spheroids, and variability was lower in the automated condition (CV 25.0%) compared with the manual workflow (CV 43.7%). This combination of higher signal and reduced variability suggests that the more morphologically uniform spheroids produced by the automated workflow may provide a more sensitive and robust readout of hepatotoxicity.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThree-dimensional liver model based on spheroid, multicellular liver microtissues, or liver microphysiological systems consistently outperform 2D hepatocyte cultures for predicting drug-induced liver injury (DILI), but their broader use is constrained by labor-intensive, user-dependent workflows that limit throughput and cross-laboratory reproducibility\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e–\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. In this study, we demonstrate that a compact benchtop automation platform, the MO:BOT, can standardize the critical steps of a HepG2 spheroid assay, cell seeding, medium exchange, compound dosing, and on-deck brightfield imaging-based quality control, while maintaining and improving the biological performance typically reported for manually handled 3D liver systems. Automated seeding produced liver spheroids that were more uniform in size and morphology and displayed higher viability than manually generated spheroids, yet retained comparable albumin secretion, indicating that the workflow improves technical consistency without compromising basic liver-like function\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eUsing acetaminophen (APAP) as a reference hepatotoxin, the MO:BOT-generated spheroids reproduced the expected concentration-dependent toxicity profile with concordant changes in viability, functional biomarkers, and injury markers across the tested dose range. The resulting EC₅₀ of approximately 40 mM is line with APAP potencies reported for HepG2 and other advanced liver models, and is better aligned with clinically relevant hepatotoxic plasma concentrations than typical 2D HepG2 monolayers\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. This agreement supports the notion that introducing automation does not erode, and may even enhance, the predictive value of 3D liver spheroid assays. The observation that bulk spheroid morphology remains largely preserved across the APAP dose range, while functional and injury readouts change markedly, is consistent with reports that molecular and functional markers of hepatotoxicity often precede gross morphological breakdown in advanced 3D liver models\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eSeveral groups have begun to address the scalability and standardization gap in 3D liver assays using large liquid-handling workstations, 3D bioprinters, bioreactors, and high-content screening (HCS) workflows. These platforms typically target specific bottlenecks but often focus on isolated segments of the pipeline, such as cell-based fabrication models (3D bioprinters and bioreactors) or endpoint assays (liquid-handlers and HSC)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. In this context, the MO:BOT unifies liver spheroid generation, on-deck imagining and downstream toxicity testing within a single enclosed system. By ensuring accurate, precisely timed execution of all workflow steps, it reduces spheroid-to-spheroid variability and increases assay robustness. Together with integrated imaging-based quality control and compatibility with standard plate-based biochemical assays, this automated 3D liver spheroid workflows is positioned as a promising candidate for inclusion in safety-testing strategies and early risk-assessment pipelines\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMechanistically, the stabilization of spheroid architecture in the automated workflow—where spheroid area remained constant between day 4 and 5, in contrast to the progressive compaction observed with manual handling—likely contributes to the more pronounced and less variable ALT signal observed in MO:BOT spheroids. This is consistent with reports that more uniform 3D liver microtissues provide increased sensitivity for detecting hepatocellular injury in high-throughput platforms\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. The lower coefficient of variation in spheroid size achieved by MO:BOT cultures, together with preserved albumin secretion and robust APAP responses, suggests that controlling physical microarchitecture is an important lever for enhancing assay sensitivity and statistical power in 3D hepatotoxicity testing\u003c/p\u003e \u003cp\u003eBuilding on the current findings, several clear next steps emerge. The MO:BOT HepG2 workflow is readily transferable to primary human hepatocyte spheroids, iPSC-derived hepatocyte models, and multicellular constructs incorporating non-parenchymal cell types, which would enable to study immune-mediated, fibrotic, and cholestatic injury mechanisms alongside intrinsic