Abstract Cancer organoids have improved our understanding of recapitulating the histology, genotypes, and drug response of patient tumors for personalized medicine. However, the existing cancer organoids are typically grown in animal-derived matrices (e.g., Matrigel), which suffer from poor reproducibility and low throughput due to uncontrollable origin of seed cells, undefined matrix, and manual manipulation. Here, we report a new strategy to massively generate uniform pancreatic cancer organoids (PCOs) in a droplet system from single cells. This system is composed of all-in-water fluids that allow to mildly encapsulate single tumor cell into isolated droplet, which subsequently proliferate and self-assemble into organoids, resembling the initial state of a tumor in the body. This high-throughput method can produce thousands of organoids in a single batch. The droplets can serve as templates for synthesizing defined microgels with proper stiffness similar to that of native tumors, facilitating functional expressions of PCOs. These organoids exhibit superior uniformity and controllability in terms of size and morphologies compared with organoids cultured in manually dropped Matrigel, due mainly to the controllable number of initiating cells and defined microgels. In addition, the established organoids maintain the key biomarkers of pancreatic tumor (e.g., KRT7, KRT19 and SOX9) and higher expression of genes associated with drug metabolism confirmed by RNA-seq and PCR analysis. Furthermore, they show distinguishing responses to four clinically used drugs in a reproducible manner in automatic pipetting workstation, indicating the feasibility of the proposed method in high-throughput drug testing. The established strategy has integrated the formation, 3D cultures, and analysis of PCOs derived from single cells in a whole system, which may provide a novel platform for advancing organoids research with standardized procedure in translational applications. Keywords: Pancreatic cancer, Uniform organoids, Droplet microfluidic, Defined microgels, Drug testing Graphical abstract [39]Image 1 [40]Open in a new tab 1. Introduction Patient-derived organoids have emerged as a preferred model for predicting cancer treatments, as they can be generated from tumor biopsy within one month and maintain the histopathological and genomic profiles of the patients [[41]1,[42]2]. These organoids retain tumor tissue heterogeneity and responses to drugs, making them valuable for high-throughput drug screening [[43]3]. Generally, organoids have many advantages over traditional 2D cultures, as they display near-physiological cellular composition and behaviors. Many organoid cultures can undergo extensive expansion in culture and maintain genome stability, which makes them suitable for biobanking [[44][4], [45][5], [46][6]]. Compared to patient-derived xenografts and animal models, organoids can reduce experimental complexity, facilitate precision genetic and imaging techniques, and, more importantly, enable the study of human development and diseases that is not feasible in animals [[47][7], [48][8], [49][9]]. A prominent example is pancreatic cancer, a leading cause of cancer-related mortality worldwide with a 5-year survival rate of 12 %, the lowest among recorded tumors [[50]10]. This cancer presents various subtypes, complex etiologies, and diverse patient responses to therapies, posing significant challenges for effective treatment. Moreover, pancreatic cancer often remains asymptomatic until reaching advanced stages due to the pancreas' deep abdominal location. Therefore, the development of preclinical tumor models with high fidelity has garnered considerable attention for advancing research into pancreatic cancer pathogenesis and developing effective diagnostic and therapeutic strategies. In this milieu, pancreatic cancer organoids (PCOs) were generated in 2013 by activating WNT signaling and expressing LGR5 in mouse pancreatic duct fragments [[51]11]. Since then, PCOs have been extensively utilized to replicate tumor progression, evaluate novel anticancer drugs, and promote precision medicine approaches for pancreatic cancer [[52][12], [53][13], [54][14]]. Despite their potential as valuable models of cancer biology, PCOs exhibit several limitations that hinder their reproducibility, throughput, and clinical translation. Typically, the initial number and condition of seeded cells