Team:TU Eindhoven/Microfluidics/Results Cell Encapsulation


Revision as of 01:21, 18 October 2014 by TomNijhof (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

iGEM Team TU Eindhoven 2014

iGEM Team TU Eindhoven 2014

Results Cell Encapsulation Device

Figure 1. Fluorescent beads trapped in
the droplet chamber.

Fluorescent Bead encapsulation

Before the devices were tested with bacteria, they were tested with fluorescent beads were performed. These beads have the same size as the bacterial cells and were therefore an excellent replacement for bacteria (Figure 1). Also they do not comply with the biosafety issues concerning bacteria. Using these beads, it was also faster since no time would be needed for culturing the bacteria. However, beads are more expensive than bacteria. Some experiments were performed with only the beads to test the encapsulation process and compare it to the cell encapsulation model. As can be seen in Figure 1 bead encapsulation was successful (see video below on the right for the encapsulation of beads).

It was found that droplets did not remain in the droplet chamber, that was unexpected. Therefore droplets were analyzed as they were captured in the outlet (see Figure 2).

Figure 2. Encapsulation of fluorescent beads. From Figure 2 it can be
observed how many droplets will contain one, two, three or zero beads.
The results are listed in Table 1 for an overview.

The results can be compared with the model for cell encapsulation and a lambda of 0.3 which is shown in Figure 3.

It is shown in Figure 3 that there is a low significance difference between the model and the experimental Poisson distribution.

There were several problems that needed to be overcome before beads and bacteria could be encapsulated successfully. First of all, it was difficult to regulate and fine-tune all the flows together. The flows were regulated accordingly; first the continuous phases were turned on until they reached the cross-junction. Immediately after the oil phase was turned on with a higher flow rate to prevent the continuous phase pushing back the oil phase. Once both flows came together at the cross-junction the flow rates were set to the right flow rates to get the desired droplet size. Because the oil phase is more viscous than the continuous phase, it flows more easily and therefore faster through the device thus making it more difficult to regulate the flows. Secondly, since the density of the beads and the bacteria is higher than the density of the solvent, the particles and cells will sediment to the bottom of the syringe. Sedimentation inside the droplets caused unstable droplets which led to breakage of the droplets and releasing the beads (Figure 4). Initially this problem was solved by using a self-made remote syringe with a stirring bean inside (see video below on the left).

Table 1. Distribution of droplets that contain beads.
Figure 3. Poisson distribution of droplet containing cells with a lambda
of 0.3 compared with experimental data. The Modelled Poisson distribution
is shown in blue and the Experimental Poisson distribution is shown in red.
Figure 4. Droplet instability caused by sedimentation
of the beads.
Figure 5. Clogging of the channels by beads. As can be seen in the left photo, the beads coagulate to the walls of the fluid resistor.

This stirring bean was moved by a computer fan with another magnet bound to it. However, this method could be potentially harmful for the bacterial cells by damaging them. For the beads this would not impose a problem but it would still not prevent sedimentation inside the device. In the end this problem was solved by adding a density matching solution to the solvent. In this research OptiPrepTM was used.

OptiPrepTM has a density of 1.32 g/mL. Using the formula given in the protocol oil and water phase (Encapsulation), the amount of OptiPrepTM that was needed can be calculated. It was found that adding too much resulted in clogging the channels as can be seen in Figure 5. Also the beads are made from polystyrene which is known to be hydrophobic. Since the devices were treated with Aquapel, the environment in the channels was hydrophobic. This causes the beads to interact with the channel walls which are also apparent from the photos in Figure 5. Also this was seen during the tests with the microscope.

Next to this, the droplet chamber did not work properly. This was due to the fact that it was too small. Because of the fast flow rates the droplets did not have time to stay in the droplet chamber but flowed right through. Flow rates could not be further reduced, because it was already at its lowest. To optimize bead and bacterial cell encapsulation, further investigation to the design of the cell encapsulation device has to be performed.

iGEM Team TU Eindhoven 2014