From fast photometry to FuraLED in three easy decades

By 1st September 2017Presidents Log


This particular blog came into existence because James our Marketing Director asked me to write something about our FuraLED illuminator, but I realised there was potentially so much to say on the subject that I couldn’t just do that on its own.  In practice the only way I could write the sort of propaganda that he had in mind, was to do that in partnership with this much larger backstory, involving Nobel prizes, the Eurovision song contest, and also my professional survival of an early encounter with the now sadly late Roger Tsien.

The common link here is that although I’m not sure quite where my fascination with the Eurovision Song Contest came from, I do know the origins of my fascination with Nobel Prizes.  The personal link between those two events is that while I have harboured no particular aspirations in either direction, I am in the possibly unusual situation of having known people who have gone on to win both, although to be fair they were different people.  And although Roger’s prize was ostensibly for his work on fluorescent proteins, it must also surely have constituted some acknowledgement of his development of chemical fluorescent indicators such as fura2 for calcium, the exploitation of which has in large part contributed to the establishment and growth of Cairn.

My Nobel Prize fascination came about because of some truly awful physical chemistry lectures to which I was subjected as a Cambridge undergraduate.  The lecturer managed to make a difficult subject even less comprehensible than it should have been, so the attention of the audience did tend to wander somewhat.  We therefore tended to pay more than usual attention to the less technical aspects of his lectures, from which we noticed that he frequently referred to leading scientists in the field in ways indicating that he knew them personally.  This was particularly the case when he told us they had won “Nobel Prizes”, whatever they were.  After a while we suspected that the real issue was that he hadn’t got one, which somehow made this poor man’s lectures even more boring.  This culminated in him mentioning in passing that he had had dinner with one Nobel laureate (Debye of Debye-Hueckel  theory) “just before he died, actually”, whereupon the entire audience, suspecting a causal relation, burst out laughing.  All very unfair, but if his lectures had been better I would perhaps have survived my first Roger Tsien encounter with a bigger safety margin than was in fact the case.

For that story, I need to leap forward a decade or so, specifically to the late 1970s.  At that time I had a postdoctoral position at Boston University School of Medicine, where as described elsewhere I was having a satisfyingly successful time measuring calcium concentration changes in molluscan neurones using the absorbance indicator arsenazo III.  This involved use of a spinning filter wheel to rapidly change the illumination wavelength, so that the (very small) absorbance changes could be measured differentially between wavelengths with greater resolution.  Fura2 didn’t exist yet, but this type of system was going to be perfect for the differential fluorescence excitation of that indicator, and which turned out to be what Cairn was going to be doing some years later.

But at the time of the story, I was still in Boston and had briefly returned to give a few talks, one of which was at the Department of Physiology at Cambridge, where Roger had I think recently completed his Ph.D. By that time I’d got a nice physiological calcium story to tell, but I’d also discovered something rather strange about the indicator, and which I felt it was definitely better not to talk about. For those of you who know the works of the spy thriller writer John Le Carre (whose latest novel has just been published), he had recently written the wonderful “Tinker Tailor Soldier Spy” novel, where British spymaster George Smiley had sent his agent Jim Prideaux on an undercover mission to Czechoslovakia. Prideaux had ostensibly been sent for other reasons, but his real mission was to attempt to uncover the identity of a “mole”, or double agent, deep within the British Secret Service. If captured, which he was, he should attempt to keep that secret at all possible costs, but this is where the Russian spymaster Karla started his interrogation.

And so it was with Roger Tsien. I thought my own secret would be safe, but he interrupted me in mid flight to ask “Have you looked at the stoichiometry of the interaction of the dye with calcium?”, so all I could do was to make an immediate confession. We already knew that the absorbance change was pretty linear with calcium, but the dye measures calcium by binding to it, so the actual calcium changes could have been reduced by its buffering effect. I controlled for that by changing the dye concentration for a known calcium concentration change, but where I discovered an apparently opposite effect. Doubling the dye concentration should have doubled the absorbance change if there were no significant buffering, but in fact it went up by a factor of four! In spite of that poor lecturer’s best efforts, I had still picked up enough physical chemistry to realise that, unlikely as it sounded, one calcium ion was binding not one but two dye molecules at once. And in retrospect I could see how that was going to be physically possible, but I would never have predicted it.

Not so Roger! He had indeed predicted it, so he smiled as he quoted a German article in (very appropriately) a Czech journal, stating that a similar dye had exactly this stoichiometry for calcium, so we were both very happy about that! But the near miss for me was that I had done those controls only a few weeks before, so I could so easily have been wrongfooted. I hadn’t realised how close his own interests were to this area, but it was only going to be a matter of time before the world found out.

Which finally brings us to Fura2! It’s now over 30 years since the paper by Grynkiewicz, Poenie & Tsien, describing fura2 and other calcium indicators was published, and apart from the later development of some lower-affinity analogues, it still pretty much represents the state of the art. In fact, Cairn was established in that very same year (1985), and indeed the Plymouth courses in which we have been so pleased to play a part had started just the year before, so it’s good that we’re all still going strong. In fact, Cairn’s first customer for a fura2 system doesn’t seem to be doing too badly either, so what did we actually sell him?

The absorbance system I’d put together in Boston used a small lightweight wheel that could be spun at several hundred revolutions per second(!) by compressed air, but for Fura2 we didn’t need to go quite so fast, and use of compressed air isn’t always so convenient, so we switched to a belt driven electric drive for that first Fura2 system. The development of that and of our subsequent filterwheels has been described elsewhere, so here we want to talk a bit more about the electronics side of things.

