Rapid filamentous fungi gene knockout identification through high-throughput droplet microfluidics

Abstract

Fungi are a diverse group of eukaryotic organisms, with over 1.5 million species inhabiting ecosystems worldwide. Many fungi grow as filaments (hyphae) and play critical ecological roles, both beneficial and harmful. Understanding their functions often requires generating knockout mutants and performing comparative analyses with wild-type strains. However, traditional methods for screening knockout mutants are labor-intensive, time-consuming, and limit the rapid identification of successful transformants. Here, we present a high-throughput droplet microfluidics platform capable of screening and sorting fungal transformants at single-cell resolution, significantly improving efficiency compared to conventional methods. The workflow involves encapsulating individual fungal transformants in pico-liter-volume water-in-oil emulsion droplets, culturing them in the presence of antibiotics, and identifying and sorting droplets containing transformants that exhibit hyphal growth. Transformants that grow in the presence of antibiotics are flagged as potential knockouts and then sorted out for confirmation through sequencing. This approach offers several advantages, including a 3-fold reduction in time for Fusarium graminearum protoplast growth until they can be distinguished from those exhibiting no growth, a screening throughput of up to 28,800 transformant-containing droplets per hour, and single-spore phenotyping to minimize post-processing requirements. Using this system, we successfully screened 24,000 F. graminearum transformants containing droplets, identified five potential transformants that exhibit growth on agar plates, of which two were confirmed via sequencing as true knockouts. These results demonstrate the utility of this droplet microfluidics-based platform as a powerful tool for accelerating fungal functional genomics and advancing our understanding of the ecological roles of fungi.

1. Introduction

Filamentous fungi play a crucial role in human welfare, both as pathogens to plants, animals, and humans, and as valuable sources of industrial enzymes and commercially important metabolites [1]. Gene knockout, which involves inactivating or knocking out a specific gene to study its function, has been proven to be a critical tool in elucidating the functions of genes involved in enzyme production, stress responses, and biosynthesis of commercially relevant compounds in filamentous fungi [2–4]. This technique is also essential in understanding fungal diseases and pathogens [5–7]. However, achieving precise gene knockouts is challenging due to the low efficiency of homologous recombination, which is often overshadowed by the competing non-homologous end joining repair pathway [8]. Hence, identifying “true” knockout mutants typically requires screening large libraries of transformants, which presents a significant challenge using traditional methods.

Traditional screening methods, such as agar plate or microtiter plate-based assays, involve manually plating and observing visible fungal colonies over 12 days, thus are labor-intensive and time-consuming, with throughput limited to screening 10⁴ cells/h [9, 10]. These methods are not only time consuming but also limited in selecting single transformants during the colony-selection process, requiring additional downstream steps. Another promising method, fluorescence-activated cell sorting (FACS), can process cells at up to 10⁶ cells/h [11, 12]. However, it presents its own limitations when applied to filamentous fungi, since the larger size and filamentous structure of fungal mycelia often prevent them from passing through the small nozzles of FACS instruments, thus restricting its applicability for high-throughput screening of filamentous fungi [13].

Recently, droplet microfluidics-based systems, where single fungal cells are encapsulated in water-in-oil emulsion droplets and cultivated within such droplets, have demonstrated the capability to enable high-throughput screening assays by significant increasing single-cell-resolution assay throughput [14–18]. For example, such systems have been used for screening and selecting high lipid producing microalgal cells [19], enzyme activity screening [20], and screening of hybridoma cells for antibody production [21]. Droplet-based microfluidic system is also suitable for screening enzymes or extracellular metabolites that cells produce as they are retained within the droplets that contain the cells that produce those molecules [22, 23].

