- You were the first in producing humanised antibody, using rational engineering technologies: how much or in which contexts do you think selection methods could be more advantageous, compared to the rational engineering methods?
In the context of antibodies, let me explain why the rational approach was important and I think still continues to be important. There are a lot of antibodies out there with a potential therapeutic value to be made by the mouse monoclonal antibody technology and the technology we developed: you can take those antibodies, almost everybody has antibody working in their freezers, and if you want you can just convert them into a humanised form, so you can access the entire field of mouse antibodies. If you want to use selection technologies, and if you want to make them from human repertoires, you have got to do all the work again, you have got to make the usual antibody, and there are some limitations in the repertoire technologies: they are very good if you have got a folded protein, they are not so good if you got to do with membrane antigens and if you are dealing with a protein that is perhaps not folded so well. And you have to bear in mind that perhaps hundreds of years’ worth to work have been done in obtaining some monoclonal antibodies, covering let say the entire universe, so you do not want to forget the entire universe by starting again with the selection approach. So I think the situation is this: as a generic tool, it is great for taking existing antibodies and converting into a form that seems to work as well as genuine human antibodies. On the other hand, starting from repertoire technologies could be appropriate if you did not really have to hand the appropriate mouse antibodies, you have a good folded target, and in fact, you go ahead in doing the selection work. In my own view, it depends on what the starting point is, where are you starting from.
- We recently published a method to select intrabodies against PTM. Do you think that these type of intrabodies could have a future as therapeutics besides their use as specific research tool? In other words, do you think antibodies and intrabodies in particular will be ever used as gene therapy tool?
Let’s start with the last part of your question first: do I think intrabodies will be used in therapy? Yes, I do. I think that it has been a long time coming but I think it will come. The people have already started to use gene therapy in limited situations, however I think you have to choose and thinking about therapy: we cannot think about therapy in general, you need to think about what disease we need to go for. And so therefore do you start … where would you look? I suppose first of all you can try major diseases that are lethal, where you can justify applying gene therapy. The other possibility is that you can look for very rare diseases, where the genetic defect is known, in particular diseases where perhaps you have an overactivation of an enzyme activity that you can block. In general, it could be useful in certain situations where you end up with, for whatever reason, a pathway that is too active and you have to shutting it down. But you have to identify the disease first and what I would do is starting from that point: what is a good disease we have a good mouse model for, let’s pursue that mouse model and validate it.
The second issue is about post-translational modifications. If you demand about post-translational modifications, and the extent to which that is particularly relevant in any particular disease. Is it causative or not? I don’t know, I am sure someone does know, so if you are asking me a question “Well, is this relevant?”, I would be saying “You need to tell me the disease that could be important in and why do you want to tackle it”. With all of this kind of work, if you want to take it forward into the clinic you have to say what is the disease, why would you do that disease, what would be the ethical issues with doing it, what would be the regulatory issues with doing it and you have to have that in your mind; at the same this advances you in whatever model system you are advancing.
I do not even know if I would focus on PTMs at the beginning: my own view is that intracellular antibody is well enough validated. It has been proven to inhibit enzymes, to block viruses’ assembly from early work that Antonino Cattaneo did many years ago when I knew him in Cambridge, that is in the case of plants and so you have a potential anti-infectivity, you can imagine natural cell proteins regulating their activities. I would focus on whatever therapeutic model is going to be the easiest rather limiting yourself. It is difficult enough to find a really good therapeutic model where you can justify intervention, so it has got to be something where people is going to die and you have got enough patients to treat and I think this is the way the field will advance, by proving it in one or two key clinical applications.
- In a review of 1991, you and Milstein, quoting Paul Ehrlich, said “We see a jungle of technologies, old and new, stimulating each other: in the immediate future, most of them start with immunized animals”. This was the state-of-art at that time,
1991? Did I say that? I was really prophetic!
A long way has been made from then, passing from the monoclonal antibodies, to humanised or fully human antibodies using new methods of selection such as the phage or ribosomal display etc. Do you think that nowadays antibody-based therapies are outdated and that the future will be in the antibody mimetics?
I personally think antibodies have got a long way to go still, is necessarily going to be into the development of technology. There are also some technology clean improvements: one can imagine making an antibody for example giving it a very long half-life in human serum. I think this could start transforming our used medicines: rather than taking a pill everyday maybe you can get an antibody once a year and I think that could be quite an exciting use particularly among old people who forget to take that pills every day. For people who are systematically taking statins or other blood pressure related drugs, that have a pretty much constant regime, as I am (I have my blood pressure that is controlled), it would be great if I could just forget about it and just take those antibodies once a year. Also, for example, for vet labs or vaccinations.