cannot be precisely controlled in the traditional tumor organoids, leading to the random development of organoids. In addition, to support the generation and 3D culture, pancreatic cancer fragments or cells are usually encapsulated in murine Engelbreth-Holm-Swarm (EHS) matrix (e.g., Matrigel) by manual manipulation. Although such matrices can provide a rich milieu of tumor-derived extracellular matrix (ECM) components, growth factors and cytokines, they exhibit remarkable batch-to-batch variability and contain xenogenic impurities that can unpredictably influence organoid phenotype [[55]15]. Furthermore, the Matrigel, with relatively low stiffness (typically less than 1 kPa), fails to reconstruct biophysical microenvironment of pancreatic cancers that showed a Young's modulus of 4∼6 kPa or even higher [[56][16], [57][17], [58][18]]. Also, the manual manipulation of PCOs increases the demand of primary tumor tissues for organoids establishment and limits the compatibility of organoids in rapid and high-throughput drug screening. Recently, droplet microfluidics have been explored to tumor organoid field, usually in terms of organoid culture, formation and drug testing. The combination of organoids and droplet microfluidics partially addressed the challenges of existing PCOs, such as realizing high-throughput generation and manipulation of organoids, and controlling the components and morphology of organoids ECM. By adding prepolymers and patient-derived cells into droplets, various microgels can be produced to form tumor organoids after cellular proliferation and self-organization [[59][19], [60][20], [61][21], [62][22], [63][23], [64][24]]. These organoids have been used for chemotherapy [[65]19,[66]22] and immune therapy [[67]21] testing to show their potential in precision medicine. Nevertheless, the existing technologies often fail to control the initial number of cells in each droplet. This have led to significant variation of the generated organoids, or even generation of more than one organoid in each droplet, which need to be addressed in this filed. Traditionally, alginate hydrogels were used as 3D cell scaffolds in tissue engineering due to their flexibility and biocompatibility that allowed the survival and growth of cells [[68]25]. In addition, alginate is well suited for droplet microfluidics because of its fast and mild cross-linking progress triggered by divalent cationic ions (e.g. Ca^2+) [[69]26]. Also, alginate was reported to be used for precisely encapsulating one single cell in each microgel produced in droplet microfluidic device [[70]27], which holds great potential in increasing the controllability and uniformity of the existing tumor organoids. Here, we provide a new strategy for scalable generation of uniform PCOs in a droplet system from single tumor cells. The clinical biopsies are dissociated and filtered for the construction of PCOs ([71]Fig. 1A). These organoids, derived from single cells, are cultured in defined microgels composed of alginate matrices with tailored stiffness, facilitating high-throughput organoid production and manipulation ([72]Fig. 1B and C). The utilization of single cells would enhance the controllability of seeded cell numbers, and replicate the initial state of a tumor in the body. The developed strategy effectively increases the uniformity of organoids in terms of their size and morphologies and decreases the number of cells demanded for organoids establishment. We systematically characterize the expression of tumor-related biomarkers and genes in organoids by using immunofluorescence and qPCR, comparing them with organoids cultured in Matrigel or primary tumor tissues. We also investigate and validate differential gene expression of PCOs through RNA-seq analysis ([73]Fig. 1D). Finally, we employ PCOs in high-throughput anti-cancer drug testing with automatic pipetting workstation, demonstrating the feasibility of the proposed strategy for preclinical applications. Fig. 1. [74]Fig. 1 [75]Open in a new tab Schematic diagram of droplet system for rapid generation of uniform PCOs from single cells. (A). The process of collecting dispersed single cells from clinical biopsy. (B). The droplet system to effectively encapsulate single tumor cells. (C). The 3D culture, manipulation, and high-throughput drug testing of the encapsulated homogeneous PCOs developed from single cells. (D). The output data acquired by