Due to these advantages, these droplet microfluidic systems have also been applied to screening filamentous fungi, where single filamentous fungal cell have been compartmentalized into pico/nano-liter monodisperse water-in-oil emulsion droplets to perform fluorescence-based screening at a throughput of 7,000~10,000 cells/h to identify strains that can produce enzymes of interest [24, 25]. These droplet-based screening method has advantages over traditional agar plate-based assay as confining a single cell in droplets enables generating clonal population of the encapsulated cells, which can then be interrogated much sooner than having to wait for visible colonies to be formed on agar plates, which typically takes days [23]. It also enables isolating single transformants due to the single-cell-resolution nature of droplet microfluidics, in comparison to potential overlaps of colonies on agar plates due to the random nature of streaking, which then require subsequent single spore isolation procedures [26].

In this work, we developed a droplet microfluidics-based system to screen and sort transformants of Fusarium graminearum, a filamentous fungus, with the goal of identifying a true gene knockout strain. The system utilizes the distinct growth phenotypes of mutants compared to the wild type (WT), enabling the identification of mutants based on their growth characteristics (no growth vs. hyphae formation) at single-transformant resolution prior to polymerase chain reaction (PCR)-based confirmation. This approach offers a high-throughput alternative to the conventional agar plate-based assay, as illustrated in Figure 1. Initially, the growth conditions of fungal transformants in water-in-oil droplets were optimized. We then demonstrated the screening capability of the droplet microfluidics sorting system by testing a defined mixture of F. graminearum mutant and wild-type strains at a known ratio. Subsequently, an actual library of F. graminearum transformants was screened, with the accuracy of the sorting process validated through PCR analysis. This multi-step process ensures precise and efficient screening of fungal mutants, enhancing the accuracy and throughput of the screening platform. The traditional method shown in Figure. 1, where researchers must plate protoplasts on agar plate and wait about 5–10 days for the transformants to grow, is a time-consuming process [9]. In addition, since single colonies need to be manually picked and transferred to individual plates, this can lead to inaccuracies due to the potential mixing of protoplasts during incubation as well as overlapping colonies requiring a further step for single-strain isolation. In contrast, using the developed droplet microfluidics-based screening method allows for efficient, high-throughput, and automated isolation of single transformants while minimizing reagent consumption and reducing the chances of error.

Figure 1.

Schematic for isolating Fusarium graminearum knockouts using the developed droplet microfluidic method (top) in comparison to the conventional agar plate-based method.

2. Results and discussion

2.1. Workflow and design of the droplet microfluidic system

The droplet microfluidics-based filamentous fungi knockout screening workflow shown in Figure 1 is realized in the droplet microfluidic chip shown in Figure 2a. First, F. graminearum WT strain was genetically modified to include hygromycin B phosphotransferase (HPH) gene, as depicted in the knockout library design. The HPH gene confers resistance to hygromycin, facilitating the selection of successfully transformed transformants by culturing them in the presence of hygromycin. The genetically modified transformants are suspended in tryptic soy broth (TSB) media containing hygromycin and calcofluor white (CFW) fluorescent stain. CFW is a fluorescent stain commonly used in fungal biology to visualize fungal structures, and it can bind to chitin or cellulose making fungal cell walls visible under a fluorescence microscope [27]. Next, they are encapsulated into droplets using a T-junction droplet generator microfluidic device. The droplet generation process achieves single transformant encapsulation when the concentration of the transformants is at 1.4×10⁶ mL⁻¹ and the volume of each droplet is 0.7 pL (110 µm in diameter). After single-transformant encapsulation, approximately 31.8% of the droplets contained a single transformant (Supplementary Figure 1). Following encapsulation, droplets are incubated for 24 h to allow for fungal growth. The droplets are then reflowed into a droplet sorter microfluidic device where laser-excited fluorescence signals are detected. The fluorescent intensity of each droplet indicates whether the droplet contains calcofluor white-stained fungi or not. The fluorescence signals, if above a certain threshold due to fungal hyphae growth, trigger an electrode pair to generate a dielectrophoretic (DEP) force [28, 29]. The sorted droplets, identified as “hits”, are then dispensed onto agar plates at single-droplet resolution for the recovery of the transformed fungal colonies (each droplet is expected to result in one colony). To confirm whether the “hits” are correct knockouts, PCR analysis is then conducted on the recovered colonies. Figure 2b shows the enlarged top view of the microfluidic chip used here, and Figure 2c presents the optical path and components of the droplet fluorescence detection and sorting system.

