Reminiscences on the "Classic" 1976 FRAP Article in Biophysical Journal

Main Text

Daniel Axelrod

Scientific progress is often described as a series of breakthroughs driven by individual brilliance. This view may be partly motivated by the prevailing socioeconomic system’s emphasis on the contributions of individual “entrepreneurs,” and it can be accurate in some cases, especially for theoretical work. But more often, progress results from collaborative work based on previous collaborative work coming from different directions and with an admixture of luck (either good or bad). For the development of fluorescence recovery after photobleaching (FRAP) this is especially true. Poo and Cone (1), the real pioneers of photobleaching/recovery, used a photobleaching “pump” coupled with a nonfluorescent absorption “probe” to measure the lateral diffusion of rhodopsin in visual rod outer segments. Peters et al. (2) subsequently studied the motion of fluorescence-labeled erythrocyte membrane proteins after localized photobleaching with a focused image of an aperture illuminated by a mercury arc. Jacobson et al. (3) and Edidin et al. (4) used a similar approach on nucleated cells employing a focused laser beam. Our work provided a mathematical basis for interpreting the data. Numerous studies by us and many others were written on diverse cellular properties and processes, e.g., membrane fluidity, transport across nuclear membranes, transport between cells, etc.

Our work on photobleaching originated from another perspective, that of fluorescence correlation spectroscopy (FCS), which Magde, Elson, and Webb (5, 6, 7) at Cornell had shown could measure chemical reaction rates and diffusion by autocorrelating spontaneous fluorescence “noise” (molecular concentration fluctuations) in systems in equilibrium. (Even FCS, as an optical noise autocorrelation technique, was not the first. It came after “quasi-elastic light scattering,” the autocorrelation of noise in scattered coherent light to measure molecular diffusion rates in solutions.) Then, after successfully employing FCS to measure diffusion rates of fluorophores in solution moving in and out of a small laser focus with nonmicroscope optics, the Cornell group tried to apply FCS to fluorescence-labeled living cells viewed in a microscope, with a focused laser spot providing the localized fluorescence excitation. With Elson and Webb, two of the inventors of FCS, and postdoctoral researcher Koppel, a brilliant and careful scientist who knew everything about statistical noise theory in FCS, the odds were favorable that it would work. All that was required was an illumination bright enough to produce a good chance of capturing a photon from each molecule in each sample time, so the molecule could “report” its presence. But in fact, it did not work very well, because the focused spot kept producing an unwelcomed shallow photobleached “hole” in the fluorophore distribution on the cell surface. Even small sample motions, ubiquitous for living cells, would then produce large fluctuations that obscured the tiny number fluctuations arising from molecular diffusion. When we turned off the laser illumination and headed for lunch, the “hole” would fill in during the lunch hour. It was completely obvious that the rate of that in-filling was determined by the diffusion rate of the fluorophore, exactly what we were trying to measure with FCS. But FCS was (and still is) much more elegant and much more appreciated and admired by us mathematical types, so we kept banging away at it for many more weeks before deciding to refocus on what was essentially FRAP. If we could not beat photobleaching, we would join it. At that stage, Yossi Schlessinger provided the biological expertise needed to apply our physics to cells.

I followed a slightly different approach vector than those above. In my previous graduate school work at Berkeley, I tried for years to measure an optical signal that was submerged in noise. This was never successful and always frustrating. Noise was the enemy. So, when I happened to read about FCS from the Cornell group in 1972, I was thrilled that random noise could be your mathematical friend. That is why I applied to the Cornell group as a postdoctoral researcher. When at Cornell, we eventually switched from FCS to photobleaching recovery (as described above); my mathematical side was initially disappointed by the switch but then cheered to discover that photobleaching recoveries could be described by a nice exotic function (an incomplete γ function) found deep in Abramowitz and Stegun’s (8) massive (and completely application-free) Handbook of Mathematical Functions. Elson’s infinite series solutions of simple functions was probably more practical for a pocket calculator. But nowadays, it does not matter. People tackle such problems (and far more complex ones) by commercial mathematics software in a tiny fraction of the time, starting with just the differential equations as an input. The equations never need to be “solved.”

