Re: Basic live cell imaging question...

Posted by Andreas Bruckbauer on
URL: http://confocal-microscopy-list.275.s1.nabble.com/Basic-live-cell-imaging-question-tp5687698p5710782.html

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi John,

I recently tested the Olympus 100x NA 1.65 objective which uses special coverslips and oil with refractive index 1.78 against our 100x NA 1.45 lens. For single molecule TIRF directly at the coverslip i get 2-3 times higher signal with the 1.65 NA lens. This was in good agreement to a calculation based on classical electrodynamics which gave 2.9x more signal in the case of the 1.65 lens. However when i calculted the case for focussing one micron into the cell, the 1.65 lens had a slight disadvantage, more reflection due to the higher difference of refractive indices. This calculation does not account for PSFs or aberations it just deals with plane waves and dipol excitation/emission.  However it shows quite clearly that It is really the coupling of the near-field of the oscillating dipol to the propagating waves in the lens which gives rise to increased signal in the high NA case, this does not work if the molecule is more than 1/2 wavelength away from the interface glass/cell.

best wishes

Andreas

 

 


 

 

-----Original Message-----
From: John Oreopoulos <[hidden email]>
To: [hidden email]
Sent: Fri, 5 Nov 2010 18:31
Subject: Re: Basic live cell imaging question...


*****

To join, leave or search the confocal microscopy listserv, go to:

http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy

*****



Hi Jim,



I'm wondering if you can clear up a few things related to this topic. You said

in your last email that the higher NA rays in a water-glass situation do not get

into the objective, even in the case of TIRF. There is a very nice paper by Jorg

Enderlein's group that uses a very simple technique to measure the actual NA of

an objective which involves viewing the distribution of light rays in the back

focal plane of an objective with the use of a Bertrand lens:



http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-13-23-9409



And another similar paper by Mattheyses and Axelrod:



http://spiedl.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JBOPFO000010000005054007000001&idtype=cvips&gifs=yes&ref=no



Both of these papers clearly show that a large percentage of the light actually

does propagate into the objective lens beyond the critical angle. However, you

did also allude to "thin aqueous biological" in vitro specimens, the ones that

are typical of single-molecule experiments where the material under

investigation has been separated from the cell, and both of these papers made

their measurements under similar conditions, so maybe this is only a special

case.



My question is, does this situation only occur for fluorescent objects that are

very near the glass-water interface? That is to say, if I were to view a

fluorescent cellular sample the same way (with a Bertrand lens, widefield

illumination, focused slightly into the cell), would I see that most of the

light propagates into the oil immersion objective BELOW the critical angle (and

hence the effective lower NA)?



As for a "quantum mechanical explanation" of why the other situation works, it

seems that Hellen and Axelrod came up with an explanation some time ago that

does not involve quantum mechanics at all, and it can be explained using

classical electrodynamics... but I've had trouble following this paper because

it gets rather complicated with the math:



Hellen, E.H. and D. Axelrod, Fluorescence emission at dielectric and metal-film

interfaces. Journal of the Optical Society of America B-Optical Physics, 1987.

4(3): p. 337-350.



Axelrod, D., Emission of fluorescence at an interface, in Methods in cell

biology, T. Langsing and Y. Wang, Editors. 1989, Academic Press: San Diego. p.

399-416.



There is a very simple diagram in the second book chapter that basically shows

very oblique light rays emitted by a fluorescent molecule near an interface (the

near-field) can turn into propagating evanescent waves that get coupled back

into the higher index medium. This website talks a bit more about this (see

Figure 6):



http://micro.magnet.fsu.edu/primer/techniques/fluorescence/tirf/tirfintro.html



John Oreopoulos







On 2010-11-05, at 11:33 AM, James Pawley wrote:



> *****

> To join, leave or search the confocal microscopy listserv, go to:

> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy

> *****

>

>> I always wonder if a water immersion lens is better than an oil immersion

lens for live cell imaging. Both have the wrong RI for cells. Water too low, oil

too high. However, the NA of an oil lens is higher than from a water lens, so

the oil objective should be more light efficient and your resolution should be a

little better. Or am I wrong....?

>>

>> kees

>

>

> Hi Kees,

>

> Yes, I am afraid that you are wrong. The higher NA rays, that an NA 1.4 lens

might accept if it were used with an oiled specimen, are totally reflected at

the water-glass interface and do not get into the objective. Oil lenses work

well for TIRF because the "specimen" is just the thin layer near the interface

in which fluorescence is excited by the evanescent wave. But the cone of light

that actually forms the image (i.e., not counting the part that is probably used

for the high-angle excitation) will still be about NA 1.25.

>

> The other "thin aqueous biological specimens" that were productively viewed

with an oil objective were the various fiber systems (isolated microtubules,

actin filaments etc. lying on a slide, in media) that were visualized using

video-enhanced contrast DIC by Shinya Inoue and Robert Allen back in the early

1980s. Although the exact quantum-mechanical explanation of the interactions

near this specimen is beyond me, it seems likely that these structures were so

close to the glass that they were essentially part of it in terms of maximum

angle at which scattered rays could enter the objective. In any case, the raw

resolution of the recorded images was about that of a normal  NA 1.4 objective

with the green light that they used. The ability to visualize much smaller

structures was more a "detection of their presence on a clear background" than a

matter of resolving them.

>

> The rationale for using oil in those days was that there really weren't any

good, high-NA water objectives available.  (Here is a good place to reemphasize

Guy's point about the absolute necessity of carefully adjusting the coverslip

correction collar. At NA 1.2, aim for an accuracy of <+/- 2µm of

water-replaced-by glass. This is less important with oil objectives because oil

and coverslip have about the same RI.)

