A self-purifying microfluidic system for identifying drugs acting against adult schistosomes
Data files
Nov 29, 2022 version files 1.39 MB
-
ART_M199_RPMI.JNB
-
ART_M199_RPMI.xlsx
-
ART_µflu.JNB
-
ART_µflu.xlsx
-
Data_for_adhesion_depending_of_worms.xlsx
-
Data_for_adhesion_on_differentes_surfaces.xlsx
-
Data_for_Media_and_Serum_comparision.xlsx
-
Data_for_Media_and_Serum_comparison.xlsx
-
Data_of_ART_for_IC50_in_microfluidic_chips.xlsx
-
Data_of_ART_for_IC50_in_petri_dishs.xlsx
-
Data_of_PZQ_for_IC50_in_microfluidic_chips.xlsx
-
Data_of_PZQ_for_IC50_in_petri_dishs.xlsx
-
PZQ_M199_RPMI.JNB
-
PZQ_M199_RPMI.xlsx
-
PZQ_µflu.JNB
-
PZQ_µflu.xlsx
-
README.TXT
-
Stat_table_for_adhesion_depending_of_worms.xlsx
-
Stat_table_for_adhesion_on_differentes_surfaces.xlsx
-
Stat_table_for_Media_and_Serum_comparision.xlsx
Abstract
The discovery of novel antihelmintic molecules to combat the development and spread of schistosomiasis, a disease caused by several Schistosoma flatworm species, mobilizes significant research efforts worldwide. With a limited number of biochemical assays for measuring the viability of adult worms, the anti-schistosomicidal activity of molecules is usually evaluated by microscopic observation of worm mobility and/or integrity upon drug exposure. Even if these phenotypical assays enable multiple parameters analysis, they are often conducted during several days and need to be associated with image-based analysis to minimized subjectivity. We describe here a self-purifying microfluidic system enabling the selection of healthy adult worms and the identification of molecules acting instantly on the parasite. The worms are assayed in a dynamic environment that eliminates unhealthy worms that cannot attach firmly to the chip walls prior to being exposed to the drug. The detachment of the worms is also used as a second step readout for identifying active compounds. We have validated this new fluidic screening approach using the two major antihelmintic drugs, Praziquantel and Artemisinin. The reported dynamic system is simple to produce and to parallelize. Importantly, it enables quick and sensitive detection of antischistosomal compounds in no more than one hour.
Methods
Microfluidic chip design
The geometrical features of the chips were designed by using COMSOL MULTIPHYSICS (Comsol AB, Stockholm, Sweden) to obtain a single microfluidic chamber, the fluid flow was calculated with the microfluidic module using laminar flow conditions assuming a very low Reynolds number (see Supplementary Table S3). The 3D FEM model is made of about 2 106 192 degrees of freedom using the predefined ‘extra fine’ mesh refinement. In the fluid flow model, non-slip initial condition was imposed to all surfaces corresponding to a solid/liquid interface. A fixed flow rate (between 0.05 and 3.4 ml.min-1) was used for the inlet and a fixed pressure (P = 0 Pa) for the outlet. The equations were solved in 900 s requiring 7.5 Go using an Intel Core i7-7500U CPU cadenced at 2.7 Ghz with 16 Go RAM configuration.
Analysis of viability in various media and sera
The viability of the worms was evaluated by observing the parasites with a macroscope (OPTIKA, SZ6745TR, Italy) and using a visual ’severity score’ to determine health status. We have defined a 3 levels scale: S2 score is for worms having a normal movement, gut peristalsis and normal tissues and tegument integrity, and attachment to plate with sucker; S1 score corresponds to worms with reduced movement but still gut peristalsis and/or degraded tissues or tegument; while S0 corresponds to dead worms having no movement or with strong tissue degradation. Each condition was observed for at least 30 s. Data were corrected when an S0 score (dead) was followed by an S1 or S2 score (still alive) and thus to avoid falsely counting worms that were still alive as dead. The data are presented as percentages, mean values and standard deviations. Statistical significance was calculated with ANOVA followed by Tukey’s test with an α risk of 5% (p value < 0.05) between human, calf and horse sera. Assumptions of ANOVA were verified by using a plot of residuals in function of predictions and a normal qq-plot. All these data analyses were done using R Software (R Core Team (R Foundation for Statistical Computing, Vienna, Austria, 2022) in the RStudio environment (1.1.463 release, RStudio Inc., Boston, USA).