Elliot L. Elson

Our original intention for FCS was to look at conformational fluctuations, in particular, of DNA, but this turned out to be beyond the capabilities of FCS at this very early stage. There appeared to be an opportunity to apply FCS to diffusion in cells. But that also was too difficult, foiled by cell motions, as Axelrod has described. This led us to our version of fluorescence photobleaching recovery and an unexpected excursion into cell biology that has not yet completely ended, at least for me. Joseph (Yossi) Schlessinger had just joined the laboratory and took the lead in the biology. He prepared for this by taking the Cold Spring Harbor (CSH) (New York, NY) summer course in cell biological experimental methods. Koppel and I took the course the next summer. The basement of Clark Hall at Cornell, inhabited by members of the physics and applied physics departments, was treated to some odd behavior, including a wild chase of an escaped white rat by Yossi and I and a clever method for measuring the laser beam width using a fluorescence-labeled strand of spider web. Luckily, Clark Hall was well supplied with spiders, but not so many white rats, however.

Joseph Schlessinger

I joined the laboratories of Elson and Webb at Cornell University in February 1975 after completing my PhD degree at the Weizmann Institute of Science in Israel. I met Elson a few months earlier in the summer of 1974 during an international conference that took place at a resort in “Ayelet Hashahar,” a kibbutz located in the north of Israel (Fig. 1). Elson presented a very exciting lecture on the subject of FCS (5, 6, 7) at a conference on the subject of proteins and enzymes. In addition to the excitement caused by the excellent presentations from many well-known scientists participating in the conference, there was an additional reason for excitement unusual at most scientific conferences. The conference took place in a scenic resort located next to the Golan Heights ∼15 miles away from the border between Syria and Israel. Since the ceasefire agreement after the Yom Kippur War had not yet been fully implemented, the participants of the international conference watched and heard with amazement as several heavy artillery exchanges occurred between the feuding military forces. As an active participant in the Yom Kippur War with Syria, I also found myself on top of the neighboring hills serving as a tour guide to an international group of scientists, describing the battles that I had participated in the Golan Heights only six months earlier.

Figure 1.

From Yossi Schlessinger: Elson and I have very similar memories from the meeting in Ayelet Hashahar. I have attached a picture I found of the guest house (that does not exist anymore) where the conference took place and where we first met. I recall that Harold Scheraga attended the meeting and that Elkan Blout took color Polaroid pictures, which was very impressive. I knew this area very well because I served there before and after the Six Days War as well as during and after the Yom Kippur War at the Golan Heights as a reserve officer from October 1973 until March 1974; for two months my post was 20–25 miles away from Damascus. It was real tough to go back and do science after the war, and I was very happy to join and do science in your laboratory; it was among the finest chapters in my life. From Eliot Elson: I vividly remember the meeting at Ayelet Hashahar and watching the explosions on the hills not that far away. I also was impressed that every evening a number of young soldiers would arrive at the kibbutz, park their rifles, and eagerly ask what were the interesting talks of that day. Also impressive was the arrival of the President of Israel (Ephraim Katzir, a renowned biophysicist) who listened to the talks just like all the other participants. He arrived in a modest black car, as I recall, accompanied by a couple of tough-looking young men that we speculated were his body guards. At least, they seemed not so interested in the talks.

As soon as I heard Elson’s lecture, I was intrigued by the intellectual elegance of FCS measurements. They offered an opportunity to determine diffusion coefficients and reaction rates of small numbers of fluorescently labeled molecules in solution or even in living cells via quantitative analyses of “noise” measurements. In my conversations with Elson at the meeting and during his subsequent visit at the Weizmann Institute, we agreed that I would join his laboratory as a postdoctoral fellow and take part in applying FCS to determine the lateral diffusion coefficients of fluorescently labeled proteins and lipids on living cultured cells to shed light on dynamic processes that take place in cell membranes.

