Priest, Richard C (rp15456@GlaxoWellcome.co.uk)

Dear Flowers, I'm interested in getting a Calibur but seem to remember seeing some e-mails a while ago indicating spillover occurred when using the carousel. Can any of you end users give me an idea of how bad a problem this is please. Another concern is that it got through sheath tank fluid by the gallon.. is this a reality ? The other machine we are considering is the Excell which seems to do pretty much the same from what I can see. If anyone has strong opinions for either machine I'd be interested to hear them. Thanks Richard

Dax Arguello (Dax.Arguello@hci.utah.edu)

Hello, There was some discussion about this a while ago. I've used both instruments with carousels, and I much preferred the XL. Spillover was not a problem as it was with the Calibur. The spillover on the Calibur can be solved by adding a wash tube between samples, or increasing the time between tubes. This was a luxury we didn't have. As far as the sheath fluid goes, the tank on the XL is also much larger than the tank on the Calibur. With the XL in a high volume clinical lab, we had to refill the sheath perhaps once a day. The tank on the Calibur needed to be filled 3 or 4 times a day, and sheath is used on the Caliburs even while the SIP is open. Something to think about. Even in a research core facility like I work in now, the BD tanks manage to get filled 3 or 4 times daily.

Both are capable machines, but if it's automated sample processing you're looking for, I'd go with Coulter.

Dax Arguello

Dax Arguello (Dax.Arguello@hci.utah.edu)

I have noticed a problem with carryover on BD instruments, even when running manually. We do mostly stem cell work here, and it's always necessary to run bleach and water before the CD34 tubes, since we experience carryover of the CD33 PE from the previous tube if we don't. It's not inconvenient, but presents a problem with automation using the loader, which may be why we haven't gotten one. Before I moved to HCI, I was using a Calibur with a loader, and noticed similar carryover. I resorted to running manually, and leaving the sample arm open for several seconds between tubes. Pretty easy solution, but it does add time to the run.*

As far as automated analysis goes, from what I've seen, I much prefer the Coulter MCL. I used that for over 2 years for HIV, and leukemia immunophenotyping with no noticable carryover. It was quite nice to load up 4 or 5 leukemia panels, and then walk away. It sure made it easier to do multiple runs, as I could stain while the first batch was running, and so on.

I'd look into the MCL if you're interested in running from a loader.

Dax Arguello Huntsman Cancer Institute Flow Cytometry Core Facility Salt Lake City, UT dax.arguello@hci.utah.edu (801) 581-8641

*John Altman (altman@microbio.emory.edu)

Can't you do the same thing with Loader Manager by adding mixing time between each tube?

Keith Bahjat (kbahjat@ufl.edu)

As I understand it, with no tube on the SIP, sheath fluid flow backwards out the SIP, where it would normally drip out, but is then caught by the DCM. Giving the instrument a couple of seconds with the DCM running should get enough sheath fluid backflushing to clear cells from the previous sample.

Clinical labs utilizing cytometers are supposed to verify that insignificant carryover between samples occurs. To test your instrument, run a tube of PBS at various times after removing a tube contaiing stained cells. We found that the typical 3 to 5 second delay implemented by the loader was enough to flush sample from the SIP, but that sometimes users manually placing tubes on the SIP would not wait long enough for this backflush to complete.

Additionally, when initially placing a sample tube on the SIP, a small amount of sheath fluid drips back into the tube. If you do not allow time for the SIP to flush fully, this drop will contain cells from the previous sample.

I use a 10 second mix between tubes, and see no carryover. During intense investigation of this matter while running thiazole orange, I was unable to detect rbc in subsequent PBS samples. I had suspect RBC's would cling to the outer metal of the SIP, but they apparently did not.Keith Bahjat

Dax Arguello (Dax.Arguello@hci.utah.edu)

I found that a longer mix kept carryover to a minimum on the BD instruments. I have also found that carryover still occurs at times. Sometimes from the mix or wash tubes themselves. We've also found a wash necessary before running any tube with rare cells, regardless of the mix time.

In my experience at the clinical lab I used to work in, this extra time was a luxury we didn't have. I have also found that despite extensive attempts to educate investigators here in the research core lab, most seem to find it inconvenient, or unnecessary to properly flush the SIP when they run samples, let alone clean the instrument correctly. This phenomenon has proved to cause a larger problem with carryover.

I guess my point is, I would do whatever it took to make any lab I were running the best it could be. Personally, I wouldn't be afraid of doing some comparison shopping to ensure that happened. I would hope that most of the flow people out there aren't so restricted to one side of the BD/Coulter fence that they'd be afraid to do that. I do understand how that happens, though. I've seen it first hand. Investigators afraid to use PC5 because they've read it's "messy", while using CyChrome or Tricolor in some of their assays. It's all the same thing. If a product is good, I say use it, no matter who makes it.

Dax Arguello Huntsman Cancer Institute Flow Cytometry Core Facility Salt Lake City, UT

Unserer Erfahrung nach haben beide Geräte, der XL und der Calibur, ein merkbares Carryover. Kleinere "abnorme" Zellpopulationen habe sich bei näherer Betrachtung schon des öfteren als Probenverschlepung herausgestellt.

Sarah Lawson (isaac.lawson@btinternet.com)

Dear All, I am relatively new to flow cytometry and have a query regarding Coulter Immunotech Intraprep reagent. Has anyone got any experience with this product? We are finding a quite marked change in scatter properties following its use, although we can obtain the same scatter pattern by increasing FS and SS total gain. Does this matter?

Any comments would be gratefully received. Kind regards, Sarah Lawson

Fischer, Randy (RFischer@therimmune.com)

Sarah, A helpful point about intracellular staining is that in general all the fix/permeabilization procedures available, either commercial or home brew, will cause a change in the scatter properties of the cells. Slight adjustments in your settings on the controls I know you run will bring back your populations and this is perfectly acceptable.

Have fun, Randy T. Fischer

Jayaraman, Sundararajan (SJayaram@med.miami.edu)

Sarah, I fully agree with Randy Fischer about the changes in scatter properties after fixation/permeabilization. I have used BD's Lysis solution and permeabilization solution to determine IL-2. IL-4, and IFN-gamma intracellularly in cultured human T cells. I found significant reduction in the forward scatter after permeabilization. This is true in the human T cell leukemia cell line, Jurkat also. Jay

Susan Zunino (szunino@biologie.uni-erlangen.de)

Carla G Hill (Carla.Hill@bms.com)

I am asking for help in determining the conditions for stimulating T cells with Staphylococcal enterotoxin B in Na heparinized whole blood. I've tried 4and 6 hour incubation at 37C using up to 40uL (10ug/mL) SEB, but no increase in CD69 expression.

Keith Bahjat (kbahjat@ufl.edu)

We routinely use SEB to study AICD in murine T cells, and I think the dynamics of activation are similar.

We culture 1 x 10^6 cells (T cells and at least 1 class II positive APC population) and 1 ug SEB and 10U IL-2 per ml of RPMI (RPMI+10% FCS + PSN + 5 mM 2-ME) in 24 well plates at 1 ml per well (IL-2 is necessary for prolonged culture periods). We see CD69 expression as follows: 24 H: 28%+, 48 H: 47%+, 72 H: 57%+, and 96 H: 46%

We analyze only the target T cell population (in mouse, mainly VB8+ T cells), and make sure we don't deplete the sample of APC's (SEB must bind class II on an APC to interact with the TCR). In humans, SEB will activate T cells bearing VB3, 12,14,15,17, and 20. Narrowing the population you analyze (using specific VB antibodies) will improve your sensitivity. Along the same lines, analysis of a non-target VB population can be used as a nice control.

Maciej Simm (simmmmer@yahoo.com)

Hello everyone. I decided to undertake a backup project and took 5 zips for a spin: I formatted them in PC format in my mac quadra 650. Then, copied a bunch of flow files from cellquest for a few months back or so. My CD-burner is at home, in a computer with winNT server and win98 installed. I copied the zips to a temp. directory on a drive formatted in an NTFS format. I noticed right away the folders had names beginning with ! and some other things were not right as far as naming. I renamed the folders and burned one CD. I took it to work and I can see the files but cellquest doesn't recognize it.

Meanwhile when I take those same zip disks and copy them to another PC at work that runs win95, thus fat-16 FS, and then copied them from the HD to a zip w/o renaming they worked fine on the mac. What is the problem: is it the file renaming, file system issues, or I just can't burn a CD with mac files? I'll do some more testing, and post an answer if I come up with one, meanwhile I would appreciate any and all input from people who've tried similar backup procedures.

BTW I got a cdrw for my new facs acquisition station which isn't here yet so I can't use it :( that's why I took it home to try it on PC.

Thanks for your time,

maciej

Jens Fleischer (jfleischer@knuut.de)

Hi Maciej, the problem is that the "identifier" (the MAC gurus would call it different...) gets lost during CD burning on the PC. So the MAC doesn't recognize that the documents belong to CellQuest. Workaround: Use the FACSConvert Software on your MAC and convert all the files from CD to a place on the harddisk. In fact, nothing is converted, but the files thereafter are again "connected" to CellQuest. I don't know whether there is any way to avoid this while working with Mac-Data on an IBM platform CD writer. Maybe the gurus know?

Jens

Maciej Simm (simmmmer@yahoo.com)

well here's how it works: 1. the mac files when copied to a PC are renamed and contain extra information in a folder called resource.frk, plus other files. These cannot be deleted or modified or else it won't work. 2. Windows NT server can be persuaded to allocate some space for MAC-like FS. 3. Burning mac files renames them (usually) (stupid windows) 4. But the CD still works when files are run through FACS convert (wohoo!)

in sum: one can back up mac files through a PC but in the end one will have to use facsconvert and naming on files and folders will be twisted beyond recognition. ex. "DAIDS A" was renamed to daid.sa etc. not to mention all the "~". hope this helps anyone stuck in a small hard drive. Thanks to all the folks at BD for walking me through this Maciej

Alice Alice L. Givan Englert (Alice.L.Givan@Dartmouth.EDU)

Hi Flowers, What are people's experiences with red diode lasers? I have been told that the manufacturers rate their lifetimes at 15,000 hours --- but that they seem to need replacing in cytometers after about 6 months. Is this true? Any ideas as to what is going on ? Thanks! Alice Alice L. Givan Englert

Bgreig2@AOL.COM

Dear Alice, We've had 3 red diode lasers in 13 months. The first one lasted about 6 months. The second lasted about 3 weeks; it was bad from the beginning. The one we currently have is about 6 months old and seems quite stable. I've heard similar reports from other local users. Sincerely, Bruce Greig Immunopathology Vanderbilt Medical Center

Mark Shlomchik (mark.shlomchik@yale.edu)

We too have seen many of the previously mentioned problems among the two FACScaliburs I use at Yale. Frequent burnouts, sometimes after a few months, although this has happened with less frequency recently. Still, lasers continue to burn out.

We have also seen the fluctuation which seemed to stabilize. This was a prodrome for burnout and although you may have thought it was running beautifully, it seems worrisome for data integrity. In my view, BD should replace such lasers even out of warranty as they have acknowledged this is a severe flaw and problem with laser 2 in the FACScalibur. Hopefully, they are listening here and will be continuing to find a source for higher quality and more reliable lasers, as some have even been suggested in other posts. Mark Shlomchik

L Affleck (l.affleck@adpro.co.uk)

Hi I asked my friend who works for Melles Griot his thoughts on the red diode post....

Dr. Louise J. Affleck

Hi, Diode lasers are particularly sensitive to the quality of their power supply - this can effect not only the usual laser parameters (such as power, wavelength etc.) but also lifetimes. What may be going on here is thus a power supply problems. Alternatives are an issue with the bare diodes themselves. But certainly a six month lifetime is unacceptable - even though it should be covered by warranty. Andrew Turner Sales Engineer Melles-Griot Ltd. Any suggestions oh laser-jock? Dr. Louise J. Affleck

John Sharpe (JohnS@Cytomation.com)

Cytomation sells a red (635nm) diode laser for use on MoFlo instruments. I took great care in selecting a vendor with a device that carried sufficiently long lifetimes for this application. This laser carries a 2 year warranty. Over the last 14 months I have been carrying out several accelerated lifetime tests and am yet to see a failure.

John Sharpe, PhD Laser & Optics Development Cytomation Inc.

Frederic Preffer (preffer@helix.mgh.harvard.edu) Sat, 13 Nov 1999 15:07:03 -0500

Dear Will Thanks much for your input. I think it would be marvelous if you can fix this problem! I do not mean to impugn the quality of these diode lasers, nor do I claim to know the overall 'engineering' answer to the problem. I do strongly feel that is the vendors job. I am merely representing the view of a lab director who is weary of the thousands of dollars spent, and related downtime lost, due to the necessity of BD coming in and needing to replace this component in the FACSCalibur. So, I will modify my statment; I honestly do not know that the lasers 'stink', but the numerous repairs required related to their failure certainly do. I know the manufacturer has nothing to gain related to this failure- I am certain they do not wish it on themselves or us. I am a long time user and fan of BD products; but this instrument is the first inwhich I was forced to conclude that an outrageously costly service contract was required, due to this problem. If another component in the instrument is causing this laser burnout, as you postulate, I feel we have been working with a beta, not a 'finished' product, and should be reimbursed as such. As you can see, our laboratory is not the only one with this problem. I hope you can help.

Regards F Preffer

Donald E. Mosier (dmosier@scripps.edu)

We have a 1 yr old FACSCalibur on its THIRD red diode laser from a THIRD supplier. The first lasted for 200 hr, the second for 2 hr (as in two hours), and the third has been up and running for 30 minutes. I would call this a reliability problem.Donald E. Mosier

Dixie Polakoff (dpolakoff@pdl.com)

I have a question concerning red diode lasers that falls in with this discussion. I have a slightly over one year old FACSCalibur with a red diode laser. For the last month (right after the instrument went off of warranty) the diode laser has been going off and on for about a hour after which it runs beautifully. I've spoken to BD and they have no idea what is going on. Any information would be greatly appreciated. Also, does turning the diode laser off when it's not needed prolong it's life? Thank you, Dixie Polakoff FACS operator

David.McFarland@mcmail.vanderbilt.edu

We've had a FACSCalibur with a red-diode laser for over two years and it still works. And we use it very infrequently. David McFarland

Frederic Preffer (preffer@helix.mgh.harvard.edu)

Hello Alice, and everybody. I know exactly what is going on. The red diode lasers available stink. Average lifetime in our instrument is about 5 months. My recommendation is that no one purchase a FACSCalibur without demanding a 2 year warrantee on the red laser; better yet, 15,000 hours of time.F Preffer

Will Roberts (wroberts@laser66.com)

Alice: I'm not sure who's laser (manufacturer) are needing replacement only after 6 months. I was an optical design enginner for a biomedical instrument manufacturer in Maryland. We had red diode lasers in service for years. These were top name, US made modules. Now I am owner of Creative Technology, a diode laser module distributor, and manufacturer. Even our imported red modules have been in service a long time ( a few years). Your information on lifetimes seems too short. Do you have any further info. on the model, manufacturer, etc? Will Roberts, Owner Creative Technology Creative Technology - Your source for diode lasers and laser modules at green, red and other wavelengths; laser pointers; and, optical lenses

Darf ich unsere eigenen Erfahrungen ergänzen: roter Laser seit ca. 1.5 Jahren in (werk)täglicher Verwendung, noch keine Probleme.
Nachtrag am 25.2.2001: Diodenlaser noch immer o.k., also seit fast 3 Jahren.

Elena Kalina (uzsjt9@uni-bonn.de)

Ladies and gentlemen, it would be very kind of you to give me a detailed explanation to followed points: 1. It is possible to use simultaneously R-PE and Cy-Chrome with dual laser flow-cytometer (FACS Calibur BD)? 2. And APC with Cy-Chrome? 3. What is the problem by dubble-exitation of Cy-Chrome in detail (physic)? 4. How different are the emission-spectra of Cy-Chrome used with single and dual flow-cytometer? Your expected answer would give us a grate help. Sincerely Yours Elena Kalina

Joe Trotter (trotter@scripps.edu)

Dear Elena, I'm sure you'll get a lot of interesting responses. In short: The tandem conjugate PE-Cy5 can be used with with both R-PE and APC in a FACS Calibur *if* you set it up reasonably. Cy5 gets excited by the red laser even though it is coupled with R-PE, so you compensate it out of FL4 by sampling the FL3 signal excited by the primary 488nm laser. The key is to have the HV "balanced" among the detectors such that the CyChrome (+) population is brighter in the FL3 channel than in the FL4 channel, otherwise it will not compensate. You can expect some cross-talk with the FL2 channel as well, but the compensation is straight forward if the detectors are set with similar sensitivities as judged by where the (+) and (-) populations fall. Since the PMTs are not all the same, you must use detector output to guage where to set HV rather than the indicated numerical values in the control panel. Since it is harder to get photons as you go out towards the red, you can expect to use more HV there, especially if you want to try to get backgrounds on scale. The emission wavelength is essentially the same for Cy5 when excited by the 633nm laser and when there is energy transfer from the R-PE. In the Calibur, the 670nm LP in FL3 may see it a little better than the 661/16 BP in FL4 because it passes more light. Since Cy5 emission is ~ 670nm, you get a strong signal in both FL3 and FL4, however. We see differences among batches of CyChrome, but have been able to use them all with FITC, R-PE, and APC in a FACS Calibur. I'll leave it to others to elaborate more. Mario's graphic at http://www.drmr.com/abcon/allspec.html may shed some light on it for you after reading his notes on resonance energy transfer dyes: http://www.drmr.com/abcon/noteRET.html.

Good luck, Joe

Mauricio Caetano (caetanom@biof.ufrj.br)

Hi people!!!! Does anybody knows any protocol to sychronize T cells in order to measure the cell cycle time? thank you Mauricio Caetano

Andrew D. Wells, Ph.D. (adwells@mail.MED.UPENN.EDU)

Mimosine (250 microM) will arrest T cells in G1. There are certainly a myriad of drugs that one could use, but I've found mimosine to have certain advantages; cells can remain arrested for several days without appreciabe loss in viability, and the arrest is reversible.

