Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Our own lab has found that modifying confocal microscopes for 2-photon
is a tricky business. As mentioned before, an AOM will cause problems
with the laser. If there is any way to bypass the AOM in your system,
or simply remove it, this will help.
Another issue is optics and objectives. Most optics and objectives
are designed for the visible range rather than the NIR. From my own
experience certain medium-high-end Nikon and Olympus lenses tend to
have the best NIR performance, and most others are terrible; absorbing
90% or more of the light. A crude test is to use a laser power meter
to measure the power after your objective. Place the detector head
some distance away where the light is diverging but still all of it
hits the detector surface; the focal point can damage your detector.
Compare this to the power before the microscope to get your system
losses. Don't be afraid to ask your manufacturer for the NIR
transmission spectrum for your lenses.
Craig
On Dec 7, 2007 5:06 AM, "José A. Feijó" <[log in to unmask]> wrote:
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
> thanks George, it's good to hear about good experiences (and definitely
> you raise out some good points). But again 10x and 770nm may not be what
> most people expect out of a 2P. And sure enough, Hoechst and DAPI are
> "the" perfect 2P dyes!
>
> George McNamara escreveu:
>
> > Search the CONFOCAL archive at
> > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> >
> > Hi Jose,
> >
> > Thanks for the great post!
> >
> > On City of Hope's multiphoton rig, with a Chameleon split between two
> > LSM510's (see Brian Armstrong's e-signature), I had no problem imaging
> > hundreds of um into freshly excised mouse brain stained with Hoechst
> > for 30 minutes (perfusion fixed - handy to flush the blood out). 10x
> > objective lens. Try 770 nm excitation, 10 ug/mL Hoeschst (do initial
> > dilution from 10 mg/mL bottle into dH2O), then can dilute into
> > whatever the brain is in. 770 nm is also optimal for several (not
> > all!) Alexa dyes according to a JBO article by Mary Dickinson - so
> > Alexa Fluor 488 tomato lectin could be used to label blood vessels
> > immediately before sacrificing the mouse.
> >
> > Edrun - This was with fairly low power, but don't be afraid to crank
> > up the power! Also, try more than one objective lens, and test
> > different wavelengths.
> >
> > If working with fixed tissue: If you need to make a fixed specimen
> > transparent, there are several recipes for clearing tissue - see
> > Robert Zucker's papers using BABB for example. An interesting mounting
> > medium is 2,2'-thiodiethanol, published by Staudt et al in January.
> >
> > best wishes,
> >
> >
> > George
> >
> >
> >
> > At 06:28 AM 12/7/2007, you wrote:
> >> Search the CONFOCAL archive at
> >> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> >>
> >> the problem I believe is not 2P physics, but the Zeiss 510. We've
> >> been trying to keep a low profile about this, but it's certainly
> >> frustrating that we can't really get much out of 510 under 2p, and
> >> it's painful that we still get much better results with a
> >> side-by-side Bio-Rad 1024MP, despite the 10yrs+ difference in terms
> >> of design and technology (do mind, sharing the same laser and optical
> >> path, we just diverge form one to the other with a simmple
> >> reflector!!). During the purchase process I made a number of
> >> inquiries about a number of physical parameters affecting the
> >> negative GVD performance of the system, namely the AOM and the
> >> external detectors path, and NEVER got any satisfactory, even less
> >> quantitative answer from anyone from Zeiss. The AOM sucks, it
> >> destroys the power and the pulsewidth, one good reason for you to see
> >> such small difference, penetration is pretty much dependent on both
> >> of these. We've tried a simple hand operated ND filter, but
> >> unfortunately the fly-back of the beam in the 510 turned out to be
> >> nasty for most samples. The external PMT's optical design sucks^2,
> >> and does not help either.
> >>
> >> In our hands, and after 2 years of fighting the beast and trying to
> >> have some answers from Zeiss, we are changing our initial plan of
> >> downgrading the Bio-Rad 1024 to confocal, and focus on the Zeiss for
> >> 2P to make use of all the nice software features and (oh insane
> >> naivety!) eventually the META detector, and now we find ourselves
> >> buying video amplifiers from China to repair the Bio-Rad amplifiers
> >> and try to keep it running, and eventually downgrade the Zeiss to
> >> bare confocal (eventually we will have tow external for sale, in case
> >> anyone's interested...).
