Debugging a sensitivity problem like this is difficult especially when you do not know how well your DiVa should be performing. If you want to take the time, here are some suggestions. 1) Do a comparison between the DiVa and your Calibur for PE (and other colors, if you want). Use an antibody that should give a peak well-separated from the background, such as CD8 on PBLs. On the DiVa, collect the data in both digital and analog mode to compare to the Calibur. The setup of the digital electronics is complex, so you want to verify that is is correct. If the separation between the peaks on the Calibur is greater by 1/2 log or more compared to the DiVa, then you know you shouldn't see your tetramer staining on the DiVa. A half log difference between the instruments on PE is not that unexpected. The PE channel on the Calibur has exquisite sensitivity for two reasons. First,the collection optics are much more efficient than jet-in-air sorters. Second, the PE fluoroxchrome is bumped into some unavailable state with even modest laser powers, so its absolute fluorescence is less on a sorter with high powered lasers. The low power, long laser transit time, and efficient optics on the Calibur are hard to beat on a jet-ion-air machine. You can verify this by running the CD8 PE stained sample on the sorter at different laser powers, adjusting the PMT voltage to place the stained peak in the same position. You will find that the separation improves as the laser power is lowered. Unfortunately that is not true for FITC and for PE-Cy5. Thus a compromise has to be reached. On an ad-hoc basis, I use between 150 and 200 mW of 488 nm to excite first laser dyes. After doing this, the following are some tricks to enhance sensitivity (works on almost any sorter). 1) As pointed out, optimize the PE detection filter. If you are not using the farther red detection channels for the first laser, remove the beamsplitter. 2) Slow down the jet velocity. This is a very easy way to gain on light collection. Run at a sheath pressure of 10 psi. 3) Optimize the fluorescence obscuration bar. The standard one supplied by BD is relatively fat. You can either replace it with a skinnier one or turn it at an angle so it presents less shading of the objective. However, you need to verify that you are still able to detect 488 nm side scatter appropriately. You might also need to upgrade your FITC filter as well, since the standard 530/30 supplied may not block scattered 488 light at a high enough level. If needed, look into getting an HQ style filter from Chroma Optics. An HQ525/30, HQ525/40, or HQ525/50 will do better than the 530/30 on FITC and block 488 nm light effectively, 4) Replace your PMT. The standard PNT BD supplies on FITC and PE is a 928. The red sensitive 3896 PMT is, on the average, about twice as sensitive as the 928 for wavelengths longer than 500 nm. You could move one of your red channel PMTs to the PE position to test this. Have fun!! Marty Marty Bigos Director, Flow Core Gladstone Institute of Virology and Immunology Building 3 SFGH Rm 509 415-695-3832 >As a general case, knowing the emission spectrum of the dye that you want to >detect is a good start. (Emission spectral shifts after conjugating to >protein occur, but for a first approximation you can ignore the shift if you >have only the dye spectrum as excited by your laser line.) There are lots >of sources of this information (Molecular Probes, Chroma Technologies, the >java app that Joe Trotter did, etc.) > >Having said that, you always have a limited number of candidate filters for >a given dye, so simply try each one and use the filter that gives you the >best siganl. On ocassion, I've been surprised that the "best" filter is not >the one you might think . (But, I have a hard time believing that a 575/26 >bandpass filter is really good for Texas Red since the dye's emission peak >when attached to protein is around 619nm and is skewed toward the red, if I >recall correctly.) > >Where experience contradicts theory (that's when science gets fun, afterall) >you might wonder why? >Consider the entire light path from laser/stream intercept to the detector >that reports the dye fluorescence. > >1. How many filters are there? Remember that even a good filter attenuates >about 10% of the light. So after a few good filters, you've lost half of the >signal. Do you need all the filters? If you are not using all detectors, >you may be able to remove some dichroics. > >2. Are the dichroic mirrors really optimal? How