A few months later, I arrived with my family in Ithaca for postdoctoral training in Elson and Webb’s laboratories and got to experience our first real blizzard shortly after arriving. It was a very exciting time, and I was happy and enthusiastic about embarking on a new project of FCS experiments using living cultured cells at Cornell University. Together with two supremely talented postdoctoral fellows, Koppel and Axelrod, who had already joined the laboratory, we applied FCS for diffusion measurements of fluorescent “lipid-like” molecules embedded in the plasma membrane of living cells as well as analysis of diffusion of cell membrane components labeled with fluorescently tagged concanavalin, a lectin that binds to a broad spectrum of carbohydrate-containing cell surface molecules.

After several months, it became clear that progress in our project was hampered by significant technical issues. An important limitation that Koppel, Axelrod, and I faced employing FCS for diffusion measurements of fluorescently labeled cultured cells was the limited computational power of computers available in the mid-1970s, which were not fast enough to perform the autocorrelation calculations of the diffusion measurements. This was further exacerbated by the instability of noise analysis caused by cell movement and photobleaching of the fluorescent probes used for labeling live cells.

Another more trivial limitation originated from our ignorance on how to grow and maintain cultured cells, which were essential for carrying on FCS and later on FRAP experiments. Ken Jacobson from Roswell Park Memorial used to periodically bring a supply of cultured fibroblasts that were used in our biophysical experiments. A few days after Ken returned to his own laboratory, I asked Eff Racker why the cultured fibroblasts that Ken Jacobson brought were growing more rapidly and why they were so much smaller than the original cells that he brought. Looking through the microscope and laughing, Eff Racker told me that I was growing a yeast cell contamination rather than cultured fibroblasts. Humiliated by my ignorance, I spoke to Elson about an urgent need to have our own cell culture laboratory and subsequently enrolled in a CSH practical course on cell culture. Several weeks later when I returned to Ithaca, we established a tissue culture laboratory in a small space kindly provided to us by George Hess’s laboratory. The CSH course in cell biology served as an important turning point in my scientific education and career, as I met Graham Carpenter from Vanderbilt University and Gordon Sato from the University of California, San Diego at the CSH course who introduced me to epidermal growth factor and the fledgling field of growth factors and their receptors, which became the main topic of my research for the next 40 years.

Axelrod and Elson described in their own reminiscences the reason the Elson/Webb laboratory transitioned from using FCS to using photobleaching experiments that we initially called “FPR” for fluorescence photobleaching recovery. The Elson/Webb laboratories enabled groundbreaking development of FCS and FRAP instrumentation (9), quantitative approach for FRAP experiment analysis (10), and a variety of quantitative analyses of dynamic properties of lipid-like molecules, membrane proteins, and surface receptors occupied by their ligand in living cells (11, 12, 13, 14).

I am indebted to Elson and Webb for providing a fantastic scientific and intellectual environment and fostering a true collaborative spirit among a group of scientists who not only loved science but also had a good time together. In addition to the science carried on in Ithaca, an additional milestone occurred in my family. My younger son Avner, who is now a member of the faculty of Mount Sinai School of Medicine was born at Ithaca Hospital overlapping with the birth of FCS and FRAP as new tools for modern biomedical research.

Dennis E. Koppel

In the mid-1970s, a number of laser fluorescence techniques were being developed to provide the capabilities of characterizing molecular transport over distances down to the order of a wavelength of light. In 1972, the laboratories of Elson and Webb were the first to publish an elegant technique, which used FCS to study thermodynamic fluctuations. In fact, the reason I came to Cornell as a postdoctoral researcher in the first place was to continue working on the FCS technique. At the same time, we (and others in two laboratories) were looking at a simpler method of FRAP. In 1976, we published our first photobleaching article on myoblast cell membranes, and the rest is history!

Editor: Jane Dyson.