Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Dear Joost,
If I understand your problem correctly, it seems very much related to the
discussion on FRAP of two-three weeks ago. En empirical FRAP method which is
based on estimating the diffusion coefficient from a t_1/2 value might run
into some difficulties.
It was discussed that, for the t_1/2 method, it is necessary to obtain
exactly the same situation after photobleaching for each sample. This might
be a problem because: 1) the amount of photobleaching can be different
depending on the mobility of the molecules that take part in the
photochemical bleaching reaction, 2) the 3D environment might be different
(e.g. cell vs. solution) and 3) the diffusion during bleaching can be very
different depending on the diffusion speed.
I understand from your question that you have two kind of samples, one with
a thickness of 1 micron and an other one with a thicknes of 6 micron. If you
want to compare t_1/2 values for both samples directly, this might be
problematic because the 3D environment is different. Compared to the thick
sample, the thinnest one might show an artificially decreased diffusion
coefficient because diffusion in the axial direction is much more limited
... unless you are using a fairly low NA objective lens for which both
samples can be considered to be 2-D. Furthermore, as you have indicated
yourself, long bleach times will be very much a problem as well and should
definitely be avoided.
Thanks for mentioning the Braga et al. article, I hadn't read it before. As
far a I understand, their method actually might be a solution to your
problem, although it is important to realize it is a semi-empirical method
as well (albeit a very different one). Because it is a semi-empirical
approach (i.e. based only in part on a physical basis) it might be difficult
to get a grip on what exactly is allowed and when. It seems though that the
authors have done a good job in carefully sorting out the conditions for
which their method is appropriate. So if you study this paper carefully, it
might actually be a viable solution to your problem.
Finally, with regard to our FRAP article, in the derivation we have tried to
adhere as much as possible to the physical basis of the entire FRAP
experiment. However, to arrive at a single final equation, a number of
assumptions had to be made, which we have tried to explain carefully in the
text and experiments. But as long as you adhere to the rules set out by the
mathematical derivation (and which are summarized in the text, don't worry)
you should be fine in any situation.
And as I have said a couple of times before, whatever FRAP method you are
using, be sure to test it thoroughly before applying it to your actual
samples. For example, make a viscosity series of your fluorescent probe(s)
and make sure you can find the Stokes-Einstein relation for a broad range of
diffusion coefficients. If possible, also try to compare your FRAP diffusion
coefficients with values obtained from independent measurements. If all this
is alright, I think then and only then you can use your method with
confidence.
I hope this helps.
Best regards,
Kevin Braeckmans
Lab. General Biochemistry and Physical Pharmacy
Ghent University
Belgium
> -----Oorspronkelijk bericht-----
> Van: Confocal Microscopy List
> [mailto:[log in to unmask]] Namens Willemse, Joost
> Verzonden: donderdag 9 december 2004 16:46
> Aan: [log in to unmask]
> Onderwerp: To kevin braeckmans about FRAP
>
>
> Hello Kevin,
>
> sorry to contact you trough the confocal list like this, but
> i have a question.
>
> I am doing CSLM FRAP studies with a high NA lense. I bleach a
> region of 1 micrometer^2 in my system where my Z-size
> differs. I have both images where the Z-size is about 1
> micrometer and images where the Z-size is about 6 micrometers.
>
> The curve fit formula I(t) = I(0) +
> (I(infinite)-I(0))*(-e^(-kt)) still fits very well but i was
> wondering if the
> D can still be calculated through T1/2 = r^2/4d
>
> I have been reading your article in the biophysical journal
> and the recent articel of Braga et al (Mol Biol. of the Cell,
> Vol 15, 4749-4760, october 2004) but I cannot manage to get a
> hold of what it is that is exactly described there. I
> understand that due to the fact that I am using a CLSM there
> will already be some recovery during the bleaching phase but
> I do not understand how to implement the results shown there
> in my calculations.
>
> My bleach power is 80% of 25 mWatt
> my scanning power is 1-5% of 25mWatt
> Zoom factor is 10 or 20.
> Pixel size is 0.04 to 0.18 micrometers
> pixel time is 6.4 microseconds
> Bleach time is 0.06 s for an experiment of 1.4 seconds
> And for some is have 0.47 bleach time and a tau of 0.27
> seconds these need the corrections during the bleaching phase
> i would say.
>
> I hope you can elaborate a bit on the questions I have.
> If someone else wants to comment feel free. I am open to all
> suggestions.
