Please could you tell us why it is important to be able to isolate and concentrate cells from whole blood?
There are several reasons for this. One is because analysis involving different kinds of blood cells may require you to be able to separate red blood cells from white blood cells, or plasma from whole blood etc.
Another reason is so that you can separate circulating tumor cells. These are cells that can be shed from a tumor and are circulating around in the blood. They are often larger in size than red or white blood cells, so there is an interest in using those as markers either for early detection of cancer or to monitor progression of the disease.
One of the main problems with this is that they are present in very small quantities, such as one in a million, or one in a billion, so there is interest in isolating those components so that they can be used more effectively for those kinds of tests.
What method has previously been used to do this?
There are different ways to do this depending on if you are separating based on size, like we are doing, or based on density, such as what would be done by using a centrifuge.
What is different about our method is that it is a continuous flow method; whereas other methods are more like batch processes. For example, previously you took a sample and did analysis on that quantity of sample; whereas we are doing a more continuous process. This continuous process means that there is the potential to analyze larger volumes.
Another reason a continuous approach could be beneficial in the future is for, what has been described as, dialysis-like therapeutics. The idea is that if you have a pathogen, such as a bacterium or some other component that is in the blood stream, that could be filtered out as part of a treatment. This is something that would only be possible in a continuous approach. You wouldn’t be able to take blood, centrifuge it, and then re-transfuse it back into the bloodstream.
How did your research into this subject originate?
We had been doing work in the past about trying to create branched (tree-like) networks of microchannels in these biodegradable substrates for tissue engineering applications. One of the barriers of making tissue-engineering scaffolds is having the internal vasculature to enhance transport of nutrients and waste.
We developed the ability to make this using an electrostatic discharge process, but for the bioapplication, to actually feed cells within those channels, the networks we were producing directly from this process had an average diameter that was too small. This meant it wasn’t optimal, so we would have increased pressure to inject the cells into the network to do the seeding. It was not optimal conditions for the cells to survive.
So we were looking for ways to widen the channels and one way we looked at was this chemical method - to use enzymes to break down the biodegradable polymer. We were using polylactic acid as the model for that.
Ultimately, this method wasn’t ideal for that application, but the idea that the enzyme could be used to break down the polymer in a controlled way gave us ideas for other applications and that’s what led to this idea. Because the etching is directed by the flow, that introduces opportunities to pattern the structures or features on the surface in a way that may be simpler than conventional methods at the moment. It may be easier to make certain kinds of structures, like the filter that we showed.
It is not that these structures can’t be made by lithography or other conventional methods, but it is a little more challenging especially when you have different parts of the channels with different depths. Along the flow direction we have a weir or a barrier along the center line and on each side of the channel is a different depth. So that is a bit more challenging to make using existing methods. The enzymatic process, because it is a flow-based process, gives us some advantages in terms of control to make those kinds of features.
What did your new research suggest as a way of isolating and concentrating cells from whole blood?
We tried to compare our method to other methods in the literature. For a size based separation what is generally used is a porous membrane, or a filter, where you inject the sample directly through the filter and the larger sized components would be retained as it would not be able to pass through the pores of the filter. The problem with this method is that it will clog the filter, because you are directing the fluid directly through the filter. It also takes a lot of pressure to force the suspension through the tiny pores of the membrane.
What we’ve been able to do is to make a filtration structure, where instead of being perpendicular to the flow, like where you inject the flow right through the pores, we’re injecting it parallel to it. So the barrier is parallel to the flow direction.
Then when we make this in a curved channel, i.e. the path of the microchannel follows a curved trajectory, that curvature introduces a secondary flow, called a Dean flow. This is well known in any curved pipe. It is like a centrifugal force because you get the fluid on the inside becoming pulled outward due to the centrifugal force. So this creates a recirculation pattern across the pipe, or microchannel in our case.
The idea then is that we have this secondary flow which transports the components across the filter barrier, but the primary flow direction is still parallel to the barrier.
The advantages are that it reduces clogging and it reduces the pressure drop. This is because the primary flow direction is still parallel to the filter, but the transverse flow due to the curvature of the path creates a driving force to more gently transport the components across the filtration barrier.
Another advantage is the physics of this Dean flow mean that the secondary flow becomes important at higher flow rates. The faster the fluid is being pumped through the channels the stronger these centrifugal forces will be. The effect is maximized at higher flow rates. This is good news for high-throughput. Many other methods that have been tried have been restricted to lower flow rates because you have such a high pressure drop. Also going to higher pressures can damage the cells, so if you want to be able to recover the cells in order to analyze them, you are also restricted to lower flow rates.