using the loaded organoids, including PCR, immunofluorescence, RNAseq, and drug testing, etc. 2. Materials and methods 2.1. Establishment of the droplet system The multi-layered microfluidic chip was fabricated using standard soft lithography and micromolding approaches. Briefly, the polydimethylsiloxane (PDMS, Dow Corning Corporation, 02085925) at a weight ratio of 10 : 1 was molded from a photoresist (SU-8 3035, MicroChem Corporation, Y311074) templates after being cured at 80 °C for 2 h in an oven to form structured top and middle layers of the chip. Subsequently, the top layer was bonded to the middle layer that was bonded to a plain PDMS block after punching with a diameter punch of 1.5 mm and oxygen plasma treatment. The height and width of the microchannels in top layer are both 300 μm, while the height and width of the microchannels in middle layer are both 100 μm. In addition, the thickness and diameter of pneumatic valve are 200 μm and 1 mm, respectively, which could control the generation of droplet templates. 2.2. Cell culture Human pancreatic cancer cell lines, 1.1B4 and Panc-1, were purchased from Meisen Chinese Tissue Culture Collections and Shanghai Zhong Qiao Xin Zhou Biotechnology Co.,Ltd, respectively. These cells were cultured using Dulbecco's Modified Eagle Medium (DMEM, high glucose, Gibco, 11965092) supplemented with 10 % fetal bovine serum (FBS, Gibco, 16170078) and 1 % streptomycin and penicillin (Sigma-Aldrich, P0781). They were cultured in a humidified incubator composed of 5 % CO[2] at 37 °C and passaged until achieved 80 %∼90 % confluence. 2.3. Human samples Pancreatic cancer surgical samples were collected at the First Affiliated Hospital of Dalian Medical University after informed patient consent under a protocol approved by the Ethics Committee of the First Affiliated Hospital of Dalian Medical University (Approval number: PJ-KS-KY-2023-162). 2.4. PCOs culture PCOs were established from resected pancreatic cancer surgical samples and protocols were followed by a previous report [[76]28] after modification. Briefly, tumor tissue was minced and digested with collagenase II (5 mg/mL, Sigma, 1148090) in human complete medium (see below) at 37 °C for 1 h on an orbital shaker. For conventional organoid formation, passage, and conservation, the collected cells were directly embedded in pure Matrigel (Corning, 354230) drops (30 μL/drop) and seeded on 6-well culture plates (Corning, 353046). For the subsequent encapsulation of single cells in microgels, the collected cells were embedded in pure Na-alginate (NaA, 50 kDa, Pharmaceutical grade) solution (1 % (w/v) NaA dissolved in physiological saline) and dropped into 1 % (w/v) CaCl[2] (Aladdin, C118704) to form alginate hydrogels (30 μL/drop). Here, the NaA was customized from Qingdao Hyzlin Biology Development Co., Ltd ([77]http://www.hyzlin.com/), which was screened with 80 mesh and its viscosity was 55 cps at the concentration of 1 % (w/v). The physiological saline was prepared by dissolved NaCl (Tianjin Damao Chemical Reagent, analytically pure, 20200908) in water at the concentration of 0.9 % (w/v). Then, the cells were culture with complete medium containing 50 % AdDMEM/F12 medium (Thermo Fisher Scientific, 12634010), Glutamax (1x, Gibco, 35050061), B27 (1x, Gibco, 10889038), R-spondin 1 (100 ng/mL, R&D, AF4645), Noggin (100 ng/mL, R&D, 6057-NG), Wnt3a condition medium (1x, MBL, J2-001), N-acetyl-L-cysteine (1 mM, Sigma, A9165), Nicotinamide (10 mM, Sigma, N0636), epidermal growth factor (EGF, 50 ng/mL, Peprotech, GMP100-15), fibroblast growth factor 10 (FGF10, 100 ng/mL, Prepotech, 100-26), and A83-01 (0.5 μM, Selleck, S769201). The cells formed organoids within 1∼2 weeks to remove cells other than tumor cells, such as immune cells, fibroblasts, or other cells by starvation in the complete medium before single cell encapsulation. For passage, organoids within Matrigel drops were first resuspended in the culture plates by using a pipette before transformed to the centrifuge tube and centrifuged for 10 min at 1500 rpm (Eppendorf, 5804). Then, the precipitate was digested with TrypLE (Gibco, 12605010) and resuspended with AdDMEM/F12 medium before centrifuged for 10 min at 1500 rpm. Finally, the digested cells and clusters were embedded in pure Matrigel drops and seeded on 6-well culture plates for the next culture using the complete medium. Organoids within alginate hydrogels were first digested with alginate lyase (Sigma, A1603) at a concentration of 10 U/mL for 10 min to remove the hydrogels before transformed to the centrifuge tube and centrifuged for 5 min at 1500 rpm. Then, the precipitate was digested with TrypLE and resuspended with AdDMEM/F12 medium before centrifuged for 5 min at 1500 rpm. Finally, the digested cells and clusters were embedded in pure alginate drops and seeded on 6-well culture plates for the next culture using the complete medium. 