Multicolour TIRF

[Steffen Steinert]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hello all,

I´m about to add TIRF to my home-built scope but due to lack of experience
with TIRF I´d rather ask a few questions in advance.
The goal is to do multi-colour TIRF with 488nm and 650nm through a 100x1.65
objective.

Due to the high difference in RI, i think the critical angle is not the
issue here. But I´m wondering whether the differences in penetration depths
between the two lines is critical when doing 'ordinary' live imaging of
yeasts, hela cells etc. Assuming both lines have an angle of ~65°, the
resulting penetration difference is ~25nm. From my perspective, 25nm
doesn´t sound much when normally dealing with diffraction limited
distances. On the other hand the nominal penetration of the 488nm would be
around 75nm and then 25nm difference is quite a lot. The question I like to
adress: Do you generally correct the difference in penetration depths in
multi-colour TIRF? If so, do you alter the inclination of the laser in
relation to the ojective or do you rather enter the objective in a parallel
manner but at distinct lateral distances from the objective´s center?

Second question aside. Does someone happen to know about a method how to
clean and re-use the extremely expensive and fragile Sapphire coverslips?
Is there maybe another, cheaper source than Olympus?

Many thanks in advance for any hints and best regards,

Steffen Steinert, Dipl.-Ing.

--------------------------------------------------
Universität Stuttgart
3. Physikalisches Institut
Pfaffenwaldring 57
70550 Stuttgart

Tel.: [phone]

http://www.pi3.uni-stuttgart.de/en/
--------------------------------------------------

[John Oreopoulos]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hi Steffen, 

No one has answered your post, so I'll give it a try. Generally speaking, unless you're trying to do quantitative TIRF microscopy where you measure changes in vertical distance based on the intensity of an object in your TIRF images, you probably don't have to worry about the difference in the penetration depths for the two wavelengths. Both should give a much cleaner image compared to regular epifluoresence as long as you meet the TIR condition with both laser lines. If you are trying to do quantitative TIRF imaging, then you should be concerned about this difference, but realize there are probably other factors they may cause the penetration depth to be different from the calculated values, namely scattering in the objective, polarization, local differences in fluorophore concentration, filter quality, non-uniform illumination,  etc. I'm sure there are other factors as well - see Jim Pawley's "39 Steps" which equally apply to TIRF microscopy and not just confocal. You'll notice if you scan the literature that there are not many papers out that address this particular problem (at least, none come to mind for me), probably because of the difficulties associated with quantitative measurements of the evanescent fluorescent intensity that I just mentioned. There are creative ways to measure the penetration depth, however, and I mentioned those in another post a while back:

http://listserv.acsu.buffalo.edu/cgi-bin/wa?A2=ind0702&L=confocal&T=0&P=19195

Using Axelrod's method described in the above link, I think you could "tune" your laser alignment for both wavelengths so that their penetration depths were nearly equal; just be sure you have sufficient opto-mechanical stages on your microscope bench to allow for these kinds of adjustments (I'm assuming of course that you're building from scratch and not using a commercial TIRF attachment - it's much easier to do with a free-spaced laser and some mirrors on translating/tilting stages). Both methods you mentioned should work (changing the inclination or the lateral distance of the beams in the back focal plane of the objective). I'm not sure how the commercial multi-color TIRF systems do this (if at all), and maybe someone out there from one of the companies can comment on that. 

As for cleaning the special sapphire coverslips, I don't know of a good method other than using hard solvents (in a sonicator perhaps) and doing some plasma cleaning afterwards. One of the things I never liked about the 1.65 NA objective was the fact that you have to re-use these expensive coverslips and you have to work with that really dangerous oil. Try to use a 1.45 NA objective if it's available to you. There's less space to align the beams in the periphery of the objective, but you still have access to angles between ~60 and 64 degrees, and this might be enough to "tune" the penetration depths for your application.

John Oreopoulos, BSc,

PhD Candidate

University of Toronto

Institute For Biomaterials and Biomedical Engineering

Centre For Studies in Molecular Imaging

Tel: W:416-946-5022

On 22-Nov-07, at 10:16 AM, Steffen Steinert wrote:

Search the CONFOCAL archive at

http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hello all,

I´m about to add TIRF to my home-built scope but due to lack of experience with TIRF I´d rather ask a few questions in advance.

The goal is to do multi-colour TIRF with 488nm and 650nm through a 100x1.65 objective.

Due to the high difference in RI, i think the critical angle is not the issue here. But I´m wondering whether the differences in penetration depths between the two lines is critical when doing 'ordinary' live imaging of yeasts, hela cells etc. Assuming both lines have an angle of ~65°, the resulting penetration difference is ~25nm. From my perspective, 25nm doesn´t sound much when normally dealing with diffraction limited distances. On the other hand the nominal penetration of the 488nm would be around 75nm and then 25nm difference is quite a lot. The question I like to adress: Do you generally correct the difference in penetration depths in multi-colour TIRF? If so, do you alter the inclination of the laser in relation to the ojective or do you rather enter the objective in a parallel manner but at distinct lateral distances from the objective´s center?

Second question aside. Does someone happen to know about a method how to clean and re-use the extremely expensive and fragile Sapphire coverslips? Is there maybe another, cheaper source than Olympus?

Many thanks in advance for any hints and best regards,

Steffen Steinert, Dipl.-Ing.