Figure 2. (a) Schematic diagram of the droplet microfluidics-based screening workflow for identifying filamentous fungi transformants.

(1) Designed partial HPH gene (HP and PH) fragments were amplified from pBS15 plasmid. FgVe1 5’ and 3’ flanking region fragments were fused with PH- and -HP fragments by single-joint PCR, respectively. The single-joint PCR products were then transformed into WT fungal protoplast to generate the knockout library. (2) The transformants are cultured in TSB media containing hygromycin and calcofluor white fluorescent stain. (3) Single transformant encapsulation into droplets. (4) Droplets are incubated at 25°C for 24 h to allow for fungal growth. (5) Post-incubation, droplets are reflowed into a droplet fluorescence detection and sorting chip, where laser-excited fluorescence signals above a certain threshold trigger an electrode pair to generate a DEP force, selectively pulling only droplets that have above-threshold fluorescent intensity. (6) Sorted droplets are then dispensed onto agar plates at single-droplet resolution and then cultured for colony formation. The colonies are then picked for PCR confirmation of the knock-out mutant. (b) Image of the droplet microfluidics chip, black part indicating the three-dimensional electrode locations and blue part indicating the microfluidic channels. (c) Optical path and components of the fluorescent droplet detection and sorting system.

2.2. Filamentous fungal germination and growth in pico-liter droplets

In this screening assay, growth differences between WT and mutants were used to identify possible fungal knockout mutants, since in the presence of hygromycin, non-transformed fungal protoplasts cannot survive and exhibit stalled germination while successfully transformed protoplasts grow into filamentous hyphae [23]. However, filamentous fungal hyphal tips can pierce through water-in-oil emulsion droplets easily if they grow too much, leading to uncontrolled droplet coalescence as those hyphae can be tangled up [24, 25]. Therefore, antifungal concentration, fluorescent staining concentration, droplet size, as well as cultivation time need to be optimized to allow sufficient growth of hyphae to a detectable level while preventing hyphae from overgrowth and subsequently clogging of the microfluidic channels. Here, spores of wild-type F. graminearum and hygromycin-resistant F. graminearum mutant were used to optimize these conditions. Spores were chosen over transformants due to their convenience of preparation, which assisted the experimental procedure. Additionally, the growth conditions of fungal spores and transformants are comparable [30], ensuring the results are representative and applicable to both conditions.

A concentration of 150 µg/mL hygromycin was identified as optimal for fungal screening, as it provided sufficient selection of hyphae while minimizing the adverse growth inhibition observed at higher concentrations (Supplementary Figure 2). Additionally, the optimal concentration of calcofluor white staining was determined to be 10 µg/mL. Lower concentrations resulted in insufficient fluorescent signals, potentially impacting the sorting process, while higher concentrations inhibited fungal growth (Supplementary Figure 3). The size of the droplets also played a critical role in the performance of the microfluidic system. Encapsulation in droplets that were too small (less than 66 µm in diameter) resulted in fungal hyphae piercing through the droplets during culture within 24 h, disrupting the sorting process. Conversely, larger droplets (over 260 µm in diameter), while better accommodating fungal growth, were incompatible with the microfluidic sorting device, reducing throughput and efficiency. Supplementary Figure 4 illustrates how fungal spore growth varied with droplet size, emphasizing the importance of droplet size optimization. In addition to droplet size, the cultivation time significantly influenced the sorting throughput and accuracy. Insufficient culture time failed to produce a discernible fluorescence signal difference between those showing growth versus no growth, while extended cultivation time led to hyphae piercing the droplets, complicating the sorting process. Figure 3a highlights an example of successful fungal spore encapsulation and subsequent growth within droplets, demonstrating the critical interplay between droplet size and culture conditions in optimizing the droplet microfluidics-based screening process.

Figure 3. Growth of fungal hyphae in droplets.