2.5. Pretreatment of single cells before encapsulation Pancreatic cell lines and the formed PCOs within alginate hydrogels were dissociated by 0.25 % trypsin-EDTA (Gibco, 25200056) and TrypLE, respectively, then filtered into isolated cells using cell strainer (Falcon, 352340) and treated with CaCO[3] nanoparticles (40–80 nm, XFNANO, XFI11-1) according to the previous report [[78]27] with minor modification. Briefly, CaCO[3] nanoparticles were suspended in CaCl[2]-free DMEM medium (Gibco, 21068028) at a concentration of 20 mg/mL, which was treated with ultrasound using an ultrasonic cell disruptor (SCIENTZ-IID, SCENTZ) and filtered to remove the free nanoparticles with a 0.22 μm syringe filter (Millex Syringe Filter, SLGPR33RB) and obtain CaCO[3] saturated DMEM (named “DMEM-Ca”). Then, 20 mg/mL CaCO[3] suspension was prepared using DMEM-Ca after ultrasonic treatment. Next, the cell suspension was washed with DMEM-Ca by centrifugation at 200×g for 3 min and resuspended with DMEM-Ca at a concentration of 2 × 10^6 cells/mL, which was mixed with 20 mg/mL CaCO[3] suspension at a ratio of 1:1. The mixture of cells and CaCO[3] was incubated for 15 min at cell culture condition on a shaker before centrifuged at 40×g for 5 min. The cells loaded with CaCO[3] were washed with DMEM-Ca and centrifuged at 40×g for twice to remove the free CaCO[3]. The pre-treated cells were resuspended in core flow solution with a density of 2.5 × 10^5 cells/mL for encapsulation experiments. 2.6. Encapsulation of single cells using microgels The core flow solution was prepared by mixing cell suspensions with a concentrated NaA solution (22.5 % (w/v) dextran (DEX) (500 kDa, GE, 17-0320-02) and 1.5 % (w/v) NaA dissolved in physiological saline) at a ratio of 1:2. The final solution contained 15 % (w/v) DEX and 1 % (w/v) NaA. And 15 % (w/v) polyethylene glycol (PEG) (20 kDa, Aladdin, P103730) and 15 % (w/v) PEG with 0.4 % (v/v) acetic acid (Aladdin, A298827) were also dissolved in physiological saline as the middle and shell flow, respectively. All these solutions were filtered with a 0.22 μm syringe filter before being pumped into the microfluidic chip device. The core, middle and shell flows were pumped into their inlets of the pneumatic valve-integrated chip by using pressure pumps (FLOW-EZ, Fluigent, Beijing E-Science Co., Ltd). In this microfluidic system, the pneumatic valve was pulled periodically as the constant negative pressure (−60 kPa) was connected and unconnected under the control of external solenoid valve, thus leading to a droplet templates generation at the end of a switch cycle at the first junction of chip. Then, droplet templates were delivered into shell flow by middle flow at the second junction, wherein the Ca^2+ of CaCO[3] was released by contacting to the acetic acid, and electrostatic complexation of NaA and Ca^2+ occurred to form microgels. The generated cell-laden microgels were collected in physiological saline/culture medium before further gelation with 1 % (w/v) CaCl[2] for 10 min. The cells in microgels were finally transferred to cell culture medium within a 96-well culture plate (Corning, 3474) and cultured in a humidified incubator with 5 % CO[2] at 37 °C with the fresh medium being changed every other day. As control group, the formed PCOs within Matrigel were digested with TrypLE and filtered to be isolated cells, resuspended in pure Matrigel drops (30 μL/drop), and seeded on 6-well culture plates at the concentration of 2.5 × 10^5 cells/mL. Then, the culture plates were placed upside down in a humidified incubator with 5 % CO[2] at 37 °C to keep the morphology of dome Matrigel drops that were solidified after 30 min. Finally, culture medium was added in the culture plates with the fresh medium being changed every other day. 2.7. Diameter measurement of microgels and organoids The implementation of size measurement of microgels and organoids are using Image J software ([79]http://rsb.info.nih.gov/ij/). Specifically, the long axis and short axis of microgels/organoids are manually measured using the optical images, and the average of long axis and short axis is used as the diameter of the given microgels/organoids. 