Marc Jacobsen (jacobsen@mailer.uni-marburg.de)

Dear flowers, I`d like to induce apoptosis in PBMCs using monoclonal antibodies against CD95 (clone DX2) and Protein G. I used freshly thawed PBMCs and IL2 activated LAK-cells but until now failed to get cells positively stained with Annexin-5-FITC. Even though I tried different antibody concentrations (5-20µg/ml) and different incubation periods (4-24h) no apoptosis was induced. I would be very grateful if anyone could help me. Marc

Plett, P Artur (pplett@iupui.edu)

Hi Marc If you get no A-V staining, since anti-CD95 should work, there may be the problem of no Calcium in the media. ( I hope I don't insult your intelligence, but this has happened to me and a couple of other people). Once the Ca++ is in, the A-V is on. I believe there was a discussion about this a while ago in this list and I remember some good suggestions on there. Alternatively, you can also coat wells with CD95, rather than using Protein G. (Cover wells with 10ug/ml CD95 in PBS for about two hours and then take off PBS-CD95 (you can re-use this solution a couple times), then wash with complete media. After the wells are coated you can incubate your cells for the same time as you do with the Protein-G--CD95 aggregates.Hope this helps Artur

Gerstein, Rachel (Rachel.Gerstein@umassmed.edu)

Dear Flowers, My students were asking about detection of placental alkaline phosphatase by FACS - why use this instead of GFP and are people detecting the enzymatic activity, and if so how, or simply treating it as a surface protein marker ? Can anyone provide some references or personal experience ? Medline didnt turn up much. Thanks, Rachel

Lucy Brown (lbrown@smtplink.Coh.ORG)

Hi Rachel, We have a user that uses an antibody against PLAP to determine expression of his vector. He has purchased it from DAKO.Lucy Brown

Kirk Watkins (kwatkins@earthlink.net)

I know there must be a few Coulter Profile users out there on the list. Has anyone tried analyzing data files generated on the Profile with Cellquest or any other Mac based flow software? Thanks, Kirk

Tina DeCoste (decoste@eohsi.rutgers.edu)

Kirk, I'm also interested in this type of conversion. Verity has a program called 'Pro2FCS' (www.vsh.com) and I believe that Ray Hicks is currently working on adding this feature to his FCS Assistant Software (www.fcspress.com). I have not yet tried to convert any Profile II files, but maybe someone else on the list has and can share some additional information. Please let me know what turns out to work best for you. Regards, Tina

Hans Jürgen Hoffmann (AKH.GRP02S.hjh@aaa.dk)

WE use a programme from DAKO called Flowmate to convert files from the profile format to FCS 2.0. It is a rarely used feature - one year after a new version of Flowmate had become available we were the first to complain to DAKO that it idd not work. We promptly received a patch to make it work.

Hans Jürgen Hoffmann Allergy Research Lab, Aarhus University, Denmark

Helen_Tryphonas@hc-sc.gc.ca

Dear Flow users: Can anyone direct me to a good reference regarding normal peripheral blood values for T cell subsets in human, both children and adults? I would appreciate your help.

Helen Tryphonas

There is a very usefull list with absolute and relative values of some of the major CD's from Comans-Bitter W.M., de Groot R., van den Beernd R., et al. Immunophenotyping of blood lymphocytes in childhood. J Pedriatics. 1997 March; 130(3):388-393 at http://www.ma.uni-heidelberg.de/inst/ikc/ikc-tab-lym-subp.html It is in German. Values are means (5/95 percentile). Regards Gerhard

Maciej Simm (simmmmer@yahoo.com)

from personal experience, these are the ranges for immunophenotypes:

I have 3 that I use frequently:

1. lymphocyte populations as a function of age by irene hannet et al from immunologytoday/june 92 (I can fax a copy if you give me a #) 2. blood v94 #5 sept. 1 1999 pp1803-1813 (after BMT, adults only) 3. thorax 1999; 54:697-700 (adults and elderly only)

Gaffin, Stephen L USARIEM (Stephen.Gaffin@na.amedd.army.mil)

Dear Flowpersons: Is a rise in this ratio from 1.74 to 2.39 (p<0.05) in women a small and unimportant, or a large and important change with serious consequences in the function of their immune systems? Can anyone supply a reference to the importance of different CD4/CD8 ratios on bacterial or viral infectivity or survival during sepsis/infection? Thanks, Steve.

Maciej Simm (simmmmer@yahoo.com)

In HIV patients inversion in CD4/CD8 ratio (when it becomes lower than 1 ie CD8>CD4) is used in our lab to asses well-being of patients. Also important is the naive/memory populations of each subset. There are many described ways to asses those, ranging from CD45RA/RO vs. CD4/8, adding CD62L etc, and even CD8/57. Dr. Mario Roederer recently had presented taking those subsets to the next level with the LUC technique, but I don't have those references.

Jason Cohn (jcohn@oci.utoronto.ca)

I am doing some competitive binding studies with some monoclonals and I would like to know what people would suggest as the ideal method to display and analyze this data. I am competing a conjugated monoclonal against a non-conjugated over a range of antibody concentration and want to know what others would suggest! Thanks!

ckuszyns@UNMC.EDU

Ask Bob Ashcroft, he and others published a paer looking at the EGF receptor and used competition to generate Scatchard plots of the binding. Charles A. Kuszynski

Dagna Sheerar (dagnas@primate.wisc.edu)

Hello Flowers: We are looking into the possibility of enumerating reticulocytes on our FACSCalibur. We have determined that we can use ReticONE (a Reagent kit offered by Beckman Coulter). We are now looking for a calculation to determine the Reticulocyte Maturity Index. If anyone could direct me to specific research in the literature, or has experience with this number and would like to share, I would be very grateful.

Thank you very much, Dagna

Keith Bahjat (kbahjat@acceleration.net)

Bruce H. Davis, M.D. (bdavis@beaumont.edu)

The RMI or it's analogous parameter, the immature reticulocyte fraction (IRF) are discussed in the following publications: 32. Davis BH, DiCorato M, Bigelow NC, Langweiler MH: Proposal for the Standardization of Flow Cytometric Reticulocyte Maturity Index Measurements. Cytometry 14:318-326, 1993. 34. Davis BH, Bigelow NC, Koepke JA, Borowitz MJ, Houwen B, Jaccobberger JW, Pierre RV, Corash L, Ault KA, Batjer JD: Flow cytometric reticulocyte analysis: Multi-institutional inter-laboratory correlation study. Amer J Clin Path. 102:468-477, 1994 35. Davis BH, Ornvold K, Bigelow NC: Flow cytometric reticulocyte maturity index (RMI): A useful laboratory parameter of erythropoietic activity in anemia. Cytometry 22:35-39, 1995 clinically useful laboratory parameter by any other name. Lab Hematolol 2(1):2-8, 1996 41. Davis BH, Bigelow NC, Van Hove L: Immature reticulocyte fraction (IRF) and reticulocyte counts: Comparison of CELL-DYN 4000, Sysmex R-3000, thiazole flow cytometry, and manual counts. Lab Hematol 1:145-152, 1996 43. Davis BH, Spier CM, Kachin J, Cornbleet PJ: College of American Pathologists' Reticulocyte proficiency testing program using surrogate blood: Insights into contemporary clinical practice. Lab Hematol 3:83-90, 1997 45. Davis BH: Report on the ISLH-sponsored immature reticulocyte fraction (IRF) workshop. Lab Hematol, 3:261-63, 1997 30:304-312, 1997 50. Koepke JA, Broden PN, Corash L, Davis BH, Horton AF, Jacobberger JW, Kanter RJ, Pierre RV: Methods for reticulocyte counting (flow cytometry and supravital dyes); Approved guideline. NCCLS Document H44-A 17(15), 1997 3. Davis BH: Flow Cytometric Analysis of Red Blood Cells. In Bauer KD, Duque RE, Shankey TV (Editors), Flow Cytometry: Principles and Clinical Applications, Williams and Wilkins, Baltimore, 1993. pp 373-385 5. Davis BH, Bigelow NC: Flow Cytometric Reticulocyte Analysis and the Reticulocyte Maturity Index. In Landay, AL, Ault KA, Bauer, KD, Rabinovitch, PS (Editors) Clinical Flow Cytometry Ann. N.Y. Acad. Sci. 677:281-292, 1993. 6. Davis BH, Bigelow NC: Reticulocyte Analysis and Reticulocyte Maturity Index. In Darzynkiewicz Z, Crissman HA, Robinson JP (editors), Methods in Cell Biology. Flow Cytometry, second edition, Academic Press, Inc., 42:263 - 274, 1994. 7. Davis BH and Bigelow NC: Automated Reticulocyte Analysis: Clinical Practice and Associated New Parameters. In Hyun BH (Editor), Hematol/Oncology Clinics N. Amer. 8:617-630, 1994.

Happy reading. Best regards,

Dear Collegues, I want to comment as we have done some work on retics and flow to validate the method. The standard flow method is to gate on red cells in a log log forward/sidescatter plot and to do a one dimensional histogram for green fluorecence (eg. as done by the BD reticount software). This is prone to many errors. The dyes in use are thiazol orange, auramin O, DEQTC or acridin orange, depending on the patent situation. The absorbing dyes (HeNe excitation) should not be used.

The false positive cells (non reticulocytes) include: leukocytes (cave CLL and other leukemias) nucleated red cells (normoblasts, eg. thalassemia without retics) large platelets parasite infected red cells and inclusions like Howell Jolly bodies (as already pointed out)

The dual parameter plot FSC vs fluorescence (as done in the Sysmex R3000) improves this situation a lot.

Our solution we found optimal is to counterstain with the laser dye LDS751 (styryl 8), first introduced by Leon Teerstappen (published in 1988 in Cytometry 9, 548-556). This vital DNA stain emits in the red (FL3 channel in the FACScan etc., 488 nm ecitation) and can be plotted versus green (FL1). By this way it is possible to discriminate retics from the cells mentioned above.

The method has beeen published: Bertsch T and Nebe CT, Flow-Cytometric Analysis of Reticulocytes, in: Aspects of the Flow-Cytometric Analysis of Red Blood Cells, K. Gutensohn, H.-H. Sonneborn and P. Kühnl (eds.), 69-80, Clin. Lab. Publications (1997)

I fully agree that the analysis of reticulocytes improves the differential diagnosis of anemias a lot. It is therefore important not only to measure the increse but also the decrease of retics and to report absolute numbers with high precision. This requires dual fluorescence staining and multidimensional gating. Some new hematology analyzers present a good compromise and we include a review of a blood smear in problem cases (diagnosis).

I appreciate the discussion on this subject.

Regards Thomas Nebe

Dr.med. C. Thomas Nebe

Gerstein, Rachel (Rachel.Gerstein@umassmed.edu)

PE may have a higher quantum yield, but APC often gives better resolution in "real life". This could be due to better efficiency of conjugation or the signal to noise at the excitation/emission wavelengths used. Also,think about amplification using biotin-avidin. Or, sometimes a polyclonal goat anti-rat (if your mABs are rat) can be much better than biotin-avidin -we have had great luck with a reagent from Caltag. Or you can use anti-PE, and as did my colleagues: J Exp Med 1995 Jun 1;181(6):2029-36 Fas antigen stimulation induces marked apoptosis of T lymphocytes in human immunodeficiency virus-infected individuals. Katsikis PD, Wunderlich ES, Smith CA, Herzenberg LA, Herzenberg LA

Mario Roederer (Roederer@drmr.com)

This question simply begs a host of others. The principal one being, against what background? In a vacuum of background fluorescence, PE would be the choice, since it carries more fluors than does APC. However, if you are measuring against a cell, then APC would probably be the winner, since the background autofluorescence is much lower.

For that matter, you may find that a tandem dye, like Cy5PE or PE-APC (both commercially available) may give even better sensitivity when measured against cells. This is because the autofluorescence spectrum at the large Stoke's shift exhibited by these molecules is even smaller than that for APC by itself.

Then, why are you limiting yourself to one molecule? Why not construct multimers using branched DNA or dextran as the backbone? For that matter, you could use phycobilisomes, which are huge "ex vivo" complexes of phycobiliproteins.

Finally, the new "quantum dot" crystals will almost certainly be the winner--for the simple reason that they have a nearly unlimited absorbance, given the right excitation wavelength. It is theoretically possible that you might be able to visualize a single crystal conjugated to an single antibody in a tube--given an x-ray excitation wavelength, probably. But that's still a few years in the future.

By the way, you also mention that you are trying to detect a "rare" molecule.... leading me to think that you are using a conjugated antibody. Remember that you will have far greater background from nonspecific binding, plus all the unwanted chemistries that went into an antibody, than would allow for single molecule detection, much less a few dozen. But my point is that you should worry about a host of other factors as well: how good is the chemistry that went into the conjugation, what kind of multimer complexes exist in the conjugate, how much unconjugated antibody is left, how much background does the antibody and even the fluorochrome exhibit, etc. And that doesn't even touch the question of detection: how good are your filters, your laser, your PMT, etc. All of these are critical in determining sensitivity.

Bottom line: try it with both, and decide for yourself: be empirical!

mr

Howard Shapiro (hms@shapirolab.com)

Mario Roederer, in a well thought-out and comprehensive answer to this question, pointed out:

> >In a vacuum of background fluorescence, PE >would be the choice, since it carries more fluors than does APC. >However, if you are measuring against a cell, then APC would probably >be the winner, since the background autofluorescence is much lower. >

That's true if you use 488 nm excitation for PE; however, if you use 532 nm excitation (doubled YAG laser), there is almost no autofluorescence (from mammalian cells), and the pendulum swings back toward PE.

-Howard

Dr. Robert Ashcroft (cytomat@netcore.com.au)

Interesting comment there Howard. It may allow some people with multi-line Argon lasers to use the 528.7nm line for PE, losing FITC of course. Bob

Marty Bigos (bigos@stanford.edu)

A consideration left out in this discussion is the type of instrument used to do the measurements. PE saturates at rather low laser power. Thus, PE measured on a BD-Facscan type instrument with low (~15 mW) laser power, efficient light collection optics, and (relatively) slow flow is much better that PE measured on a jet-in-air sorter with higher (several hundred mW) laser power, inefficient optics, and fast flow. APC, to my knowledge, does not have this problem. This may influence the final results obtained with the two dyes. Also, in addition to the doubled YAG, you might try the 568 nm line of a krypton laser. This should hit PE right near its excitation peak, and may make it a better choice than APC.

Marty Bigos Operations Manager

Dana Levasseu (Dana_Levasseur@microbio.uab.edu)

Hello All, We're working with several different transgenic knockout lines of mice which exhibit a severe peripheral blood reticulocytosis as well as hypercellular bone marrow. It would be nice to find a reliable enrichment or purification protocol to look at retics and stress retics in these animals to verify the numbers I am seeing with thiazole orange and other nucleic acid-staining dyes like SYTO. The literature mentions exploiting ion gradient characteristics but I want to try to keep the mouse cells in a reasonably physiological buffer as they are a little less hardy than human RBCS. Anyone have any experience or ideas? Much appreciated!

Robert C. Leif, Ph.D. (rleif@rleif.com)

You can enrich normal reticulocytes using linear BSA gradients (1). In any event, erythrocytes and presumably their precursors require serum albumin as a membrane stabilizer. Even small differences in tonicity will effect the buoyant density of cells.

(1) R. C. Leif and J. Vinograd; „The Distribution of Buoyant Density of Human Erythrocytes in Bovine Albumin Solutions". Proc. N.A.S. 51, pp. 520-528 (1964).

Maryalice Stetler-Stevenson (stetler@box-s.nih.gov)

I have an interesting case. We studied a 5 y year old Chinese boy who had hemolytic anemia secondary to hemophagocytic syndrome in infancy, a fibrosarcoma of the chest wall that was successfully resected, and refractory, persistent Salmonella infection of the bone, lymph nodes, and blood. Despite this he is functional and active. Family history is notable for a sibling who died at age 2 in China from diarrhea. We get peripheral blood on the kid- WBC is 3.29 k/ul with 28.3% lymphocytes. 72% of the lymphs are NK cells. T-cells are normal, no increase in CD56+, CD57+ or CD16+ T-cells and only 5% CD8+CD3+. The B-cells are normal. I am concerned at this number of NK cells- the Asian genetic backgroud adds to this concern and I wonder about an NK- LGL lymphoproliferative disorder. How can I assess clonality in these NK cells? Before I get much of a chance to worry about this- we review a previous lymph node biopsy which has a B-cell lymphoproliferative disorder- namely monoclonal(determined on paraffin sections), plasmacytoid B-cells. Morphology is c/w a marginal zone lymphoma. The monoclonal B-cells are likely surface light negative. They are CD20 negative. Maybe we missed them in the blood but our numbers are good- we can identify what is there- so maybe we didn't miss it. I would like feed back on 2 issues from the clinical flow community.

1. What do you do if you think you have an NK-LGL lymphoproliferative process? I haven't had a specimen before that I considered this diagnosis before.

2. Have you seen this peripheral blood profile in a patient with marginal zone lymphoma or other plasmacytoid neoplasm? MaryaliceMaryalice Stetler-Stevenson

Liu, Te Chih (liutc@pacific.net.sg)

„Have you seen this peripheral blood profile in a patient with marginal zone lymphoma or other plasmacytoid neoplasm?„ No. We don't usually immuno-type the blood unless there's a suspicion that the blood or BM is involved though. In cases where there is involvement, the malignant (marginal zone/plasmacytoid) cells appear in the blood. Te-Chih.

anpo@mb.ks.se

Hi Maryalice, 1. What do you do if you think you have an NK-LGL lymphoproliferative process? I haven't had a specimen before that I considered this diagnosis before. I would always check the TCR gamma/delta rearrangement, it may be rearranged even if CD3 is negative. 2. Have you seen this peripheral blood profile in a patient with marginal zone lymphoma or other plasmacytoid neoplasm? I haven't seen a similar case but we have recently had two patients with monoclonal lymphoplasmocytoid NHL that also had in the BM and lgl clonal proliferations (PRC confirmed) of CD8+ T cells. We have also seen an elderly woman who was diagnosed with LGL (NK phenotype) lymphoproliferative disorder and treated with steroids. A month later NK cells disappeared and AML developed. We thought that the NK cells were a sign of response towards developing leukemia. Best wishes Anna

mohamed elkhalifa (melkhali@hotmail.com)

According to the numbers you gave, this patient has an absolute lymphocyte count of 920 cell/ul. The number of NK cells is 663/ul, with the remainder (258/ul) being T and B cells. This represent a profound decrease in T cells. It is suggestive of an immunodeficiency, perhaps a variant of severe combined immunodeficiency (SCID) with relative and absolute elevation of NK cells. This would explain the increased susceptiblity to malignancies (fibrosarcoma, hemophagocytic syndrome and lymphoma) and the persistent salmonella infection. The history of a sibling dying at age two with a diarrheal disease is also in favour of an inherited immunodeficiency. Further investigations along this line may be of help. Mohamed Elkhalifa, MD, PhD University of South Alabama

Nebe, Thomas C. (thomas.nebe@ikc.ma.uni-heidelberg.de)

Hello Maryalice, why do you think its not reactive? Missing light chain and lymphoma? Hemophagocytosis and (esp. T) lymphoma occurs. LGL lymphoma in children is very rare. We do thre colour CD2/5/7, 57/56/8, > 3/56&16/8 to look for clonality (similar combinations work as well). I didn´t had a case like this yet. Regards Thomas Nebe

Fischer, Randy (RFischer@therimmune.com)

Hi fellow flowers, First, thanks to all who responded to my query about setting up a FACSCalibur/flow cytometer for GLP/GMP, it was indeed enlightening.

