ITI Life Sciences, Dundee, UK
Name: Fergus McKenzie
Position: Program Manager, ITI Life Sciences, Dundee, UK, since 2004
Background: Various positions, Amersham Biosciences (now GE Healthcare), 2000-2004; faculty member, State University of New York, Stony Brook, 1998-2000; professor and postdoc, University of Nice; PhD, biochemistry, Glasgow University
ITI Life Sciences, a Scottish government-backed initiative dedicated to expanding Scotland’s global presence in the life sciences market, this week launched a £9.5 million ($18.8 million) research and development program to develop an automated process for producing high-quality human stem cells for drug discovery.
As part of the three-year program, ITI will collaborate with scientists from the University of Glasgow and with Swedish biotech Cellartis, which will set up an R&D and manufacturing facility near ITI’s Dundee facility.
CBA News covered ITI Life Sciences’ first initiative, a £3.7 million effort with Scottish biotech firms Edinburgh Instruments, Hannah Interactions, and CSS-Albachem, to develop 3D fluorescence lifetime cell-based assays for drug screening (see CBA News 2/15/2005).
According to ITI’s website, that program failed to meet its commercial objectives with the prescribed timeframe or funding, but produced intellectual property that has been licensed by individual entities for further exploration. In addition, ITI has since initiated three additional projects related to automated scientific text mining, transgenic animals, and cardiac biomarkers.
This week, CBA News caught up with ITI Life Sciences Program Director Fergus McKenzie to discuss his vision for ITI’s new stem cell initiative.
Can you provide some additional details about the program announced this week? What types of methods do you envision for improving the automated production of stem cells?
At the moment, human embryonic stem cells are the realm of dedicated and very specialized scientists. It requires a very high-tech ability. You can’t just say “I want to go and grow human ES cells.” If, for example, I sent a scientist some human ES cells, I guarantee they’d kill them for months until they became used to growing them.
The other problem is that they’re always grown to a very small scale. If you want to grow human ES cells and produce assays from them for drug screening, you’ve got several problems. One is that they are not robust enough at the moment – you can’t produce them to scale. The second thing is that it’s not easy to differentiate them into your chosen or favorite cell type. You can do this in the corner of the lab, and it might take a few weeks to produce a population of neuronal cells, but not all the cells will differentiate. They wouldn’t all differentiate the same way every single time, so the process to get there is not in place.
We’re interested in putting the building blocks together to automate the production of human ES cells as well as their derivatives. In the first instance, if you take human ES cells, they’ll grow on a feeder layer of cells, in something like an in vitro fertilization dish, which is maybe five centimeters in diameter. We want to take this up to growing them in a T-175 cell culture flask, with a robot, so there is minimal human intervention. That’s not really been done before, so that’s a big step forward – to automate production.
The second thing we’re then going to do is produce 96-well plates of human ES cells that are ready for screening. We’re going to undertake a screening program looking at chemicals, compounds, and growth factors that can either maintain pluripotency of the cells; or allow them to grow and divide a bit better; or, at best, entice them to differentiate along the chosen path we want them to take.
That’s going to eventually produce a range of standard operating procedures, and in a few years’ time, you can look at our SOPs, take your cells, and you too, in the comfort of your own lab, can automate the production of, for example, cardiomyocytes on a regular basis. We feel that is what is lacking at the moment. People want to take stem cells through to cellular therapy and put them into patients, but before you can do that, you have to have the bioprocess in place. That bioprocess is probably going to be the same for drug discovery you’re doing with your cells, or for cellular therapy.
We are interested in drug discovery – in getting these cells into big pharma, so that they can screen them. That’s what we’re after.
ITI said that it will own all the IP related to this. Can you tell me how the different components of this program – Cellartis and the University of Glasgow – are contributing to this project?
Cellartis are the experts in the production of human ES cells. They have generated upwards of 30 different ES cell lines to date, and are well acquainted with the husbandry of these cells – how to keep them alive, to ensure they are karyotypically normal, that they are going to differentiate, or that they are growing in an undifferentiated manner. So Cellartis will contribute all of the quality control and validation.