2.8. Young's modulus measurement of microgels The microgels fabricated using different concentrations (0.5 %, 1 % and 2 %) of NaA were collected and fixed at the bottom of confocal dish (Biosharp, BS-20-GJM) coated with 1 % (w/v) chitosan (100–200 cps, Aladdin, C105799) that was dissolved in water with pH = 5 adjusted by acetic acid. Then, the Young's modulus of microgels were measured by using a Piuma Nanoindenter (Optics11) with a probe of appropriate range. Each sample was measured at least in triplicate. Finally, the primary tumor tissues and Matrigel were also tested on their Young's modulus for comparation with that of the produced microgels. 2.9. Scanning electron microscope (SEM) analysis of microgels Hydrous microgels were freeze-dried in a lyophilizer (SCIENTZ-10ND, China) for at least 4 h. Then the microgels were coated with a layer of gold for 60 s with a sputter coater (SBC-12, KYKY) at the electricity of 8 mA. The size and surface morphology of the microgels were characterized using a SEM (Hitachi TM3000, Japan) at 15 kV. To investigate the internal structure of microgels, the core flow solution without cells was manually dropped into a collection tank of 1 % (w/v) CaCl[2] to form large microgels that were dissected before SEM analysis. In addition, pure NaA solution (1 % (w/v) NaA dissolved in physiological saline) was prepared and dropped into 1 % (w/v) CaCl[2] to form pure alginate hydrogels and Matrigel was dropped into empty culture plates for gelation at 37 °C, respectively. Here, the pure alginate hydrogels and Matrigel were used as control groups to microgels. The implementation of porosity/sectional area of pore measurement of microgels, Matrigel and pure alginate hydrogels are using Image J software ([80]http://rsb.info.nih.gov/ij/). 2.10. Endotoxin testing of core flow solution with different concentrations of NaA The content of endotoxin in core flow solution with different concentrations of NaA was tested by using endotoxin assay kit (GenScript, L00350C). Core flow solutions with 15 % DEX and varied NaA, including 0.5 %, 1 %, 2 % and 3 %, were prepared and the endotoxin was tested according to the user's manual, and the absorbance at 545 nm was measured with a microplate reader (Tecan Infinite M Nano). 2.11. Live/dead testing of encapsulated cells The proportion of viable cells within microgels was evaluated by using live/dead assay. The cell-laden microgels were incubated in cell culture medium with ethidium homodimer-1 (red, dead, 1:500) and calcein-AM (green, live, 1:1000) (LIVE/DEAD viability/cytotoxicity assay kit, Gibco, L3224) at 37 °C for 25 min. Then, they were rinsed with physiological saline before imaging under a laser scanning confocal microscope (FV3000, Olympus). The viability percentage was determined by counting the number of live cells (green fluorescence), which was divided by the total number of live (green fluorescence) and dead (red fluorescence) cells at a given focal plane. In addition, the viability percentage of cells before encapsulation was measured using the same method. 2.12. Fluorescence immunohistochemistry of organoids Pancreatic cell spheroids and PCOs in microgels, as well as PCOs in Matrigel were soaked in 4 % paraformaldehyde (PFA) (MeilunBio, MA0192) for 20 min at room temperature to be immobilized after cultivation of 10 days. Then, they were permeabilized with 0.2 % Triton X-100 (Amresco, SH11087) for 10–15 min and blocked in a blocking solution (ZSGB-BIO, ZLI-9022) for 1 h at room temperature, successively. Next, these cells were incubated at 4 °C overnight with the required primary antibodies that are listed in the supplementary information ([81]Table S1), which were diluted with antibody diluent (ZSGB-BIO, ZLI-9028) in the experiment. Then, the second antibodies of IgG H&L (Alexa Fluor®, ab150077, ab150080, ab150113, and ab150116, Abcam) were used to treat the aforesaid cells at a dilution rate of 1:500 for 2 h at room temperature. The cell nuclei were stained with DAPI (1:4000, CST 4083) for 15 min at room temperature. All the images were acquired with the laser scanning confocal microscope. 