Second, I have been asked to summarize the responses so others can possibly do the same thing, so here they are. The first consensus is that there is no consensus/standard procedure for validating the instrument like Analytical Chemists do for their machines. So, here are some excerpts from the responses as it seems like individual users need to tailor their set ups and validation protocols.

1. All clinical flow cytometers have a "certificate of validation" that validate the hardware and software. I know this because I was required to produce it for the GLP auditors. Our instrument set up only involves running a bead to confirm alignment (DNA Check), and a second bead to shown that each day they fall within the same fluorescence channel (Standard Brite). (If you're doing relative fluorescence intensity, then probably showing amplifier linearity and multiple level fluorescence checks might be needed. Like a Rainbow Beads. I have never used these.).

The rest of our protocol QCs each reagent (how can you detect a defect in lysis, how do you know each antibody stains as it should, etc.) We use Cyto-Trols and CD Chex (preserved PMBC with assay values) to QC each antibody in our panel. Then there is the whole issue of calibrating pipets, scales, etc. Also important was recording lot numbers and expriation date on every reagent we used. We use a sheet for each day's run that records all of this information and becomes part of our results.

2. I have been doing GLP studies in flow cytometry for 5 years. Depending on what your GLP study is, will depend on how in depth your QC needs to be!!! Example: Immunophenotyping - Calibrite beads including APC monthly QC for alignment Daily check off list (usually provided by vendor) Antibody titers and parallels for new lots Basically you need to show that your detectors and lasers are working properly and consistently from day to day.

3. We are running a FACSCalibur for our quality control.(GMP) Prerequisites are quite simple: 1. a service contract for the instrument 2. regular performance of internal quality control by FACSComp 3. validation of methods (CD34 quantification) and essential reagents (antibodies) 4. standard operating procedures (SOP)

4. We routinely use a FACSCalibur for GLP studies. We used to run both GLP and non-GLP studies on the same machine. In that instance each assay was audited by QA. You will have to maintain records of when the machine was used, user name, whether the user was trained and certified for FACScalibur use, daily QCing (FACSCOMPing basically), printouts from the FACSComp, SOP for the assay you wish to run under GLP, reagents date of mfr, expiry, records of reagent use, so on and so forth.

5. f you are just starting out, visit http://www.cap.org/html/ftpdirectory/checklistftp.html and download the Flow Cytometry Checklist and others.

6. I have used our Coulter XL to perform GLP studies. I think what the chemist is referring to is that your QC procedure must include analyzing known standards that span the expected range (multiple levels) of your specimens. For a chemist, this would mean, for example, analyzing control reangents known to contain 10, 100 and 1000 IU of alkaline phosphatase, to show that your assay matches these values. In flow cytometry, for example, you can assay CD-Chex which have assay values (in the normal range) for the major lymphocyte subsets and CD-Chex LOW which have assay values for where CD4 counts are below normal. I've gone to battle with GLP inspectors because there are few QC reagents for flow and less with "mulitple levels".

However, for most of our studies, the GLP inspectors were content that we run an assay with each monoclonal antibody tested on one level of control cells (Cyto-trol or CD-Chex) and show that our results fall within the expected range.

Again, as it is obvious to even those of us just starting down this path, there are no real guidelines/requirements which we can easily follow. The CAP checklist does give a little help in providing the questions to be answered, but no answers to the question of what standards/controls to use, only to use them. If any more information becomes available, I will of course post it to the list.

Randy Fischer

Jaramillo, Richard (RJaramil@lrri.org)

I have an investigator that has designed an experiment that lends itself to using Texas Red as the fluorochrome of choice. She has acquired an antibody congugated with Texas Red for an assay that is specific to her needs. Unfortunately, she failed to inform the operator of our Facstar Plus about her choice of fluorochrome. Our instrument is equipped with a dual laser system with one laser limited to a single 488nm line. The second laser has multi-line capabilities, but can only be tuned to the maximum of 514nm. The optimal excitation wavelength for Texas Red as I have discovered is 590nm. I have discussed this with her, but she insists on analysis at 514nm. I cannot envision getting any suitable results. Does anyone out there have any experience with Texas Red that could shed any light on this problem?

Thank You

Richard J. Jaramillo

Steve G. HIlliard (steve@ancho.cb.uga.edu)

Richard, Those of us who run core facilities face this problem quite frequently. And the best part is that she will insist on doing it, and then go away saying that a) you don't know what you're doing or b) flow doesn't work. "Sometimes you just wanna smack em!"

I would try to refuse to run them, but that depends on your political situation. Figure 4 in Dr. Haugland's chapter in Methods in Cell Biology V42, PtB, (page 647) suggests that you might get a bit of a signal, but I would imagine it would take a highly expressed target. I would show her that figure and do your best to explain the limitations of this approach. And either do it for free, or make it clear that you will charge her no matter what (depending on your budget and her attitude). There's no getting around the physics of fluorescence. (Maybe we need an international organization dedicated to an increased understanding of fluorecence--we could call it the Rainbow Coalition.Sorry, I'm buggy from proofreading! ;-)

Marty Bigos (bigos@stanford.edu)

Richard- I have enclosed excitation and emission spectra of TR as PICT and TIFF files; hopefully you can open one of them. At 514 nm the excitation efficiency of TR is only a few %. Thus it will be very difficult to see much antibody expression with this choice of fluorochrome. We use a dye laser set to 595 nm for TR measurements (this has been done here for over 10 years). At that excitation wavelength TR is a reasonably good fluorochrome. Hopefully your user can be educated to choose a more suitable fluorochrome.

Dr. Robert Ashcroft (cytomat@netcore.com.au)

Why not try out the 528.7 line by borrowing some mirrors from Coherent? There is a 5x increase in absorbance (Eric's numbers) and if it works, you will only be up for the cost of the mirrors. [You will lose your FITC possibilities though] Then you will have to work out who plays the two to three grand. It's a lot of change from $35k or so. Bob

faisal rawas (frawas@hotmail.com)

Hello to all, I would like to know if any one out there been having problems recovering plasma cells from bone marrow aspirates when they are submitted to be phenotyped by flow cytometry.It seems that I always recover 20-30% less than what the pathologists are estimating with there diff.This is also causing me a problem when i try to do CD38/PI staining to estimate the S-phase of the plasma cells specially when you end up with less than 10% by flow.Any ideas or suggestions to how I can minimize my loss or enrich the recovery of plasma cells to match the diff are welcomed. Faisal Rawas

anpo@mb.ks.se

Hi, We have noted the same phenomenon. Plasma cells are fragile and the best recovery can be obtained if the sample is processed as soon as possible after it is taken and using amonium chloride based lysing and no wash. The counts obtained by morphology are usually too high because people count only 200-500 cells and start counting in areas where they see plasma cells. If one counts 1000 or more BM cells the numbers go down. Best wishes Anna

Darf ich noch einen Gedanken einwerfen: der Pathologe und auch der mikroskopierende Labormediziner, der ein mittels Adhäsionsmethode ausgestrichenes Knochenmarksstück (vulgo Quetschpräparat) betrachtet, zählt den Anteil der Plasmazellen im Knochenmark. Wir Flower bekommen immer ein Gemisch aus Knochenmark und mehr oder weniger peripherem Blut zur Untersuchung. Da im Blut meist keine oder sehr wenige Plasmazellen sind, werden unsere Ergebnisse auch aus diesem Grund oft deutlich unter den mikroskopisch ermittelten liegen. Je schlechter das gewonnene Material umso mehr werden unsere Ergebnisse abweichen.

Jeff Miller (jemiller@invivoscribe.com)

Dear Sharon, This is a very interesting discussion and a difficult case. However, I believe that I be able to provide a rapid, coherent and scientifically sound approach to diagnosis for this case, and other cases of suspect monoclonality.

As you know, approximately 90% of follicular lymphomas display a characteristic bcl2 t(14;18) translocation that can be rapidly and specifically identified using PCR. Now a PCR-based series of tests have been developed and validated that simultaneously identify both bcl2 translocations and clonal rearrangements of the B and T cell antigen receptors. The performance characteristics of Multi-Loci Clonality Testing were recently peer-reviewed and published [Molecular Diagnosis 4(2):101-117, 1999]. New assays using this advanced technology will be presented this year at the annual meeting of the Association for Molecular Pathology (Nov 4-7, St Louis), and the International Society for Laboratory Hematology Meeting (April 12-16, Banff).

This technology, which uses a PCR-based approach, will identify clonality and translocations testing as little as 50ng (50 x 10-9g) of genomic DNA, an amount of material easily obtained from blood, bone marrow, fresh (aspirate), frozen or formalin-fixed paraffin embedded tissue, or even a single glass slide. The monoclonality detection rate, determined in a method comparison study with Emory University Medical Center, was determined to be 100%, the sensitivity approximately 1 cell in one hundred normal cells, and the specificity was determined to be 100%.

One of the major advantages of this approach is that, in addition to testing for bcl2 translocations, both the immunoglobulin heavy chain gene and T cell receptor gamma chain genes are tested. Therefore, the technology identifies cross-lineage clonal rearrangements that are often missed following lineage assignment and testing for either B cell clonality or T cell clonality. I admit a bias for this approach; it is our company's technology. Turnaround time for the assays is generally one day, and the results can be posted on our SECURE Internet Client Data Access Site.

Best regards,

Christopher S Boyce (csboyce@beckman.com)

Hi All,

I am wondering what your thoughts are on this:

Suppose you calculate a ratio between two separate FL distributions (the median, mode, or mean) on the same cell line (ratio 1). The acquisition for each cell line being set on the autofluorescent sample. Now suppose you want to compare ratio(1) with the ratio(2) of another cell line stained with the same set of mabs (or multiple cell lines for that matter). What are the considerations here?

1) How valid are the calculated ratios given the type of scale used to measure the range of FL? Linear conversions of log data? Channel number only?

2) There are obviously some limitations here with comparing very different cell lines? Say for example, kidney cell vs. liver cells, rather than liver cell samples from multiple animals?

3) Would a subtraction of the two distributions being compared, then a K-S test comparing the subtractions of two different cell lines be an alternative analysis? Somehow this doesn't seem right.

4) Is setting the instrument on the unstained sample for each line really useful? Why not just set on the highest signal of the whole bunch, and let the chips fall for all cell lines (if of course all cells being tested fall within the same analysis gate for scatter signals)? Different cell types would make this approach difficult.

5) And statistics? What is the comparison we are making? Ratio of two distribution's medians compared to the ratio of two distribution's medians--- ad nauseum for every cell line? What might be the best statistical test to approach the measure of significance? How would variation fit into the picture?

Any ideas, any comments would be greatly appreciated. Thanks, Christopher Boyce Hybritech

biocytex@biocytex.com

Dear Dr Boyce : I am not sure K/S statistics is really needed is such a case. However, it seems that quantitative cytometry may be of help. If I understand well, the aim of the proposed comparison is to determine whether different cell lines express different amounts of a given antigen / receptor. Am I right ?

In case you would not have easy access to flow cytometry, what else could you do ? A binding assay with radiolabeled antibody (Ab*) would be a classic alternative. Ab* binding is measured as counts per mn (cpm) per (let say) 10e6 cells. Then, correcting the test-Ab cpm with the background binding (e.g. cpm in excess unlabeled Ab) , dividing by the specific activity of the Ab* and by the number of cells per test gives you the mean number of Ab molecules bound per cell (AB/C or ABC). That number is an absolute measurement and thus can be compared to a similar determination made on the same cells an other day. AB/C may also be compared among different cells provided you stress that the unit is the whole cell and not a real unit of cell surface (for example µm²). ( In case you really want to compare the results as Ab molecules per surface unit, then you have to assume the cell is a perfect sphere and measure its volume using the Coulter principle to determine the surface).

Measure AB/C on an absolute basis is typically what quantitative cytometry can easily do (and much more). As an example, the expression of endothelium- associated cell-surface antigens has been compared on various cells of endothelial origin (see George F. et al. (1995), in: Schlossman SF et al. (eds), Leucocyte Typing V, pp1798-1801). This showed , for example, that the typical AB/C for anti-endoglin MAbs (CD105) were very different on HUVEC ( ~ 1,200,000 molecules/cell), on the cell line Eahy 926 (~200, 000) and on the cell line ECV 304 (~20,000). Although the scatter signals were not exactly similar between cell types , this did not impede comparison of the calibrated mean fluorescence intensities.

As an other example, one can also compare the expression of an antigen common to cell types of very different size . DAF ( CD55 ) has logically very different expression levels on HUVEC (2 to 300,000 molecules/cell, same ref. as above, fig.1) and on human erythrocytes (4,800 +/- 1,400 , unpublished internal data). Although these measurements were made at very different periods (several years) using different calibrators and protocols (indirect IF with washings on cultured cells versus a 2-step, whole blood, no-wash quantitative assay), we can compare them. Both experiments (or series of) have used saturating doses of the same MAb and the calibrators are matching each other, although they have a different size and cover very different ranges of AB/C.

Once "normal" value ranges have been established for a given antigen, then cell (blood) samples can be compared for their levels of expression. As previously suggested (Ulrik Sprogoe-Jakobsen on oct. 12), the important issue is the repeatability of the measurement on the same sample. Values falling out of a range of repeated values (e.g. mean +/- 2SD) can then be considered as significantly different.

Hope that helps. Philippe Poncelet BioCytex

Christopher S Boyce (csboyce@beckman.com)

Hello Phillipe, I appreciate your ideas very much. My questions are more hypothetical to try and understand more about the proper way to set up a flow experiment from my perspective. I am looking intracellularly for the anitgen(s) rather than cell surface. I am specifically interested in the relationship between the two antigens, and not so much the actual ABC because I am doing high throughput screening of mabs. However, quantitative ABC will be the issue later in my studies when desiring to know the level of expression. I do recognize that flow is probably not the best format for measuring ABC, and the radiolabled (cpm) approach would be useful. The ratios I describe are being used to try and get a handle on the "normal" range of the relationship between the two antigens against the permutations of cell lines and multiple mabs---before quantification. Thanks so much for your references and experience with the subject. I have seen the K-S stat used for things it probably shouldn't have been used for, and I wanted to throw out a good example for everyone to "chew on," and see what happens.

I intend to summarize everyone's responses and get them out later at a later time.

My Regards,

Christopher

Ulrik Sprogøe-Jakobsen (ulrik.sprogoe-jakobsen@OUH.DK)

My two cents...

The use of K-S or other appropriate statistics to the comparison of two histograms (e.g. sample A versus sample B) will reveal the true statistical difference (i.e. either a p-value or a confidence interval). Given the high number of events (typically 5 - 10,000 or more), the uncertainty of the mean (aka. standard error of the mean) is very very small, even though the distribution of each histogram is broad and overlapping. Consequently, even the smallest difference in mean arbitrary fluorescence of sample A versus sample B, is perceived as statistically significant. Therefore, as reported to this list, running the same sample twice, will often lead to statistical significance when comparing results of the two runs by the K-S test.

However, this is not what we really want to know !!!

What we want to know is: as evaluated by measurement of fluorescence, does sample A belong to population of cells (or individuals) different from that of sample B ?

To answer this question we have to know the variation in mean (geometric mean or median) fluorescence when running the same sample multiple times on the flow cytometer. Next, to compare the difference of mean fluorescence of sample A and sample B with this uncertainty of the mean. If the former is bigger than the latter, it may be concluded that sample A and sample B belong to two different populations of cells. Formally, this may be expressed as:

Mean fluoresc. sample A - mean fluoresc. sample B > SQRoot 2 x standard error of the mean (of fluorescence measurement)

To conclude, the relevant comparison is not that of 2 means from 2 distributions of events (each from a single run), but the comparison of 2 means from 2 distributions of means (from multiple runs). In the latter case, knowledge of a general uncertainty of measurement of mean fluorescence (CV% or standard deviation), may substitute for the repeat measuring of the 2 samples, according to the abovementioned formula.

Please note: The above does not take into account the variations introduced by staining, lysing, washing, etc, only the crude variation in fluorescence measurement by the flow cytometer.

Please comment,

Ulrik Sprogoe-Jakobsen, M.D.

James Liou (jliou@bu.edu)

Hi all, I am also interested in determining statistical difference between the histogram distribution of two samples (i.e. -/+ Rx), but find this K-S stuff very confusing. My simple question is: Is it possible to just do the experiment 3 independent times and use basic standard deviation/t-test statistics? If so, does one use the mean fluorescent intensity or peak fluorescent intensity from the experiments for calculating the difference? Thanks for any/all answers and help.

Sincerely, James Liou

Candace Enockson (enockson@musc.edu)

James, what my researchers are using are overlays of the two histograms for a picture and then a statistical comparison of the geometric means of the two populations, usually a simple "flourescence intensity of population B is n times greater than that of population A." Sometimes it gets a bit more complicated when each population has its own negative that needs to be subtracted, but I first do subtracted overlays for the individual populations and then overlay the subtracted peaks to compare the two and use the geometric means of the subracted peaks for statistics. This is all done in CellQuest on a FACSCalibur. I hope this helps. (Geometric mean is important to use for logrithmically acquired data.)

Candace Enockson Medical University of South Carolina

--On Tue, Oct 12, 1999 5:32 PM -0400 James Liou <jliou@bu.edu> wrote:

>

Kenneth Ault (AULTK@MAIL.MMC.ORG)

I can't disagree with the discussion about comparing means, but in my mind that's not where the K-S test is useful. The D value is sensitive to differences in the shape of two distributions, regardless of their means. We have used it in situations in which we think we are seeing a difference in the shape of the distributions, for example one distribution may have a suggestion of two populations but they are not resolved. The D value is essentially the magnitude of the maximum difference in the integral of the two normalized distributions. An additional parameter that can be derived is the location (i.e. channel) at which that maximum difference occurs. This gives you an idea not only about how different the two distributions are, but where the difference is (mostly) located.

Ken Ault

Derek Davies (daviesd2@icrf.icnet.uk)

Dear Ulrik, Thanks for that post, which, I for one, think is eminently sensible! The variation of means of the same sample re-run on the cytometer is something that few people seem to take into account and seems to me that, especially if you are looking for small fluorescence shift, that this is an important measurement.