Adhesion studies on different surfaces
For each experiment, 30 couples were introduced into the chips with different glass surface coatings: w/o coating, coated with MDCK cells, with type I-A collagen or with both MDCK cells and type I-A collagen. Chips were placed on a heating block set at 37°C (Major Science, MD-02D, USA), and connected to 15 mL tubes placed in a water bath set at 37°C. A camera (Microsoft, Q2F-00015, USA) was used to monitor the microfluidic channel at the beginning of the experiment in order to identify worms that lost adherence and to measure the duration of their residence in the microsystem before being flushed out by the flow. To modulate flow conditions, the peristaltic pump was programmed to pull the culture medium at different flow rates: 1) 50 μL.min-1 to 300 μL.min-1 with a step of 50 μL.min-1 every min; 2) 300 μL.min-1 to 1 mL.min-1 with a step of 100 μL.min-1 every 2 min; 3) 1 mL.min-1 to 3.4 mL.min-1 with a step of 200 μL.min-1 every 3 min. The percentages of worms still attached to the chip were plotted against the flow rate. A 2-segment linear regression analysis was performed using SigmaPlot v14.5 with Piecewise fitting curve option to identify the transition value (T1). We next performed a linear regression using the data between 0 and 1.6 mL.min-1 for glass, MDCK and collagen + MDCK, while the data between 500 µL.min-1 and 1.6 mL.min-1 values were used for collagen. This analysis furnished the slope of the first segments, the standard deviation and the 95% confidence intervals. Experiments were performed in duplicate (n = 2) with worms from the same age after perfusion and a 48h maximum duration of culture. To compare gripping between males, females or couples, 30 worms or couples were introduced in a collagen-coated chip and were perfused at increasing flow rates as described above. Experiments were performed in triplicate (n=3). Statistical significance was calculated with ANOVA followed by Tukey’s test with an α risk of 5% (p-value < 0.05) between males, females and/or couples. Assumptions of ANOVA were verified by using a plot of residuals in function of predictions and a normal qq-plot (see source data files).
Drug testing under flow conditions
Praziquantel (PZQ) (SIGMA, P-4668, USA) was prepared as a 0.5 M stock solution in DMSO and diluted in RPMI Horse serum from 900 nM to 25 nM. Artemisinin (ART) (Tokyo chemical industry, A2118, Japan) was prepared as a 0.1 M stock solution in DMSO and diluted from 900 µM to 12.5 µM in RPMI Horse serum. Stock solutions were stored at 4 °C. Once the worms were introduced in the system as previously described, the flow rate was increased up to 1 mL.min-1 by 200 µL.min-1 increments every 5 s in order to favor worm attachment to the collagen-coated chips. The recording was started and the 1 mL.min-1 flow was maintained for 5 min. This lag period allowed the elimination of weakened worms. Next, PZQ or ART was added to the tube containing the culture medium. The medium was mixed and infused into the device. The video recordings were used to count the worm couples present in the microfluidic chamber as a function of time. Experiments were performed in triplicate (n = 3) with worms from the same age after perfusion and with a maximum duration of culture prior experiment of 48h.
Drug testing under static conditions
Typically, 10 worm couples were incubated in 12 well plates at 37°C for 120 h in RPMI Horse supplemented with PZQ or ART (0 to 900 nM for PZQ or 0 to 900 µM for ART, 3 mL per well). The media were removed and renewed every 24 h. After this step, each well was immediately recorded for 15 s under a macroscope (OPTIKA, SZ6745TR, Italy) equipped with a camera (Microsoft, Q2F-00015, USA). In this experiment, we used a more resolved ‘severity score’ than those classically used for drug evaluation against helminths in static conditions (35). We designed for this assay a 5-level scoring since it gives a more precise evaluation of phenotypes, in particular paralysis or tegument alterations (S4 = Normal mobility and bloodsucker adhesion; S3 = reduction of mobility and/or loss of bloodsucker adhesion; S2 = minimal mobility or occasional movements; S1 = no mobility except intestinal movement; S0 = total loss of mobility, no movement, death). Experiments were performed at least in triplicate for each concentration (n > 3) with freshly perfused worms (< 12h).
Usage notes
Programs required to visualize and analyse data:
Videos: mp4 file player
Tables: Excel 2019 or any software compatible .xlsx file
Scripts: Rstudio (1.1463)
Fluidic simulations: Comsol Multiphysics (v.5.6)
IC50 data and graphs: SigmaPlot (v.15)