But at a some point, the half-life of the antibodies becomes very important, so that is built into the design of antibodies. It is not to say that you could not do it into the design of other things, but antibodies are quite a beautiful system for building that property in having a long half-life. There might be other ways you could use antibodies and probably the advances are going to be less in the technology, for example extending the half-life, and much more in the targets you go for. There is so much biology you can interfere with, and then antibodies are wonderful reagents for doing it. In particular, because of their half-life, they could not give the kind of blockade to reset the ligand interactions that you could easily achieve with a small molecule, which gets cleared very quickly. I set up a small company called Bicyclic Therapeutics, which deals with small antibodies mimics. Those are the future and, in my view, it is going to be slightly different from the way which we use antibodies, as you are looking at different kind of approaches.
So antibodies still will have a long life!
I think antibodies will go on for a long long time, and of course we also have to bear in mind that, unlike the chemical antibody mimics, which we are involved in making, is not impossible that people could find ways of generating antibodies in vivo for example by transfecting into the cells, having the cells producing antibodies. And in fact you have continuous production of antibodies and again my guess is one day that might happen as well. You might not rely on this physical dose but at that time maybe they will take a subset of stem cells and some antibody genes and they will continue producing them, and you may have monoclonal cells that you can use to kill only things that are out of control. There is quite a lot to play for. My own guess is that what will happen in these technologies will wax and wane and change moreover many years and I will be very surprised if antibodies don’t continue to be in somewhere or another a major player.
- As you said before, in the last years, you focused your attention on bicyclic peptides, their selection and use as inhibitors: what are the interesting features of these molecules, compared to the Abs (as you said they have different kinds of applications) and, in terms of expanding the chemical variability, have you ever thought of incorporating unnatural amino acids in them, by using the recent technologies available at LMB?
Yes, the bicyclic peptides have the advantage, at the moment, you can select them on the surface of phage and we can get good binders or good lead binders to particular targets. Because they can be synthesised chemically, we can of course incorporate different amino acid analogs, and we do that quite routinely to avoid cleavage of the peptide or sometimes you can play with some sidechain variants to stop cleavage.
I think we have the ability to chemically derivatise and do screening of bicyclic peptides with novel amino acids incorporated but that of course means that we have a selection process and then we have a screening process. The ideal will be to go with a selectional way and use technologies such as that developed by Jason Chin in the LMB to incorporate novel amino acids directly. But, first of all, that technology at the moment does not have the ability to introduce anything you want, meaning there is a limited number of variants you can introduce in, my guess is they may find ways of incorporating a much greater range of variants. There are some other issues with it, from our point of view, in the sense that you usually get a less efficient display if you try to use that, because quite often they are using amber stop codon and you get some stoppage and then….. So you are trying to achieve a high level of display and then quite often you also end up with problems because actually do not get the read-through. In the end I think it will be useful and indeed we did consider it and we had a discussion with Jason about the possibilities, but because we have a company, and you have got to make something in like 2-month time and that is going to have a higher potent affinity. In term of proceeding now, would I use that method? No, because it is going to take me too long to develop it, we have got too much tightened timelines. Is it worth that someone develop it? Yes, it is and I think in the end it could be a very useful adjunct technology, possibly even a primary technology for generating a large repertoire but you will have to generate a large number of different variants to make it competitive. I have to say: do you know you have 20 amino acids? Is not that enough for you? I can see in the case of a reporter group, it might be very useful, or in the case perhaps of putting a chelating agent built in directly into your structure, but usually with the bicyclic peptides we can just hang it off the backside of the molecule, so we can do drug conjugates. We can do all sorts of things, and link them on chemically, after having selected the affinity lead. The question is: is it good enough for generating things with binding affinity? Yes. Do they get cleaved by proteases? Yes, but not so fast and you can make other changes if it turns out to be an issue.
- You spent your career in the Academic, but at the same time, you founded and directed three different biotech companies. How difficult is, for a scientist, to conciliate the intellectual freedom of the academic research with marketing demands of a company? And moreover, what are the main skills that a scientist needs to acquire in order to be able to manage successfully a company?
The first part of the question relates to reconciling the way one works in a company and the way one works for the lab and I do not. When I’m being an academic, by work in a lab and focused on my things as pure intellectual, I try to follow my curiosity and be a proper intellectual: write up papers and I try to focus on important problems that I think it is worth solving (because if you are going to spend your life you might as well focus on something that is important). Many years ago, when I was very young, I remember saying to my own supervisor: “I think that’s a really interesting problem” and he said “Bugger interesting: is it important?” And he was right. If it was not that important, it was really interesting but he said: “focus on important things and then secondly don’t speculate, if you think it is important put a stake through its heart, do it properly”, which again was a very good advice, and that is what I try to do in my own basic research.
As far as a company is concerned I am kind of a red in tooth and claw in a company so I focus and I worry about money. I just switch to a different mentality so I am like a Jekyll-and-Hyde person: you just behave in different ways in the different contexts. Actually, I am not as quite as extreme as that because in the end you have to be able to link the research to what is going on in the company so I have to be sensitive to the way academic bodies are going to behave and the way that they expect to be treated. For example, sometimes companies can treat academics in a condescending manner and, because of what, I know what goes on in Academia and how that can be understood. I am sensitive to the ways in which you have to treat academics. From the other side I’m also sensitive to understanding the issues that companies have with academics who never want to focus on the problem, but always want to say: “oh yeah I want to do this, I want to do that, I want to do some other things on the side”. It is a matter of trying to get some mutual understanding of the other side, from different perspectives and then trying to adjust myself to being able to wear in one hat and then taking off and putting on another hat.