I often have users who want to claim a significant difference between the cumulative distributions of the control and test samples on the basis of a small shift in mean channel value in one experiment. I always say that they need to repeat to verify that but that they also need to be aware that there will be a variation in fluorescence level even between negatives run at different times. At some point, a muliple run of the same sample must be done so that the variation can be quantified. If this is then smaller than the variation between the control and the test samples, then a further statistical test can be performed to determine the level of significance. Let us also not forget that the change of a continuous distribution into a dicrete channelised distribution will also involve loss of resolution so I am always wary of small changes in any one experimental run.

As for KS statistics, as they always seem to indicate a significant difference (for the reasons stated by previous repsondents) I have never encouraged their use.

{Apologies for quoting the whole of the previous text, but I think it needs it!}

Derek

Jacqueline Saleh (jacqueline.saleh@ariad.com)

Hello Flowers, Is there a way to measure beta-galactosidase using flow cytometry? Thanks in advance for your response.

Mario Roederer (Roederer@drmr.com)

Jackie, while no longer in vogue (due to the popularity/ease of GFP), beta-gal remains an excellent choice for reporter gene detection. It is considerably more sensitive than GFP; we were able to detect as few as 5 molecules of beta-gal per cell: probably a single mRNA molecule's worth! However, the beta-gal assay requires just a little bit more work than GFP. A full description of the methodology can be found in a few references. Also see the chapter in Current Protocols for Cytometry, which updates a review of these articles plus a variety of articles that have used the method in the past 10 years.

1. Nolan, G. P., et al. (1988). Fluorescence-activated cell analysis and sorting of viable mammalian cells based on beta-D-galactosidase activity after transduction of Escherichia coli lacZ. Proc Natl Acad Sci U S A 85:2603. 2. Fiering, S., et al. (1991). Improved FACS-Gal: flow cytometric analysis and sorting of viable eukaryotic cells expressing reporter gene constructs. Cytometry 12:291. 3. Roederer, M., et al. (1991). FACS-Gal: Flow cytometric analysis and sorting of cells expressing reporter gene constructs. Methods: A Companion to Methods in Enzymology 2:248.

mr

lbrown@smtplink.Coh.ORG

Hi Jackie, We had a user that followed the following reference for Beta-Gal Activity with good success: Cytometry 12:291-301 (1991) Hope this helps. Lucy Brown

Baxter, Monica (mobaxt@pahosp.com)

We have noticed that our anti-CD13 antibody has been staining cells in the lymphocyte gate. All the references we've read indicate that CD13 should only be present on cells of myelomonocytic lineage. Has anyone else noticed this problem. (The staining is dim, but definitely there). Thank you,

Olindo Assis Martins Filho (oamfilho@cpqrr.fiocruz.br)

Hi Monica, I suggest you to check if your anti-CD13 is a mouse IgG2a. I always have problems of unspecific binding of this subclasse in my experiments. The way to avoid it is to incubate the cell suspention with some mouse normal serum (1-5%). Another possibility is that your settings are bringing some neutrophils to the lymphocyte gate. Did you check were (on the SSC x FSC graphs) your positive cells are located within the lymphocyte gate? Good Luck, Olindo

Bgreig2@AOL.COM

Monica, There is an interesting article in "Immunology Today", Feb.1999, vol. 20, No.2, pp 83-88 that describes CD13 being found on early B and T precursors. We see the CD19+/CD13+ phenotype in occasional early B leukemias. It's a nice phenotype for follow-up because of its rarity. Hope this helps. Sincerely, Bruce Greig.

Ich nehme an, M.Baxter hat ausgeschlossen, daß es sich bei diesen Zellen um Basophile handelt. Die sind ja auch im (Scatter-)Lymphogate.

Sharon Vogt (svogt@bellsouth.net)

Hi all, We have a clinical case that's interesting to the point of frustration. Any comments?

53 y male, 3 years ago and now (recurrent) lymphoma, B-cell type. Phenotype is similar in both flow cytometric studies:

CD19+ CD20+ CD22 dim, CD5 dim before, partial now, CD25+, CD23 partial, CD11c partial, with kappa light chain restriction. Negative for CD10. We do not stock FMC7 (but will soon), CD21 or CD24.

By morphology and flow (considering dim CD5 as negative), the diagnosis of follicular center cell lymphoma (noting that 20-30% of these are known not to express CD10) was made. None of us are entirely sure of that diagnosis now, and there appears to be a tad more CD5 expression (recent specimen was FNA of cervical node); it was suggested that this may be a mantle cell lymphoma.

OK, simple enough - immunohistochemical stains bcl-2 and cyclin D1 should answer that question. Those are difficult stains to optimize, however, and are affected by variations in fixation. In other words, they didn't help much. CD20 nor kappa light chain expression is remarkable for staining intensity, but I'd favor a CLL based on the rest of the phenotype. The pathologists say small cell, low grade, but not really CLL-like and definitely not PLL. ?? Thanks for any discussion.

Sharon

Brent Dorsett (brentd@nyct.net)

Just make sure that you do the bcl2 and cyclin D1 immunohistochemistry after high pH anitigen retrieval. It makes a big difference.

Brent Dorsett Lenox Hill Hospital --- NYC

anpo@mb.ks.se

Hi, This could be either sc atypical CLL or mantle cell lymphoma. I would make a decision considering: cytology of the cells (are they "cleaved"?) and CD23 expression - dim? how large fraction of lymphoma cells is positive? The pattern of Follicle Dendritic Cells in the lymph node biopsy (if available) can also help in differential diagnosis. Best wishes Anna

Maryalice Stetler-Stevenson (stetler@box-s.nih.gov)

This could very well be a CLL or small lymphocytic lymphoma- SLL (a not yet leukemic neoplasm with same features as CLL). The CD22 dim, CD5+ (at least partial), CD10 negativity and CD23 +(at least partial)are consistent with CLL.The partial CD11c (as long as it is dim) and CD25 + is common in CLL. Mantle cell is almost always CD22 moderate or better. This would also go with the morphology. Lymph nodes in CLL/SLL show an inflitrate of small lymphoid cells with larger lymphoid cells forming pseudofollicles, which are proliferation centers. Thus CLL/small lymphocytic lymphoma can be confused with follicular lymphoma based upon morphology. The small lymphocytes can be irregular resulting in confusion with mantle cell lymphoma. On the other hand, if the CD5 is really negative, one can see CD10- and CD23+ follicular lymphoma. It all depends upon the CD5 to distinguish between the 2. Have you tried blocking stainin with CD5 with non-labeled CD5? If you have true CD5+ I don't like follicular lymphoma and I don't like mantle cell with CD23+ and CD22 dim. Your cyclin D1 and bcl-2 may be difficult to interpret because they are both negative. It is impossible to come to a strong final diagnosis based upon the data reported, but I would favor CLL/SLL. This is not a typical anything, and its biological behavior is likely to reflect that. Maryalice

Nebe, Thomas C. (thomas.nebe@ikc.ma.uni-heidelberg.de)

Dear Sharon, I appreciate the comments from Maryalice and Anna.

We had many of these discussions among collgues here in Germany. For the benefit of the patient and clinician: The way to go is to extirpate the lymph node in total and send it (or pass on) to a reference pathologist who is well trained (belongs to the REAL group). They hate fine needle biopsies as the architecture of the lymph node gets lost. Your given immunophenotype (by "feeling", missing morphology and conjugation) does not suggest a CLL and lymphoma diagnosis by flow is a difficult game even including morpholgy and experience. The discussion of your data is complicated by the fact that the missing fluorochrome is crucial: PE gives positive signals eg. for CD25 or CD11c while FITC conjugates give negative results.

Lymphoma classification and treatment of lymphoma is based upon lymph node pathology. The extended marker panel you anticipated we use for a while. It helps to exclude B CLL but does not give a safe diagnosis of other B-NHL (except HCL, may be Immunocytoma). The so called atypical B-CLL has never been defined and even very experienced collegues refused to give a presentation on that subject.

SLL from the working formulation is not a well defined clinical entity and therefore did not exist in the KIEL classification. Harald Stein (Berlin, REAL group) just presented at the german hematologist meeting in Jena the new concept of a B1 and B2 CLL, where the latter is CD38 positive and forms lymphoma. (The concept behind is based on B cell development in the lymph node). This provoqued a lively discussion esp. among those who used CD38 in the past. We also found CD38-FITC surface positive (dim compared to immunocytoma or plasmocytoma) in those CLL patients with spleen and lymph node infiltration (total CLL n=85). Possibly cytoplasmic staining resolves the controversial issue. I dont want to be bouring but it would be helpful to have a minimum report format (combinations / fluorochrome / surface or cytoplasmic / intensity compared to normal counterparts) when discussing such cases. Best regards Thomas Nebe

Dr.med. C. Thomas Nebe Universitaetsklinikum Mannheim Zentrallabor

JUAN LUIS CASTILLO NAVARRETE HI FLOWERS: Today I have a immunphenotyping of a patient,a male,4 years, the sample was bone marrow (als periphereal blood), and the diagnosis was Kostmann`s neutropenia. Also the patient is under treatment with G-CSF. In the immunophenotyping I found a few things that not are clear for me: the netrophil population have a normal FSC but have a low SSC, and also have normal markers for myeloid (CD33, CD13, MPO, TDT), but also have a expression of CD61 (M7?). Also I found a little population of cells with low SSC and Low expression of CD45 (the classical region of blast). This cells was CD19, CD20, HLA-DR, CD10(CALLA), MPO dim, TDT(-), CD61(-), CD34 (-), CD33 (-), CD13 (-), CD14(-). I search information about Kostmann`s syndrome, and the next is the principal:

Severe congenital neutropenia (SCN) or Kostmann's syndrome is characterized by a stop in differentiation of myeloid progenitor cells at the myelocytic or promyelocytic stage. with absence of neutrophils in bone marrow (BM) and blood. The pathophysiology of SCN is still unclear. (Exp Hematol 1999 Jun; 27(6):1038-45. Hypotheses of the pathophysiology of SCN include (1) defective production of granulocyte colony-stimulating factor (G-CSF), and/or (2) defective response to G-CSF. (Blood 1991 May; 77(9):1919-22). The administration of granulocyte- colony-stimulating factor was shown to be safe and effective also in reducing infectious episodes in these patients. (Acta Paediatr 1992 Feb; 81(2):133-6). Thus, it is hypothesized that the underlying defect responsible for SCN is based on an abnormal G-CSF-induced signal transduction pathway .eutrophils from SCN patients show an increased autophosphorylation of JAK2 (a nonreceptor tyrosine kinase involved in the signaling pathway of G-CSF )in comparison with that of neutrophils from healthy volunteers. (Blood 1995 Dec; 86(12):4500-5).

So, the patient : May be a leukemia ? , M7 ? , lymphoid ? or Does the normal production of bone marrow upon treatment of G-CSF ? Any idea will helpme Thanks

Nebe, Thomas C. (thomas.nebe@ikc.ma.uni-heidelberg.de)

Dear Juan, Neutrophils express CD64, the high affinity receptor for IgG upon G-CSF treatment, resulting in high background and unspecific staining (hard to get rid off by blocking steps). The other population you described are normal B cell precursors in the bone marrow. Regards Thomas Nebe

anpo@mb.ks.se

Nancy Gin (nancy_gin@yahoo.com)

Hello, Would like to know if any of you have successfully identify Anaplastic T-Cell Lymphoma by using CD30 by flow cytometry. Any information on manufacturer/clone, procedure used would be very much appreciated. Thanks in advance. Nancy Gin - United Hospital, St. Paul, MN

Leslie and Len Brown (lenbrown@powerup.com.au)

Dear Nancy I have had good results using DAKO's conjugated CD30 antibody on the rare cases of anaplastic large cell lymphomas that I've encountered. Normal surface staining using established whole blood methods or usual procedures for handling FNA or lymph node tissues proved no problem. As an adjunct, I would be wary of refering to anaplastic lymphomas as being of T cell orign!

Regards Len Brown Brisbane

Abby Kelliher (allena@helix.mgh.harvard.edu)

I have combined all of the responses to my Cyto Chex queary and have included those responses (with names removed) below. In summary, most people responded positively but there were two people who have had poor results with this product. There seems to be conflicting results regarding use of this product with whole blood.

My own testing, while progressing slowly is encouraging. I have tried it on a positive B-ALL bone marrow with no differences one week later. I have also tried it on a CSF specimen and the specimen held in Cytochex looked better than the one we processed fresh.

I thank everyone for their detailed experiences with the product and hope that this turns out to be the answer to my weekend prayers!! I am also interested in further experiences with this reagent.

Abby Kelliher Clinical Flow cytometry Lab Mass. General Hospital Boston, MA 02114

Positives: ---------------------------------------------------------------------------- -------------------------- We have used Cyto Chex over weekends on blood and it works well. ---------------------------------------------------------------------------- -------------------------- I usually don't keep blood or bone marrow specimens for a week before flow analysis (usually it's 2-3 days, due to weekends and my being off one day for working the weekend in the coag section of our lab), but the vast majority of the time, I get good results with keeping blood/bone marrow samples in Cyto-Chex--as long as we get the specimen into the Cyto-Chex the same day that it has been obtained. I get VERY good results with keeping tissues in Cyto-Chex, as long as I disaggregate the tissue the same day that it has been received by the pathologist assistants. I think the occasional junky-looking specimens have more to do with the specimen itself, not the Cyto-Chex, per se. ---------------------------------------------------------------------------- -------------------------- We have been using Cytochex for a couple of months now and are pretty happy with it. For the most part we get very good light scatter and Ag expression. We use it on late Friday afternoon specimens and use it only on non-bloody specimens. I would not recommend it for whole blood since we have had some problems with it too (as one person described it today on this list). If you want to do bone marrows or bloody specimens, you need to lyse it first and then store it in Cytochex. The only problems we have encountered so far, are: 1. DNA ploidy is often near-diploid even on benign samples after Cytochex storage. 2. CD 10 expression seems to be often non-specifically positive. But if you compare against your internal negatives, you are able to determine whether the positivity is true or non-specific. 3. Problems with peripheral blood as described. We really try to only use it when it is too late on Friday to analyze but it is not a stat specimen. Over all, it is not the perfect solution but pretty close! ---------------------------------------------------------------------------- --------------------------

We have been using the cyto chex for over a year, but our validation shows a good correlation up to 4 days, we used over weekends or holidays. Do not check viability by PI or 7AAD after cyto chex some permeabilization is present. ---------------------------------------------------------------------------- -------------------------- I have carried out some stability work on cytochex, but it was limited to whole blood and ORTHO TRIO reagent i.e CD3/4/8/16/19. We tested over an eight day period at room temp and 4degC. and the results were very good. If I remember rightly, the manufacturers state that the sample should be viable up to 7 days at 4 deg C for both transport and storage. The only issue to keep in mind is obtaining accurate dilutions for absolute counts. ---------------------------------------------------------------------------- -------------------------- I also have had excellent experience with this product. Although I was one of the original "testers" of this product I have taken it further and used it in my lab for my own purposes. For example, I stimulated normal peripheral blood with CD2/2R to obtain high expression of CD69. When the blood is put in Cyto-Chex the specimen can be used as a positive control In our activation assays for at least one week with no loss of antigen expression. We're pursuing other uses for this great product. Also , when we need a "normal" control (fresh blood) we draw only on Monday, put the blood in Cyto-Chex and use it the rest of the week so we don't have to draw someone daily. I hope you keep experimenting with other uses for this great product. I have nothing but good to say about it. ---------------------------------------------------------------------------- -------------------------- We have looked very extensively to the Streck reagents: -Platelet Chex (unfortunately discontinued) -CD Chex -CD Chex Plus -Cyto Chex

This Streck company has some great products. We evaluated their controls and stabilizers for our quantitative flow cytometry application kits. You must know that quantitative flow is not very forgiven and we found great performances for these products on various measurements. It did not work for everything. ---------------------------------------------------------------------------- -------------------------- Negatives: ---------------------------------------------------------------------------- -------------------------- I tried it out in my lab in Massachusetts to permit fedexing samples from the South. A single sample of human blood (mine) was treated with CytoChex on day zero and then analyzed daily for the next 4 days. The results of CD4+ and CD8+ were significantly different on day 1 and subsequent days. As a result, despite our great hopes according to claims made, we did not adopt its use. ---------------------------------------------------------------------------- -------------------------- I hate to comment because I haven't finished my own evaluations. But, it's not at the top of my list of to do's because I wasn't overly pleased with the initial comparison data.

I did have better results for subsets than the one who got varying subset values on days 2, 3, 4. We use a CD45 v. Side gating strategy for subsets and got comparable percentages on days 2 and 3 - I didn't test beyond that. The scatter, however, looks pretty bad as grans don't seem to hold up well. While it's true that that is of no great consequence with my particular gating strategy, it could lead to a very messy specimen for those using light scatter gating, since the grans frequently fall into the gated area. And it just looks bad - I hate reporting out stuff that is so obviously aged, even if it has been validated on some other bunch of samples.

Also, we tried Cytochex on some lymph node and bone marrow specimens. Same story - ratty looking cells. Worse, though, was that it appeared that the surface light chain expression was undemonstrable after Cytochex storage of the single cell specimen preparations. If I can't demonstrate clonality if present, I might as well not run the specimen.

I knew that there was some literature out there using this preservation method, so asked my new Streck salesman for it. He zipped it right over to me - you should ask them for it via fax - or I'll send it to you if you want. What the paper _apparently_ says (ok, I haven't read it all the way through yet, myself, but a co-worker reported the findings to me) is that the reagent would keep the neutrophils from upregulating or otherwise changing the CD11b status of the specimen. The pictures I did look at(!). It looked to me like a large part of the specimen had turned to debris after three or however many days. While it may be so that the percentage of CD11b on the grans was comparable to that of the original test, I don't see how the measurement is even valid after losing a large part of the population.

Ray Hetser wrote: Several times over the last few months when analyzing cells that have been stained with propidium iodide we have seen, using pulse processing, cells that have an FL2W signal that would seem to qualify them as doublets and yet their FL2A is similar to that of a singlet. E.g., in the example below, if xxx represent G0/G1 single cells, yyy are G2/M single cells, and zzz doublets, what are the cells represented by ?

