Overview
One of the vital elements of a Do-It-Yourself bioreactor is the hardware design which should be feasible as well as reliable. The most important point to be kept in mind while constructing a bioreactor is the sterility. At every step of the development of the bioreactor, we have fulfilled these criteria.
Key Achievements
- design and characterization of DIY peristaltic pumps with a dynamic range of 0.5-1400 ml/h
- development of a semi-continuous OD measurement unit
- calibrated magnetic stirring from 100-1400 rpm
- integrated circuit for controlling three pumps and stirring
- designed DIY constructions to hold circuit, pumps and bioreactor setup
- parallelization of multiple reactor control units
- characterization of DIY chemostat bioreactors
Pumps
We built upon a design from thingyverse [1] that uses a 4-wired NEMA 17 stepper motor with 1.8 °/step to construct a peristaltic pump. The original design however did not fit our requirement of sterile assembly, because the tubing was pushed through a hole in a 3D-printed structure. Using Autodesk Inventor, we designed a different structure to accomodate rapid prototyping and sterile assembly.
Principle
For our bioreactor, we need a peristaltic pump and it is necessary to use sterile tubings which requires the pumps to be assembled and disassembled in minimum possible time. Hence the pumps were built with different layers and a 3D structure that could be partly disassembled to position the tube. The tube was pressed between the 3D structure and the walls thereby generated the required pressure to pump the liquid from one side to another.
A detailed construction manual is available at pumps.
Stirrer
We use a DC motor and a 3D printed structure holding 4 magnets as our stirrer. The Voltage to the circuit is controlled using Analog output values thus controlling the speed of the stirrer. The magnets are placed in such a way so that half of the structure is north and the other half is south. This increases the area of magnetism thus holding the stirrer firmly into its magnetic pull.
Contruction Manual can be found on stirrer.
Continous OD Device
How can we find out if there is really bacteria growing in our reactor? The common way is doing OD measurements. But if there is only 10 ml of reactor volume, every measurement would disturb the conditions drastically and thereby influence the experiment itself. Therefore, we are required to take minimum possible sample volume.
It is also important that the samples are taken in a sterile manner. Furthermore the OD should be measured automatically so that the experiment can run without the need of a person taking samples. Finally the device to measure to OD has to be cheap and easy to assemble while having a precision to compete with commercial spectrometers.
Measurement Principle
Light having wavelength 605 nm is emitted by a LED and is emitted in the medium. On the opposite side, we have a phototransistor to measure the intensity of the light reaching it, through the sample.
We have developed on iGEM Aachen 2014 OD device measurement principle to get live data, and built a device using optimum materials, along with use of calibration file to get as precise values as possible.
To obtain the cell concentration in the medium, a special transistor that converts light to frequency, TSL 235R[2], is used.
To keep the whole measurement sterile, the medium with the bacteria flows through silicon tubes and a transparent glass structure through the device without getting in contact with the surrounding air, while the device itself is mounted around this tube.
A detailed construction manual is available at biomass sensor.
To test if our continuous OD device can also be used to monitor batch fermentations, we set up a simple batch culture. We used a preculture of Eschererichia coli BL21 Gold (DE3) with plasmid pSB1KRDP in M9 medium with 40 mM glucose that was grown in a stirred flask. A pump was used to cycle the culture from the vessel through the continuous OD sensor and back into the vessel. This setup was far from being perfect from a bioprocess perspective, but the goal of this experiment was testing the OD sensor and not the generation of biologically relevant samples.
As shown in the figure above, our DIY online OD sensor capable of recording growth curves at a very high resolution. By building multiple sensor units it is certainly possible to increase the throughput and data quality of strain characterization experiments.
Assembly
The assembly of the bioreactor electronics was the biggest challenge that we faced. Each device had a defined pin protocol which was required to be maintained and tagged to the controlling firmware.
Outlook
Since the current version of our project uses all digital ports of the Arduino Uno, it will be necessary to change to a bigger controller, for example an Arduino Mega, for future designs.
Then it will be possible to include further analysis units. We already took some steps towards an offgas analysis with CO2 and O2 sensors, and CHX analysis could be interesting for various other experiments. An integration of these components would be easy. Furthermore we cooperated with a community lab, the Technik Garage, who developed a pumping and heating system for the water bath for us. To include this in the whole system had given us many more parameters to control the experiment.
Biological experiments often run for more than a week. It is time-consuming to monitor the experiment all the time. It would be the most convenient to control your experiments via a local net or even the Internet. Imagine lying on a meadow in summer with a cool drink in your hand and while you check the OD instead of visiting the lab just to take a sample.
As the number of opportunities increase, so do the challenges. Hence the modularity and open source features of our system serve as a platform for future development.
References
- ↑ http://www.thingiverse.com/thing:642192
- ↑ http://www.ti.com/lit/ds/symlink/tsl235.pdf