--------------------------------------------------

Universität Stuttgart

3. Physikalisches Institut

Pfaffenwaldring 57

70550 Stuttgart

Tel.: [phone]

http://www.pi3.uni-stuttgart.de/en/

--------------------------------------------------

[Steffen Steinert]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Dear John and others,

Thanks for the input!
For those interested in the cleaning issue of sapphire coverslips. I got two simple protocolls off-list:
a) overnight washing with 50% nitric acid and extensive rinsing (thanks to D.Toomre)
b) wash them with ethanol and Sparkle (thanks to J.Borejdo)
I´ll probably go for method b) with subsequent plasma cleaning.

As to the different penetration depths, I´m going to correct the discrepancy in penetration depths, although the difference may not be so crucial. Not having the intension of doing surreptitious advertising, but actual commercial TIRF setups seem to favour the compensation as well. The guys from the company that starts with "L" and ends with "eica" apparently introduced their new TIRF MC which has automated scanner optics to place the laser at the corresponding wavelength dependent BFP-position. Not having checked the other companies but they´ll probably offer similar features, I guess.
The reason I asked about the method how to adjust the correct angle was because I think that changing the lateral distance from the objective´s center with a parallel beam to the optical axis is the better way compared to changing the inclination. The difference may not be too significant (and that may be the reason why it doesn´t matter), but altering the convergence of the incoming beam changes the focus and thus results in a different TIR fluorescence area, doesn´t it?

Anyway, thanks for the pointers and best regards,

Steffen

At 21:40 23.11.2007, John Oreopoulos wrote:

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal


Hi Steffen,

No one has answered your post, so I'll give it a try. Generally speaking, unless you're trying to do quantitative TIRF microscopy where you measure changes in vertical distance based on the intensity of an object in your TIRF images, you probably don't have to worry about the difference in the penetration depths for the two wavelengths. Both should give a much cleaner image compared to regular epifluoresence as long as you meet the TIR condition with both laser lines. If you are trying to do quantitative TIRF imaging, then you should be concerned about this difference, but realize there are probably other factors they may cause the penetration depth to be different from the calculated values, namely scattering in the objective, polarization, local differences in fluorophore concentration, filter quality, non-uniform illumination,  etc. I'm sure there are other factors as well - see Jim Pawley's "39 Steps" which equally apply to TIRF microscopy and not just confocal. You'll notice if you scan the literature that there are not many papers out that address this particular problem (at least, none come to mind for me), probably because of the difficulties associated with quantitative measurements of the evanescent fluorescent intensity that I just mentioned. There are creative ways to measure the penetration depth, however, and I mentioned those in another post a while back:

http://listserv.acsu.buffalo.edu/cgi-bin/wa?A2=ind0702&L=confocal&T=0&P=19195

Using Axelrod's method described in the above link, I think you could "tune" your laser alignment for both wavelengths so that their penetration depths were nearly equal; just be sure you have sufficient opto-mechanical stages on your microscope bench to allow for these kinds of adjustments (I'm assuming of course that you're building from scratch and not using a commercial TIRF attachment - it's much easier to do with a free-spaced laser and some mirrors on translating/tilting stages). Both methods you mentioned should work (changing the inclination or the lateral distance of the beams in the back focal plane of the objective). I'm not sure how the commercial multi-color TIRF systems do this (if at all), and maybe someone out there from one of the companies can comment on that.

As for cleaning the special sapphire coverslips, I don't know of a good method other than using hard solvents (in a sonicator perhaps) and doing some plasma cleaning afterwards. One of the things I never liked about the 1.65 NA objective was the fact that you have to re-use these expensive coverslips and you have to work with that really dangerous oil. Try to use a 1.45 NA objective if it's available to you. There's less space to align the beams in the periphery of the objective, but you still have access to angles between ~60 and 64 degrees, and this might be enough to "tune" the penetration depths for your application.


John Oreopoulos, BSc,
PhD Candidate
University of Toronto
Institute For Biomaterials and Biomedical Engineering
Centre For Studies in Molecular Imaging

Tel: W:416-946-5022


On 22-Nov-07, at 10:16 AM, Steffen Steinert wrote:

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hello all,

I´m about to add TIRF to my home-built scope but due to lack of experience with TIRF I´d rather ask a few questions in advance.
The goal is to do multi-colour TIRF with 488nm and 650nm through a 100x1.65 objective.

Due to the high difference in RI, i think the critical angle is not the issue here. But I´m wondering whether the differences in penetration depths between the two lines is critical when doing 'ordinary' live imaging of yeasts, hela cells etc. Assuming both lines have an angle of ~65°, the resulting penetration difference is ~25nm. From my perspective, 25nm doesn´t sound much when normally dealing with diffraction limited distances. On the other hand the nominal penetration of the 488nm would be around 75nm and then 25nm difference is quite a lot. The question I like to adress: Do you generally correct the difference in penetration depths in multi-colour TIRF? If so, do you alter the inclination of the laser in relation to the ojective or do you rather enter the objective in a parallel manner but at distinct lateral distances from the objective´s center?

Second question aside. Does someone happen to know about a method how to clean and re-use the extremely expensive and fragile Sapphire coverslips? Is there maybe another, cheaper source than Olympus?

Many thanks in advance for any hints and best regards,

Steffen Steinert, Dipl.-Ing.

--------------------------------------------------
Universität Stuttgart
3. Physikalisches Institut
Pfaffenwaldring 57
70550 Stuttgart

Tel.: [phone]

http://www.pi3.uni-stuttgart.de/en/
--------------------------------------------------

Steffen Steinert, Dipl.-Ing.

--------------------------------------------------

Universität Stuttgart

3. Physikalisches Institut

Pfaffenwaldring 57

70550 Stuttgart

Tel.: [phone]

http://www.pi3.uni-stuttgart.de/en/

--------------------------------------------------