Thomas Delohery (t-delohery@ski.mskcc.org)

Subject: Re: DNA: Double wides but single brights

We routinely look at FSC vs FL2A (integral fluor) as well as FL2A vs FL2W. The "double wide" G1 cells are better resolved with FSC than width. This occurs in primary epithelial cells in response to agents that induce differentiation; eg, TGF-b. Curiously, the smaller cycling cells can be fully resolved from the huge G1s with nuclei prepared via the Nusse protocol but when intact cells fixed with EtOH are used, we see one broad distribution on both FSC and FL2A. I speculate that it has to do with cytokeratin production during differentiation. td

Derek Davies (daviesd2@icrf.icnet.uk)

Hi Ray, You dont say what type of cells you are looking at. If they are large epithelial-like cells this could be a bit of non-specific cytplasmic uptake but that is unlikely to be confined to the G1 population. Or do you see a general width increase as well?

Alternatively it could be FACSRubbish. Are you triggering on FSC or FL2? If you trigger on FSC and look at the AvW plot does that wide 'G1" population look as if it is heading off towards the origin? I have seen this sort of thing when there is a lot of debris or alternatively when the culture from which the cells are taken has been contaminated.

In either case, it would probably be productive to have a look down the microscope and see if you can determine what could be causing it. Hope this helps etc Derek

Michael Ormerod (Michael_Ormerod@compuserve.com)

The obvious interpretation is that they are 'fat' nuclei in G1. Put another way , they are nculei from cells in G1 but for some reason they are much biger diameter.

You do not say how the cells were treated prior to analysis or how they were prepared for analysis. It is possible that either some drug treatment has caused the nuclei to expand or that this has occurred during preparation. You should look at the stained nuclei under a fluorescence microscope to see if any of them are appreciably larger. Michael Ormerod

Carl-Magnus Högerkorp

Dear All, does anyone have any experience in analyzing preparations stained with Ab's conjugated with PE, PE-Cy5 and PE-Cy7 at the same time. Are there any problems associated to this kind of stainings?

Carl-Magnus Högerkorp Stem cell laboratory Department of Internal Medicine University Hospital of Lund Sweden

Mario Roederer (Roederer@drmr.com)

Dar Carl-Magnus: We have extensive experience with these dyes. There are no new "problems" associated with these stainings, only complications. Nearly all complications have to do with the compensation difficulties.

FIrst of all, you should carefully compare Cy7PE conjugates from different manufacturers: they can be significantly different in quality and spillover requirements.

Second, you should recognize that most different "lots" of Cy7PE tandems (i.e., nearly every different conjugate) will require a different compensation setting. This means you will have to collect compensation controls for each different reagent. (Note that this can be true for Cy5PE reagents as well!). You might want to design your panels so that you only use 1 or 2 different Cy7PE reagents so as to keep the different compensation settings to a minimum.

Third, you will have to compensate the sample using software. You can do partial compensations on the instrument, but you will not be able to completely compensate the data. There are a couple of software products that can do the compensation. On the Mac, we use FlowJo as it keeps track of the different compensation settings necessary for each different panel, and automatically applies the correct compensation as needed.

Finally, you will find that some combinations work better than others (i.e., if you are doing CD3, CD4, CD8, CD19 on the FITC, PE, Cy5PE, and Cy7PE dyes, you may find that some combinations of antibody:fluorophore work better than others). We typically put together several different combinations and evaluate them before settling on one for more extensive work. mr

Dr S Sreckovic (Sasa.Sreckovic@bristol.ac.uk)

Hi Carl, I did a few experiments using FITC, PE, PE-Cy5 and PE-Cy7 like 4-color analysis from one laser line (488nm) on a FACSVantage. Compensation for overlaps between PE and its conjugates is a bit tricky but it's not more then 30% either way and you just need nice monocolor controls to set it right. We are using Streptavidin-Pe-Cy7 which tends to be a little sticky and careful titration would be necessary.

Regards, Sasha

Marty Bigos (bigos@stanford.edu)

Hi-We have run these 3 dyes in our facility for quite a while. There are several technical issues involved. The first is to make sure that the two tandam conjugates have good energy transfer. This can vary from lot to lot. Second, for the Cy7PE especially, but for the others as well, you want to make sure that you have the optimum light collection setup. This means that a) you need red-sensitive pmt's (hamamatsu R3896 or equivalent) and b) optimal filters. With those 3 dyes a 575/25 bandpass for PE, 700/80 bandpass for Cy5PE, and a 750 long pass for Cy7PE are good choices. Also the beam splitters need to match the above filter set-up as well. The reason for all the above hulla-baloo is that when you are collecting out in the far red, the actual amount of light there is relatively low, much lower that for FITC, even though the separation between the unstained background and stained cells may be quite good. So although single stains with each reagent may look fine, when doing multiple stains, the compensation interactions and the inherant statistical variation in the low level light signals can cause problems (smearing of the backgrounds, mainly). Below is the spillover values for these three colors as well as FITC. Fluor PE Cy5PE Cy7PE Fluor 1.0000 0.3532 0.0712 0.0044 PE 0.0091 1.0000 0.2588 0.0150 Source Cy5PE 0.0004 0.0241 1.0000 0.0549 Cy7PE 0.0043 0.2138 0.0822 1.0000 For all the PE-based reagents, there are enough interactions between them so that full 3x3 compensation is usually necessary for obtaining interpretable results. The addition of FITC further complicates the interactions. In general, after compensation you will find that the the background on all three measurements will increase.

Newsom, Brian S. (BSNEWSOM@txccc.org)

I have an interesting/puzzling case for the group. A 16 month old Female, (non-Downs) was diagnosed with M7 seven weeks ago. She was not doing well on therapy and we got the case. Her morphology shows large blebbing blast-like cells but her cytogenetics look very pre B-ALLish, WBC is 3.2X10^6/mL. Her flow is as follows:

Positive for CD38, CD42a, CD56 and CD117

Partially/Dimly positve for CD33 and CD61

Negative for CD3, CD4, CD5, CD7, CD8, CD11b, CD13, CD14, CD15, CD16, CD19, CD20, CD34, CD95, Kappa, Lambda, TdT, MPO and HLA-DR

There also is a population of about 0.5% of the BM that is CD10+CD22+ (pre B-ALL fraction of two clones?) which falls in the same CD45vs.SSC gate but seems negative for all other markers. Any ideas? Brian Newsom

Nebe, Thomas C. (thomas.nebe@ikc.ma.uni-heidelberg.de)

Ress, S, Stan, Dr (sress@uctgsh1.uct.ac.za)

Hi Folks, This question has been posted before by myself and others, but I'm still in the dark, so here I go again... Does anyone have a protocol for freezing whole blood that would permit subsequent flow cytometry on the thawed sample? We have a study starting but the ab have not all arrived. Alternatively, is there any data/personal observations on comparison of frozen buffy coats vs ficoll gradient PBMNC with regard to flow cytometry? There must be some information on this, someone must have tried it!

Any data or advice would be greatly appreciated.Thanks, Stan

Candace Enockson (enockson@musc.edu)

Stan, I have a user who routinely freezes his cells prior to staining for surface markers and cytokines. However, he does not freeze the whole blood as you asked. He processes the blood as he would for non frozen samples - stimulates, lyses, fixes - then suspends them in 10% DMSO and 90% FBS and freezes. When he's ready to run, he thaws and stains. Works fine.

Candace Enockson Medical University of South Carolina

Sharon Vogt (svogt@bellsouth.net)

Hi all,

In the above scenario, the cells are fixed before freezing. While certain antibodies (to surface antigens) will bind after fixation, others won't - true? Is it really 90% FBS?

I should start by saying I don't have _much_ experience with this, but have followed some protocols for freezing wbc from a ficoll prep with DMSO (8-9%) in PBS w/ some small percentage of FBS. I limit the number of cells per ml to 1-2x10e7, thinking that some limit is necessary, but arbitrarily picking that range.

Assuming that all of that is ok, why can't you do the same thing with whole blood? It must not be as simple as the freezing or thawing process acting to lyse the rbc, because the question is raised from time to time. Is there too much protein or cellular material released? Since grans don't tolerate freezing very well, the percentages of wbc populations would be altered, but mononuclear cells would be alright? Maybe I'll try it, but I don't really have the need to since we don't routinely freeze cells in my lab - I'm just curious. If I'm completely missing some basic concept, please feel free to let me know.

Candace Enockson (enockson@musc.edu)

Yes, it is 90% FCS. It comes from the Pharmingen catalog Cytokine Analysis protocol under "5. Alternative Fixation and Permeabilization Protocol". It does fix before staining so surface abs have to be able to withstand the fixing process.

Te Chih Liu (liutc@pacific.net.sg)

Fiebig et al. Cytometry 1997; 29:340-350 describes a simple method that gives comparable results for lymphocyte analysis to analysis on fresh blood.

Te-Chih

Te-Chih Liu, MRCP(UK), MRCPath Haematology Division National University Hospital Singapore

adrian rubio (genius@nmt.edu)

Make a regular RPMI and 40% Fetal Calf Serum mix with 10% DMSO Freeze your pbl's in it and they will be fine for flow.

Ress, S, Stan, Dr (sress@uctgsh1.uct.ac.za)

I want to thank everyone who responded for their helpful comments, for advice and protocols received.

The reference from Te Chih (see below/above) was extremely helpful and validates freezing WB for subsequent performance of CD4 and CD8 lymphocyte markers. I am interested in monocte markers and just hope that the same will hold.

I would also just reiterate what others have said: this is an incredibly active and helpful list. People are very generous with sharing experience and protocols, and I appreciate all the helpful responses I received recently, in response to this and to several other queries.

Thanks again,

Sarah Lawson (isaac.lawson@btinternet.com)

Dear All, I have read with interest the correspondence regarding freezing whole blood rather than Ficoll separated MNCs for subseqent flow analysis. Following this I have several queries.

1. Has anyone got experience with freezing whole bone marrow and its subsequent analysis after thawing, especially at diagnosis of acute leukaemia? Does the immunophenotype remain stable? I have a similar problem to Dr Stan Ress in that I am receiving samples but am awaiting a delivery of antibodies.

2. I am having difficulty obtaining the quoted 'Cytometry' reference - Fiebig et al 1997, 29, 340-350. Does anyone know if this journal is kept anywhere in the area of Birmingham, UK? Alternatively, can anyone send or e-mail me a copy? (isaac.lawson@btinternet.com)

Thanks for your help.

[SMTP:raswp@mahidol.ac.th]

Rhodamine 123 is a distributional membrane potential probe. The dyes distribute according to the nernst equation, but once inside the cell they find an intracellular binding site they hang on to acting as an intracellular sink. This allows more fluorochrome to enter the cell, leading to a substantial signal enhancement. As nicely described in Howard Shapiros book, in mamalian cells it can be used at high concentrations to act as mitochondrial MP measurement. When the dye is washed away the dye is retained in the mitochondria as it is in distributional balance with the remaining dye in the cells cytoplasm. However, if you read Howards book you will find better ways to do that, (especially in his recent work on distributional MP staining) and also how it all works. Remember that the signals are dependent on the MP, and the number of sinks (e.g. volume of the cell and concentration of binding sites)

In bacteria the problem is that most of them have efflux systems for small cationic lipophilic molecules. Ethidium Bromide and RH123 are effected by those pumps, thus do not label pumping cells. Dyes with higher influx kinetics or used at very high extracellular concentrations can overcome the efflux pumps. Propidium Iodide (PI) is so far the most reliable dye exclusion stain, e.g. it does not go in intact (potentially reproductive growing) cells. If you mix a heatfixed and a hapy non-pumping lot of cells and stain them with PI you can counterstain with different concentrations of RH123 to see at what concentrations only the intact cells pick up the RH123 and when both populations become green fluorescent. I got the pictures in my tutorial if you need them

Richard McFarland PhD, MD (mcfarland.richard@pathology.swmed.edu)

Dear Flow Community: Does anyoen know if a antibody to Wiskott-Aldrich Syndrome Protein (WASP) is commercially ( or non-commercially) available?

Thanks Richard McFarland

biocytex@biocytex.com

Richard: You may look at the following paper: Wiskott-Aldrich Syndrome/X-Linked Thrombocytopenia: WASP Gene Mutations, Protein Expression, and Phenotype. Qili Zhu , et al. Blood , Vol 90, No 7, 2680-2689 The authors used rabbit ant-WASP antibodies raised against 3 hydrophobic peptides sequences selected from the WASP sequence reported by Derry JMJ et al. Sincerely Michel Canton.

biocytex@biocytex.com

Richard, I would suggest a look to the following paper : D.M. Stewart et al.,(1996) Studies of the expression of the Wiscott-Aldrich Syndrome protein., J. Clin. Invest., 97:2627-34. WAS can also be investigated using quantitative flow cytometry on platelets. In case you are interested in WASP, platelets and X-linked thrombocytopenia (XLT), an other WAS disorder characterized by small platelet size, I suggest you have a look to an interesting study of XLT presented at the last I.S.T.H. meeting in Washington : N Schlegel, M-F Hurtaud-Roux, O Fenneteau, C Kaplan. Is X-linked thrombocytopenia (XLT) a pure platelet variant of WAS ?

Hope that helps. Philippe Poncelet

Howard Shapiro (hms@shapirolab.com)

Jose Stoute writes:

>Does anyone know the address of a company in the US or Europe where I >can purchase LDS-751? I need address, telephone and fax numbers that are >not toll free. If there is a web site, that too would be helpful. >Any help will be greatly appreciated. >Thank you, >--

The LDS-751 originally used by Loken, Terstappen, et al came from Exciton, in Dayton, Ohio - they have a rudimentary web site, www.exciton.com, which gives their phone numbers and e-mail addresses but does not list products. Fisher Scientific (www.fishersci.com) sells LDS-751 at 10 mg for $68.00; it's in their Acros Organics catalogue. However, I have been told that there are at least two different dyes being sold as LDS-751; one of them works as a vital (but not stoichiometric) nucleic acid stain, as originally reported, and the other is said not to. I've used Exciton dye successfully, but I bought it years ago.

-Howard

Jeffrey M Scott (jeffreys@bcm.tmc.edu)

Fellow Flow-ers I have an investigator looking at apoptosis by PI alone. I have told him about TdT and Annexin V. What he is wondering is how can one know that the sub-G0/G1 peak ( which looks quite good ) is actually apoptosis and if one assumes that it is; How can one know that those events in the sub-G0/G1 peak are not necrotic. I have talked with him at length about this topic and wondered if anyone had any other ideas about how to address his question Jeffrey Scott Baylor College of Medicine Houston Texas

Steve G. Hilliard (steve@habanero.cb.uga.edu)

Jeffrey, At the risk of sounding grumpy, his best approach is to listen to you! The apoptosis experts have said over and over again that it is a mistake to rely upon one assay for apoptosis, and I believe a few recently pointed out the pitfalls of single-color PI on this list. I still suggest it for quick pilot studies, but I'm constantly after my users to back it up with other assays. Occasionally one will surprise me and take my advice.

Steve (a core manager who often feels like his explanations fall upon deaf ears)

Michael Ormerod (Michael_Ormerod@compuserve.com)

When asaying for apooptotic cells, you should always observe the nuclear morphology by light microscopy. The simplest way is to look at the PI stained nuclei using a fluorescent microscope. Alternative stains are Hoechst 33342 or acridine orange. The fragmented nuclei of apoptotic cells are generally quite characteristic

bambam (bmalon01@fiu.edu)

Michael, considering that the cells you are looking at are very small and they show a natural fluorescence in all the wavelengths, have any ideas on how you could squash the natural and just pick the staining?

Thanks

Barb

Michael Ormerod (Michael_Ormerod@compuserve.com)

I am sorry, we seemed to have moved on from the original query. As I understood (misread?), the original query concerned the sub-G1 peak, whether this represented apoptotic cells and whether there could be necrotic cells present.

It is now being suggested that small, highly fluorescent cells are being studied. What are these cells? How do you manage to observe a sub-G1 peak from PI staining by flow cytometry if you are precluded from making visual observation of the fluorescence?

Steve Woodard (steve.woodard@ibb.gatech.edu)

What is the best way to prepare a stock solution of Thiazole Orange ? Has anyone compared preparations from powder to "ready to use" solutions? I know that there's a considerable difference in price

Van Bockstaele, Dirk (Dirk.Van.Bockstaele@uza.uia.ac.be)

Hi Steve, Make your stock solution, using methanol as solvent (1 mg/mL). Working solutions consist of a 1/10,000 dilution of this stock in PBS. Easy, simple and cheap. It gets even cheaper if you use the thiazole orange analogue DEQTC that we described years ago: Cytometry 10:214-216, 1989. It is available as product 36811, fluka catalogue.

Maryalice Stetler-Stevenson (stetler@box-s.nih.gov)

We have an interesting case of a pleural fluid from a patient with a questionable diagnosis of T-cell lymphoblastic lymphoma (relapsed 6 months post treatment). No additional history available. The cells are surface CD3+, CD4+, CD8+, CD7+, CD2+, CD5+, CD38+, CD56+, CD57+, TCR alpha beta +, CD25-, CD16- CD34- and dim CD10. No B-cell antigens. The CD57+ and CD56+ seems strange for lymphoblastic lymphoma. Anyone have experience with these antigens in lymphoblastic lymphoma? Unfortunately the correct history came long after the specimen and we were told a mature T-cell lymphoma so the TDT was not done. We are trying to get more material for a TDT. Anyone seen this before? Maryalice Stetler-Stevenson

Anna Porwit-MacDonald (Anja.Porwit@mb.ks.se)

Hi Maryalice, I have seen CD56 both in T lymphoblastic NHL and in T-ALL. We do not test for CD57 routinely. The most important for differential diagnosis would be TdT. Best wishes Anna

darber@smtplink.Coh.ORG

Maryalice, I am aware of two older series looking at CD57 expression in T lymphoblastic lymphoma. They describe a total of nine cases that are positive and are Kaplan et al. Am J Hematol 1986;22:355, and Shebani et al. Cancer 1987;60:183. I am not sure if they were studied for CD56. We no longer do CD57 routinely on acute leukemias, but our recent experience is that 6% of our ALLs are CD56 positive.Dan Arber

Paul Kuon (pmk3351@usl.edu)

How does one pursue an acute leukemia using three color analysis? How do you know where to look for the blast population, if the blasts are < 10% of the total? Do you use a CD45 Vs LSS histogram setup? We are new at this particular aspect and would appreciate any help available. Also any specific panels that are run for ALL ,AML,AMML etc. Thanks in advance.

Reply to: pmk3351@usl.edu or the group.

Anna Porwit-MacDonald (Anja.Porwit@mb.ks.se)

Hi, We are routinely using three color analysis and have a leukemia panel that is modified depending on patient age - for children more ALL oriented and for adults more AML oriented. You can find specific panels that we use at http://www.ki.se/onkpat/res_area/patologi/sff/leukly.html At diagnosis we gate in blast population either on FSC/SSC or find it using SSC/CD45 in triple 2 or 3.

