APPS

Objective and context

In laboratory, routine tagged-protein (His or strep tagged proteins) purification is usually made manually with the the help of resins. This procedure is often, so to say, boring and repetitive. Automatic alternatives exist but are costly (8000 - 15000), bulky and often too specific for such simple but long task. The idea is to develop a small, (semi-)automatic, cheap(er) and modulable protein purification system for most of the routine task.

Routine Protein Purification

In the world of protein engineering, enzymology, drug development and synthetic biology, amounts of pure proteins are required to study and evaluate the potential of an enzyme to be a candidate for future development. One routine example of protein production and purification would be the following: A bacterial strain is transfected with a plasmid containing the protein of interest (POI) tagged at its C and N termini with His and Strep tags. The POI would be expressed by the strain (i.e. e.coli) in liquid culture. The bacteria in the cultures would by lysed and the homogenate purified. A usual purification would involve to prepare a column and pack it with a resin such as the His purification resin. The column would be (i) equilibrated with a first buffer, (ii) the homogenate would be slowly passed through the column, (iii) a second buffer would be flown through the column to remove all the remaining unwanted bacterial debris. All these step are made with the help of a peristaltic pump to apply suction but the performer has to stay by the system to ensure that the resin never becomes dry. (iv) A last buffer would be used to unblock the protein from the resin and collect it in a gravity dependent manner. These purification steps will be then performed again with a second type of resin to ensure total purity of the protein.
This procedure occupies a large amount of the time of a scientist and could be in this way automatized.

Protagonists

The idea was launched by Yann P. when he got fed up by manual protein purification.
He was quickly backuped by Gustavo S. and his impressive skills in prototyping and electronics.
Together we work in pair, Yann pointing out the requirements of a system for a decent protein purification and Gustavo to propose quick and clever alternatives to satisfy those standards.

System design

alt text

Schematically the system would be composed of several modules: A buffer holder where several 50ml tubes can be stored, a column holder where various sets of columns could be fitted, a 280 LED based optical cell, a peristaltic pump, a simple trash/sample collection system and a controller (Arduino or Raspberry Pi) + display.

UV system

The UV system will be composed of 280nm emitting LED and a UV detector. The optical chamber will be made as a 1cm U shaped pathway (described below)

280 nm LED

Several European and American LED providers exist but propose prices that are way above reasonable price for such LEDs.
The following LED providers were investigated. (accessed 08.2015)

Provider	prices	links	Comment
Roithner lasers	Expensive	link1, link2
Intl UV LEDs	Around 34.5 $ / LED	link
ThorLab	Expensive	link
Seoulviosys	Expensive	link
Qphotonics	Expensive	link
TME	Around 20/LED (can buy single ones)	link	three units were ordered here Datasheet
Alibaba	- - -	link	Company verified, 6yr business, Iso compliant.
Taobao	1.20.- per LED	link	280nm , 0.5 - 30. mW
Ledwv	- - -	link

UV detectors

We will use for starters UV detectors compatible for arduino to facilitate the development. These are often cheap and can be bought for testing.

Provider	name	price	links	Comment
Adafruit	GUVA-S12SD	6.50$	link Datasheet	22% of maximal theoretical absorbance at 280nm
Adafruit	Broad spectrum SI1145 Digital UV Index / IR / Visible Light Sensor	9.95$	Datasheet link

UV Transparent material

We need UV transparent material for the flow cell. We will first try to cut small portion of those 1cm UV spectrometers optical cells.
Quartz pieces could be used as well.

Hydraulics

Since the flow system is based on a peristaltic pump, low pressure tubing could be used. We will base the system on what is the most standard luers and connectors to optain an easily plugable watertight system.

Luers and connectors

Several large scale providers were investigated and often large batches match in these companies match the prices of single units in specialized companies (seriously Bi**ad... )

Provider	price	links	Comment
Cole-Parmer	10-15 $ for 25x transparent luers units, 10$ for manifolds	link
Nordson	- - -	link	Large batch orders of luers, fittings and tubes.

Columns

We will base our system on standard polypropylene columns but will expand to larger varieties of columns afterwards. Several providers were investigated.

Provider	price	links	Comment
Quiagen	193 per 50 units	link	1ml resin / 6 ml total
Quiagen	250 per 50 units	link	5ml resin / 15 ml total

A decent alternative for closing the column top would be the following product (95$ / unit, big nope but you get the idea) For the purification of 1 to 5 mg of protein 1ml resin is usually enough.

Pumps

The system should be operating under low pressure and usually tubings should be in silicon. The pumping unit will be a peristaltic pump. Several providers were investigated.

Provider	price	links	Comment
Adafruit	25$/u	link	likely compatible with arduino (example)
Sparkfun	120$/u	link	pretty expensive...
Welco	~70euros	link	Several compact types of pumps
Elecrow	4$/u	link	cheap, 9v, 1.2m suction.
Ali Express	7euros / u	link	12 V pump, the design and size looks adequate

Ali Express has a good quantity of different pumps available.

Tubings

Usually we will use silicon tubings and luers adaptators. We are working at low pressure so this type of setup should work.

Valves

We need to have automated cervo valves. We will either use plastic manifolds attached to a cervo or directly electronic valves if they are affordable.

Housing

Design 1
The final goal is to have a modular housing. The idea is to have a cubic device where each of the "walls" has a functionality. Therefore, you would have a wall for the column dispense, a wall for the pumps, a wall for the optical cell and lastly a wall for the collection. On the top you would have the housing for some tubes. In the end the device should be easy to dismantle and rebuild. It would be interesting to have different grade of quality for each walls (let's call them modules for now). For example for the column module, it should be easy to replace any column easily and adapt bigger columns without too much struggle. As you can see below, the whole device looks as presented above. The design was made with Freecad. The whole system would be held together by four metallic bolts (orange line in dismantled fig). The following parts are shown as: green : pumps, dark blue : 3-way valves, light blue : propylene column, orange : UV cell, yellow : 50ml plastic tubes. The scale of every object was made according to technical data but will be confirmed upon material arrival. Several parts were not taken into account in this first design and should change the overall size of this first approach : Electronics such as plugs, boards, Raspberry Pi & screen; Hydrolics such as tubings, luers, fittings; The size of the UV-Cell might be underestimated.

• 

  side 1

• 

  side 2

• 

  top

• 

  dismantled

Experimental

In this subsection we list out everything that has been tried for each of the modules. Consider this section more as a lab-book than as a final document.

Valves

proto 1 Gustavo put together micro electro-valves using 3 way plastic microvalves glued to a cervo and held together, in a version by wood, and in the other version by a 3d printed PLA structure (pictures below)

• Woodvalve1.png

  Wooden valve

• Woodvalve2.png

  Wooden valve

• Plavalve1.png

  PLA valve.

Costs

Here we summarize all the elements we bought to have an idea of the monez we already invested in the project.

Item	Provider	Price/u (CHF)	Number	Shipping	Total	Added Total
LED UV 280nm	TME	19.-	2	17.-	55.-	55.-
UV detector	Adafruit	33.- (total)	2 + 2	35.-	68.-	123 .-