From Glasgow University, we’re pulling in a couple of signal transduction experts, because we’re trying to do things a bit differently. We’re trying to apply signal transduction techniques to ES cells, and saying, “let’s take a look at all the GPCRs or tyrosine kinase receptors expressed on the surface, and let’s intelligently attack the signal transduction pathways to try and turn these cells into a different cell type.” Rather than just a random fishing trip with chemicals, we want to do a more intelligent approach, to try and modulate signal transduction pathways that we know exist and are active in ES cells.
As the research program progresses, we anticipate another announcement in a few months because we’ll have to pull in a screening arm and a medicinal chemistry arm. We have these all lined up, and we wanted to make sure we could transfer the technology to other labs that are not stem cell experts. And if we can [do that], we know we can go ahead with the screening and medicinal chemistry. I would expect in another three or four months another announcement from us taking the program forward to the screening stage.
That would be with another partner?
Yes, it would be at least another partner. We’re not trying to make labs that don’t screen into expert screeners in three months. We would rather see labs that know what they’re doing, and get them interested in our program and pull them in. We’re obviously fairly well-advanced in those discussions, and should pull that in quite shortly.
Will there also be partners for the robotics and instrumentation?
We kind of already have that expertise in that we’re going to have people on board who are experts in robotic systems. We’d be happier if we could pull people on board who have already used the robot systems that we’re going with. We’re going with The Automation Partnership in Cambridge, UK. My background is with Amersham Biosciences and GE, so I’ve worked with TAP in the past and am aware that they are possibly considered the best automation system out there. Many companies such as Tecan and Hamilton also have excellent systems out there; it’s just that we’re aware of the TAP system and quite comfortable with it. We’d like to hire the appropriate expertise so we have people who are good at stem cells, and people who are good at robotics, and then marry the two.
You mentioned that your initial production of stem cells would be feeder cell-dependent …
At the moment, the other thing we have to consider is cost. If you grow human ES cells on a feeder layer, it is actually cheaper than it would probably be to grow them in a feeder-free environment. The reason for this is that if you grow human ES cells in feeder-free systems, you require a great many growth factors, but particularly fibroblast growth factor, which is expensive. So it’s actually more cost-effective to grow cells, for bulking up at least, on a feeder layer.
You can imagine that as time progresses, you would like to get off of the feeder layer, because you want to make the system for bulk-up and automation as simple as possible. We could envisage heading in that direction. But one thing we have to be quite careful of is that there are quite a few competing companies that have intellectual property in the area, and we have to make sure we’re absolutely above board. If somebody has any IP in the area, we will try our very best to license that IP.
Geron comes to mind. CBA News recently reported that Invitrogen has established a partnership with Geron for feeder-free methods of stem cell production, because feeder layers can be time-consuming and possibly contaminating.
That’s true. It depends on what you want to do with your cells. If you’re heading toward clinical development, then you cannot grow cells on a mouse feeder layer, because then they’re coming into contact with another species. If you’re going toward clinical use of the cells, you have to get to a xeno-free system, where the cells have never seen animal products, like fetal calf serum or mouse feeder cells, and they are ready to go into the clinic. In Europe, there are no cells corresponding to those criteria, which would be called GMP, or good manufacturing practice. To my knowledge, such cells do not exist in Europe at this time. Geron says that it has at least one cell line which conforms to GMP criteria.
Bear in mind that what we want to do is drug discovery – to produce cell model systems for big pharma for new medicine discovery. We are not heading to the clinic at all, mainly because we do not have any GMP cells at this moment in time. We can imagine that in a few years, once we’ve got the process nailed down, then one could use our process with GMP cells to go into the clinic. But that’s not where we’re heading at the moment.
Invitrogen also has announced a partnership with Cellartis for engineering differentiated human ES cells. Is that similar to what you’re doing with Cellartis?
It’s slightly different. What they are doing for Invitrogen is producing a range of ES cells with fluorescent markers, and Invitrogen, I suspect, wants to use them as research tools to follow cell differentiation along specific pathways. That can seem quite similar, because you require the ability to handle human ES cells, and manipulate them genetically, and maybe insert DNA into them – but after that they should stop.
One of the biggest issues here is, of course, that we need to separate out what Cellartis is doing for themselves and for other companies, compared with what they’re doing for us. We’re very careful to try to wall off what is happening in our program – because we want IP – against what Cellartis may do with other companies. Hopefully there is enough room for everybody, but that means we have to be careful and look at contracts so there is no duplication.