>

> As the RI of water is about 1.33, one might assume that one could use an

objective with an NA of up to 1.33. The reason this doesn't help much is that,

although the rays between 1.2 and 1.33 may not be totally reflected at the

water-glass interface, they are still highly reflected. (Just remember how

bright the image of the sun is when it reflects off a pane of glass at glancing

incidence. Although some sunlight is still being refracted through the glass

into the building, it will be dim because most of the light was reflected.)

>

> And finally, as mentioned by others, there is the matter of spherical

aberration (apparently my favorite topic!). How is this important? If you are

looking at large fluorescent objects (say >1µm), then it isn't quite so

important. Assuming it is not lost to reflection at the glass-water interface, a

larger NA lens will accept more light and if it is not absorbed or reflected

while passing through the optics, this light will end up in the image somewhere

close to its proper location, making the blob appear brighter than it would be

otherwise. However, if you are hoping that the higher NA of the oil lens will

yield  better resolution on a water specimen, forget it. As Rimas Juskaitis

makes clear in Chapter 11 of The Handbook, even under optimal conditions (i.e.,

the right oil, temp etc) the best 1.4 objectives then available were only free

from phase error up to NA 1.35 (i.e., the rays between 1.35 and 1.4 were passing

the objective but not being focused properly and hence not contributing the a

reduction in the imaged size of a point object.) I don't know how the newer

1.45, 1.49 and 1.65 objectives would perform under his very stringent test

conditions but I would like to point out that he only got the old ones to work

at 1.35 by controlling the oil temperature to +/- 1deg C.

>

> Once SA is present, the image of a point object gets bigger in a complicated

way (it is hard to define the PSF of an aberrated image with a single number.).

This means not only that the resolution is reduced, but that that the brightest

part of this image will be dimmer than it would have been without the

aberration. i.e. The high-NA oil image of a point object will be dimmer than the

aberration-free image from a water lens with slightly lower "faceplate" NA. The

best plan is to always include small (<.3µm) fluorescent beads in your

preparations and before you start imaging "for real," focus up and down on the

beads "by eye" to make sure that the slightly-out-of-focus image seen above

focus, looks very much like that seen the same amount below focus. If this is

true, then SA should not be a problem.

>

> What I am trying to say is that resolution isn't always proportional to the

number on the side of the objective. Everything else has to be "perfect" and it

seldom is. As you point out, cells are neither water or oil. It is worse than

that. Their internal RI is extremely variable, which is why DIC and phase show

strong contrast of subcellular details.

>

> But that is another story for another hour...

>

> Regards,

>

> Jim Pawley

>

> *********************************************************************************

> Prof. James B. Pawley,                                Ph.  608-263-3147 Room

223, Zoology Research Building,                              FAX 608-265-5315

> 1117 Johnson Ave., Madison, WI, 53706  [hidden email]

> 3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada

> Info: http://www.3dcourse.ubc.ca/ Applications still being accepted

>          "If it ain't diffraction, it must be statistics." Anon.

>

>

>

>

>> -----Original Message-----

>> From: Confocal Microscopy List [mailto:[hidden email]] On

Behalf Of Gert van Cappellen

>> Sent: 04 November 2010 20:38

>> To: [hidden email]

>> Subject: Re: Basic live cell imaging question...

>>

>> *****

>> To join, leave or search the confocal microscopy listserv, go to:

>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy

>> *****

>>

>>  I'm quite sure the cells will also survive an oil immersion lens and

>> normally this gives enough information for single cells. However a water

>> immersion lens is better but certainly not necessary.

>>

>> Best regards,

>> Gert

>>

>> Op 4-11-2010 2:03, Axel Kurt Preuss schreef:

>>> You need a water immersion object or have to build one

>>>

>>>

>>>     Cheers

>>>

>>> Axel

>>> -----

>>> Axel K Preuss, PhD,

>>> Central Imaging, IMCB, A*Star, 61 Biopolis Dr, 6-19B, Singapore 138673,  

sent from 9271.5622

>>>

>>>

>>> On Nov 4, 2010, at 4:06 AM, Gert van Cappellen<[hidden email]>  

wrote:

>>>

>>>> *****

>>>> To join, leave or search the confocal microscopy listserv, go to:

>>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy

>>>> *****

>>>>

>>>>   Culture your cells on a round coverslip. Take an object glas glue a

>>>> square piece of non-toxic rubber with a round hole on it. Fill this with

>>>> CO2 satured medium somewaht more as the volume of the hole. Put your

>>>> coverslip on it, with the cells to the medium off course. Press it

>>>> gently down and the glass will seal itself to the rubber ring. Now your

>>>> cells will survive for a couple of hours, so you can do the first

>>>> imaging. For real experiments you have to find a way to heat the object

>>>> glass to 37C.

>>>>

>>>> Good luck, Gert

>>>>

>>>> Op 29-10-2010 21:00, Dolphin, Colin schreef:

>>>>> *****

>>>>> To join, leave or search the confocal microscopy listserv, go to:

>>>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy

>>>>> *****

>>>>>

>>>>> We would like to do live cell imaging - mammalian cell lines - but only

have direct access to an upright Olympus BX61. We don't really need complicated

perfusion chambers, etc just something simple. We're real neophytes so all

suggestions gratefully received.

>>>>>

>>>>> Colin

>>>>>

>>> Note: This message may contain confidential information. If this Email/Fax

has been sent to you by mistake, please notify the sender and delete it

immediately. Thank you.

>

>

> --

> James and Christine Pawley, 21 N. Prospect Ave. Madison, WI, 53726   Phone:

608-238-3953