hepatotoxicity\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. The establishes APAP model establishes a clinically relevant proof-of-concept under an acute 24 h exposure and future studies can extend this to larger panels of hepatotoxic and non-hepatotoxic compounds, as well as repeated-dose and longer-term paradigms, to benchmark predictive performance against established 3D liver systems and liver-on-a-chip platforms\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFinally, while this work primarily addressed biological and technical performance, and systematically evaluation of economic and operational metrics represents an important next step. Quantifying labor savings, error reduction, and cross-site reproducibility using standardized performance indicators would clarify how automated spheroid workflows such as MO:BOT can support regulatory initiatives that promote human-relevant \u003cem\u003ein vitro\u003c/em\u003e methods under frameworks inspired by FDA Modernization 2.0 and 3.0\u003csup\u003e8,18\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eRegulatory initiatives that enable or encourage the use of human-relevant \u003cem\u003ein vitro\u003c/em\u003e approaches in place of animal studies require not only biologically predictive models, but also documented assay reliability, standardization, and scalability\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. By demonstrating that a compact automation platform can generate uniform 3D liver spheroids, perform controlled compound dosing and support multi-parametric hepatotoxicity readouts with performance equivalent to or better than manual workflows, this study addresses several of these practical prerequisites. This is further supported by the observation that bulk morphology and functional damage can dissociate in 3D spheroid models, where viability loss and molecular stress markers often precede gross structural breakdown\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e, underscoring the value of the multi-parametric readout strategy employed here.\u003c/p\u003e \u003cp\u003eLooking forward, leveraging the advantages of the MO:BOT with more advanced analytical modalities, such as high-content phenotypic imaging, transcriptomics and metabolomics, could further increase the mechanistic depth and translational value of the generated datasets\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Expansion of the workflow to additional liver-relevant cell sources and to other tissues such as kidney, heart, or intestine would support construction of interconnected multi-organ platforms for systemic toxicity assessment\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Finally, multi-site studies comparing automated and manual workflows across laboratories will be essential to establish generalizability of the observed gains in reproducibility and to support formal qualification of automated 3D liver spheroid assays under regulatory frameworks inspired by FDA Modernization 3.0 and related initiatives\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eMO:BOT, the 3D cell culture automated device\u003c/h2\u003e\u003cp\u003eThe MO:BOT, a modular benchtop automation system developed by mo:re GmbH, was employed for the automated generation of human three-dimensional (3D) liver spheroids and subsequent drug exposure studies in high-throughput 96-well plates. This platform integrates multiple interchangeable plug‑and‑play modules, without the need for cables or screws, across ten configurable work fields to support specific cell culture workflows.\u003c/p\u003e\u003cp\u003eThe system includes the following functional modules: 1) MO:CROSCOPE, a brightfield imaging unit used to collect and analyze morphological parameters of individual 3D cell aggregates, 2) MO:HEAT, a heating module capable of reaching up to 40°C for applications such as sample incubation or enzymatic reactions, 3) MO:COOL, a cooling module capable of maintaining temperatures down to − 5°C to ensure thermal stability for temperature sensitive experiments or short‑term sample storage, 4) MO:TILT, a tilting module enabling gentle mixing to enhance medium exchange and homogeneous distribution of cells and reagents, and 5) MO:SHAKE, a shaker operating up to 1,000 rpm for vigorous mixing and ensuring reaction uniformity. Additionally, the MO:BOT features several passive modules: 1) MULTIRAK, a reagent rack compatible with various tubes sizes ranging from 1.5 mL to 50 mL, 2) TIP BOX MODULE, used to register the placement of pipette tips of different volumes such as 250 µL and 1000 µL, and 3) PLATE HOLDER, a module designed to accommodate various plates formats (U-Bottom, V-Bottom, F-Bottom plates) and sizes from 6-well to 384-well plates, with optional adapters for organ-on-a-chip devices.\u003c/p\u003e\u003cp\u003eThe MO:BOT can be equipped with single and multi‑channel positive‑displacement pipette capable of handling both viscous and non‑viscous liquids, supporting precise cell seeding, medium exchange, and compound dosing to promote consistent culture conditions. In addition, a robotic gripper allows automated transfer of plates between modules within the workspace.