2.13. Real-time quantitative PCR (qPCR) The whole mRNA was extracted from PCOs in microgels and Matrigel with the Trizol reagent (TAKARA, 9109) after alginate lyase treatment and mechanical stirring, respectively, to detect the associated genes expression level. The whole mRNA was also extracted from primary tumor tissues with the Trizol reagent after mechanical lapping and used as the reference for PCOs in microgels and Matrigel. The final concentration of mRNA was adjusted to 50 ng/mL. The cDNA was then synthesized by reverse transcription polymerase chain reaction (RT-PCR) using a reverse transcription kit (TAKARA, RR037A). Subsequently, real-time quantitative PCR was implemented with a SYBR Green kit (TAKARA, RR820) under the following reaction conditions (40 cycles): denaturation at 95 °C for 30 s, annealing at 58 °C for 30 s, and extension at 72 °C for 30 s. The primer pairs were synthesized by Sangon Biotech and listed in the supplementary information ([82]Table S2). Quantification was performed using GAPDH as the reference gene. 2.14. RNA-seq analysis of PCOs The PCOs cultured in microgels and Matrigel drops (1 week and 2 weeks) were treated with alginate lyase and mechanical stirring, respectively, for cell retrieval and the mRNA was extracted using the Trizol reagent according to the supplier's information. For each sample, 500 ng total RNAs were used for stranded RNA sequencing library preparation using KC-Digital Stranded mRNA Library Prep Kit for Illumina (Catalog No. DR08502, Wuhan SeqHealth Co., Ltd., China) following the manufacturer's instruction. An average of 49 million reads per sample was uniquely aligned to the hg38 human reference genome (GRCh38) STAR software (version 2.5.3a) with default parameters. The kit eliminates duplication bias in PCR and sequencing steps, using unique molecular identifier (UMI) of eight random bases to label the preamplified cDNA molecules. The library products corresponding to 200–500 bps was enriched, quantified, and finally sequenced on Hiseq X 10 sequencer (Illumina). Alternative splicing events were detected using rMATS (version 3.2.5) with an FDR value cutoff of 0.05 and an absolute value of Δψ of 0.05. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for differentially expressed genes was implemented by KOBAS software (version: 2.1.1) with a corrected P-value cutoff of 0.05 to judge whether it has statistically significant enrichment. The RNA-seq data reported in this paper were accessible in the SRA with the accession code PRJNA1110825. 2.15. Anti-tumor drug testing of PCOs PCOs grown in microgels for 1 week and 2 weeks were exposed to 4 selected compounds for a further 24 h to test the cell viability. Briefly, approximately 200 organoids in each single well of a 96-well plate containing 200 μL culture media were added with Gemcitabine (MeilunBio, MB5386), Erlotinib (MeilunBio, MB1734-1), Olaparib (MeilunBio, MB1700), and 5-Fluorouracil (MeilunBio, MB1273) at 6 concentrations of 0, 0.1, 1, 10, 100, and 1000 μM in the automatic pipetting workstation (TIANGEN, AP400). As the frequency of generating single-cell-laden microgels can be precisely controlled by the pneumatic valve (the reciprocal of valve period is the generation frequency of microgels), the number of organoids for drug testing can also be controlled. Then, the resulting organoids was incubated using a cell counting kit-8 (CCK-8, MeilunBio, MA0225) assay according to the user's manual, and the absorbance at 450 nm was measured with a microplate reader. Here, the viability of organoids treated by drugs at the concentration of zero was set as 100 % viability, and the other viability was normalized by the 100 % viability. In a separate experiment, the resulting organoids that were treated by drugs were labeled with Ki67 and TUNEL Apoptosis Assay Kit (Elabscience, E-CK-A321) according to the manufacturer's protocol, respectively, to validate the proliferation and apoptosis of organoids. All the images were acquired with the laser scanning confocal microscope. 