Sparks, Scottie (SSparks@unch.unc.edu)

We are interested in performig HLA-B27 testing by Flow. We are evaluating BD's B27 Kit and it seems to work well. Our questions are concerning controls: How often are positive controls run by those of you perfoming this test clinically? Do you have a way to preserve positive control blood that has worked for you?

Also, what have you learned about cross-reactions (like B7)?

Thank you in advance for your help! Scottie

Snider Denis (sniderd@fhs.csu.McMaster.CA)

Scottie, We run the B-D HLA-B27 test routinely now as a screen test. We do approximately 200 tests per month, as a regional test center. In the initial stages we ran 2-300 samples with parallel serologic histocompatibility testing. This established a "gray-zone" surrounding the B-D positive/negative cutoff, in which we found almost all of the B7 and other cross-reactions, as well as the very rare (<< 1%) false negatives. We now only send about 5% (gray-zone) of our samples for serologic/molecular typing. A comparison to the Lambda One antibody in the "gray-zone" was also done. This antibody performed well in ruling out B7 cross-reactions. However, it was not cost effective to add it to the screening protocol.We run occasional positive controls (1 or twice a month). They are always positive, because they are selected from known patients in the histocompatibility lab.

Hope this helps.

Denis

Denis Snider Ph.D., F.C.A.C.B. Associate Professor

CATHARINE FRITSCHI (CSFRITSC@SENTARA.com)

We evaluated BD's package a few years ago. Caution: We ran several double dose crossreactive patients (both HLA-B epitopes were B7-CREG antigens) and found that the system incorrectly typed our patients as B27.

Te Chih Liu (liutc@pacific.net.sg)

>Our questions are concerning controls: How often are positive controls run >by those of you perfoming this test clinically? Do you have a way to >preserve positive control blood that has worked for you? >

>Also, what have you learned about cross-reactions (like B7)?

We've been running it in parallel with our traditional serology tests for a short time - approx 100 samples. Works well. Just had a single B7 cross-reaction which the Ab/software resolved clearly.

We don't get that many positive B27s here in Singapore and we'll not be able to put in a positive patient control with our test runs. So far we've depended on BD's supplied positive control. Te-Chih

Sebastian, J, Gelderbloem, Mr (sjg@samiot.uct.ac.za)

Daer Flowers I have recently started using flow for test for PNH. We are using CD55 and CD59 and looking at granulocytes and erythrocytes. I would like to hear your experiences with this test. I also have a problem identifying the different types of PNH. At the moment we are only set up to identify PNH abnormalities. Please help.

Yours sincerely Sebastian Gelderbloem

Medicine (lipsel01@doc.mssm.edu)

I would be interested in receiving your responses. I am following a patient now with ?PNH but with equivocal results on flow for CD59. What is the sensitivity/specificity of the finding?

Jean-Pierre Delville (jdelvill@ulb.ac.be)

Dear Lewis and Sebastian and people involved in this discussion

You can also try to find the others GPI linked molecules defect on white blood cells. For example using a double staining with CD15 and CD16 on neutrophils, CD64 and CD14 on monocytes. CD15 and CD64 molecules are not GPI linked and so, can be used to gate respectively the PMN and monocytes cell populations. It is now very easy to see if the CD16 and CD14 are or not, expressed in cells of your PNH patient. Take care that some monoclonal antibodies against the CD16 molecule, recognise a molecule not GPI linked. This is a very simple alternative to CD55 and CD59 quantification on red blood cells, for example, following a red blood cells transfusion. Success Jean-Pierre Delville

Nebe, Thomas C. (thomas.nebe@ikc.ma.uni-heidelberg.de)

Dear Collegues, from our experience I want to comment on the ongoing discussion on PNH: The patients in question either present with a hemolytic anemia (primary PNH) or secondary to aplastic anemia (PNH syndrome). The non flow laboratory data should support one of these conditions (LDH, bili, urobili, haptoglobin, retics etc). Only about 25% of the primary PNH show nocturnal hemoglobinuria (PNH = paroxysmal nocturnal hemoglobinuria). Esp. young patients may show spontaneous remission (about 10%?). Most of the patients (esp. those who are not suspect) show a mosaic of normal and affected cells. In aplastic anemia the proportion may be very small in the beginning (5% or less).

In preparation of a "consensus" protocol (what ever that may mean) I discussed with Dr. Schrezenmeier (who has seen the most patients in Germany) the following: As the acquired defect may be differently affect the blood cell lineages, red cells, platelets and leukocytes have to be checked separately. Reticulocytes are of special interest as patients might have received blood transfusions. CD55 and CD59 with phycoerythrin (orange fluorescence) versus thiazol orange (green fl.) is a good combination. Each lineage should be checked by two different CDs. Immunological gating is preferred.

The problem of false negative cells is a major issue: CD14 and esp. CD16 are maturation dependent and not recommended. Scatter gates of these cells produce negative cells as they include lymphs in the mono gate or eosinophils, myelocytes or less mature granulocytes in the neutrophil gate (CD16 negative). In aplastic patients and those with infections this represents a significant proportion. CD16 on neutrophils in addition undergoes alternative RNA splicing which ends up in a polymorphism and you need a monoclonal antibody that recognizes a conserved epitope (eg. clone 3F8, CALTAG). CD16 is monomorphic on NK cells. CD24 and CD48 are better in that respect. CD59 versus side scatter allows to differentiate among the major leukocyte subsets. Non red cells in the erythrocyte scatter gate (platelets, debris etc.) and red and debris in the platelet gate are significant and require immunological gating. The testing of CD34 stem cells (for autologous transplantation) with CD59 (expressed on all maturation stages) requires some experience.

Diagnosis should be made in the peripheral blood as bone marrow esp. shows the problems mentioned above. Monitoring every 6 month (?) is meaningful as the proportion of affected cells varies over time.

Strict quality control (including always a normal control sample) and education are important because of the severity of the diagnosis. Each laboratory needs to set up its own threshold value of GPI negative cells.

I want to stop here as the message might has got already too long. I think you will have got a feeling about the laboratory diagnosis of PNH. Therefore, everybody needs to decide on his own whether he is able to perform such a test esp. when the frequency is low.

More details are published in: Nebe CT, Paroxysmal Nocturnal Hemoglobinuria: Clinical Aspects and Flow Cytometric Analysis, in: Aspects of the Flow-Cytometric Analysis of Red Blood Cells, K. Gutensohn, H.-H. Sonneborn and P. Kühnl (eds.), 81-94, Clin. Lab. Publications (1997)

Regards Thomas Nebe

Dr.med. C. Thomas Nebe

Voorn, John (J_Voorn@CLB.nl)

Dear Sebastian,

I will first briefly explain the situation within our institute to clarify why we perform the PNH test as described.

CLB (Central Laboratory of the Dutch Red Cross Blood transfusion services), situated in Amsterdam, is with more than 800 co-workers active in the following fields:

1. Preparation of products: therapeutic products derived from blood and production of blood grouping and immune reagents (CD's, cytokines, a.o.).

2. Specialised laboratory services for Dutch hospitals, including: . blood typing; . transfusion reactions due to the administration of blood and blood products; . identification of abnormalities in blood cells and plasma proteins.

3. Applied and fundamental research in the following areas: . homeostasis and thrombosis; . transfusion medicine and haematology; . inflammation and sepsis; . clinical and experimental immunology; . product development.

4. Academic and postgraduate teaching of students in medicine, biochemistry and biology. 5. Reference tasks include: "WHO International Laboratory for Biological Standards with special reference to human blood products and blood derivatives" and the "European Bank of Frozen Blood of Rare Groups".

You can imagine that this combination of fields for PNH has given us the opportunity to develop a very accurate flow cytometry assay. This is published by Ellen van de Schoot at al. in Blood, 1990, 1853-1859 : Deficiency of PI-linked membrane glycoproteins of leukocytes in PNH, Description of a new diagnostic cytofluorometric assay.

In this study it is shown that the expression of PI markers is most reliable measured at granulocytes and there fore we use at present a panel of two colour antibody combinations. We use CD66e-PE as granulocyte-identification marker and CD16, CD24, CD66b and CD59 as PI markers next to a negative control and two leukaemia related controls: CD15 and CD11b (all FITC). Of course all antibodies are in-house made, well characterised in the international workshops and commercially available. Some times we had a remark that so many antibodies makes the PNH test very expensive, but if you realise that all antibodies can be diluted 1:20 and thus you can perform 2000 test with one vial which makes the cost almost negligible.

For information on our products and distributors you can check our web site, for clinical information you can contact Dr. Ellen van der Schoot directly at CLB tel +31-20-5123377. Or you can (as many hospitals do) send your patient sample to CLB-diagnostic services.

John Voorn CLB Reagents

S. Komura, M.D. (skomura@pol.net)

Two questions to toss out to the group:

1. Looking for suggestions on how to handle platelet adhesion to leukemic blasts when performing immunophenotyping using CD41 and CD42b on whole blood/bone marrow aspirates.

2. When doing surface kappa/lambda analysis on whole blood/bone marrow aspirates, we wash the cells prior to staining with a three color combination (CD19-Cy5/lambda-PE/kappa-FITC or CD20-Cy5/lambda-PE/kappa-FITC) to remove plasma and prevent false negative results. On some cases, we have noticed a much lower percentage of CD19 positive cells compared to results obtained on our initial screen, which is a stain/lyse/wash procedure using CD19-PE and CD20-PE or CD20-Cy5. Anyone have any ideas why?

Two questions to toss out to the group:

1. Looking for suggestions on how to handle platelet adhesion to leukemic blasts when performing immunophenotyping using CD41 and CD42b on whole blood/bone marrow aspirates.

2. When doing surface kappa/lambda analysis on whole blood/bone marrow aspirates, we wash the cells prior to staining with a three color combination (CD19-Cy5/lambda-PE/kappa-FITC or CD20-Cy5/lambda-PE/kappa-FITC) to remove plasma and prevent false negative results. On some cases, we have noticed a much lower percentage of CD19 positive cells compared to results obtained on our initial screen, which is a stain/lyse/wash procedure using CD19-PE and CD20-PE or CD20-Cy5. Anyone have any ideas why?

Anna Porwit-MacDonald (Anja.Porwit@mb.ks.se)

Hi, 1.Platelet adhesion is very commeon, especially if the sample is processed few hours after it is taken. If we get a very high positivity with CD61 or CD41 we perform cytoplasmic staining in BM slides or cytospins with APAAP method to see if the staining is specific. AML M7 are rare and I think that diagnosis should be confirm by this kind of staining.

2. We use similar procedure but we haven,t noticed such phenomenon. The only explanation I can figure out is that you loose cells while washing. Did you check other subpopulations (T cells?). Do they remain stable?

Best wishes Anna

At 18:57 1999-08-19 -1000, you wrote

craig.turner@nbs.nhs.uk

Dercksen et al in Blood suggest flow cytometery in the presence of EDTA which reduces the effect of P-selectin and integrins. I've found that total removal from large volumes of PBMCs is rare, but is definitly aided by washing with anti-coagulant (EDTA/acidfied citrate).

robert gniadecki (rgniadecki@hotmail.com)

Hello everybody! I am currently treating a patient with total skin inflammation (erythroderma). We suspected Sezary's syndrome, so we performed flow. We found the following phenotype: CD3-CD4+CD8-CD30+CD45RO+ negative for CD25,26,27,28,29,CD45RA Peripheral blood is normal which excludes the anaplastic large cell lymphoma/leukemia. Skin biopsy did not show any leukemic/lymphoma cells, though few, normal CD30+ lymphocytes were present. What is the diagnosis? Has anybody seen such a phenotype before?? Any input will be appreciated.

Regards, Robert Gniadecki, MD, PhD, DMSci Dermatology, University of Copenhagen Denmark

Bgreig2@AOL.COM

It sounds like a Ki-1 (CD30) lymphoma to me and these usually have a large (even extra large) cell display on light scatter that is quite from normal lymphocytes. I'm curious about the "normal" CD30 you are seeing on the lymphocytes. In our experience CD30 is rarely found on normal lymphs and when seen in Ki-1 lymphoma it is usually bright i.e. the 3rd decade of the log amp. Is this what you are seeing?

Hope this helps.

Regards, Bruce Greig Flow Cytometry- Immunopathology Vanderbilt Univ. Medical Center Nashville, TN.

Nebe, Thomas C. (thomas.nebe@ikc.ma.uni-heidelberg.de)

Ress, S, Stan, Dr (sress@uctgsh1.uct.ac.za)

Hi all, Does anyone have normal range values for gamma-delta subset from birth to adulthood?

I checked ISAC Leukocyte ref data on their site: reference is made to Clin Immunol + immunopath 1994 70, 152-8, but from the abstract, I was unclear if GD+ subset was included.

Any help appreciated.

Thanks,

Stan

Stanley Ress

Maryalice Stetler-Stevenson (stetler@box-s.nih.gov)

The normal ranges for gamma delta T-cells differes based upon racial background and the environment one was exposed to as a child (i.e. native country). White North Americans of Northern European descent have less than 3% of lymphs gamma delta T-cells. Black North American have higher values, as do Americans of Italian descent (I don't know why). South Americans (even of Northern European descent) and Asians have much higher numbers (up to 20% of lymphs). You will need to determine your own normal ranges.

Maryalice

Karel Drbal (drbal@leuko.biomed.cas.cz)

Tom Mc Closkey (thomasm@nshs.edu)

Can anyone clarify for me the definition of memory and naive lymphocyte populations as detected by CD45RA and CD62L? Our system is lysed whole blood labeling using CD45RA and CD62L in combination with CD4 and CD8 for evaluation of memory and naive cells in HIV infected children.

1] Are all RA+ 62L+ cells naive cells?

2] What are RA+ 62L- cells? I have seen these referred to in the literature as memory cells and also as naive or activated naive [as opposed to RA+ 62L+ which were called resting naive]. Have these cells seen antigen?

3] What are RA- 62L+ cells? My understanding was that 62L was lost prior to the RA to RO conversion. Why would cells which no longer express RA still express 62L?

4] Do memory cells eventually become RA- 62L-? ie the most mature cells are the double negatives?

5] As CD4s and 8s encounter antigen is the sequence of changes in expression of Ra and 62L the same for both subsets?

Thanks for any info.Regards, Tom

Mario Roederer (Roederer@drmr.com)

Hi Tom--we have worked extensively on this, building on the pioneering work by Louis Picker and others. We have a couple of publications in print that will help, one appearing in November in International Immunology (Mitra et al.), and one that is in preparation, all on this topic.

Before I give some answers to your question, I'd like to mount my high horse on the topic of RA/RO again (it's an annual peeving, as long readers will note). There is no information that can be gained by using both CD45RA AND CD45RO in the same stain (on peripheral blood samples). RA and RO stain (nearly) completely inversely. Choose one or the other, and combine it with stains that provide immense additional information, like CD62L, CD11a, or CD27. Among these, CD62L is the costain of choice for fresh cells; CD27 for frozen cells. (They all identify different memory subsets, but are roughly equally good for indentifying naive T cells. The phenotype of naive T cells is (in terms of commonly-used and useful markers): CD45RA+ CD45RO- CD62L+ CD11a-dull CD11b- CD27+ CD28+ CD57- CDw60- CD95-. Note that naive T cells express somewhat LESS CD28 than do memory T cells, and, interestingly, they express very low (but detectable) levels of CD38--they are negative for other so-called activation markers such as CD25, HLA-DR, CD69, CD71...

Here are some quick & dirty answers:

(1) About 95-98% of RA+62L+ cells are naive. There is no indication that naive cells are found in other phenotypes.

(2) RA+62L- cells are definitely memory T cells. That are neither naive nor activated naive. These cells have seen antigen. We find these to be the most restricted (in terms of repertoire), and to contain what appear to be terminally-differentiated effector cells.

(3) RA-62L+ are also memory cells. Among CD4, they contain the principal population of IL-4 producing cells. We find them to be closest to naive in many functional aspects. Among these cells, the CD11a-dull cells are the most Th2-like of all CD4 subsets (clinically, this latter subset is elevated in PBMC of people with Th2-diseases such as Atopy or Lepromatous leprosy, and diminished in Th1-disease like tuberculoid leprosy).

The expression of CD62L is very complex. It is immediately (within minutes) clipped off by a surface protease after activation by PKC-activated signals (and, incidentally, during all freeze-thaw procedures I have ever tried!). Within 24-48 hours, it is then expressed at higher levels than resting. Over the next several days, as the cells return to "resting", it may or may not be expressed depending on which subset the resting cells eventually become.

(4) RA-62L- cells are also memory cells. Among these, the CD11a-bright cells make the most IFN-g and can be considered the most Th1-like of all subsets (and its representation is opposite, clinically, from the RA-62L+CD11a-dull cells: these are elevated in Th1 diseases and diminished in Th2 diseases).

(5) We have no real data to understand the differentiation progression of these subsets. For example, we don't know if (i) cells can interconvert between these subsets; (ii) there is an ordered progression from one to the next; (iii) if these subsets are related at all by differentiation. So I can't answer the question directly. However, it is interesting to note that there is a high degree of parallelism between the functions of any given phenotypic subset of CD4 and CD8 cells. For example, the only IL4-producing cells within CD8 are also RA-62L+... and terminally differentiated CTL can be found in the RA+62L- subset.

For our references, see:

Watanabe, N. Blood 1997 90:3662-3672

Roederer, M. Cytometry 1995 21:187-196

Rabin, R. L. J Clin Invest 1995 95:2054-2060

Roederer, M. J Clin Invest 1995 95:2061-2066

Roederer, M. Semin Immunol 1997 9:389-396

In addition, see the excellent work by L. Picker's group and citations to related work in these references.

darber@smtplink.Coh.ORG

Although there are some reports of CD117 positive ALL, we have used it on several hundred acute leukemias and found it to be very specific for myeloid lineage.

Dan Arber City of Hope

Frederic Preffer (preffer@helix.mgh.harvard.edu)

Dear Jorge When it is present I consider CD117 very helpful in discriminating 'early' monomyeloid linneage cells from lymphoid cells. Although literature i have seen suggests its presence on early lymphoid cells too, it has not been my experience to see it there. In terms of your specific question, I am unaware of any 'definitive' [what I would consider <underline>clonal</underline>] marker of acute myeloid malignancy. CD117 is 'normally' expressed.