\u003c/p\u003e\u003cp\u003eFor the present study, the integrated single‑channel pipette, MO:CROSCOPE (QM, mo:re GmbH), Plate holder (HM, mo:re GmbH), Tip box module (PM, mo:re GmbH), and Reagent rack (MM, mo:re GmbH GmbH) were utilized. The “Liver spheroids generation and hepatotoxicity testing” protocol, available in the MO:BOT software (moreOS), was selected for both spheroid formation and subsequent drug stimulation steps. The protocol comprises three routines executed at different timepoints: day 0, cell seeding, day 4, medium exchange with or without the hepatotoxic compound, and day 5, sample collection or endpoint analysis. As part of the day 4 and 5 routines, brightfield images of each well were acquired and analyzed by the MO:CROSCOPE immediately before and after medium exchange. Additionally, the protocol “Liver spheroids APAP dose titration” comprising two routines (cell seeding and medium exchange) and it was performed to determine the acetaminophen (APAP) EC\u003csub\u003e50\u003c/sub\u003e.\u003c/p\u003e\u003cp\u003eUpon protocol initiation, the system provides guidance for correct placement of modules and consumables across designated workfields. Each module is automatically recognized and activated at predefined steps of the workflow. The automated pipetting actions were executed with high precision and accuracy to ensure uniform cell distribution and reagent handling. Data generated by the MO:CROSCOPE module were used to monitor and analyze spheroid morphology and culture evolution throughout the experimental process.\u003c/p\u003e\u003ch3\u003eLiver cell culture\u003c/h3\u003e\u003cp\u003eHepG2 cells (ATCC, HB-8065) were maintained in Dulbecco's Modified Eagle Medium (DMEM) (DMEM, high glucose, Sigma-Aldrich, D0819-500ML) supplemented with 10% Fetal Bovine Serum (FBS) (ThermoFisher Scientific, A5670701) and 1% Penicillin–Streptomycin (P/S) (ThermoFisher Scientific, 15140-148), the DMEM complete medium. Cells were incubated at 37°C and passaged at 70% confluency using Trypsin-EDTA Solution (Himeda, TCL033) to maintain exponential growth. Cells were routinely screened for mycoplasma contamination using a PCR-based detection assay (ThermoFisher Scientific, M7006) and all cell cultures tested negative throughout the study.\u003c/p\u003e\u003ch2\u003eAutomated liver spheroid generation\u003c/h2\u003e\u003cp\u003eFor all experiments, HepG2 cells were used within a defined low passage range (passage 18–25), and the same passage number was applied across biological replicates and assay runs to ensure standardized handling. Before seeding, cells were harvested, counted using a hemocytometer, and resuspended in complete DMEM medium at the desired concentration. Cell seeding was performed with the MO:BOT and an additional plate was seeded manually to evaluate the impact of automated handling on spheroid quality and uniformity.\u003c/p\u003e\u003cp\u003eFor automated spheroid generation, the MO:BOT protocol “Liver spheroids generation and hepatotoxicity testing” routine 1, was selected from the integrated protocol library. Required consumables were positioned on the corresponding mo:re modules (described in section 4.1), including a 96-well U-bottom ULA plate (Facellitate, F202003), a 15 mL tube containing the prepared HepG2 cell suspension, and a box of 1000 µL positive-displacement tips (mo:re GmbH, MO:RE 1000 µL piston tips), and placed on the MO:BOT workfield. Upon protocol initiation, the MO:BOT’s single-channel pipetting tool automatically picked up a tip, aspirated the defined volume of cell suspension, and dispensed 2,000 cells in 150 µL of medium per well to support uniform spheroid formation. After seeding under both handling conditions, plates were incubated at 37°C, 5% CO\u003csub\u003e2\u003c/sub\u003e for 4 days to allow spheroid aggregation. Further details about the MO:BOT system are provided in \u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e and at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://more.science\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e\u003ch2\u003eMO:CROSCOPE, automated bright-field imaging module\u003c/h2\u003e\u003cp\u003eSpheroid morphology was assessed using the MO:CROSCOPE imaging system integrated into the MO:BOT. Cell culture plates were loaded into the MO:CROSCOPE carrier, and brightfield images were acquired under standardized illumination and focus settings using a 5X ZEISS A-Plan objective, IMX477 sensor (Arducam B0279) with an image resolution 2160 × 2160 pixels (center-cropped from full sensor) and pixel size of 0.79 µm/px. The MO:CROSCOPE automatically captured brightfield images of each individual spheroid across a full 96-well plate and analyzed in real-time key morphological parameters, such as size, roundness, and ellipticity. Imaging of a full 96-sample plate occurred within 10 minutes, limiting the time of the plate outside of the incubator. The MO:CROSCOPE served as a quality control check for spheroid growth and was used before and after medium exchange and drug stimulation to monitor potential morphological changes.