do you know? Do you have >transmission spectra of the filters? The manufacturer often supplies them >(you kept the sheets and put them in a binder when you received the filters, >right?) Note that the region of reflection or transmittance, may not be best >for your laser/fluorochrome combination. Moreover, depending on filter >design, there can be lots of light coming from outside the region of >interest. In addition, the laminations in filters can degrade over time. >It's fairly easy to run a transmission spectrum of you filters and determine >what the detector is really seeing. > >3. Lastly, as Dave McFarland noted below, is your system really aligned >well? Very small shifts in dichroic position can give 10-fold differences in >sensitivity. A regular calibration check of your instrument is the only way >to know. > >As a final remark, the optimal filter configuration will depend on the >amount of fluorescence emission overlap that exists among the dyes that >excite with the same laser. A narrower bandpass centered about the emission >peak may exclude more of the emission of other dyes and give you a better >signal (desired emission) to noise (undesired emissions) ratio. Hence, you >may have to give up some desired fluorescence to discriminate among dyes. >The empirical approach will win here as well. > >Dave >======================================== >David M. Coder, Ph.D. >Consultant in Cytometry >email: d_coder@msn.com or dcoder1@hotmail.com >tel./messages: 206-499-3446 > > > > > >----- Original Message ----- >From: David.C.McFarland@gsk.com >To: cyto-inbox >Sent: Wednesday, May 29, 2002 6:30 AM >Subject: Re: How to enhance PE signal? > > > >Jeff, > >Pardon me if this is too obvious, but I would also suggest tweaking the PE >signal by slightly adjusting the dichroic between FITC (FL1) and PE (FL2). >My thought is that it MUST be something pretty obvious though, because PE >(FL2) is one signal that I've never had a problem with! And as far as I >know, the 575/26 has always been in place. > >And Joe, FYI the 575/26 is listed as the standard PE (FL2) filter for all >configurations listed in Appendix B (pp308-312) of the FACSDiVa User's >Guide. I for one had no idea that this was intended for Texas Red and that >a 585/42 was suggested for PE. (I just took a quick look and I do indeed >have a 585/42). > >Cheers, > >Dave > >David McFarland >GlaxoSmithKline >----- Forwarded by David C McFarland/DEV/PHRD/SB_PLC on 29-May-2002 >08:46 ----- >JTrotter@PharMingen.com > >24-May-2002 13:11 > > > > To: "Cytometry Mailing List" > > cc: > Subject: Re: How to enhance PE signal? > > > > >Jeff, > > Usually, when this occurs, the Vantage user has PE band pass filter >designed for use with Texas Red in the FL2 channel (a 575/26) , and the >Calibur uses a 585/42. Have a look and replace the filter if necessary, you >should also have a 585/42 for the Vantage in your kit that will let in a >lot more PE signal. You didn't mention what laser power you use, 100 - 200 >mW on the sorter? > > > Joe > > > > > >"Jeffrey Rice" <rice@aecom.yu.edu> on 05/23/2002 07:05:47 PM > >Please respond to rice@aecom.yu.edu > >To: cyto-inbox >Subject: How to enhance PE signal? > > > > >Hello, > I'm running into a problem here with my sorting. I've got a >peptide-tetramer > using biotinylated >peptide with SA-PE. On our Calibur, I get between half a log and a log >difference >between my negatives >and positives, but on our Vantage+DiVa I lose most of that. I really need >to enhance >the PE signal from >my PE positives to improve my sort. > Now, I know things like Oregon Green labeled anti-FITC exist. Do any > FL2-labeled anti-PE >reagents exist? Or, FL4-labeled anti-APC would be OK too. > > I've tried switching fluorochromes... I get good labeling on the >Calibur with > a direct labeling of the >protein using Alexa Fluor 647 (FL4), but again I lose the separation >between my >populations when I try to >sort on the Vantage. I could use a brighter SA-fluorochrome conjugate for >my tetramers, >or a brigher >fluorochome to label the protein. But I would think that using SA-PE or >SA-APC for >tetramers, or Alexa >Fluor 647 for direct labeling, are among the brightest available. > > Any suggestions would be greatly appreciated, as I'm fairly well stuck >at > the moment. > >Jeff > >---- >Jeffrey Rice >rice@aecom.yu.edu > >Diamond Lab >Microbiology and Immunology >Albert Einstein College of Medicine >Bronx, NY --
This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:26:12 EST