Magnetophoresis in Centrifugal Microfluidics at Continuous Rotation for Nucleic Acid Extraction

Abstract

Centrifugal microfluidics enables fully automated molecular diagnostics at the point-of-need. However, the integration of solid-phase nucleic acid extraction remains a challenge. Under this scope, we developed the magnetophoresis under continuous rotation for magnetic bead-based nucleic acid extraction. Four stationary permanent magnets are arranged above a cartridge, creating a magnetic field that enables the beads to be transported between the chambers of the extraction module under continuous rotation. The centrifugal force is maintained to avoid uncontrolled spreading of liquids. We concluded that below a frequency of 5 Hz, magnetic beads move radially inwards. In support of magnetophoresis, bead inertia and passive geometrical design features allow to control the azimuthal bead movement between chambers. We then demonstrated ferrimagnetic bead transfer in liquids with broad range of surface tension and density values. Furthermore, we extracted nucleic acids from lysed Anopheles gambiae mosquitoes reaching comparable results of eluate purity (LabDisk: A260/A280 = 1.6 ± 0.04; Reference: 1.8 ± 0.17), and RT-PCR of extracted RNA (LabDisk: Ct = 17.9 ± 1.6; Reference: Ct = 19.3 ± 1.7). Conclusively, magnetophoresis at continuous rotation enables easy cartridge integration and nucleic acid extraction at the point-of-need with high yield and purity.

Keywords: centrifugal microfluidics; magnetophoresis; nucleic acid extraction.

Figure 1

Schematic of the principle of gas-phase magnetophoresis under rotation. (A1): Two microfluidic chambers are located radially outwards from a magnet. During rotation, the magnetic beads undergo the centrifugal force, ${\overrightarrow{F}}_{\mathrm{cent},\mathrm{bead}}$ and the magnetic force, ${\overrightarrow{F}}_{\mathrm{mag}}$ (notably, only the radial component of the ${\overrightarrow{F}}_{\mathrm{mag}}$ vector is indicated here, as it is the one that influences the magnetophoresis between the chambers). The A-A’ cut (A2) shows the distance, Δz, between the magnet and the top of the cartridge, in the z-direction. (B1,B2): When the rotation takes place at an angular velocity below a critical value, ω_crit, the magnetic force is higher than the centrifugal force, thereby the pellet is pulled radially inwards. (B3): Azimuthal transfer of the magnetic beads to a downstream chamber. This movement is supported by the bead cluster inertia depending on the direction of rotation.

Figure 2

Basic arrangement of magnets under investigation.

Figure 3

(Left): The updated configuration (and the final one, compared to Figure 2) in which a second pair of magnets is added in order to enhance the area over a full rotation along which magnetic force is exerted on the magnetic beads. (Right): A special shape of the microfluidic chamber in which an inclined top side of the chamber between r_tilt,in and r_tilt,out enables the z-component of magnetic force to contribute to the radial bead motion.

Figure 4

Net (simulated over one full rotation) total (magnetic, centrifugal and surface) force in r-direction plotted for a range of different radial positions and for various rotational frequencies. The plot shows the simulated force according to Equation (5). The net force increases in general, if the rotational frequency is decreased. In addition, the geometry shown in Figure 2 leads to an additional net force in the range of the tilted chamber wall, caused by the term F_mag,z→r. Below the magnet at low radial positions, the net force decreases due to lower magnetic gradient in this area.

Figure 5

Bead cluster speed versus rotational frequency for different commercially available magnetic beads. AJ: Analytik Jena; MMD: MagnaMedics Diagnostics BV (currently magtivio BV); GE: GE Healthcare. The fit MMD and fit AJ gave R² = 0.9837 and R² = 0.9928, respectively (n = 3; 1× standard deviation). Ferrimagnetic beads respond very strong to the magnetic field setup. We attribute this to the clustering behavior of ferrimagnetic beads under an external magnetic field.

Figure 6

Design of the nucleic acid extraction microfluidic module. #1: Lysis chamber. #2: Binding chamber. #3: Washing chamber 1. #4: Washing chamber 2. #5: Elution chamber. Stiffening structures that provide higher mechanical stability are visible on top of the large structures that host the stickpacks. The chambers #2–5 were designed according to the findings from simulation, i.e., with tilted top chamber walls and radial positions to achieve magnetophoresis at the determined critical frequencies.

Figure 7

Stroboscopic images indicating the steps of continuous rotation magnetophoresis. The yellow frame marks the bead cluster. (A): Magnetic beads are rehydrated and remain in the binding chamber. (B)–(G): Stepwise reduction of the rotational frequency in order to collect the beads, create a cluster, pass the liquid-air interface and achieve magnetophoresis. In (G) the cluster is blocked by the pocket above washing chamber 1 in order to prevent further proceeding azimuthally. (H): Upon rotation at 30 Hz the cluster is sedimented in the subsequent chamber (this case, washing chamber 1).

Figure 8

Results comparing the performance of the nucleic acid extraction using the LabDisk (n = 4) and the manual reference method (n = 3). (A): Ct values of RT-PCR on RPS7 gene (PCR on all eluates, derived from both the LabDisk and manual extraction, were performed in tube). (B): Spectrophotometric determination of eluate purity by means of protein contamination assessment.