\u003c/p\u003e\u003ch2\u003eViability assay\u003c/h2\u003e\u003cp\u003eTo assess spheroid quality, cellular ATP content was measured using the CellTiter-Glo 3D Cell Viability Assay (Promega, G9681). The assay was performed following manufacturer’s instruction. For each handling condition, 50 µl of culture supernatant containing the spheroid was transferred from the culture plate to an opaque 96-well assay plate (Corning, 3610) prior to assay reagent addition. Spheroid plates were equilibrated at room temperature (RT) for approximately 30 min prior to reagent addition to optimize cell lysis and signal stability. CellTiter-Glo 3D reagent was added at a 1:1 ratio to the transferred 50 µl spheroid-containing supernatant in each well, and the assay plate was shaken for 5 min at RT at 1,000 rpm to ensure efficient lysis of the 3D structures and thorough mixing of reagent and sample. Plates were incubated for an additional 25 min at RT to allow complete ATP extraction and development of a stable glow-type luminescent signal. Luminescence was measured on a ClarioStar plate reader (BMG Labtech, 1 s integration), and for APAP stimulation values were normalized to DMSO vehicle controls and expressed as percentage viability.\u003c/p\u003e\u003ch2\u003eFunctional assay: Albumin secretion\u003c/h2\u003e\u003cp\u003eAlbumin secretion, a key marker of liver-specific functionality, was quantified in spheroids generated manually and with the MO:BOT using the Human Albumin ELISA Kit (Abcam, ab108788). Kit components and samples were brought to RT, and albumin standards were prepared by serial dilution according to the manufacturer's instructions. After thawing and centrifugation (1,500 rpm, 10 min, 4°C), 50 µL of each supernatant or standard was added to the pre-coated wells and incubated for 1 h at RT. Wells were washed five times with 1X wash buffer, followed by sequential incubations with 50 µl biotinylated anti-albumin antibody and 50 µl streptavidin–HRP conjugate (SP conjugate) for 30 min each, with wash steps between incubations. After a final wash, TMB substrate was added and incubated for 30 min before adding 50 µL of stop solution to each well, and absorbance at 450 nm was recorded using a ClarioStar plate reader. Albumin concentrations were interpolated from the standard curve.\u003c/p\u003e\u003ch2\u003eAutomated Acetaminophen (APAP) dose titration\u003c/h2\u003e\u003cp\u003eTo determine the half-maximal effective concentration (EC₅₀) of acetaminophen (APAP), a dose-response titration experiment was conducted. A 5 M APAP stock solution was prepared in DMSO (Sigma Aldrich, D2438-5X10ML) and subsequently diluted in fresh complete DMEM to the desired working concentrations ranging from 1 mM to 100 mM. Vehicle control wells received complete DMEM containing the corresponding DMSO concentrations without APAP.\u003c/p\u003e\u003cp\u003eThe protocol “Liver spheroids APAP dose titration” was selected from the MO:BOT software library. Cell culture consumables were arranged in the designated MO:BOT work fields, including the liver spheroid culture plate, 1000‑µL tips, seven tubes containing culture medium with different APAP concentrations, one tube containing the vehicle control medium, and the liquid and tip waste modules. Upon protocol initiation, the MO:BOT automatically acquired a fresh tip, aspirated 80% of the culture medium without disrupting spheroid integrity or 3D morphology, and dispensed the respective APAP treatment solutions according to the predefined plate layout, using new tips between conditions to avoid cross‑contamination.\u003c/p\u003e\u003cp\u003eFollowing 24 h incubation under standard conditions, spheroid responses were evaluated based on morphological parameters and cell viability. Dose-response curves were generated from normalized viability data, and EC₅₀ values were calculated by nonlinear regression (log[concentration] vs. response, variable-slope model), yielding an EC₅₀ of 40 mM that was subsequently used in downstream experiments.\u003c/p\u003e\u003ch2\u003eAutomated Medium Exchange for EC\u003csub\u003e50\u003c/sub\u003e APAP stimulation\u003c/h2\u003e\u003cp\u003eAfter establishing 40 mM APAP as the EC₅₀ in liver spheroids, spheroid responses were characterized in both MO:BOT-generated and manually generated spheroids. On day 4 post-seeding, the MO:BOT performed routine 2 of the “Liver spheroids generation and hepatotoxicity testing” protocol. The MO:BOT aspirated the culture medium from each well and replaced it with either complete DMEM containing the corresponding DMSO concentration (vehicle control) or complete DMEM containing 40 mM APAP (treatment), following the workflow described in section 4.7. Plates were then incubated for 24 h before endpoint analyses were performed. Morphological, functional, and viability readouts were assessed in spheroids generated with the MO:BOT and compared with those generated manually.