2.16. Statistic In this work, all experiments were performed at least in triplicate. The PCR data were presented as mean ± Standard Error of the Mean, others were presented as mean ± Standard Deviation. The ΔΔCt method was used for PCR data. Statistical significance was declared when ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001 as determined by one-way ANOVA and student's t-test. 3. Results 3.1. Development of droplet system for the generation of PCOs from single cells Current tumor organoids are limited by their variation partially due to the uncontrollable origin of each organoid and undefined physicochemical properties of matrices. To improve the uniformity of the resulting tumor organoids, we set up a droplet system to synthesize PCOs within defined microgels from single tumor cells. The droplet system was constructed according to our previous study [[83]29,[84]30]. We used a multilayered and valve-integrated microfluidic chip ([85]Figs. S1A and S1C) to fabricate the defined alginate microgels in all-in-water fluids. DEX solution with NaA and PEG solution were utilized as core and middle flows that pumped into the chip, respectively, for the formation of droplet templates. Since alginate has been previously used as 3D cell scaffolds in tissue engineering due to their flexibility and biocompatibility [[86]25], we chose it as the matrix of organoids. In addition, alginate is well suited for droplet microfluidics because of its fast and mild cross-linking mechanism [[87]26], as well as holds great potential in precisely encapsulating one single cell in each microgel [[88]27], which lines with our needs in this work. The single tumor cells derived from clinical biopsy or the formed PCOs were mixed in the core flow after attached with CaCO[3] nanoparticles that can release Ca^2+ once contacting the acetic acid in the shell flow for the gelation of NaA to form Ca-alginate (CaA) microgels within the chip ([89]Fig. 2A). Then, the unloaded droplets were filtered out by gently wash, and effective encapsulation of single cells can be realized. In this system, the negative pressure driven by a solenoid valve controller was utilized to activate the normally closed pneumatic valve for controlling the generation manners of droplet templates ([90]Fig. S1B). Herein, we used PEG solution as the continuous phase to establish the all-in-water microfluidic system, instead of using traditional oil phase (e.g., mineral oil). In this system, although the throughput of droplets decreased slightly, we directly obtained cell-laden microgels by using the pretreated cells without additional demulsification step (e.g., adding demulsifier and using vigorous centrifugation). Also, the all-in-water system increased the biocompatibility of the progress of producing cell-laden microgels. All of these would ensure the cell viability within microgels after encapsulation. Fig. 2. [91]Fig. 2 [92]Open in a new tab The characterization of defined microgels. (A). Schematic diagram of the massive and controllable generation of microgels with single cells. (B). Microgels generated under serial rates of core flow, ranging from 0.1 to 0.5 μL/min. Scale bar: 100 μm. (C). Microgels generated under serial valve cycles, ranging from 0.1 to 0.7 s. Scale bar: 100 μm. (D). Microgel diameter as a function of core flow rate, as well as the ratio of long and short axis of microgels (L/S) as a function of core flow rate. (E). Microgel diameter as a function of valve cycle, as well as the ratio of long and short axis of microgels (L/S) as a function of valve cycle. Quantitative analysis of the diameter of microgels were performed on at least 50 microgels. (F). SEM images of intact microgels and cross section of microgels. The scale bars in the images are 500 μm, 30 μm, 2 mm, and 500 μm from left to right. (G). SEM images of the surface and cross section of Matrigel dome. The scale bars in the images are 1 mm, 500 μm, 1 mm, and 500 μm from left to right. (H). The porosity of freeze-dried microgels, Matrigel and pure alginate hydrogels. (I). The average sectional area of pore within freeze-dried microgels, Matrigel and pure alginate hydrogels, which represents the pore size of corresponding matrices. (J). Young's modulus of primary tumor tissues, Matrigel, and microgels fabricated with different concentrations of alginate. Microgel-0.5, microgel-1, and microgel-2 represent microgels fabricated with 0.5 %, 1 %, and 2 % NaA, respectively. (K). Determination of endotoxin contamination of the alginate prepolymers with different concentrations. FDA acceptable level was indicated with dotted blue line (0.5 EU/mL). Statistical significance was declared when ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001 as determined by student's t-test for [93]Fig. 2H, I and 2J. (For interpretation of the references to color in this figure legend, the