Infectious Disease Research Center, Université Laval
At a Glance
Name: Régis Peytavi
Title: Project Leader, Infectious Disease Research Center, Université Laval, Québec
Professional Background: 2002 — present, Project Leader, Infectious Disease Research Center, Université Laval, Québec ; 1999 — 2002, Postdoctoral fellow at University of California San Diego.
Education: 1999 — PhD, Université de Montpellier, France; 1992 — MS, Université Paul Sabatier Toulouse, France.
A paper published in this month's edition of Clinical Chemistry, entitled "Microfluidic Device for Rapid (less than 15 min) Automated Microarray Hybridization" and co-authored by researchers at the Infectious Disease Research Center at Quebec's Université Laval, takes aim at increasing the speed of hybridization in microarray-based experiments.
Led by project leader Régis Peytavi, the group — which also included Quebec's Central University Hospital and the department of mechanical and aerospace engineering at the University of California, Irvine — invented a microfluidics based-tool that it claims can reduce hybridization from hours to minutes.
The tool comprises a microfluidic flow cell built into a polydimethylsiloxane substrate where the substrate is bound in reverse to the microarray slide to form a microfluidic unit. The microfluidic units are in turn placed onto a disc-shaped support fixed on a rotational device and use different speeds to disperse the liquids in the experiment.
To learn more about the utility of the new invention, BioArray News spoke with Peytavi last week before he headed to London to present at the inaugural Advances in Microarray Technology conference.
Why did you decide to publish your paper in Clinical Chemistry?
Why? Because we wanted to target more companies and we thought that Clinical Chemistry was a very good [publication] for that.
So what led to the development of this new tool?
We wanted to develop a very rapid diagnostic for disease and we raised CA $12.6 million (US $10.7 million) to develop a test for sepsis and respiratory viruses to deliver the results within an hour. Of course, when we are using passive microarray hybridization [it] takes too much time to be able to do all the tests in one hour. We have our collaboration with UCI in California, so we could access their microfluidics technology and we wanted to speed up and automate the process.
But what is your expertise in microfluidics?
I have two years of working with [UCI], so I help them to solve problems in more biological terms, because I am more of a biologist. But I have [gained] enough experience with microfluidics also. I am the core technology coordinator for the [entire] project. We are eight teams around the world — most of them are in Canada, one in California — and there are physicists and engineers and molecular biologists and sociologists and physicians and so on.
If the need exists for high-speed hybridization, how come this kind of tool hasn't been created yet?
Because there are a lot of problems. I think it is not that easy to make something that works very well and most of the time the people [that are working on it] are more like physicists, not biologists. They use oligonucleotides to make a proto-concept, but oligonucleotides are very short and in passive hybridization systems they can diffuse very fast and there is not very much difference between using hybridization of oligonucleotides in a passive system and using it in microfluidics. Where you see a big difference is when you use amplicon because an amplicon is much more difficult to get a good signal in five minutes, for example, than in oligonucleotides, which are very easy [to get a signal from].
I think we have developed a good expertise in the coating of the slide and also in making a very sensitive probe that is able to detect a very good signal with an amplicon. Because that's real life. Oligonucleotides are not really real life. You want to do a molecular assay — you need to have amplicons or longer DNA fragments.
Does this work for biomolecules other than DNA?
I don't have so much experience on that but I would say yes, because you increase the chemical rates of reaction using microfluidics. Especially when you use flow-through, you force the liquid in the target over the probes, so the distance in diffusion is reduced using this technology, [which] is why we can reduce the time of hybridization down to five minutes.
How does the invention actually work?
We have already mastered the glass slide microarray technology in the lab, and we have all the devices to make the arrays and to read them, the spotter, the scanner, et cetera.
So we wanted to do something that could be removable [in the lab]. So the idea was to take classical standard glass slide microarray to print the array using the spotter, the classical coating, et cetera. By chance the coating is kind of hydrophobic and we needed that for the centripetal valve.
In California they designed the flow cell, they made these small channels into the polydimethylsiloxane … that can stick to the glass without the need of any glue or chemical bonding. So that is the beauty of the system because you juxtapose the PDMS flow cell above the array.
The hybridization current is just above the microarray, the reservoir for the labeled sample is 2 microliter, then there is the10 microliters of washing buffer reservoir, 10 microliters of rinsing buffer, and everything is connected by a central canal with very tiny canals 15 micrometers wide and 20 micrometers deep, and this canal goes through all the way down the hybridization current.
You fix the support for the chamber on top of a motor which is controlled by computer, and using different speed of rotation of the disc you can disperse the difference of the centripetal valve to sequentially open first the sample, which will hybridize for five minutes. After we increase the spinning rate of the disc with the program, the washing buffer will flow through the hybridization chamber and wash the chamber, and then we will increase the speed again and the rinsing buffer will flow through the chamber. The whole process takes fifteen minutes including one minute for drying.
This is done at room temperature. So we set up all the protocol to be specific enough to be done at room temperature and still be able to detect [a signal]. The beauty of the system is that as soon as you dry the slide, you can peel off the flow cell from the glass slide microarray and the microarray is ready to be scanned by a standard scanner.
You mentioned in the paper that there are other microfluidics-based systems but you said that their use is prohibited by being "too complex and too expensive." How does your invention compare with regards to complexity and price?
The main problem I see with the other systems is the valves. Our valves just rely on the tension surface and the tension surface is overcome by G-force by just increasing the rotation of the disc. The other systems are more complex to manipulate the valves and for diagnostics and for something to make hybridization you need very cheap flow cell. PDMS is very cheap to manufacture, it is very cheap to fabricate. We can use it for one time and throw it away for any contamination. We use our own glass slides — we do our own coating — so it's maybe 10 cents per slide as well.
Both the microfluidics are very cheap and the arrays themselves are very cheap, but we think we may go to plastic later — we are working on that, to make it in a larger quantity.
There are a number of commercial products mentioned in the paper. Reagents from Sigma Aldrich, spotters and reagents from Telechem International, a Packard Biochip Scan Array. Did these companies help you at all in your work?
Absolutely not. We just tried the products and used the ones that were best for us. That was it.
To test the results of the hybridization you used Staphylococcus aureus. Why did you choose that species and how did the test work out?
We work in an infectious research disease center, so our lab is really dedicated to the diagnostics for infectious diseases and we have a lot of data. We have a bank of pathogens so we used what we have and we used this species, but now we have developed probes for 68 pathogens you would find in sepsis, for example. We have much more data and now we are using other probes.
When I arrived three years ago and started using these probes they worked very well and we knew that those probes were very sensitive and that's why we used them.
There are other companies that are working on these rapid tests for infectious diseases. What would be your advice to others that are working in this area?
I think that of most of the companies I have seen, like Motorola or ST Microelectronics, ST was right to get a partner with a company [with clinical connections] like Mobidiag (see BAN 5/18/2005).
That's the way to go, because with Motorola, if you do not have any biologists or any good people in the field, you will have something totally useless. The way ST has done it is much better than Motorola, in my opinion because [Motorola] did not make any contact with the biologists.
How can you optimize your new invention? How can it be improved upon?
We can reduce for sure the time for washing. The rate of hybridization I think is pretty optimized. I think we can reduce it to 12-13 minutes total.