\u003c/p\u003e\u003ch2\u003eCytotoxicity assay in APAP-stimulated liver spheroids\u003c/h2\u003e\u003cp\u003eThe cytotoxicity of spheroids stimulated with APAP from both handling conditions was analyzed using the LDH-Glo Cytotoxicity Assay (Promega, J2380) following the manufacturer’s instructions. The cell culture plates were equilibrated to RT for approximately 30 min before reagent addition to stabilize LDH activity and luminescent output. For each condition, 50 µL of culture supernatant and 50 µL of LDH-Glo detection reagent were combined in an opaque-walled 96-well plate (Corning, 3610), mixed briefly (500 rpm) on a plate shaker, and incubated for 60 min at RT protected from light. Luminescence was recorded using a ClarioStar plate reader, and signal intensity was proportional to extracellular LDH levels.\u003c/p\u003e\u003ch2\u003eMetabolic function: Alanine Aminotransferase (ALT) activity\u003c/h2\u003e\u003cp\u003eDrug-induced hepatocellular toxicity was further evaluated by measuring Alanine Aminotransferase (ALT) activity in culture supernatants using the Human ALT SimpleStep ELISA Kit (Abcam, #ab234578). Reagents, standards, and samples were equilibrated at RT, and the ALT standard curve was prepared as described in the kit protocol. For each well, 50 µL of sample or standard and 50 µL of the antibody cocktail were added, mixed gently, and incubated for 1 h at RT to allow in-well immunocapture. Wells were then washed three times with 1× wash buffer, followed by addition of 100 µL TMB substrate and incubation for approximately 10 min at RT protected from the light. The reaction was stopped by adding 100 µL stop solution, and absorbance was measured at 450 nm using a ClarioStar plate reader. ALT concentrations were calculated from the standard curve and expressed as fold change relative to DMSO vehicle controls, providing a metabolic toxicity readout complementary to viability and LDH release.\u003c/p\u003e\u003ch2\u003eMO:CROSCOPE spheroid image segmentation\u003c/h2\u003e\u003cp\u003eSpheroid boundaries were delineated using the MIT-B0 SegFormer semantic segmentation model fine-tuned on organoid bright-field images deployed as an ONNX model (ONNX Runtime v1.23.0). Morphological features were computed from the segmented binary masks using OpenCV (v4.11.0) and an in-house Python (Python v3.10 ) library. Spheroid area was computed as the number of pixels enclosed by the contour boundary and converted to mm² using the calibrated pixel size. Roundness was calculated as the isoperimetric quotient R = 4πA / P², where P is the contour perimeter, yielding R = 1 for a perfect circle. Ellipticity was defined as the minor-to-major axis ratio of a least-squares fitted ellipse (E = a_minor / a_major), with E = 1 indicating a circle and lower values indicating elongation.\u003c/p\u003e\u003ch2\u003eData analysis\u003c/h2\u003e\u003cp\u003eExperimental data were collected from plate reader output files and MO:CROSCOPE raw data exports and processed using custom R scripts (R v4.4.2). Prior to quantitative image analysis, all MO:CROSCOPE images were screened for artefacts such as out of focus wells, debris, or edge located spheroids, and wells failing these basic quality checks were excluded from further analysis. Raw luminescence and absorbance values were first background corrected using medium or blank wells, and technical replicates were averaged to yield a single value per biological replicate for each condition. Imaging-derived morphological parameters (area, roundness, and ellipticity) were exported from the analysis pipeline described in section 4.4 and linked to treatment metadata for downstream statistical comparisons. Where applicable, area measurements were normalized within each plate by dividing by the median of vehicle control wells (0 mM APAP), yielding fold-change values relative to untreated controls. Dose-response curves were fitted with a four-parameter logistic (4PL) model using log10[concentration] as the independent variable to derive EC₅₀ values and associated 95% confidence intervals.\u003c/p\u003e\u003ch2\u003eStatistical analysis and Manuscript preparation\u003c/h2\u003e\u003cp\u003eGroup comparisons for functional readouts (viability, cytotoxicity, albumin, ALT) were performed using one-way ANOVA when more than two conditions were compared, followed by Tukey's HSD post-hoc test to correct for multiple pairwise comparisons. For two group comparisons (e.g. MO:BOT vs. manual at the same treatment condition), unpaired two tailed Welche’s t test (unequal variances) were applied after confirming approximate normality and homoscedasticity of residuals. Normality was assessed using the Shapiro-Wilk test and homogeneity of variance using Levene's test; when assumptions were violated, non-parametric alternatives (Kruskal-Wallis test with Dunn's post-hoc) were used. Dose-response relationships for APAP titration were modelled by nonlinear regression (log[concentration] vs. response, variable slope) to estimate EC₅₀ values and 95% confidence intervals. Coefficient of variation (CV) values were calculated for each condition as CV = (SD / mean) × 100 across biological replicates to assess assay variability and robustness. Statistical significance was evaluated with a p value \u0026lt; 0.05 considered statistically significant. Significance levels: *p \u0026lt; 0.05, **p \u0026lt; 0.01, ***p \u0026lt; 0.001, ****p \u0026lt; 0.0001.