(a) Spore with hygromycin resistance germinate in droplets and grow into hyphae (0 to 24 h, bright field and DAPI channel showing CFW-stained hyphae). (b) Fungal hyphal growth and fluorescence intensity of CFW-stained hyphae in droplets over the 24 h cultivation period. p < 0.05, N = 10, scale bar: 30 µm.

For hyphae growth to be identified in a droplet microfluidics system, fluorescent staining of hyphae is needed to distinguish those that show growth from those that do not grow based on fluorescent intensity differences. A fluorescent microscope was used to image the DAPI channel with the sample stained with CFW. The DAPI channel (Ex 358 nm, Em 461 nm) was selected due to its compatibility with the fluorescence emission of CFW (Ex 347 nm, Em 433 nm), ensuring optimal visualization. The spores began to germinate after 6 h, and the fluorescent intensity reached 20 times than its original intensity after 24 h before the hyphal tip punched through droplets. Mutant spores that have hygromycin resistance could germinate into hyphae while WT strain spores could not survive in the presence of hygromycin and stall at the single-spore stage. The CFW added to the droplets could successfully identify fungal growth based on the increased fluorescent intensity, enabling transformant screening by measuring fluorescence signal intensity in mutants compared to WT (Figure 3b).

2.3. Platform validation by screening a mock library having a mixture of wild-type and hygromycin-resistant mutant spores at defined ratios

To validate the droplet microfluidics screening system, fluorescence signal detection was first performed using droplets containing WT and mutant F. graminearum spores after droplet cultivation. Droplets encapsulating WT fungal spores and mutant spores after 24 h incubation were reflowed into the droplet microfluidic sorting device at a throughput of 8 droplets/sec. Representative bright field (BF) and DAPI fluorescence images of the encapsulated fungi provided direct visualization of growth differences between WT and mutant strains. A WT spore encapsulated in droplets (Figure 4a) showed no growth and a weak fluorescence signal in the presence of hygromycin. In contrast, a hygromycin-resistant mutant spore (Figure 4b) exhibited significant hyphal growth, as indicated by strong DAPI fluorescence and clear hyphae growth within the droplets. These results confirmed the ability of the droplet microfluidic platform in distinguishing WT and mutant strains based on growth differences and fluorescence intensity differences.

Figure 4. Droplet fluorescent intensity distribution and visualization of fungal growth in droplets.

(a) BF and DAPI fluorescence images of wild-type F. graminearum encapsulated in droplets, showing no growth. (b) BF and DAPI fluorescence images of hygromycin-resistant mutant (MU) in droplets, displaying growth and high fluorescence intensity. (c) BF and DAPI images of droplets before sorting, showing a mixed population of WT and mutant type droplets. (d) BF and DAPI images of droplets after sorting, demonstrating the droplets with high fluorescent intensity were sorted. Scale bar: 30 µm. (e) Fluorescence intensity histogram of droplets containing WT (blue) and hygromycin-resistant mutant (orange). Approximately 6,000 droplets from both conditions were analyzed. (f) Violin plots show the distribution of fluorescence intensity for wild-type and mutant strains, with mutants exhibiting higher fluorescent intensity due to active growth. There is a significant difference between WT-containing droplet fluorescent signal compared to mutant-containing droplet. P < 0.001, n = 6000.

Next, a mock library of WT and mutant spore mixtures were prepared with a predefined ratio of 1:10. Cells were then encapsulated into droplets at a dilution that maximized single-spore encapsulation (31.8% of droplets had single spore, Supplementary Figure 1). After cultivation for 24 h at 25°C to allow for detectable growth differences in droplets (Figure 3), droplets were reinjected into a droplet sorting device (Figure 2b). Droplets with fluorescence signals higher than the threshold value of 8,000 (arbitrary unit) were pulled by turning on an electrical field for dielectrophoretic droplet sorting while the rest continued to flow into the waste channel. Figure 4c shows the images of droplets before sorting, where a mixed population of WT and mutant protoplasts exist. Figure 4d shows the bright field and fluorescent images of droplets after sorting, where most droplets exhibited a high level of hyphae growth and thus high fluorescence signal. This further validates the platform’s capability in sorting only droplets that contain spores growing into hyphae. The fluorescence intensity distribution of the droplets was assessed to compare fungal growth (Figure 4e). The resulting fluorescence intensity histogram revealed two distinct populations: droplets containing WT exhibited significantly lower fluorescence intensity, whereas those with mutant displayed significantly higher fluorescence intensities (3-fold difference in peak intensities). This difference can be more clearly seen in Figure 4f. This demonstrates that the droplets encapsulating WT and hygromycin-resistant mutant strains can be effectively distinguished using the droplet microfluidics sorting system, validating its capability for accurate and high-throughput fungal screening.

