Gustavo,

The amplitude of each species in the FCS curve is influenced by both their fractions in the equilibrium and by the amount of dye-labeled enzymes per polysaccharide.

Can you run the same experiment on the same setup but at single molecule conditions (that is, freely diffusing single molecule)?

If you can, that could solve the problem. You can perform the measurement and then perform a burst search. Each burst of photons you will identify will have a duration (from first to last photon) and a size (total amount of photons), as well as a measure of the instantaneous photon rate (the amount of consecutive photons in your sliding window for the burst search, divided by the duration of this number of consecutive photons). The burst duration will be a function of the diffusion time through the effective excitation volume. The duration of bursts should be distributed exponentially for each species. In your case, following what is clear from the FCS curve, you should have two well-defined species: brief bursts and very long ones. Therefore, it'll be easy to distinguish the two.

However, your question, namely, the number of dyes per each polysaccharide is related to the molecular brightness of your species. The molecular brightness of your species can be approximated:

1. Either by the ratio of the burst size and burst duration

2. or by the burst maximal photon rate

By calibrating your calculation to the quantities you get for the species with brief bursts, you should, in principle, be able to quantify the amount of dyes (enzymes) per polysaccharide. That is, of course, assuming, the fluorescnece from each dye-labeled enzymes is independent from others and they can simply be summed up.

Dear Eitan,

Thanks for your answer.

"The amplitude of each species in the FCS curve is influenced by both their fractions in the equilibrium and by the amount of dye-labeled enzymes per polysaccharide."

Then my reasoning is incorrect? FCS will provide the correct proportion of enzymes in the bound state, regardless if more than one is associated to the polysaccharide in average? If this is the case, then the problem is partially solved.

"Can you run the same experiment on the same setup but at single molecule conditions (that is, freely diffusing single molecule)?"

If I understood you, it's already done. You can see it in the figure above. The pink line is the FCS curve of the free enzyme.

About the burst search: is it an actual data acquisition or is it just an analysis of the photon counts over time? In the former case, the problem is that I did these measurements more than 2 years ago and I'm far away from the place where I did them. So, unfortunately, I cannot do any experiments right now. In the latter case, is there any program that assists in this process, or is it based merely on visual inspection of the photon counts over time? I'm currently using QuickFit.

Best regards,

Gustavo

Gustavo, I guess you did not get me correctly.

You performed the FCS experiments with labeled enzymes at a concentration that yields a noisy fluorescence read that allows for calculation of the autocorrelation function (probably a few nM). However at even a lower concentration (what I called single molecule concentration) what you get is mostly background, and once in a while a burst of photons (< 100 pM). This is because at even a lower concentration of dye-labeled enzymes than your working concentration molecules most of the time do not cross the confocal illumination. They do only once in a while which yields a burst of photons.

I did not suggest only a different analysis of your existing data but rather a different measurement where you record fluorescence with dye-labeled enzymes at

Gustavo:

I would suggest using the Photon Counting Histogram (PCH) ; see the seminal papers below.

PCH and FCS are determined on the same time series of events. To extract the distribution of the duration of the fluctuations we use a math based on calculation of the correlation function, or FCS. To extract the distribution of the amplitude of the fluctuations, we use a math based on the PCH distribution.

FCS provides the diffusion coefficient D and the number of particles in the observation volume; PCH provides the brightness ε ( the number of photons emitted in the unit time per molecule) and the number of particles in the observation volume. The brightness may change in your system depending upon the number of fluorophores you have.

1. The Photon Counting Histogram in Fluorescence Fluctuations Spectroscopy"; Chen Y1, Müller JD, So PT, Gratton E. ; Biophys J. 1999 Jul;77(1):553-67.

2. Characterization of Brightness and Stoichiometry of Bright Particles by Flow-fluorescence Fluctuation Spectroscopy. Johnson, J.1., Chen, Y., Mueller, J.D. ; Biophys J., 2010, 99(9), 3084-92.