\u003c/p\u003e\u003cp\u003eA large language model-based assistant (Perplexity, powered by GPT-5.1) was used to support manuscript text editing and phrasing. All scientific content, data interpretation and final text were reviewed and approved by the authors.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor contributions statement\u003c/h2\u003e\n\u003cp\u003eD.He. designed and conducted the experiments, analysed the data, and wrote the manuscript. D.S.Z. developed hardware and software for morphological parameter analysis and reviewed the manuscript. F.E. and M.A. developed the moreOS software. P.D., A.S.O., J.K., T.P.H. and D.Ha. designed and developed the MO:BOT hardware. F.K.S. was involved in project communication and dissemination. L.G. supervised the budget and reviewed the manuscript. J.V. led the project and reviewed the manuscript.\u003c/p\u003e\n\u003ch2\u003eCompeting Interests\u003c/h2\u003e\n\u003cp\u003eAll authors are employees of mo:re GmbH, a for-profit organization that develops an automated 3D cell culture and organoid platform, namely the MO:BOT, related to the subject of this research. The research reported here was conducted as part of the authors\u0026rsquo; employment at mo:re GmbH, which provided salary support and research funding. These relationships may be considered competing financial interests. The authors declare that they have no additional financial or non financial competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding Declaration\u003c/h2\u003e\n\u003cp\u003eNo funding was received for conducting this study.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eD.He. designed and conducted the experiments, analysed the data, and wrote the manuscript. D.S.Z. developed hardware and software for morphological parameter analysis and reviewed the manuscript. F.E. and M.A. developed the moreOS software. P.D., A.S.O., J.K., T.P.H. and D.Ha. designed and developed the MO:BOT hardware. F.K.S. was involved in project communication and dissemination. L.G. supervised the budget and reviewed the manuscript. J.V. led the project and reviewed the manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eWe thank Prof. Dr. S\u0026oslash;ren Gersting (University Medical Center Hamburg-Eppendorf, UKE) for valuable scientific input and technical expertise related to cell culture methodologies. A BioRender account was used to create schematic illustrations for this work.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe datasets generated and analyzed during the current study are available from the corresponding authors on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLoewa, A., Feng, J. J. \u0026amp; Hedtrich, S. Human disease models in drug development. \u003cem\u003eNat. Rev. Bioeng.\u003c/em\u003e \u003cb\u003e1\u003c/b\u003e, 545\u0026ndash;559 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeist, M. et al. The biological and ethical basis of the use of human embryonic stem cells for in vitro test systems or cell therapy. \u003cem\u003eALTEX\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 163\u0026ndash;190 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, S., Ooka, M., Margolis, R. J. \u0026amp; Xia, M. Liver three-dimensional cellular models for high-throughput chemical testing. \u003cem\u003eCell. Rep. Methods\u003c/em\u003e. \u003cb\u003e3\u003c/b\u003e, 100432 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMysior, M. M. \u0026amp; Simpson, J. C. An automated high-content screening and assay platform for the analysis of spheroids at subcellular resolution. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e19\u003c/b\u003e, e0311963 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVorrink, S. U., Zhou, Y., Ingelman-Sundberg, M. \u0026amp; Lauschke, V. M. Prediction of Drug-Induced Hepatotoxicity Using Long-Term Stable Primary Hepatic 3D Spheroid Cultures in Chemically Defined Conditions. \u003cem\u003eToxicol. Sci. Off J. Soc. Toxicol.\u003c/em\u003e \u003cb\u003e163\u003c/b\u003e, 655\u0026ndash;665 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProctor, W. R. et al. Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury. \u003cem\u003eArch. Toxicol.