2.4. Platform validation by screening a real fungal knockout library

After confirming the functionality of the developed droplet microfluidic screening system with the mock library, a real knockout library of F. graminearum was screened to sort true knockouts. Following the same experiment protocol as the mock library screening, F. graminearum transformants were encapsulated into droplets at single-transformant resolution and cultivated for 24 h at 25°C. The cultivated droplets were then reinjected into the droplet sorting device and sorted at 8 droplets/s. Around 24,000 droplets were screened within 50 min. 81 sorted “hit” droplets were dispensed on a Yeast Peptone Dextrose (YPD) agar plate containing hygromycin. Five colonies emerged (Figure 5a), and those colonies were checked by PCR to confirm whether they are true knockouts (Figure 5b). Here, the first pair of PCR primers were used to test the presence of the FgVe1 gene, and the second pair of PCR primers were used to detect the inserted hygromycin gene. WT showed a band with the first pair of primers but not with the second, which indicates the presence of FgVEe1 gene that has not been replaced by the hygromycin gene. Isolates #2, #4, and #5 showed bands with the first pair of primers and none with the second, suggesting that the target gene was not successfully knocked out and the hygromycin resistant gene was not inserted. In contrast, isolate #1 and #3 showed bands with the second pair of primers but no band with the first, confirming that they are true knockouts of the target gene. This result demonstrates that screening of filamentous fungi knockout library based on their growth phenotype differences using the developed droplet microfluidic system is an effective approach in identifying true gene knockouts.

Figure 5. Confirmation of true knockouts after the droplet microfluidic screen.

(a) Sorted “hit” droplets were plated on an agar plate and further cultured to allow visual colony formation. (b) PCR confirmation to identify true F. graminearum knockout mutants. Two out of five colonies grown on hygromycin agar plate (isolate #1 and #3) are confirmed to be true knockouts.

Traditional screening techniques are limited due to their lower throughput or the inability to handle the large and complex structures of filamentous fungi. Our droplet microfluidics platform overcomes these limitations, making it a powerful tool for large-scale fungal screenings. The system’s high-throughput capability, screening over 28,000 droplets in just one hour, represents a significant improvement over the conventional method that would have taken over 3 hours [31]. Thus, the successful recovery of true mutants from non-transformed fungal cells strongly suggests that this platform is ready for use in high-throughput fungal research and biotechnology applications. It is to be noted that recent droplet microfluidic systems developed have demonstrated a throughput of 100,000 droplets/h with sorting error of less than 0.1% [32], showing that the throughput used in this study can be significantly increased by at least 3-fold.

The screening workflow, illustrated in Figure 2, is broadly applicable to high-throughput screening of fungal libraries to identify mutants with enhanced enzyme production for industrial biotechnology [33]. Additionally, this system is suitable for antifungal drug discovery by enabling the rapid assessment of fungal resistance under selective pressures [34]. Furthermore, the workflow can be adapted for genetic studies aimed at understanding fungal pathogenicity and stress response mechanisms [35]. The use of droplet microfluidic systems in filamentous fungi genetic transformation workflow is expected to significantly reduce labor, time, and reagent costs, providing higher efficiency and precision in fungal genetic research in agricultural, food, and medical industries. Thus, the platform has broad applications in multiple fields of fungal biology and biotechnology applications.