Maryalice Stetler-Stevenson (stetler@box-s.nih.gov)

I have a dumb question. I always thought that NK cells were CD4-, CD5- and CD25-. Now I am presented with a "NK-cell" leukemia that is CD5+, CD2+, CD3-, CD4+, CD8-, CD57+, CD56-, CD16- and (here's the kicker) TRCab+. This smells like a CD3- T-cell tumor to me. Any thoughts? Anyone seen this before?

Maryalice Maryalice Stetler-Stevenson Director Flow Cytometry Unit Laboratory of Pathology, NCI, NIH

I have a dumb question. I always thought that NK cells were CD4-, CD5- and CD25-. Now I am presented with a "NK-cell" leukemia that is CD5+, CD2+, CD3-, CD4+, CD8-, CD57+, CD56-, CD16- and (here's the kicker) TRC<fontfamily><param>Symbol</param>ab</fontfamily>+. This smells like a CD3- T-cell tumor to me. Any thoughts? Anyone seen this before?Maryalice

Maryalice Stetler-Stevenson (stetler@box-s.nih.gov)

I sent the first message too soon. The "NK-cell" leukemia is CD5+, CD2+, CD3-, CD4+, CD8-, CD57+, CD56- and CD16- but we don't know TRCab. The patient has had some sort of lymphoproliferative process for a long time. This smells like a CD3- T-cell tumor to me. NK cell neoplasms are usually CD56+ and aggressive. Any thoughts? Anyone seen this before?

Maryalice Maryalice Stetler-Stevenson Director Flow Cytometry Unit Laboratory of Pathology, NCI, NIH

I sent the first message too soon. The "NK-cell" leukemia is CD5+, CD2+, CD3-, CD4+, CD8-, CD57+, CD56- and CD16- but we don't know TRC<fontfamily><param>Symbol</param>ab</fontfamily>. The patient has had some sort of lymphoproliferative process for a long time. This smells like a CD3- T-cell tumor to me. NK cell neoplasms are usually CD56+ and aggressive. Any thoughts? Anyone seen this before?

Maryalice

Maryalice Stetler-Stevenson

Director Flow Cytometry Unit

woodbl (woodbl@u.washington.edu)

The expression of CD4 and CD5 would be extremely unusual for an NK cell process, and I agree are most consistent with a T-cell neoplasm. We certainly have seen CD3- T cell lymphoproliferative disorders that have unusual immunophenotypes like you are reporting. Demonstration of clonality by T-cell PCR of either gamma or beta chain would help confirm the T cell nature of the process.

Frederic Preffer (preffer@helix.mgh.harvard.edu)

hi maryalice

the fact that the TCR is negative takes some of the puzzle away. i would suggest checking for cytoplasmic CD3 if you have not do so already.Based on your existing data your case seems more like a T cell tumor and not an NK cell tumor.

Te Chih Liu (liutc@pacific.net.sg)

This is a question to the clinical users doing immuno-phenotyping in the group.

Has anyone got a comment to make on the differing phenotypes of mononuclear cells that are sometimes seen when the same sample is prepared using a lysed whole blood approach as opposed to PBMC obtained after a ficoll separation.

What about marrow samples?

I'm presently looking at a sample from a 6-year old child with CD10+ ALL. On the ficoll preparation, the plots are that of a straight forward common ALL. On the LWB preparation, there is a distinct CD5+/CD20+ population (15% of MNC), separate from the CD5-/CD20+ B-cells/blasts and the residual CD5+/CD20- T-cells. I can't think of any reason why that population should be there at all.

Thanks. Te-Chih Liu

Ronald Rabin (RRABIN@niaid.nih.gov)

This reported at a meeting of, I believe, the Clinical Immunology Society. Ficoll Hypaque decreased B cell percentage to a mean of 35%, but there can be selective depletion of T and NK populations as well. The fact that you see CD5+ B cells may simply reflect the predominance of this type of B cell in children.

The reference is Brown, M.R., Cleary, J. M.,and Fleisher, T.A. Selective lyphocyte depletion by ficoll hypaque density gradient centrifugation. I am not sure what meeting this was, or whether this was ever published in a peer reviewed journal.

DARZYNKIEWICZ ZBIGNIEW (DARZYNK@nymc.edu)

Chapple, Peter (peterc@petermac.unimelb.edu.au)

Dear Flow-ers,

I have a bunch of FCS files (MAC-BD generated) which I would like to hand over to a local college for educational purposes. The problem I have is that each file has the patient's name and URN attached to it - presumably somewhere in the header.

Simon Monard (smonard@trudeauinstitute.org)

Hi, You may be able to alter the information in the header using Ray Hicks FACS Assistant program, you will have to edit then one by one though, If the field you wish to edit is the patient id then this may not work I cant remember if he included this field. If that cannot edit the patient id field or you dont want to pay the scandalous $20 price tag then you could use a program called Ledit, its a simple text editor available free from www.sophists.com. Be very careful with this program, you must not change the number of characters in the header or it all turns to rubbish, so you could change the name from Smith to ***** or something. The safe thing to do is to make a copy of original files. When saving dont use the "save" or "save as" command but click in the little box to make the window disappear then save when prompted. It is a rather tedious job.

Simon

David Novo (dnovo@ucla.edu)

DARZYNKIEWICZ ZBIGNIEW (DARZYNK@nymc.edu)

Differential staining of RNA and DNA with AO is simple but requires stringent conditions, in particular correct AO concentration in the final staining solution. Dr. KuKuruga correctly emphasizes its advantages and possible problems. The problem with contamination of the instrument is of the same magnitude as using other strongly fluorescing fluorochromes eg. rhodamine 123. I am guessing that the problem of contamination of the instruments by AO was overblown by some companies that sell reagents. The lifetime supply of AO costs about $ 30.- , Considering that AO can substitute for some monoclonal antibodies e.g in discriminating G0 cells, analyzing lymphocyte activation, etc., it is little wonder that the companies selling these antibodies try to discourage the use of AO. We use AO continuously since 1974, in parallel with with many other probes, with different flow cytometers, recently FACScan. A rinse with bleach between the samples, as Dr. KuKuruga points out, is sufficient to remove any contaminating fluorescence. The protocols on use of AO are in Vol 41 of the Methods in Cell Biology (Academic Press, 1994) as well as in Current Protocols in Cytometry. Zbigniew Darzynkiewicz

Mark A. KuKuruga (kukuru@umich.edu)

Jeff, I think that the flow cytometry community in general has overemphasized the difficulties arising from the use of AO. Much (if not all) of this is unwarranted. In my opinion, AO is extremely useful for detecting perturbations to the cell cycle, perhaps second to BrdU. It's perhaps the best way to detect the transition from G0 to G1. Its advantage is that it is probably much easier to develop. True, AO can be problematic, but with proper timing, cell and dye concentrations, fixation, and buffers, AO staining and analysis can be very straightforward. Users also express concern over cytometer contamination with residual AO . . . we have no problem with this. Simply be diligent about flushing your sample system with 10% bleach after AO. I've found that with this, AO cross-contamination is a non-issue. If you're interested, I'll try to outline our AO staining protocol, though I'm sure there's plenty in the literature to direct you. MAK.

Mark A. KuKuruga (kukuru@umich.edu)

Thomas,What are you taking about? What fo you mean by "measure?" Do you dispute the relative change that occurs in a quiescent population upon stimulation to proliferate, as detected by AO? Have you looked at the supporting literature that validates this phenomenon? Have you done these experiments, and observed these changes first hand? I think it's time you showed us some "independent assays and data" to support your opinion that AO does not do what that extensive body of literature suggests. MAK.

Thomas Delohery (t-delohery@ski.mskcc.org).

On the contrary, the lack of data from assays independent of flow cytometry in the "extensive body of literature" is one of the problems I was attempting to point out. But even if AO can quantitatively differentiate RNA from DNA in intact cells how is this a reliable measure of G0? The AO assay relies on G0 cells having low RNA content relative to G1 cells. What about fully differentiated cells? For example, a terminally differentiated epithelial cell in a secretory gland, let's say a mammary gland of an actively lactating new mother. These cells have large quantities of mRNA and rRNA for producing proteins unrelated to G1 entry or DNA synthesis. Smaller, cycling cells in the basal layer could have less total RNA than the differentiated secretory cells.

The bigger problem is defining G0 cells. It's like Time; everybody knows what it is but can't define it. Consequently, people use the term G0 to describe any cells not in log-phase growth. The current habit of referring to "resting" lymphocytes, serum-starved cell lines and terminally differentiated cells in-vivo as being in G0 implies these cells are in the same, defined state of growth and growth potential. We need definitions based on molecular markers known to be involved in cell cycle regulation.

td

Dixie Polakoff (dpolakoff@pdl.com)

One of my investigators is looking for a marker for human eosinophils. She is presently using CD49D combined with CLA. Thanks, Dixie Polakoff

Dr. Martin R. Hadam (Hadam.Martin@mh-hannover.de)

Depending on your needs, you may use CD52, CD48 or CD81. All will differentially stain eosinophils compared to neutrophils.

DARZYNKIEWICZ ZBIGNIEW (DARZYNK@nymc.edu)

We have a problem with Cy5 binding (apparently) non-specifically to lymphocytes. Has anyone else seen this and are there any suggestions as to how to avoid this?

Thanks, Andrew

Voorn, John (J_Voorn@CLB.nl)

Eric Miller (e.miller@icrf.icnet.uk)

Here is my summary of opinions from round the cytometry group: thanks folks! PS. these are only opinions I recieved: my own views are not included!

Approximately 70% of responders favoured FACScaliber. The rest preferred Coulter, apart from one responder who used a MoFlo and liked it very much.

FACS pros:

Negative population is shown as a discreet population More 4-colour reagents are available Machine and computer are faster in use. Less of a learning curve in upgrading from FACscan Easier to use Good technical backup Upgradable to sorting and dual laser

FACS cons:

More expensive

Coulter XL pros:

Well developed data acquisition system Superior fluidics(sample acquisition can be halted without wasting sample) Gating can be altered during acquisition Good technical and service support Good interface for clinical samples Cheaper than FACS

Coulter cons:

No upgrade to sorting Not as good with dim fluorescence Negative population not displayed -leading to difficulty in setting compensation

MoFlo pros:

Speed of sorting (c. 20-40,000 cells/sec) Simple design Easy adjustments

Just to comment that you can indeed set up the XL to see the unlabelled cells. I usually set them to occupy the entire first decade. They have a broader distribution as you work in the noise range whilst I understand that BD uses an offset which leads to a smaller distribution in the first decade. Having worked on BD and Coulter instruments I would not say that they are more or less difficult to compensate, neither that one or the other is easier to use or more or less sensitive to fluorescence. It depends more on what you are used to work with. I contribute to the regular confusion by plotting forward scatter on the Y-axis and sidescatter on the X-axis and both in log - which scares the hell out of most users of either type of instrument as it is unfamiliar to them.

To be even more provocative I thought you might want to have a look at the DAKO/Partec hybrid that might feel even more unfamiliar but is said to offer full automatisation including the addition of antibodies to the sample (even if they are from a different manufacturer). I'm looking forward to the first feedback on that instrument.

Happy choice

Collette Hillier (z2178724@student.unsw.edu.au)

We have used thiazol orange to stain our parasitised red blood cells and have been unable to separate the ring stage parasite from the trophozoite stage. Has anyone had success with this?

Collette Hillier University of NSW

Mark A. KuKuruga (kukuru@umich.edu)

Thomas, These types of changes may be due to either turbulence in the flow cell, of a restriction in the sheath flow. Check the alignment of the center stream, or for a partial blockage in the fluidics PAST the flow cell. MAK.

Thomas Misgeld wrote:

> Hi there! > > Hope you don't mind a novice's question. > Using a FACSort in a simple single-staining experiment of human myoblasts I > observed that the cell population in the FSC/FL3 blot would "move" during > the course of a measurement towards higher FSC values only to stabilze after > about 10 secs. of measurement (ca. 100 events per s). In parallel the FL1 > (cells stained w. FITC-labelled sec. abx) histogramm would shift to higher > values also. > Could anyone explain why this happens? I was suggested osmotic swelling of > cells due to an imbalance between the PBS the cells are suspended in and the > PBS in the source tank. But this would not explein the FL1-shift, would it? > > Thanks in advance, > > Thomas

Moss, Delynn M. (dmm3@CDC.GOV)

To all CellQuest users: We have purchased BD's FACStation for our FACScan. The FACStation includes a MacIntosh computer with OS 8.1 and CellQuest software. So far the system seems to be working great, but currently we do not have the capability to make 35 mm slides of plots created by CellQuest. Our slide making equipment is PC based. What is the preferred software to put on the Mac so that CellQuest files can be PC converted. Would something like PowerPoint or Corel Draw be sufficient for the CellQuest files and a third party software to do the conversion? Advice welcomed and appreciated.Thanks Delynn Moss

Antony Bakke (bakkea@ohsu.edu)

From your question, I believe that you mainly want to make slides on the PC and not analyze the flow data on the PC. If that is the case, you can cut and paste Cellquest histograms into Powerpoint on the Mac and then save that file on a PC formated disk. Take the disk to a PC and load the file into Powerpoint on the PC. These files can then be made into slides. Best regards, Tony Bakke

Mika Korkeam{ki TUY (mkorkeam@ra.abo.fi)

Hi, I think that it is wisest to down load WinMDI and analyze those files you want to make slides of with your PC and use copy/paste for your slide making program. WinMDI can read CellQuest list mode files and a Mac can handle PC diskettes and ZIP disks (if it has a PC reader). Another approach is that you could buy Microsoft Office 98 to your Mac, it is compatible with PC office packs (...->97). Then you could use copy/paste for your slide making program.

Mika University of Turku Finland

Stephen (FYHOU@DELPHI.COM)

It is not that easy to preserve the color of the dots after conversion. We have tried the following approach, which works great to transfer Paint-A -Gate color dot plots to PowerPoint presentation on PC. On MAC: paint the dots in Paint-A-Gate copy to clipboard paste to MacDraw Pro save as PICT2 file, either on PC-disk or on network On PC open PowerPoint choose blank slide using menu "insert picture from file" to the PICT2 file make sure you indicate the file as all formats *.* because there is not extension associated with the PICT2 file should see the same color dot plot now

Elaine Kunze (mek4@psu.edu)

Gerhard Nebe-von-Caron (Gerhard.Nebe-von-Caron@Unilever.com)

Hi Elaine Apart from removing the milk, I assume both dyes should have a considerable overlap into the orange FL2 channel, so you should be fine in using that. Alternatively you can add an independent red marker onto the cells to make them red and green. I don't know if a red fluorescent cd45 would work in milk. Regards Gerhard

XL dual trigger

Elaine Kunze (mek4@psu.edu)

Barsky, Lora (%20Success"LBarsky@chla.usc.edu)

Hello to all, With the advice from Juan Gea-Banacloche, NIAID, I was able to download the correct drivers to get our iMac to run CellQuest. So if you're interested, here's what I did on an iMac running MAC OS 8.5.1:

1. Visit http://www.griffintechnology.com/imac/imate.html. Download the latest iMac firmware upgrade and the latest iMate driver.

2. Remove or disable old iMate drivers. If you have pre-1.0.x versions of iMate Enabler and iMate USB Module, trash them. The new iMate driver (v.1.2) replaces both with a single extension.

3. Install the iMac upgrade onto your hard-drive. This upgrade will give you MAC OS ROM v. 1.2.1. Simply install into your System Folder.

4. Shut down the iMac. I have the iMac keyboard connected into the USB #1 port. I have the iMate connected into the USB#2 port with the Cellquest dongle connected. (I suspect Modfit dongle will work as well, although I haven't tried this yet.)

5. Restart the iMac. Watch the extensions load. You should see an extension with the ADB symbol. If it isn't covered by a "red-X", Cell-Quest will work.

6. If you see the ADB symbol covered by a "red-X", try opening the iMate Troubleshooting file, and follow the suggestions. This worked for me. Just be sure to shut down and then re-boot the computer. (I had some trouble at first, maybe restarting from Special Menu didn't take the first time.)

Good Luck to all interested parties and Thank You for the web site!

Lora Barsky Children's Hospital Los Angeles Research Immunology/BMT 4650 Sunset Blvd. Mail Stop #62 Los Angeles, CA 90027

Ulrik Sprogøe-Jakobsen (usj@dadlnet.dk)

Alice L. Givan (Alice.L.Givan@dartmouth.edu)

I believe the blue and white G3s have an ADB port in the back. You can hang the security dongle on that port, with nothing connected to its distal end. This will allow the software to run. Alice Alice L. Givan Englert Cell Analysis Laboratory of the Norris Cotton Cancer Center Dartmouth Medical School Lebanon, New Hampshire NH 03756 tel 603-650-7661

Adrian Smith (A.Smith@centenary.usyd.edu.AU).

Look again, the Blue and White G3s DO have an ADB port!!! We have CellQuest running just fine on one here. Apparently there is a rumour going around that CellQuest won't run on a B&W G3 - it is not true (BD Australia was very peeved about it). As for an iMac that is a different story but you may be able to get it to run with one of the USB-ADB convertors - in particular the ones that have been designed especially for copy-protection dongles (I have heard reports about these but I have no idea if they work or even if they are actually shipping). Try the product page at http://www.macintouch.com/imac.html for links... Having never even used an iMac I can't really say anymore...

Adam Treister (adam@treestar.com)

The new G3 has an ADB jack in the back (in addition to the two USB jacks). I bought a 350 MHz G3 this weekend and set it up today. After getting quite upset that the cost of having this very fast machine was that I was going to have to live with the cute-but-not-very-functional keyboard and mouse, I just discovered the extra ADB jack and I'm now back to my big gray keyboard that doesn't put the arrow keys directly under the shift key (where I often hit them and inadvertantly moved the cursor during typing). Presumably, one could hang the CellQuest key back there.

Juan Gea-Banacloche (JGBANACLOC@niaid.nih.gov)

We tried to use CellQuest on the iMac using the iMate from Griffin Technologies (an ADB to USB adapter: http://www.nashville.net/~griffin/imac/imate_ds.html) and it did NOT work, despite downloading the (then) latest drivers and iMac firmware. This was more than 2 months ago, things may have changed.

Koratich, Mike (koratich@sri.org)

Hi all, Does anyone have a good protocol for doing phagocytic activity of peritoneal macrophages using FITC beads by flow. We are currently harvesting the macrophages by lavage and adherence to 6-well TC plates for 2 hours in DMEM w/o serum. We then wash and add DMEM w/ 10% FBS at let them incubate ON. We then wash and add 1E8 beads, which is roughly a 100:1 bead to cell ratio. This is incubated for 90 mins. and then washed to remove most of the free beads. This all works fairly well and the cells look good under the scope. Our problem is how to get the cells to detach so we can run them on the flow without causing disruption of the membrane and loss of beads. We are currently using 0.4% EDTA with poor results. We have found a protocol that uses dispase and will try that next. Any ideas on other things to try would be appreciated. Further, any suggestions on the method in general are also welcome.