Although the electronics was also rather different from before, especially as those early systems used photomultipliers rather than photodiodes, we did continue to divide up the electronics into a series of plugin modules. That was very handy as it allowed us to introduce further modules later on, rather than having to design the entire system at once. The idea behind these early systems is that you could take the analogue outputs from these modules, and digitise them in an existing data acquisition system, allowing a fluorescence capability to be added to an existing electrophysiological system, for example. This turned out to be economical for the customer, but also lucrative for us! Where for absorbance, sensitive differential measurement tends to be the order of the day, for calcium indicators like Fura2, it’s the ratio between different wavelengths that tends to be important. We soon introduced a “ratio module” to do this electronically, followed by various further modules including a fullblown computer interface and acquisition system of our own over the next few years.

At the time of that first system, I still had a fulltime job at those old corporate research labs just outside Sittingbourne, and I was still there when we secured a much larger

Optoscan Monochromator

Optoscan Monochromator

order. This marked a real “baptism by fire” for Cairn, but which we fortunately survived! The order was from Tullio Pozzan in Padova. He had known me from my Boston work, and had also collaborated with Roger Tsien in Cambridge. He was now looking to buy a Fura2 system, and this had also come to the attention of another company with interests in this area, although they didn’t have a suitable system to sell. That company, which has long since ceased to exist as such following the various mergers and acquisitions that tend to be so prevalent in our business area, had offered to help us develop our first system into a full commercial product, but they almost immediately became involved in one of those takeovers, leaving us to do everything ourselves.

In retrospect that was no bad thing, but at the time I was left feeling just a little stretched, although somehow I managed to put a pretty decent system together. Even so, I felt the chances of actually getting it to work in Tullio’s lab when I delivered it were not good, but I failed to take into account just how incompetent the rest of the world can be too. On my arrival, Tullio was frustrated that another system he had bought from the company that had promised to help us still wasn’t working properly after three months, so I realised that we might actually be no worse than them, and hence could probably get away with it if more needed to be done to our system. In fact, we turned out to be better, as to my (carefully concealed of course) surprise, ours did work straight off. In spite of the perceived improbability of this event, I had done some “back of envelope” calculations on the journey over, which showed that if I could somehow get out of the situation alive, Cairn was not only going to be a viable business, but also a self-financing one, which we have remained to this day. No venture capital for us, thank you!

So, over the following years we sold a lot of filterwheel systems, mostly for Fura2, steadily developing and improving our designs as we went along, until in the late 1990s we saw the market requirement for a fast monochromator that could also be used for exciting this indicator. This led to our development of our Optoscan monochromator, which is still on sale today and enjoying a new lease of life now that TILL (in consequence of their being a victim of corporate acquisitions) have discontinued their design.

However, the success of the Optoscan more or less killed the sales of our filterwheels, which were designed for fast continuous spinning to change the excitation wavelengths, and used a relatively small filter size (12.5mm) in order to achieve this. At this point we embarked on the design of wheels that could take the larger 25mm diameter filters for camera imaging applications, culminating (after an embarrassingly long time) in the Optospin, but that isn’t part of the Fura2 story so we won’t go into it further here.


FuraLED with infinity cube

Cairn Fura2LED Poster

Cairn Fura2LED Poster

More recently, a 340nm LED has become available, and in comparative measurements with the Optoscan in the 2016 autumn Plymouth course, we found that even though this wavelength isn’t as bright as its longer-wavelength cousins, it was still outshining the Optoscan by an order of magnitude or so. In anticipation of this, and in order to keep the cost of a dual-LED system down, we introduced our FuraLED design, in which the two LEDs are mounted in a common housing as you can see. However, for reasons which aren’t entirely clear, but basically because of pressure of other work, we made a lousy job of promoting the product, which gave a certain other company the impression that they were the first to offer this type of device. All very annoying, but entirely our fault! Hopefully we are putting this right now.

The LED illumination market is very competitive generally, and it has never been our intention to supply a mass market. Instead, we have concentrated on the specific features of very fast (submicrosecond) switching, and the use of optical feedback for absolute stability. These two features go very well together, because the optical efficiency of an LED tends to be somewhat temperature-dependent. What matters here is the temperature of the LED chip itself, not that of its housing, and if you are switching an LED on and off (which you need to do for Fura2 excitation as you switch between the two wavelengths!), then the temperature of the LED chip will be cycling up and down, even though the temperature of the housing will remain effectively constant. The timescale for this (as judged by the effect on the LED’s efficiency) is on the order of a few milliseconds, whereas optical feedback can be applied – in our hands at least – on a microsecond timescale, which is sufficient to kill this problem stone dead.

However, there are still traps for the unwary! With at least some LEDs the spectral output may also change with temperature, and the issue here is that in practice an LED needs to be used with some sort of cleanup filter to remove wavelengths outside the desired range. It’s therefore important that the optical feedback is applied downstream of the cleanup filter, and we’re pleased to say that we spotted this potential issue in advance, rather than finding out the hard way.

So, especially with these embellishments, LEDs are now indeed sufficiently bright and stable to be used for Fura2 excitation, so we’re pleased to add the FuraLED to our product range. And it’s intriguing that Fura2 itself has stood the test of the last three decades so well! The only really useful developments to it have been lower-affinity variants, such as Furaptra, which can monitor transient calcium changes in the micromolar range, but these are just minor tweaks to the basic chemical structure. More recently, an intriguing analogue named Fura8 has become available, in which the excitation wavelength range has been redshifted by 20-30 nanometers, potentially removing the need for UV-transmitting optics. We are currently comparing it with Fura2 at the Cell Physiology workshop at Plymouth, and our results so far do suggest that it may well prove to be a viable alternative, so we’ll be keeping a close eye on this analogue!

Now, what OTHER products are we inadvertently being too quiet about?????