IntroductionDroplet-based microfluidics has emerged as a powerful tool for miniaturized, high-throughput biological experimentation, enabling precise control over cellular environments, single-cell resolution, and massively parallel assays. By compartmentalizing reactions or individual cells into nanoliter-sized droplets, these platforms facilitate highly reproducible, scalable studies that would be difficult or impossible using conventional bulk methods. While this technology has seen wide adoption in biomedical and microbial research1,2, its application in plant biology, particularly for fragile, wall-less cells such as protoplasts, remains limited and does not include prolonged cultivation3,4,5.Protoplasts, isolated plant cells without their cell walls, are inherently sensitive and highly responsive to microenvironmental cues. These properties make them both technically challenging and biologically informative as a model system for studying cell viability, division, and differentiation. Beyond their utility in basic research, protoplasts serve as a versatile platform in plant biotechnology, enabling somatic hybridization, transient expression assays, and stable transformation through regeneration into whole plants. They are widely used in DNA delivery and genome editing workflows, providing a direct route to study gene function and engineer new traits without the constraints imposed by cell walls6,7,8. Traditionally, protoplast research has relied on static culture systems that offer limited control over spatial and temporal conditions, making it difficult to study heterogeneous responses at the single-cell level or in dynamic conditions9,10,11.Here, we present an improved droplet-based microfluidic platform, previously demonstrated for successful application in microspore embryogenesis investigation12. This system allows for the encapsulation, tracking, and longitudinal observation of individual protoplasts or small cell clusters within discrete aqueous droplets with precise control over droplet volume and chemical composition. Our primary objectives were to (i) assess the suitability of this microfluidic approach for maintaining plant protoplast viability over extended culture periods, (ii) evaluate its capacity to monitor early cell division events at nearly single-cell resolution, and (iii) explore the influence of low growth regulator concentrations on developmental outcomes.To address these objectives, we employed protoplasts from three dicotyledonous species selected for their complementary experimental properties. Brassica juncea was included due to its straightforward protoplast isolation and initial suitability for feasibility testing. Nicotiana tabacum, a well-established model organism in plant cell biology, provided a benchmark for system performance and experimental reproducibility. Kalanchoe daigremontiana was chosen for its natural propensity for somatic embryogenesis, offering a biologically relevant context to explore developmental potential in microfluidic culture. By using these species as a test case, we demonstrate the platform’s capacity to reveal species-specific sensitivities, growth dynamics, and viability trends, as well as seek to identify technical constraints and optimization strategies for long-term culture. This work not only advances technical capabilities for plant single-cell analysis but also establishes a framework for broader applications in regenerative plant biotechnology and precision phenotyping.Materials and methodsPlant culturePlant material used in this study included Kalanchoe daigremontiana Raym.-Hamet & H. Perrier plants (original plant was obtained in 2021 from a horticultural source in Ilmenau, Germany and maintained clonally in the TU Ilmenau greenhouse. The species identity was verified morphologically by leaf-morphology diagnostics (abaxial spotting/banding)), Nicotiana tabacum cultivar Samsun NIC 422, and Brassica juncea CR 2664 seeds obtained from the gene bank of the IPK Gatersleben, Germany. Seeds were surface disinfected by immersion in 70% ethanol for 30 s, followed by treatment with 2% sodium hypochlorite (NaOCl) containing a drop of Tween-20 for 5 min. They were then rinsed three times with sterile distilled water for 5 min each. K. daigremontiana plantlets were disinfected using the same NaOCl solution for 10 min, omitting the ethanol step. Following disinfection, seeds and plantlets were transferred to Murashige and Skoog (MS) medium containing 30 g·L− 1 sucrose and solidified with 4 g·L− 1 gelrite. Germination and cultivation were conducted in a controlled environment at 24–25 °C with a 16 h light/8 h dark photoperiod and a light intensity of approximately 5.0 × 10− 5 mol·m⁻²·s⁻¹. N. tabacum plants were subcultured every 3–4 weeks. B. juncea plants were freshly sown for each experiment and used at 2–3 weeks of age. K. daigremontiana plants were propagated vegetatively every 3 months.Isolation of protoplastsFor standard droplet cultivation experiments, protoplasts were isolated via enzymatic digestion from the top leaves of N. tabacum, B. juncea and K. daigremontiana plants of various age (N. tabacum: 3–4 weeks old, B. juncea: 2 weeks old, K. daigremontiana: 3 months old) following procedure described by13. Briefly, leaves were cut into small pieces on sterile sheets of paper and incubated in preplasmolysis solution for 1 h. The solution was then replaced with 5 mL of enzyme mixture (1.6% cellulase and 0.8% macerozyme in BNE9 solution, previously stirred, centrifuged, and filter sterilised), and samples were incubated for 15–17 h at 27 °C in darkness, followed by an additional 30 min of gentle shaking at room temperature (36 rpm). The digested material was filtered through 100 μm sieves (Greiner Bio-One GmbH) into 13 mL round-bottom centrifuge tubes. An equal volume of 20% sucrose solution (5 mL) was added, mixed gently by inversion, and overlaid with 1.5 mL of washing solution to form two phases. After centrifugation at 760 g for 5 min, protoplasts collected at the interphase were