\u003c/em\u003e \u003cb\u003e91\u003c/b\u003e, 2849\u0026ndash;2863 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, L. et al. Organoids and organs-on-chips: Recent advances, applications in drug development, and regulatory challenges. \u003cem\u003eMed\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 100667 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZushin, P. J. H., Mukherjee, S. \u0026amp; Wu, J. C. FDA Modernization Act 2.0: transitioning beyond animal models with human cells, organoids, and AI/ML-based approaches. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e \u003cb\u003e133\u003c/b\u003e, e175824 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, J. J. F. D. A. \u0026amp; Modernization Act 2.0 allows for alternatives to animal testing. \u003cem\u003eArtif. Organs\u003c/em\u003e. \u003cb\u003e47\u003c/b\u003e, 449\u0026ndash;450 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUnderhill, G. H. \u0026amp; Khetani, S. R. Advances in Engineered Human Liver Platforms for Drug Metabolism Studies. \u003cem\u003eDrug Metab. Dispos. Biol. Fate Chem.\u003c/em\u003e \u003cb\u003e46\u003c/b\u003e, 1626\u0026ndash;1637 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBircsak, K. M. et al. A 3D microfluidic liver model for high throughput compound toxicity screening in the OrganoPlate\u0026reg;. \u003cem\u003eToxicology\u003c/em\u003e \u003cb\u003e450\u003c/b\u003e, 152667 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHong, S. \u0026amp; Song, J. M. A 3D cell printing-fabricated HepG2 liver spheroid model for high-content in situ quantification of drug-induced liver toxicity. \u003cem\u003eBiomater. Sci.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 5939\u0026ndash;5950 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen, D. G. et al. Bioprinted 3D Primary Liver Tissues Allow Assessment of Organ-Level Response to Clinical Drug Induced Toxicity In Vitro. \u003cem\u003ePloS One\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, e0158674 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchofield, C. A. et al. Evaluation of a Three-Dimensional Primary Human Hepatocyte Spheroid Model: Adoption and Industrialization for the Enhanced Detection of Drug-Induced Liver Injury. \u003cem\u003eChem. Res. Toxicol.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 2485\u0026ndash;2499 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaycock, J. W. 3D Cell Culture: A Review of Current Approaches and Techniques. in 3D Cell Culture (ed Haycock, J. W.) vol. 695 1\u0026ndash;15 (Humana, Totowa, NJ, (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBell, C. C. et al. Characterization of primary human hepatocyte spheroids as a model system for drug-induced liver injury, liver function and disease. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 25187 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, Y., Shen, J. X. \u0026amp; Lauschke, V. M. Comprehensive Evaluation of Organotypic and Microphysiological Liver Models for Prediction of Drug-Induced Liver Injury. \u003cem\u003eFront. Pharmacol.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 1093 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTutty, M. A., Movia, D. \u0026amp; Prina-Mello, A. Three-dimensional (3D) liver cell models - a tool for bridging the gap between animal studies and clinical trials when screening liver accumulation and toxicity of nanobiomaterials. \u003cem\u003eDrug Deliv Transl Res.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 2048\u0026ndash;2074 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFey, S. J. \u0026amp; Wrzesinski, K. Determination of drug toxicity using 3D spheroids constructed from an immortal human hepatocyte cell line. \u003cem\u003eToxicol. Sci. Off J. Soc. Toxicol.\u003c/em\u003e \u003cb\u003e127\u003c/b\u003e, 403\u0026ndash;411 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBasharat, A., Rollison, H. E., Williams, D. P. \u0026amp; Ivanov, D. P. HepG2 (C3A) spheroids show higher sensitivity compared to HepaRG spheroids for drug-induced liver injury (DILI). \u003cem\u003eToxicol. Appl. Pharmacol.\u003c/em\u003e \u003cb\u003e408\u003c/b\u003e, 115279 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Z. et al. Generation of hepatic spheroids using human hepatocyte-derived liver progenitor-like cells for hepatotoxicity screening. \u003cem\u003eTheranostics\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 6690\u0026ndash;6705 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKulsharova, G. \u0026amp; Kurmangaliyeva, A. Liver microphysiological platforms for drug metabolism applications. \u003cem\u003eCell. Prolif.\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e, e13099 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJeon, H., Kim, G., Carpenter, J., Col\u0026oacute;n, Y. J. \u0026amp; Wang, Y. Automated High-Content, High-Throughput Spatial Analysis Pipeline for Drug Screening in 3D Tumor Spheroid Inverted Colloidal Crystal Arrays. \u003cem\u003eACS Appl. Mater. Interfaces\u003c/em\u003e. \u003cb\u003e17\u003c/b\u003e, 49210\u0026ndash;49226 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVilas-Boas, V. et al. Primary Human Hepatocyte Spheroids as Tools to Study the Hepatotoxic Potential of Non-Pharmaceutical Chemicals. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e, 11005 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShin, D. S. et al. Hepatotoxicity evaluation method through multiple-factor analysis using human pluripotent stem cell derived hepatic organoids. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 10804 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElje, E. et al. Hepato(Geno)Toxicity Assessment of Nanoparticles in a HepG2 Liver Spheroid Model. \u003cem\u003eNanomaterials\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 545 (2020).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9116861/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9116861/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eToxicity is a major cause of drug failure and post-marketing withdrawals, making sensitive and reproducible toxicity readouts essential for modern drug development. FDA Modernization Act 3.0 and related strategic roadmaps explicitly encourage the use of human-relevant in vitro models as New Approach Methodologies (NAMs) as an alternative to animal models. Yet, current preclinical drug discovery and toxicity testing with human \u003cem\u003ein vitro\u003c/em\u003e models still depend on manual three-dimensional (3D) spheroid workflows that are challenging to standardize and scale. Although 3D human models better capture tissue-like architecture, cell-cell interactions, and organ-specific functions, 3D culture protocol standardization is precluded by user-dependent protocols, variable spheroid size and morphology, and fragmented steps for seeding, maintenance, and compound dosing. This lack of standardization weakens assay robustness, complicates cross-study comparison, and slows broader adoption in decision-making pipelines. In responses to these needs, we present the MO:BOT, a modular automation platform that unifies 3D organoid and spheroid seeding, medium exchange, image-based quality control, and compound dosing, within a single enclosed system. In this article, we present one use case application focused on an automated workflow to generate HepG2 liver spheroids and screening for acetaminophen (APAP) drug-response. In direct comparison with a manual workflow, the MO:BOT reduced the well-to-well variability in liver spheroid size, stabilized spheroid area over time, and yielded spheroids with higher viability. Automated medium exchange preserved spheroid architecture, demonstrating that appropriately tuned pipetting routines can maintain the integrity of delicate 3D structures. A MO:BOT automated APAP dose–response workflow produced a clear, sigmoidal hepatotoxicity profile with an EC\u003csub\u003e50\u003c/sub\u003e in the clinically relevant range and concordant changes in viability, LDH release, and ALT activity are closely matching the responses. Together, these findings show that the fully automated 3D liver spheroid workflow implemented on the MO:BOT improves culture uniformity and assay readout precision, providing a standardized, scalable foundation for deploying human 3D human models as robust, decision-enabling tools in modern preclinical testing.\u003c/p\u003e","manuscriptTitle":"Automation of 3D Liver Spheroid generation and Acetaminophen dose– response on the MO:BOT enhances assay robustness and precision","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-24 13:21:05","doi":"10.21203/rs.3.rs-9116861/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-27T07:34:36+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-25T14:41:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"40871492793918376279800805765014350500","date":"2026-04-16T13:33:41+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-01T12:13:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"178508802717277995059928741652524448387","date":"2026-03-23T08:14:08+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-23T02:12:41+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-20T19:26:31+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-17T02:45:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-17T02:45:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-13T16:46:48+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7503e04e-a1d0-4530-93d2-118404448795","owner":[],"postedDate":"March 24th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64906390,"name":"Biological sciences/Biological techniques"},{"id":64906391,"name":"Biological sciences/Biotechnology"},{"id":64906392,"name":"Biological sciences/Drug discovery"},{"id":64906393,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-12T03:39:47+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-24 13:21:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9116861","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9116861","identity":"rs-9116861","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}