TIA,

Mike Koratich Cell Biology and Immunology Group Serquest Southern Research Institute

Jörg Hildmann (hildh000@polly.zdv.uni-mainz.de)

Hallo Mike, I worked on phagocytosis of macrophages, too. I used labelled 1µm FITC beads etc. or unlabelled beads up to 10µm. The way to obtain cells with high viability (>95%) was to use culture plates without!!! TC for adhearence (works fine) and to remove them with ice cold 1% EDTA/PBS (incubation: 10' at 4°C).

Good luck for your work Joerg

John Waitumbi (waitumbi@net2000ke.com)

Koratich, In reply to your question on phagocytosis by flow, you could probably modify the following protocol which I learnt in a flow cytometry course at Sheffield:

1. Use siliconised glass tubes 2. Add 40 ul of a 2.5% suspension of fluorescent 1 um diameter microspheres (Polysciences) to the macrophages suspended in 1.9 mls PBS. Incubate at 37 C for 60 min. 3. Gently pipette the suspension onto 1 ml fetal calf serum and centrifuge to generate a cell pellet (200g) 4. Aspirate the supernatant containing free beads and discard 5. Resuspend the cells in 1-2 mls PBS+0.1% trypsin (type IIIs, Sigma) and 5nMol EDTA (ie equal parts stock tissue culture Trypsin/Versine). Incubate at 37C for 10 min to detach adherent but non phagocytosed beads 6. Repeat step 3. 7. Wash cells X2 in cold PBS 8. Analyse by flow John

Koratich, Mike (koratich@sri.org)

Derek Schulze (flow@post.queensu.ca)

As far as bacteria stuck on the outside - I believe you can add crystal violet to your samples to quench any fluorochromes outside of the cells. The problem here is that at least in my experience quite a few bacteria will adhere to the cells in the early stages of phagocytosis and you would lose their contribution to the total numbers.

At 10:32 AM 1999-06-01 -0500, Koratich, Mike wrote: >

Gerhard Nebe-von-Caron (Gerhard.Nebe-von-Caron@Unilever.com)

Gema Acebedo Lerzundi (gacebedo@hg.vhebron.es)

The question is about the Bone Marrow sample preparation. What is the best method for Bone Marrow sample preparation prior to antibody staining ? 1. wash the sample first? 2. filter through needle or milipore filter? 3. both ? 4. neither of both. we shall be grateful for your indications.

Wing-keung LEE (hkleewk@netvigator.com)

Hello, Rachel: I don't know whether this is the best way of bone marrow specimen processing, but want to share my experience. 1. Wash the specimen twice with pH7.4 PBS. I use the Immunfuge to spin 2 min at low speed after washing. Ensure no significant cell loss during these steps. You can countercheck the cell population by prepareing a blood smear. Also alert that high centrifugal force will casue decrease in Forward scatter. Suspend the cell with protein base buffer e.g. fetal calf serum or cheaper use 10% IgG free albumin in PBS. 2. Filter is unnecessary as you use fine pipette tip to aspirate the sample. It is the cheaper and easier method. I found there is no blockage of flow path or loss of cell population. I'm afraid that cell will loss during filtration. Good luck.

Duperray Christophe (duperray@bacchus.montp.inserm.fr)

Dear flowers, I've changed my operating system on my Data analysis Mac G3 to OS 8.5. After that, it became impossible to have colour prints on HP printers (Deskjet 1600 CM). We always had error messages and printmonitor crashed. Just to precise, Cellquest version is 3.2, and 1600 M printer drivers : HP laserjet 8.3.1 and 1600 CM ppd : 1.0. Do you have this problem ? Do you have any solution ? With many thanks. Sincerely DUPERRAY Christophe

Ray Hicks (rh208@cus.cam.ac.uk)

Hi Cristophe, You can download the latest version of HP deskJet 1600M software at http://www.hp.com/cposupport/prodhome/dj1600m.html or ask them to notify you of updates (choose the latter option - the latest version is the one that you use). Have you tried using the system 8.5 apple LaserWriter driver? I've found that it's been the best option with our HP LaserJet 5MP since LaserWriter 8.0 was released, as long as the ppd is valid and visible to the driver.

Pedro Flores-Villanueva (Pedro_Flores-Villanueva@MacMailGW.dfci.harvard.edu)

Does any body know which markers are the best to define Human NK cells?

Thanks for your attention to this e-mail, Pedro O. Flores-Villanueva, M.D.

Paula Lavery (plaver@po-box.mcgill.ca)

We have always used CD57, clone HNK-1. The hybridoma is available from the ATCC, so you can (inexpensively) prepare your own.

Paula Lavery

Transplant Immunology Laboratory Royal Victoria Hospital McGill University

"If we knew what we were doing it wouldn't be called research, would it?" -- Albert Einstein

Harry D. Dawson (DawsonH@grc.nia.nih.gov)

Dear Flowers, I think CD57 is the least useful of the NK cell markers. It is expressed by a smaller fraction of NK cells than either CD16 or CD56. In addition most CD3+/CD28- cells express CD57. Why not try a combination of markers. There is significant overlap of CD16, CD56 and CD57 expression by NK cells which can serve as a basis for functional classification. The ability of NK cells to lyse target tumor cells can be distinguished based upon expression of CD16, CD56, and CD57, with CD16bright/CD56dim/CD57-, CD16dim/CD56bright/CD57-, CD16bright/CD56bright/CD57+ and CD16-/CD56bright /CD57+ NK cells having the highest to lowest NK cell activity, respectively. Sincerely,

Antony Bakke (bakkea@ohsu.edu)

Alterations in cytotoxic and phenotypic subsets of natural killer cells in acquired immune deficiency syndrome (AIDS). Journal of Clinical Immunology. 7(1):16-23, 1987 Jan. )

Harbour, John (JohnHarbour@hmhs.com)

Does anyone have a reference that displays the typical location of cells on a SSC/45 plot? I am particularly interested in "large B cell lymphoma" cells. Thanks

Cynthia Aller (caller@execpc.com)

Juan Luis Castillo Nadevrete (axelyoyi@entelchile.net)

HI, I have a student that has interest in Fetal Hemoglobin. Who can helpme, I need information about flow tecnique, and references in the literature.Adios

Ralph Bohmer (ralph.bohmer@es.nemc.org)

Two-color flow cytometry to study correlated contents of fetal and adult hemoglobin in nucleated red cells is described in British Journal of Hematology 1998, 103:351-360.

Ralph M. Bohmer, New England Medical Center, Boston

Fareed Al Gurg (algurg7f@emirates.net.ae)

Refer to a book (Man's Haemoglobins) by H. Lehmann & R.G.Huntsman published by North-Holland. It will give you all the information you need about this Hb and many other Hbs.

Bruce H. Davis, M.D. (bdavis@beaumont.edu)

My group has published our work using an anti-HbF clone and method available through CALTAG lab in Transfussion 38:749-756, 1998. Other references include Campbell et al, Cytometry 35:242, 1999, Thorpe et al, Brit J Haematol 74:125, 1994.

Artur Plett (plett@auhs.edu)

This is another question about ambiguity and the question itself may be ambiguous: What can one do with a stain in which the whole population shifts into a higher fluorescence and no discernible separation of a subpopulation ocurrs? (The population leaves a gap in the left hand side) Say, for example in the case of activation markers CD69 and CD25. After activation all the cells are on their way of expressing it, but after short term incubation only a percentage will shift away from the "background fluorescence", "isotype fluorescence" or "negative" quadrant. Can the change in MFI for the whole population be considered relevant (if indeed they are all positive?), or is the only right way to look at this by cutting off the "negative" and looking at the MFI of the "positive" population? Could a possible explanation be that cells who were initially in the low "negative" population have actually shifted into a higher "positive" fluorescence, but this shift is being masked by the background fluorescence? Any answer and /or lengthy explanations would be greatly appreciated. Thank You. Artur

Michael Ormerod (Michael_Ormerod@compuserve.com)

The situation described (of a shift in the mean fluorescence intensity of the whole population of cells on aantibody staining) is quite common.

In my view, it would be incorrect (indeed, meaningless) to subtract away the isotype control fluorescence and then to describe the remaining cells as 'positive'. The whole population is positive, the observation that the positive and negative populations overlap is irrelevant.

In this type of experiment, you are not trying to identify positive and negative sub-sets but looking at the expression of an antigen across the whole population.

In these experiments, I always express the results as either the mean or the median fluorescence intensity of the whole population.

It could be legitimate to subtract the MFI of the isotype control to express the result as the increase in MFI due to antibody binding.

Michael Ormerod 34 Wray Park Road Reigate RH2 ODE

Howard Shapiro (hms@shapirolab.com)

When cells acquire or upregulate an antigen in the same compartment in which they are measured, as is the case in in vitro activation, it is typical to see distribution shifts rather than "positives" and "negatives". From the statistical point of view, it may be more legitimate to compare shifts in median fluorescence intensity (the median being a more robust statistic than the mean for distributions which deviate substantially from normal) than to attempt to define "positive" and "negative". The latter terms make sense when you look at cells which acquire or modulate antigen outside the compartment in which they are observed, as happens with peripheral T-cells. In they thymus, they acquire both CD4 and CD8; one of these antigens is lost before the cells migrate into peripheral blood, and one can therefore observe clear positives and negatives in distributions of CD4 and CD8.

Tom Mc Closkey (thomasm@nshs.edu)

Our division saw a patient yesterday with an extremely low CD8 T cell count (200 CD8s/ul), an 11 year old who has experienced recurrent infections. From the literature I have found 2 defects which affect the 8s: ZAP70 results in complete absence of 8s and CD3 gamma defect which results also in absence of CD4 CD45RA cells (RA is normal), neither of which seem to fit. Has anyone seen anthing like this? Any thought s or comments are greatly appreciated.

Thanks, Tom

Dr. Martin R. Hadam (Hadam.Martin@mh-hannover.de)

Over the years I've seen a number of such patients. Consistent finding was the absence of bright CD8 cells, while dim CD8 / NK cells persist. All except for one clinically presented as severe combined immunodeficiencies. The single patient with exactly the same phenotype, yet only marginally immunodeficient (if at all) was a girl then of ca 8 years. For a CD3gamma defect I'd anticipate lower levels of CD3 as well.

Maurice R.G. O'Gorman

Hello Tom there is a third possibility, all related to HLA class I expression abnormalities which could ulitmately have several genetic etiologies.

TAP proteins (transporter associated with antigen processing) if mutated fail to load the Beta 2 class I heavy chain complexes with peptides and as such these "empty" complexes are very unstable, are inefficiently transported through the golgi compartment and as a result the class I heavy chains do not get sialylated and most do not get expressed on the cell surface. the end result is that CD8+ T cells do not get selected in the thymus and you end up with few CD8 positive T cells. Another cause of class-1 deficiency is Beta-2 microglobulin deficiency. This disorder is referred to as the Type-1 bare lymphocyte syndrome and you could easilily screen for class I expression by flow cytometry using any number of available anti-MCH-Class I antibodies. the reference for the original description of this disorder in 2 family members is science 265:237-240, 1994.

DIANE M SHARP 695-7248 (Diane.M.Sharp@dupontpharma.com)

Dear colleagues, I have a B-D FACScan and routinely do PI staining for cell cycle analysis. I gate on width and area (doublet discrimination). Does this guarantee exclusion of doublets? I can clearly gate out doublets that appear to be 4N DNA content with a large pulse width signal, but I also often see greater than 4N events that do not have a large width pulse. Are these truly >4N cells? The cell lines I use are adherent human tumor and fibroblast cell lines. Any input would be appreciated.

Mark A. KuKuruga (kukuru@umich.edu)

In general, the electronic approaches to excluding aggregates (BD's Width vs Area, Coulter's Integral vs Peak, also Coulter's Time of Flight) work fairly well. I would never say they guarantee this exclusion, however. I've seen artifactual staining patterns (e.g.., with squamous cells) that can cause disproportionate signal duration leading to false detection of aggregates. It's also a problem with cells that have been treated with cycle modulators/blockers. So, in cases where doublet/aggregate detection is clearly separated, it's probably safe to trust it. In cases where resolution is diminished, or synthesis progression has been disturbed, better to be conservative with the interpretation. In my experience, the Width vs Area configuration seems to be a bit more sensitive than Integral vs Peak, although Time of Flight works well in the absence of any fluorochrome. Ultimately, it's best to 1) prevent doublets/aggregates as much as possible with modifications to your labeling protocols, and 2) validate the detection of aggregates via cell sorting and/or microscopy. An aside . . . we've used Time of Flight very effectively to improve purity in cell sorting. Alternatively, many of the DNA histogram fitting routines available have the ability to "correct" for doublets/aggregates in a sample (you may hear from them in response here). These methods are very effective as well, although less so in that they are not usable in real time.

> I can clearly gate out doublets that appear to be 4N DNA content with a large > pulse width signal, but I also often see greater than 4N events that do not > have a large width pulse. Are these truly >4N cells? The cell lines I use > are adherent human tumor and fibroblast cell lines. Any input would be > appreciated.

If one trusts the doublet exclusion method employed, one then must accept that cells that lie on the diagonal line with the "singlet" region must ultimately be seen as non-aggregate. This is not really a problem, since most cell lines have abnormal DNA content, and many may be bi/multiclonal. Certainly in the context of primary tissue, aneuploidy detected along with "normal" diploid cells would be seen the same way. There are also cases where cells may endoreduplicate. MAK.

Larry Arnold (lwarma@med.unc.edu)

Diane, It sounds like you are seeing endoreduplicating cells. There is a paper in the most recent Cytometry from the Jaccoberger group about these and how to separate the 4N G1 from the 4N G2M. Larry Arnold

Derek Davies (daviesd2@icrf.icnet.uk).

I think that the short answer is that it doesnt guarantee exclusion of all doublets but will certainly get rid of a lot of them from the analysis. The Area v width plot will reveal two G1 cells that travel through the beam consecutively but not two that travel side by side. Doublets will show a greater width than true G2 cells - although this is much more applicable to small round cells - epithelial cells, where there can be a degree of cytoplasmic staining, can be difficult - keratinocytes seem to be a particular problem.

Looking at the plot with area on the y axis the trail of single cells travels upwards at an angle about 10 degress off the vertical (towards the top right). I usually put a polygonal gate to exclude the obvious doublets. However certain cell types will show endoreduplication and in these cases it is best to run all samples ungated and decide on gating strategies later as the multiploid cells will follow the angle of the single cell population. By excluding all cells above 4n it is possible to lose what are single cells with multiple nuclei by labelling them as clumps - a sample showing true reduplication will have peaks at 2n, 4n, 8n, 16n etc but none at 6n, 10n and other n's in between. Knowing what is happening biolgically will help you decide whether you are getting real changes or clumping as will sorting the relevant populations onto a slide and checking microscopically.

Barren, Phil (BarrenP@MedImmune.com)

FLOWers, I am trying to look at the binding of virus to cells. Does anyone have any thoughts on doing this without using antibody to label the virus?? I want to effect the virus/cell binding as little as possible. Thank you for your time. Philip Barren

Leary, James F. (jleary@utmb.edu)

Dear Phil, We published two papers back in 1982 (oldies but goodies!) showing how viral binding, including kinetics, could be performed using fluorescent dyes that partition into the virus. Those particular dyes are probably not commercially available, but there are a number of other fluorescent membrane-partitioning dyes that should work. We're not doing that line of work anymore so I can't give you more recent information. You do need to test to see that the dyes you use will not interfere with either the binding or the subsequent biological action that you are studying. Here are the two references which should give you some help in setting up your studies: Leary, J.F., Notter, M.F.D.: "Kinetics of Virus Adsorption to Single Cells Using Fluorescence Membrane Probes and Multiparameter Flow Cytometry" Cell Biophysics 4: 63-76 (1982).

Notter, M.F.D., Leary, J.F., Balduzzi, P.C.: "Adsorption of Rous Sarcoma Virus to Genetically Susceptible and Resistant Chick Cells - A Laser Flow Cytometry Study" J Virology 41: 958-964 (1982). I hope this helps. Jim Leary

Howard Shapiro (hms@shapirolab.com)

At least some viruses (depending on which nucleic acid they have and how tightly packed it is) can be labeled with nucleic acid dyes. The dimeric dyes (YOYO-1, TOTO-1, etc.) have high enough affinity constants that they won't come out of the viruses. This approach was used by Curtis Suttle to "phage type" straines of Synechococcus using labeled bacteriophages and fluorescence microscopy (Hennes KP, Suttle CA, Chan AM: Fluorescently labeled virus probes show that natural virus populations can control the structure of marine microbial communities. Appl Environ Microbiol 61:3623-7, 1995); using samples he sent me, I was able to detect both viruses bound to the cyanobacteria and (probably) single labeled virions using a high-sensitivity Cytomutt, but we never got around to publishing the observations. Daniel Vaulot and coworkers did the same trick more recently and did publish, also in Applied and Environmental Microbiology, within the past few months; they were detecting free virions by YOYO-1 fluorescence using a FACScan (or maybe a Calibur). Since the nucleic acid dye should not be bound to the surface of the virus, the binding of labeled viruses to cells shouldn't be affected.

In the early 1980's, Ken Rosenthal (a virologist now at Northeastern Ohio Universities college of Medicine, not the immunologist from McMaster) labeled EBV with cyanine dyes (by growing it in cells loaded with dye) to look at binding by flow cytometry (I don't recall whether we published this or not); the nucleic acid labels give substantially larger fluorescence signals.

-Howard

linda.weaver@pharma.novartis.com

Dear Phil, There is a year-old paper that may start you off with a protocol for labelling a virus directly: Leopold. P.L. et al, Human Gene Therapy, 9:367-378; February 10. 1998 " Fluorescent Virions: Dynamic Tracking of the Pathway of Adenoviral Gene Transfer Vectors in Living Cells". This group labelled adenovirus directly and used Cy3 as the fluorochrome. You may not prefer Cy3, but you may want to substitute some small molecule of your preference. According to the authors, the adenovirus retained greater than ninety percent of its titer. This is the key issue. I don't know what virus you are using; some viruses may be more susceptible than others to labelling protocols. The authors apparently had to optimize their dye-to-protein ratios to achieve well-labelled yet biologically active virus particles. I hope this helps! Cheers, Linda Weaver

Daryl Webb (dwebb@schooner.waite.adelaide.edu.au)