transferred to fresh tubes, washed twice with 5 mL of washing solution (centrifuged at 700 g for 5 min each), and finally resuspended in 2 mL of 8pm7 cultivation medium. Based on cell counts using an Anvajo cell counter, the initial protoplast culture was adjusted to a density of 1.5–4 × 10⁵ cells·mL− 1.To accommodate the distinct experimental requirements, two protoplast isolation protocols were applied: one for cross-species comparison and one optimized for N. tabacum in the dose–response experiments. For growth regulators dose–response experiments, which were conducted exclusively with N. tabacum, protoplasts were isolated from young leaves using a modified version of the protocol described by14. The initial disinfection step differed: leaves were halved and incubated in 2% PPM (Plant Preservative Mixture, Gentaur GmbH) for 1–2 h at room temperature with gentle shaking, then rinsed with sterile water. All subsequent steps – including cutting, enzymatic digestion (using 0.25% cellulase and 0.25% macerozyme in F-PIN solution), filtration, two-phase separation, and washing – were performed as described above, with minor adjustments: digestion was carried out for 14 h at 26 °C in darkness, followed by 30 min of shaking at room temperature. Isolated protoplasts were resuspended in 4 mL of F-PCN medium without growth regulators at a concentration of 1.6 × 10⁵ cells·mL− 1. For establishing a concentration range of hormonal treatments, F-PCN medium containing 4 mg·L− 1 BAP and NAA was prepared and serially diluted as needed. The composition of all media used during protoplast isolation and culture is listed in Supplementary Table S1.Microfluidic setup for cultivation in microdropletsThe experimental setup was based on previously described microfluidic systems12,15, with several modifications (Fig. 1). Droplets were generated using a 6-port manifold, with flow rates precisely controlled by a multi-syringe pump (NEMESYS, Cetoni GmbH, Germany). For the standard droplet generation system, 1 mL glass syringes (SETonic GmbH, Germany) were used to deliver the cultivation medium and any effector solutions, while a 2.5 mL glass syringe was used for the carrier phase (PP9). All elements of the system were connected through fluorinated ethylene propylene (FEP) tubing with an inner diameter (ID) of 0.5 mm and outer diameter (OD) of 1.6 mm. This configuration ensured precise flow control.To minimize mechanical stress on the protoplasts, cell suspensions were not directly injected. Instead, PP9 was introduced into a glass container holding the protoplast suspension, displacing the contents via a smaller polytetrafluoroethylene (PTFE) tube (ID 0.3 mm) toward the droplet generator. The remaining two syringes were used to deliver cell culture medium (for adjusting cell density) and, when applicable, specific effectors. Standard flow rates were approximately 20 µL·min⁻¹ for the aqueous phase and 30 µL·min⁻¹ for the continuous PP9 phase. Following droplet formation (typically 120–300 nL per droplet), they were directed into PTFE incubation tubing with thinner walls (ID 0.5 mm, OD 1.0 mm) to improve air perfusion. After formation, droplets were incubated in the sealed tubes in darkness at 24 °C.To assess and ensure comparability between experiments, the initial protoplast suspension used for droplet generation was adjusted to 1.5–4 × 10⁵ cells·mL⁻¹ (standard droplet cultivation) or 1.6 × 10⁵ cells·mL⁻¹ (dose response experiments). In the droplet-based system, precise control of cell density is inherently limited because the generation flow rate is relatively slow (20 µL·min⁻¹ for the aqueous phase and 30 µL·min⁻¹ for the continuous phase), and the suspension was not continuously stirred to avoid generation of additional shear stress, leading to small fluctuations in the number of encapsulated cells. Image-based estimation of actual cell densities within droplets yielded values between 1.06 × 10⁵ and 2.7 × 10⁵ cells·mL⁻¹, confirming that the variation remained within a narrow and biologically comparable range. For reference Petri dish cultures, a lower initial cell density (1–2 × 10⁴ cells·mL⁻¹) was applied to minimize cell aggregation and to better mimic the effective single-cell separation achieved in the droplet system.Fig. 1Experimental microfluidic setup. Droplets were generated via a 6-port manifold allowing delivery of up to two effectors. Protoplasts were supplied indirectly from a suspension container. All flows were controlled by a syringe pump, and droplet formation was monitored by digital microscope. Cultures were incubated in sealed tubing on a frame for imaging. Illustration created with BioRender.com.Full size imageReference cultivation in petri dishesIn parallel with droplet-based experiments, bulk protoplast cultures were prepared as controls under identical conditions. Freshly isolated protoplasts were transferred into 6 cm Petri dishes containing 4 mL of respective cultivation medium (8pm7 or F-PCN) or into 12-well plates with 1 mL of medium per well, supplemented with varying concentrations of the tested effectors. Cell density was adjusted and maintained in the range of 1–2 × 104 cells·mL− 1.Growth characterization and data processingProtoplast cultures were monitored at regular intervals (every 1–3 days) using a digital microscope (Keyence VHX-5000) to assess cell morphology and detect the onset of cell division. In the dose response screening experiment, respective droplets could be found back based on their initial position in the incubation tubing. For quantitative analysis, the number and diameter of cells within individual droplets were measured manually using ImageJ (Fiji) software.Statistics and reproducibilityEach experiment was conducted in duplicate, with 5–15 droplets analysed per condition in each run. Statistical analyses were performed in Jamovi (version 2.7.5).For K. daigremontiana protoplast size measurements, differences across cultivation days (0, 2, 5, 7) were assessed using the Kruskal–Wallis test, as normality was violated (Shapiro–Wilk, p