> >>
> >> I voted against the merge Zeiss/BioRad, but was kind of hoping that
> >> Zeiss people could eventually learn from their 2P design investment,
> >> and their user database experience. That is surely not the case with
> >> the 510. In short I fail to meet anyone with a different experience
> >> than ours, if there is someone out there on the list that actually
> >> managed to bend a 510 to fulfill the kind of expectations we've got
> >> used to with other machines, and have been reported in many papers
> >> out there, we would be very interested in learning how. By now I
> >> have problems seeing it coming from Zeiss, they are probably more
> >> concerned on developing the forthcoming 610 or whatever (and then
> >> perhaps propose costly upgrades "true that one didn't work, but this
> >> one will..."), but the sad reality is that I feel that Zeiss has NOT
> >> come of age in matters of 2P. Not even with the monopoly of the patent.
> >>
> >> I would be very happy to retract myself to the list if I'm proven
> >> wrong in this regard...
> >>
> >> Jose
> >>
> >> Edrun Andrea Schnell escreveu:
> >>> Search the CONFOCAL archive at
> >>> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> >>>
> >>> Dear list,
> >>>
> >>> it is a known fact that a two photon laser will penetrate deeper into
> >>> tissue, and we have done several experiments to show this, but without
> >>> success. We have used spheroids labeled with several dyes, and imaged
> >>> with our Zeiss LSM510 with 1) visible laser and descanned detector,
> >>> 2) two
> >>> photon laser and descanned detector and 3) two photon laser and
> >>> non-descanned detector. We have only found a difference of about
> >>> 10-20 um
> >>> from exp. 1) to exp. 3), whereas it should be around 100 um.
> >>>
> >>> So I'm wondering if anyone else have done this kind of experiment
> >>> and have
> >>> any tips as to how to image this increase in penetration depth? Type of
> >>> sample and dye etc.?
> >>>
> >>> Thanks a lot!
> >>>
> >>> Regards,
> >>> Edrun A. Schnell
> >>>
> >>> --
> >>>
> >>> Edrun Andrea Schnell
> >>> Divisional engineer,
> >>> Dept. of Physics, NTNU
> >>> Hogskoleringen 5, 7491 Trondheim
> >>> Norway
> >>>
> >>>
> >>
> >> --
> >>
> >>
> >> **********************************************************
> >> Jose' A. Feijo', Prof.
> >> ----------------------------------------------------------
> >> Dep. Biologia Vegetal, Fac.Ciencias, Universidade Lisboa
> >> PT-1749-016 Lisboa, PORTUGAL
> >>
> >> tel. +351.21.750.00.47/00/24, fax +351.21.750.00.48
> >>
> >> and/ e
> >>
> >> Inst.Gulbenkian Ciencia, PT-2780-156 Oeiras, PORTUGAL
> >>
> >> tel. +351.21.440.79.41/00/19, fax +351.21.440.79.70
> >> __________________________________________________________
> >> e.mail: [log in to unmask]
> >> URL: http://www.igc.gulbenkian.pt/code/research.php?lang=en&unit_id=38
> >> **********************************************************
> >
> >
> >
> >
> >
> >
> > George McNamara, Ph.D.
> > University of Miami, Miller School of Medicine
> > Image Core
> > Miami, FL 33010
> > [log in to unmask]
> > [log in to unmask]
> > 305-243-8436 office
> > http://home.earthlink.net/~pubspectra/
> > http://home.earthlink.net/~geomcnamara/
> > http://www.sylvester.org/health_pro/shared_resources/index.asp (see
> > Analytical Imaging Core Facility)
> >
>
> --
>
>
>
> **********************************************************
> Jose' A. Feijo', Prof.
> ----------------------------------------------------------
> Dep. Biologia Vegetal, Fac.Ciencias, Universidade Lisboa
> PT-1749-016 Lisboa, PORTUGAL
>
> tel. +351.21.750.00.47/00/24, fax +351.21.750.00.48
>
> and/ e
>
> Inst.Gulbenkian Ciencia, PT-2780-156 Oeiras, PORTUGAL
>
> tel. +351.21.440.79.41/00/19, fax +351.21.440.79.70
> __________________________________________________________
> e.mail: [log in to unmask]
> URL: http://www.igc.gulbenkian.pt/code/research.php?lang=en&unit_id=38
> **********************************************************
>
|
