Hello everyone,
In reply to Steve Kempf concerning a discussion on the pros & cons of
different confocal systems, I thought I could add my bit as I did spend a
lot of time about 18 months ago in assessing several confocal setups.
I've noticed most confocal microscopists are very concerned about the
resolution obtained (be that in x,y or z) when, as a molecular cell
biologist, I found that sensitivity of the machine is of paramount
importance as there is no sense in having a theoretical resolution when you
don't have an image to play with! Biological samples are notoriously fickle
to deal with, and although I'll admit Fluorescent pieces of glass have their
place in assessing the ultimate performance of the instrument, I think a
real biological sample is the best.
Our current setup is a BioRad MRC 600 confocal system with an Ar ion laser,
attached to a Zeiss Axiophot inverted microscope. We have a fast 486 PC with
a large capacity hard drive for collecting images, with a Magneto Optical
(MO) drive attached for storage of images on removable disks. We also have a
heated/cooled stage for looking at live cells.
We found there were the following basic criteria in assessing the
microscopes, of which the ultimate resolution of the machine was only one.
1. The resolution (both x,y and z).
2. The sensitivity, this must be compared at the same speed of image
collection (whether that is from a single scan or from multiple scans).
Sensitivity is very important when imaging live material, particularly
material that has some movement and so one can only collect from single
scans. Most machines can increase the laser power to quite high levels, but
again this is not a good thing when you have live material with no antifade
reagents present. Good sensitivity at low laser power is most important in
biological applications.
3. The ease of use, this is most important when many people are using the
machine.
4. Robust nature of machine, and how easy it is to re-align the laser, again
this is important on a multi user machine.
To assess the confocals I used a fixed malarial parasite at the Schizont
stage. A Schizont has about 16 merozoites (contained within a still intact
red blood cell), and at the apical end of the merozoites there are two
organelles called rhoptries that from EM studies are known to be about 0.1
to 0.2 um apart. Furthermore, the body of the merozoites are packed full of
DNA. To test the confocal systems I fluorescein labelled the rhoptries using
a monoclonal antibody known to detect a protein found only within the
rhoptries, and the DNA was stained with Ethidium Bromide. The level of label
could easily be manipulated by suitable dilution of either the primary or
secondary antibodies, or by further dilution of the Ethidium Bromide used.
In this way I could make a series of slides that varied from very bright to
barely visible by normal epifluorescence. The Leica instrument was the only
machine able to just resolve the two rhoptries, however the BioRad (MRC600)
was significantly more sensitive and only slightly worse than the Leica
machine in resolution, all other confocals weren't even in the race! They
fell down badly in terms of sensitivity, particularly marked in single scan
collection (which is needed for collection from poorly fluorescent live
material).
I'm not suggesting you use malarial schizonts as your test for confocals,
but I do think the best way to go is to pick a biological sample that is
reasonably representative of the type of work you want to do on the
confocal. Even where the machine is a multi user system (as is ours) I think
you'll find the work done on it is mostly determined by the background of
people likely to use it most.
The instruments we tested (side by side at the same time!), were the
confocals from BioRad (MRC600), Zeiss, Molecular dynamics and Leica. We came
down in favour of the BioRad at that time because it had a definite edge on
the others in terms of sensitivity, that may not necessarily be the case
today with improvements appearing all the time on the various instruments.
We have been very happy with the BioRad machine overall, but I would like to
point out a few problems that you should be aware of;
1. The Kr/Ar laser gave out after only a few months (and probably
never worked properly in the first place), we are currently running on an Ar
ion laser which is working well but is not ideal for dual labelling
applications. Many others have had problems with the Kr/Ar laser supplied by
BioRad.
2. I have never been happy with the production of hard copy of images.
The easiest way is through a screen camera set up, but the best resolution
is obtained by translating the BioRad PIC images to BMP and then
re-colouring (non of this software is provided by BioRad) and importing into
Corel Draw for making into a slide. If this sounds complicated, I assure you
it is! And even more so with a multi user machine. It would be nice if
Biorad could provide all the necessary translation and colouring programs as
an integral part of the comos software.
Good luck with your assessment of confocals! Don't forget it's the overall
capabilities of the machine that matter in the end, not just the resolution.
Unless of course your prime interest is in the confocal microscope itself,
but then I wouldn't be the person you'd want advising you.
As a final note, if you buy the BioRad machine then you have to decide on a
microscope as well! We chose the Zeiss Axiophot as we had previously
assessed a number of microsocpes and decided it was significantly better in
terms of both resolution and available light when using fluorescence,
however it does suffer from the fine focus slipping problem when using the
focus motor drive as discussed by Guy Cox recently.
If anyone would like further information please don't hesitate to ask.
Bye, Alan Hibbs
Queensland Institute of Medical Research
PO Royal Brisbane Hospital
Brisbane Qld 4029
Australia.
email [log in to unmask]
|