3. Materials and methods

3.1. Fungal strains and culture conditions

F. graminearum wild-type strain PH-1 was used for the study [36]. F. graminearum was inoculated in Fusarium regeneration broth (1 M Sucrose, 0.02% (w/v) yeast extract) [37] for protoplast generation. Tryptic Soy Broth (TSB, Becton, Dickinson) was used to culture transformants in water-in-oil droplets. Calcofluor white (CFW, Sigma-Aldrich), a fluorescent stain that binds to fungal cells so that hyphae growth can be detected through the fluorescent signal, was added at a concentration of 10 µg/mL to stain the fungi, which excitation and emission peaks are 355 nm and 433 nm, respectively. Hygromycin (Sigma-Aldrich) was added to select transformants at concentration of 150 µg/mL.

3.2. Gene deletion, polymerase chain reaction (PCR), and transformation

The constructs for F. graminearum transformation were generated following our laboratory standard procedures [38]. In brief, DNA fragments representing the upstream (5’) and downstream (3’) flanking regions of the gene of interest were amplified from wild-type genomic DNA. Meanwhile, the hygromycin B phosphotransferase (HPH) gene in the pBP15 plasmid was employed to generate HP and PH fragments. The 5’ and 3’ flanking region fragments were joined with the PH and HP fragments using single-joint PCR, respectively [38, 39]. The protoplast preparation and transformation protocols followed well-established protocols [40]. The presence of targeted gene deletion in the drug-resistant colonies forming on agar plates was confirmed through PCR analysis. The complete list of primers utilized here is provided in Supplementary Table 1.

3.3. Microfluidic device fabrication

The process of the microfluidic device fabrication steps is described in Supplementary Figure 5. To form the microchannel, the SU-8 2050 photoresist (MicroChem, USA) was spin-coated at 2800 rpm on a 3-inch silicon wafer (Universal Wafer Corp.) to get a 68 µm thickness film and spin-coated at 2200 rpm to obtain a 100 µm film, followed by standard photolithography. Next, liquid-phase PDMS (Sylgard 184, Dow Corning) was mixed at a 10:1 ratio of base to curing agent and poured onto the SU-8 master mold. After curing the PDMS in an oven at 80 °C for 4 h, the microfluidic channels were aligned and bonded onto a 50.8 × 76.2 mm glass substrate (Swift Glass Corp.) using oxygen plasma treatment (Harrick Plasma PDC-001-HP), ensuring a strong bond between the PDMS and the glass layers. The droplet generation module had a channel height of 68 µm, while the sorting device had a channel height of 100 µm to facilitate sorting and movement of droplets. To enable DEP-based droplet sorting, the 3D electrode was fabricated by filling the microfluidic electrode channels with Roto 144F Low Melt Fusible Ingot Alloy (Roto Metals, CA, USA) at 85 °C, producing 3D electrodes with no additional fabrication steps [41].

3.4. Encapsulation and generation of fungal spore in droplets

Water-in-oil emulsion droplets containing single fungal spore and culture media was generated by a T-junction droplet generator where the aqueous suspension of 2.2×10⁶ spores/mL measured by Nanodrop OD₆₀₀ (Thermo Scientific) in TSB (Becton, Dickinson) supplemented with CFW (Sigma-Aldrich) was pinched off by fluorinated oil (Novec 7500 engineering oil, 3M) containing 2% PFPE-PEG surfactant (Picosurf 1, Sphere Fluids). The oil and culture media mixture are flown into the droplet generation device at a 200 µL/h and 180 µL/h flow rate, respectively, to generate 110 µm diameter droplets at a rate of 50 droplets/s. In order to encapsulate a single spore into a 110 µm diameter droplet (690 pL volume), the fungal spore media was diluted to a concentration of 1.4×10⁶ spores/mL. Hygromycin B phosphotransferase (HPH) with concentration of 150 µg/mL was used as the selective marker, and homologous recombination was confirmed by PCR. Here, the wild-type spores do not grow due to the hygromycin added in the TSB culture media. However, mutant spores, which carry a hygromycin resistance marker, can successfully grow into filamentous hyphae. Single fungal transformants were encapsulated in droplets together with CFW. The droplets, each having a volume of 690 pL, were incubated for 24 h to allow the spores to germinate and grow into hyphae, thereby creating a detectable growth difference. The single fungal spore encapsulation and analysis results are described in Supplementary Figure 1. Droplets were collected and incubated in a microfluidic chamber for germination observation under 30 °C in an incubator. A fluorescence microscope (Axio Observer, Zeiss) was used to perform time-lapse observation of fungal spore germination in droplets.

