In last month’s blog I caught up with Professor Daniel Robert (a sensory biologist and Professor of Bionanoscience at the University of Bristol) to talk about how biologists and physicists approach AFM differently and the importance of interdisciplinary relationships.
It was a fascinating and wide ranging discussion about his use of AFM in recent research and his experiences with AFM as a biologist and it was one that we felt was well worth sharing in full with you on the blog this month.....Enjoy!
Professor Daniel Robert
(Image courtesy of University of Bristol)
Can you tell me about a particular area of research that you’ve used AFM on recently?
The most significant application we have used AFM on is the mechanical response of the antennae of mosquitoes, their hearing organs[1].
The question we were asking is: ‘Can we measure the mechanical input at the level of the neuron itself, or near the neuron itself?’
As opposed to using the scanning laser Doppler vibrometer, a light-based non-contact mechanism, which can be used further up the antenna, we wanted to know if we can bring the problem straight to where the transaction takes place, where the mechanical stimulus gets encoded into the electrical stimulus.
What James [Windmill] did was open up a mosquito system, so a microscopic system, and approach a cantilever in the area where we know these neurons are. We can see the neurons - an optical problem which can be resolved - but of course the nanomechanics of the system needs an instrument that makes the measurements for us at that level.
James could actuate the antenna and put a cantilever on the very area where the neurons are. He could have the antenna respond to that activation and see the response profile – which turned out to be a non-linear response profile.
Without going into the details, the idea was to find out: ‘can we see in the mechanical input to the neurons and the neuron output, the same non-linearity that we see in the antenna? Is that non-linearity reflected in the neurons?’ For many reasons we think it is because the neurons are also actuators, both mechanical sensors and mechanical actuators.
And these things still have to be alive for the things to work?
Exactly, the whole joke there is that you can put a laser light in of course but the laser light has a high energy density, as soon as you focus it, even if it’s red light you actually warm up the systems, so that creates problems. But the beauty of the AFM is that this is a very soft, very interference free system, as long as your stiffnesses work out.
This is a classic problem you find in normal bio-AFM, but here it’s also on a system, which is a macro system, which reacts to nanoscale events and which is quite soft. Part of the problem here was to have soft enough cantilevers and to position them properly. The positioning was a huge problem, but also working under liquid as well. It took a while and much effort, but it worked.
Now, one critique of that is, how sure are we that we are on that particular neuron? That’s where we need to be careful in our interpretation of the AFM based mechanical data; we don’t, in fact, know exactly where the cantilever measures vibrations because the cellular tissue is complicated. But by slightly moving and probing around, one can “feel” what part of the tissue is moving according to the mechanical input to the system. So, to explore the nanoscale expression of vibrations in the complex cellular matrix of an auditory system, the AFM approach was invaluable indeed.
This is a very general question that you have in probe microscopy of course, where does your interaction take place in reality? And if you talk about a system that generates information somewhere on an xy plane for instance, are you measuring where it’s generated or are you measuring where it has been propagated? This is a big opening in the analysis of the nanomechanics of biological systems; it’s not only for mosquitos, it can apply to cells, clusters of cells, organs, etc.
How does AFM further help your understanding?
I think the key message here is that one can gain understanding at the multiscale. It is about comparing AFM used perhaps in more classical biophysical assays on single molecules or maybe the active myosin systems, or if you can actually have mechanically active mechanical motors at the molecular level and read out their effects on large functional biological structures, like cells and entire organs. For us the exciting thing here is that we can apply AFM to macroscopic systems that also have a nanoscale behaviour. Approaching the analysis of biological tissue as active matter is a very interesting and insightful prospect.
How has your use of AFM changed and developed over time?
It has changed, almost by definition, because the AFM isn’t a core technique for us; you know we don’t study AFM and its applications in biology. We have questions in biology and AFM is one of the tools we can use to answer particular questions that releases us from some of the constraints of other techniques. There’s the whole trick about contact vs non-contact measurement of nanoscale events and these you can explore using AFM. Alternative techniques such as laser Doppler allow remarkable, sub nanometer sensitivity, but comes with heat transfer problems on small biological structures. I am sure key developments are around the corner, whereby perhaps less powerful lasers could be used, and more manoeuvrable AFMs could be designed. Our use will the further evolve, in turn generating opportunities for entirely novel questions.
Now, we have other questions on electrostatics in its role in the sensory systems in insects in particular. Here with the AFM we are starting to redevelop a new application for that, so in that sense, yes our application has changed. We jumped from maybe the simple contact AFM measure frequency of response of a system or an oscillation, the shape of an oscillation, it’s linearity or non-linearity to characterising whether that system carries an electrostatic surface charge. We were able to identify whether the system acquires or dissipates charge, whether it is polarisable or not. There is evidence that some biological sense organs carry electrostatic charge, some picoCoulombs worth of positive charge to be a little more quantitative. The problem here is to quantify charge distribution and dissipation on non-flat surfaces spanning tens of micrometers. Working with physicists Rob Harniman and Alex Hughes-Games in Bristol, we could image such charge distribution. This made visible a physical phenomenon, but also prompted new possibilities as to how biological sense organs could benefit from triboelectric charging. Here, the use of Kelvin Force Probe microscopy is not what comes first to the mind of a biologist. But through the looking glass of AFM, the eye of the biologist then sees new horizons to explore, trying to better understand biological organisation. This is why physics meeting biology make our research so exciting.