So did you ever have a conflict on taking a decision for the academic?
You have conflicts: you just declare them. In one case, I did have a conflict, I do not want to go into it, but I thought the company was being entirely truthful. At that point, that was probably a good commercial business for them but there’s certain things that I can’t get involved in and I think if that’s the case you have to say if you are not being truthful and in a way what you are saying may be thought to be misleading but I think it’s important that you just are not associated with that. Particularly if you are on the board of the company because that will have the ability to damage you in your academic roles that, of course, is your major job and so my view is certainly: the most important thing is honesty.
- According to your experience, how much and how the approach to the patent issues in the scientific community has changed over the last years?
When I started, we are talking about the 70s, there was a very little understanding of patents and the awareness of that has increased over the years. I still do not think that academics necessarily know as much as they should about patents, but quite a lot of that comes from experience. I think we have some issues to do with their understanding of patents, because what is tended to happen is academics have not really wanted to find out about patents because it is so complicated as legal type of stuff. Therefore what you have got is a class of middle men and middle women which will be the technology transfer department, stepping up an occupying the space between academia and companies. In some senses that is good because it means that if you do not want to talk to companies you do not have to and you can go through your technology transfer people. But the disadvantage is that academics do not feel any obligation to try to understand the company point of view and to understand what patents are about. I am setting my case: I had the advantage with Cambridge Antibody of setting it up when the Medical Research Council did not have a technology transfer section and that proved too much quicker than when I was involved in the two other companies Domantis and Bicycle Therapeutics.
Does that question that the, in my case, MRC technology was helpful? In a number of ways, but they also, in other ways, can slow down the process. I think you may be better, you may get a more refined agreement or refined understanding. It is certainly true in the case of Cambridge Antibody. Without that level of diligence, we did a number of things on a handshake that in the end came to bite us at a later date. On the other hand, we did move and we got the company established within 2 months and money in passing hands and the whole thing on the road: if that has not happened, we would be beaten. So actually sometimes you just have to make these compromises. Particularly with biotechs, speed is the essence and sometimes technology transfer departments, in their wish to do things properly, can actually slow down to a sort of very slow speed that makes it difficult to move quickly and in the end can be off putting for scientists and for companies. So it safeguards the interests of the different parties but sometimes it can lead to slowing down those interactions.
On balance, I would prefer a system where the academics are empowered to go directly to companies if they wish to. I prefer a model in which then the institution always gets to share of wherever they get, and that is simple. So yes, in some cases academics will be taking advantage of, in other cases investors may take advantage of, but I think you need something where academic is free to do it on their own. One very good thing about Cambridge University, which differs a little bit compared to the Medical Research Council, is that in Cambridge University they have a policy. (I have to be careful about the exact percentages but I want to give you the flavour). If you are working within the university and you think it is important to file a patent, you go and talk to technology transfer people and they say “Thank you, Cambridge University is the owner of this patent or will be the owner of intellectual property. You have a choice: we think it’s worth exploiting and we will pay the patent fees etc. and if we take it forward you will get a certain percentage” (I cannot figure out the exact figure, I think it is one third). On the other hand, if after 3 months of us looking at it, we have not done anything or you actually decide that you do not want to work with us you can go and do it yourself in which case you go ahead you pay all the fees than you can get two thirds. I do not know if the numbers are quite right but it is the principle: because there are so many ideas pouring out of academia they cannot possibly pursue them all and some people are actually much better in pursuing that, than in other cases someone may not wish to pursue it and would just handed to someone else. It is a great idea: I do not have time, I need to get back to my lab and I cannot be bored with this talking to the industrials.
It is important a model where really, in the end, it is the individual inventors and scientists who has some sort of control of what is emerging. The institution is also important because there had been cases were professors have taken advantage of the work of their graduate students, claiming they are the inventor and not bothered to look into who exactly had the idea and such like that. One good thing about Cambridge University is that they look in all of that and they will actually make sure they go through with diligence on working out who contributed to it and I think that is an important role in an institution that someone does not actually exploits somebody else, but that does not have to take a long time. Going back to the question “what is the understanding?” I think people have a greater understanding, it differs between universities which have a more free approach and universities where you have got a centralised bureaucracy that wants to control everything and it does affect the level of culture that you develop among scientists in the university. Also affects the culture you have in the technology transfer organisation: if a technology transfer organisation behaves in unreasonable manner, if it has a monopoly, there is nothing you can do about it. Whereas if you have the option to say “We are opting out of that, we’ll pay everything’s, I will pay also third party costs like patent, things you can cut already, and I can take the rights back and you will have a share of wherever I get” I think it is a great model.
You need to have a good model for the exploitation of the intellectual property.