Hi everyone,
I was wondering if anyone of you has looked into the background generated by autofluorescence of standard (TIRF) objectives.
I have set up a TIRF microscope for single molecule imaging with laser lines 488, 561 and 638nm. While the imaging of single emitters excited with the 561 laser (e.g. Cy3B, ATTO565) and 638nm (ATTO655) is easily possible with good signal-to-background, I have troubles getting similar quality signals from ATTO488 or Alexa488, even though the dye should perform even better on paper.
It also appears that the signal level of the 488 dye is not the problem, but the increased background is. It even spills over onto the 650-750nm emission channel.
Filters, Dichros and Immersion oil could already be ruled out, so I looked into autofluorescence of the objective and found a short paragraph in Brunstein et al:
*Furthermore, upon 488-nm excitation with low μW power and EMCCD detection, all three tested objectives displayed measurable yellow-green autofluorescence (Fig. S2). This instrument fluorescence passes undetected as a background offset when imaging the sample-plane. BFP imaging with a Bertrand lens allowed us to identify the origin of this fluorescence inside the objective.*
Besides choosing narrower emission filter bandpasses or using different modes of illumination, how are people tackling this? Please also let me know if you have no idea to solve it, but experienced the same issues 
Best,
Christian
1 Like
Could you clarify what objective(s) have you tried, and what you consider to be good signal-to-background?
In my experience Nikon’s CFI Plan Apochromat Lambda 60X Oil (1.4NA) and Olympus’s 100x 1.4NA oil objective have reasonable performance at this wavelength. These objectives can be used for TIRF, though they will require more precise alignment.
It is also my experience that the 488 dyes are not the best in practice. They photo-bleach quickly and are not as bright as Cy3 or Cy5 wavelength dyes. In addition the auto-fluorescence background from cells / tissue is much higher at this wavelength. They would not be my first choice for single molecule imaging, but I believe they have been used for 3 color single molecule FRET experiments. If you can find those papers they might have some suggestions.
1 Like
This paper uses Alexa 488 for single molecule FRET and includes some example time traces. Unfortunately I can’t find anything about what objective they used, sigh.
Chung et al, Biophysical Journal, 2010.
Thanks for your answer. I am working with a Nikon Apo TIRF 60x NA 1.49 oil immersion objective.
Signal-to-background for 561 and 638 excitation range is about 2.5-3, whereas it is about 1.6 in the 488 excitation channel, for supported lipid bilayers with the respective fluorescently labeled lipids using otherwise similar conditions.
I had a look at the FRET paper, they use a confocal setup. I could imagine that the autofluorescence would be less of a concern without the totally reflected excitation beam inducing autofluorescence a second time.
I have noticed autofluorescence and in my case oil was culprit. It was old/contaminated. You may want to check oil or try new oil. It is unlikely that TIRF lens is contributing noticeable autofluorescence. Easy to check - just put a coverslip without any sample (or dye) and you will see similar level of autofluorescence.
1 Like