Is there anything actually outside of AFM that could do similar stuff for you? Or is AFM the only technique that we have to do this with?
It’s not the only technique. It depends on the resolution and length scale required, but these sensors are pretty small. I guess in principle, one could use a simple electrometer equipped with a microprobe, and control its position... but again AFM offers remarkable spatial control, so why not take advantage of this There’s nothing that prevents you doing that. The AFM approach is therefore very practical because it offers a very stable system and can be close to the sensor’s surface. You can be knowingly close to your object of study, yet in a non-invasive way.
I guess we tend to be question-driven, and keep realising that understanding better analytical techniques is a part of that very process of driving the questions in the first place. For now, we find ourselves in awe with the Kelvin probe approach and work at perfecting it, but maybe in two year’s time we will require magnetic force microscopy, or something else – who knows?!
But what we like about it is the level of control you have on a small patch of material, and if it’s a biological material for us it is so important to make the measurement very quickly in a non-invasive, non-modifying way.
Now of course the AFM is also good at delivering some information to the system. So if we apply a potential, or drive a current, on the probe we can use AFM as a stimulus delivery system as well. This becomes then an analytical tool as well, and a very precise one at that. There’s a whole experimental approach there that we need to further consider – you can use the AFM literally as a probe for the dynamic characterisation of your system, using an active probe rather than a read out only to interrogate biological triboelectrification.
When did you first start using AFM?
It’s all Merv’s (Professor Mervyn Miles) fault actually! How it started is quite funny because we did work on that mosquito antenna system and Merv came to give a talk in biology. He was talking about AFM cantilevers and how they were, and you know the whole active control, the Q-control[2] breakthrough. We had this paper that came out that showed that mosquitos do Q-control on the mosquito antenna. We did not realise it was Q-control, in our field it is called active auditory mechanics. So, Merv and I got to talk and it became rather quickly very apparent that the mosquito antenna was very similar in its operational logic to the active Q-control that Merv presented, and Infinitesima was developing and applying. It was a key moment, that realisation that actually there was a very serious logic behind it, to enhance detection properties and that’s where the mosquito, in its own way, using its active neurons, was doing. And henceforth that discussion resulted, rather naturally, in the idea that, well, ‘why not use a cantilever to measure the mosquito antenna?’, cantilever against cantilever – it makes sense, so that’s how it started.
To what extent (and why/how) would you encourage biologists to familiarise themselves with AFM for their research?
That’s the classic problem. It doesn’t apply to AFM only, it applies to a lot of physical measurement techniques, to metrology, and the question of what is in fact measurable out there.
Well, there are several ways to go about it but how you encourage the biologist to get into that is very tricky territory because it depends on several things: The willingness of the biologist to go and learn, learn about the physical principles that often are not that complicated.
But if you put a biologist in front of a radio telescope and said ‘What can you use it for?’, it’s gonna be more tricky - unless they are astrobiologists and they say to you, ‘I can measure life somewhere else but here on earth, this isn’t the gear for that’.
But if you ask that to another physicist which is doing say, soft matter spectroscopy, the question becomes much more relevant, and the effort becomes productive. The realisation is that there are lots of processes in biology that rely on physical events. How to understand those in the context of the rather unique, and sometimes “messy” biological processes is where the crux is. Essentially, this is a path of discovery as to what aspects of the biological world can be measured, quantified in all there variability and elusiveness. ‘What is out there in the world of biology that can be measured?’.
So, yes AFM is one more, relatively recent development that deserves more attention from biologists. There is plenty more to discover, and as said before plenty more to think about once one realises the capabilities of the instrument. We are still developing new ways to use the electron microscope, decades after its invention. AFM is young in comparison, and quantitative approaches in biological research can only benefit from it.
It’s classic interdisciplinary approach which is why you’ve got 2 physicists on your paper!
Yes, and I’ve always had physicists and biologists in the lab and it’s sometimes hard to find those that can work in conjunction. We don’t make the biologists physicists and the we don’t make the physicists biologists.
Stick to their strengths and come together?
Yes, and somehow they consistently emerge in yet another area – in Merv’s terms actually, it’s not meeting in the middle between A and B, it’s meeting at C, somewhere else. He always said that. You leave home, don’t expect to be in between the interaction, but you could be somewhere completely different, and much more exciting.
And this is where the question driven research can really work well together with the technique driven research. This is where guys like you can come and open our inquisitive eyes when you say, you know, “we can do that but we can also do all sorts of different things” and that may speak to the biologist equally importantly as to the physicist or the geologist, or any other open-minded scientist.
[1] The resulting work on Frequency doubling by active in vivo motility of mechanosensory neurons in the mosquito ear was published in the Royal Society Open Science journal in January last year.
[2] The two papers that relate to the talk Merv was giving are: Piconewton regime dynamic force microscopy in liquid and High-Q dynamic force microscopy in liquid and its application to living cells.
If you liked this blog post you might also like An interview with Professor Mervyn Miles and The multidisciplinary nature of AFM
Need probes? Click here for our full range!