3.5. Droplet screening chip design and operation

Figure 2b depicts a microfluidic device designed for droplet sorting. The main structure includes an inlet for infusing the sample droplets to ensure uniform and stable droplet flow. Electrodes are positioned along the microfluidic pathway to facilitate dielectrophoretic (DEP) sorting. A bias is applied to control droplet flow toward specific outlets. Droplets containing single fungal spore were generated and collected in a 1.5 mL centrifuge tube, which can accommodate approximately 2×10⁶ droplets, and cultured at room temperature for 24 h. Cultured droplets were reinjected into the concentric-designed droplet sorting device at 60 µl/h and spaced with oil flowing at 1,200 µL/h. A bias oil flow was applied at 400 µL/h to constantly direct droplets toward the waste collection tube. A laser optical setup (Laserglow, 360 nm, 100 mW, DPSS laser system) was built to excite droplets at a 360 nm and detect fluorescent signal at 461 nm, as described previously [28, 32, 42, 43]. The detected fluorescence signal was processed by an FPGA I/O data acquisition card (PCIe-7842, National Instrument) and a custom LabView program (LabView, National Instrument). Droplets expressing fluorescence signal higher than a set threshold were sorted by applying a dielectrophoretic force generated by AC field pulses (10 kHz, 600 V_pp, 100 ms) via a voltage amplifier (Model: 2205, Trek Inc). Sorted droplets were directed from outlet through tubing and dispensed onto agar plates containing YPD (Becton, Dickinson).

3.6. Polymerase chain reaction (PCR) verification

To confirm the sorted fungi are true knockout mutants, the sorted droplets were dispensed on a YPD agar plate to form colonies at 30 °C through a 3-day culture. The cultivated mycelium was collected for genomic DNA isolation. Genomic DNA was extracted from mycelia using a fungi DNA isolation kit (Thermo Scientific, Phire Plant Direct PCR Master Mix) following the manufacturer’s manual. Nuclear-free water was used as negative control. After fungal DNA extraction, PCR assay was performed in an Eppendorf Mastercycler machine following the manufacturer’s manual. Each reaction mixture contained 0.5 μL genomic DNA, 10 μL 2X Phire Plant PCR Buffer, 1 μL of PCR forward and reverse primers each, 0.4 μL Phire Hot Start II DNA Polymerase, and 7.1 μL nuclear-free water to form a final volume of 20 μL. The PCR protocol is as follows: 5 min of preheating at 98 °C, 5 s of initial denaturation at 98 °C, 5 s of annealing at 64 °C, and 30 s of extension at 72 °C for 40 cycles.

4. Conclusions

In this study, we successfully developed and validated a droplet microfluidics-based platform designed for high-throughput screening of filamentous fungi. The system successfully screened 24,000 F. graminearum transformants-containing droplets in less than 50 min (28,000 droplets/h throughput) identified five potential “hits” that displayed growth on an agar plate, of which two were confirmed to be true knockouts based on PCR results. The platform demonstrated exceptional performance in distinguishing and selecting mutant fungal strains from wild-type strains at single-cell resolution, a capability that is critically important for research into genetic modifications and fungal strain development. This approach highlights the effectiveness and higher throughput of the droplet-based microfluidic platform compared to traditional methods, which are often labor-intensive, time-consuming, and have limited throughput. In summary, this robust tool is well-suited for fungal research and biotechnological applications, offering a streamlined process for high-throughput filamentous fungal screening.