We don’t have these limitations. This is the main benefit of the geometry of the structure that we have made – because it actually works better at high flow rates, which is the opposite of other methods.
What challenges did you have to overcome in your research?
We’ve demonstrated that this method can work. The main challenges now are adapting it to particular tests or assays or type of cells you want to isolate.
One issue is the physics of the flow pattern. Ideally, you would want to process whole blood directly without any dilution. The issue there is that you need to have the same fluid properties, such as viscosity, on both sides of the barrier. That can be designed.
We diluted the whole blood by a 1:5 ratio with phosphate buffered saline (PBS) buffer. This was so that we could equalize the viscosity on both sides, as we injected the buffer on the collection side of the channel and in the blood in the sample we injected. The separation effect is enhanced when the properties of the fluid on both sides of the barrier are as closely matched. This can be addressed in the future by adjusting the design to compensate for that.
What impact do you think your new method will have?
From my understanding of the literature, one of the big problems not just for this cancer cell analysis but developing any assays that require isolating different cellular components from blood is this throughput limitation. If you want to make a portable device, or a device that can be used outside of a conventional lab environment that doesn’t use traditional centrifuge kinds of preparation. The capability to process larger volumes of blood in a shorter amount of time has been one of the limitations in miniaturization of these kinds of technologies. A centrifuge-type operation is difficult to scale down. So it is hard to make it into a portable test.
In the cancer cell area you are essentially looking for a needle in a haystack, so you need to process larger volumes in a short amount of time in order to isolate the components. Being able to isolate those are cells more efficiently can contribute to existing technologies for analysis of cells and also being able to recover them, by preserving their viability in their natural environment. Some other methods require a lot more dilution, they may require you to re-suspend the cells and so forth. This method doesn’t have those kinds of requirements.
How do you think the future of isolating and concentrating cells from whole blood will develop?
There is this idea of having a dialysis-like therapeutic. Cancer research gives the example where there are components in the blood that have a different size or different mechanical properties than normal blood cells. Any technology that could better isolate those lays the foundation to expand the use of those kinds of biomarkers, not only for diagnostics but also to monitor efficiency of treatment.
This could help towards having personalized medicine, as people respond to different treatments in different ways. That is also true in the case of cancers. Tailoring treatments to particular people to maximize their effectiveness would require this kind of monitoring to be done. Any technology that can help collect the data with finer resolution, more frequently more sensitively, is really the key to enabling that vision to be realized.
Do you have any plans for further research into this area?
We are looking for collaborators to see if we can apply the technology to specific applications. It is really a generic filtration method, so we are looking for collaborators so that we can look at more specific areas to target and directly contribute.
Would you like to make any further comments?
The whole technique has a lot of potential because there are different directions it can be taken. There are families of enzymes that have been shown to have the ability to do this degradation for different kinds of polymers. Even for PLA, we found that this etching rate also has a selectivity to the degree of crystallinity in the material. So we showed that when you anneal the polylactic acid material at a particular temperature, the degree of crystallinity is influenced by the thermal history. So when we anneal it in a way to create isolated domains of high crystallinity and then etch it with the enzyme we can see the surface topology. So the etching is inhibited in the regions of high crystallinity.
What is exciting to me about this from a materials standpoint is that those features are templated not by lithography but they are templated by the material’s own crystalline architecture. It introduces the possibility to use the material’s own molecular architecture as a template to create surface features that could be useful and they could have nanoscale topologies that are difficult to construct using lithography, especially over a large space, like an electron beam or other kinds of lithography methods. It is kind of a new way to pattern these kinds of features on the surface.
Where can readers find more information?
http://onlinelibrary.wiley.com/doi/10.1002/anie.201204600/abstract
About Victor Ugaz
Victor Ugaz is an Associate Professor and Kenneth R. Hall Development Professor in the Artie McFerrin Department of Chemical Engineering, Dwight Look College of Engineering at Texas A&M University. He joined the faculty in January 2003.
His research focuses broadly on harnessing the unique characteristics of transport and flow at the microscale, with specific interests in microfluidic flows (both single-phase and nanoparticle suspensions), microchip gel electrophoresis, PCR thermocycling in novel convective flow devices, and construction of 3D vascular flow networks for biomedical applications.
Ugaz earned B.S. and M.S. degrees in Aerospace Engineering at The University of Texas at Austin, and a Ph.D. in Chemical Engineering from Northwestern University.
He currently serves as a Deputy Editor of the journal ELECTROPHORESIS, Past President of the American Electrophoresis Society, and Chair of the interdisciplinary Professional Program in Biotechnology (PPiB) at Texas A&M.
Key kinase involved in cell division linked to abnormal blood vessel growth in genetic disorder