>
> -----Original Message-----
> From: Confocal Microscopy List on behalf of Kevin Braeckmans
> Sent: Wed 1-12-2004 9:23
> To: [log in to unmask]
> Cc:
> Subject: Re: damage to live-cells in FRAP
>
>
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> Mark,
>
> you are right that the same light dose is needed in
> both cases (stationary and scanning beam) to obtain the same
> amount of photobleaching, assuming simple photobleaching
> kinetics and no diffusion during bleaching. The latter
> condition is intrinsic to most FRAP methods, as has been
> amply discussed over the last week. This also answers the
> second part of your question.
>
> The calculation was from a methodological point of
> view: in a typical spot photobleaching experiment one will
> generally use a higher light dose because the laser beam is
> allowed to stay stationary for quite some time, while in the
> scanning case, because the laser beam is continuously moving,
> one will usually have a lower light dose per unit of area.
>
> But you are right that by no means there would an
> intrinsic problem with the spot method. It is just that one
> often sees that people are somewhat misguided by the desire
> to effectively see a dark photobleached area in their sample
> and consequently are using a far larger light dose than
> actually necessary. If you do things quantitatively, you
> really don't need more bleaching than 20 to 50%. In
> combination with the appropriate fluorophores this really
> doesn't require a high light dose.
>
> That's all.
>
> Best regards,
>
> Kevin
>
>
> -----Oorspronkelijk bericht-----
> Van: Confocal Microscopy List
> [mailto:[log in to unmask]] Namens Mark Cannell
> Verzonden: dinsdag 30 november 2004 23:03
> Aan: [log in to unmask]
> Onderwerp: Re: damage to live-cells in FRAP
>
>
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> Perhaps I missed something but in the scanning case,
> shouldn't the local light dose be about the same since it
> still has to destroy the same molecules in the focal volume?
> Furthermore, if new molecules can diffuse into a selected
> focal volume between scan lines, shouldn't it take _more_
> energy to destroy the molecules in the focal volume?
>
> Cheers
>
> Kevin Braeckmans wrote:
>
>
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> Dear all,
>
> in the discussion on FRAP last week,
> James Pawley rightly expressed his concern, both on and off
> list, about damage to live cells during the bleach phase in
> FRAP experiments. As we don't have any practical experience
> as yet in our lab on FRAP on live cells, is there anyone else
> who can comment on that from experience?
>
>
> JP also asked to give some numbers
> about actual light doses in the bleach phase of a FRAP
> experiment. I'll give it a try:
>
> If P is the power (W) of your laser
> beam at the sample, and r the radius (m) of the diffraction
> limited spot in the focal plane, the local (average)
> irradiance is H = P/(pi*r^2), with units W/m^2 or J/(s*m^2).
> Hence, the amount of energy per unit of area is H*t, where t
> is the time of illumination.
>
> The bleach time in a FRAP experiment is
> usually about 1% of the characteritic lateral diffusion time:
> tau = R^2/(4*D), where R is the radius of the bleached area
> and D the diffusion coefficient.
>
> Now, there are roughly speaking two
> types of bleach methods: 1) with a stationary laser beam
> (spot photobleaching) or 2) with a scanning laser beam on a
> confocal microscope. An example:
>
> 1) spot photobleaching: the radius of
> the spot is approx. 1 micron, so assuming a typical diffusion
> coefficient of 10 mum^2/s, the characteristic recovery time
> is tau = 25 ms and hence the bleach time approx. 0.25 ms. For
> a typical bleach power of 1mW the total energy recieved per
> unit area is therefore 1mW/(pi*1mum^2)*0.25ms = 80 kJ/m^2.
>
> 2) scanning laser beam: To calculate
> the total amount of light received in this case, the
> calculation is a little more complicated because it now
> depends on the scanning speed and the distance between the
> adjacent scanning lines. So in stead of the formula H*t, one
> now needs to use H/(v*dy), where v is the scanning speed of
> the laser beam and dy the distance between the adjacent
> scanning lines (see our article on FRAP for details). v and
> dy depend on the zoom setting of course, but will be
> typically in the order of: v = 0.1 m/s and dy = 0.2 mum.
> Hence, the total amount of energy per unit area is:
> 1mW/(pi*1mum^2)/(0.1m/s*0.2mum) = 6 J/m^2.
>
> As this little calculation shows, it is
> much more likely to inflict damage to your live cell when
> doing spot photobleaching, compared to scanning
> photobleaching. Luckily, hardly anyone is still doing spot
> photobleaching experiments I guess.
>
> As a final note, those numbers don't
> tell anything about the actual damage to a live cell of
> course, they're just quantities of energy. To relate them to
> actual damage to a particular structure in the cell, one
> would have to take the absorption cross section (wavelength
> dependent) into account. Does anyone have any data on that?
>
> Best regards,
>
> Kevin
>
>
> Kevin Braeckmans, Ph.D.
> Lab. General Biochemistry and Physical Pharmacy
> Ghent University
> Belgium
>
>
|
