Dear group, This is part B of my intervention regarding cell surface receptors. The delay in sending it is partly due to my time consumed in preparing our french cytometry association (AFC) to be held in Marseilles on October 8-10 (getting so close now). Good science, many participants (more than 250), many posters (more than 230) and hopefully good fun. In case some of you want last minute information please connect to AFC web page (http://cytobase.cnusc.fr:8101/ , select AFC menu) I do not pretend I can deliver a general recipee to measure cell surface receptors. I would only present to your reading the questions we ask ourself at BioCytex when attempting to develop a quantitative FCM assay for a given receptor. Such development projects are based upon the QIFI technology (see part A) or related extension of the same technology. In doing such I may illustrate some points with lived experiences including targets such as CD4/HIV-R, CD21/EBV-R, APO-1/Fas R, GpIIb/IIIa, or even cytokine receptors (IL2-R, IL6-R). Included in the list of questions are the following: 1- Do we have the basic information on the receptor, especially about the biochemical configuration? Is the receptor monomeric (CD4, CD21, Fas/CD95) or dimeric (GpIIb/IIIa) or multimeric (IL2-R, IL6-R with gp80 and gp130 homodimer). 2- What is the cellular distribution of the molecule? Is this receptor to be measured on well defined cell types or populations or subset? For example CD4 expression can be easily measured on CD4+ lymphocytes as well as monocytes, GpIIb/IIIa on all platelets, whereas IL2-R measurement may require counter-staining for CD3+ T lymphocytes or CD3 -/CD20- (or CD56+) NK cells. By the way, multicolor quantitation is feasible in the QIFI approach despite the indirect staining. See Bikoue et al. (1996) Cytometry. 3- Do we have access to murine IgG MAbs which recognize the potential receptor? If yes they should be tested for saturating concentration. High affinity MAbs do it at less than 1mcg/ml (good boys). Most of the MAbs do it at 10 mcg/ml (still OK). Some will hardly do at more than 50 mcg/ml (bad boys to be avoided if possible). 4- Do we have access to target cells (preferably homogeneous cell lines) which express detectable amounts of the receptor? HUT 78 expresses about 130,000 Fas/cell. Jurkat about 12,000 Fas/cell. Daudi only 3,000/cell. Platelets express about 50,000 GpIIIa/platelet with rather low inter-individual variations. 5- Do various MAbs measure the same number of sites? Yes, for CD4, using Mabs directed against different domains. No, for Fas, one CD95 MAb will reveal 12,000 sites /Jurkat. Another, 10,000 sites. Athird one, 7,000 sites and a fourth one will only bind 2,500 sites. Antigenic sites are not always accessible to MAbs at the same extent. This has been well described for CD34 MAbs in the literature. We would prefer the most accessible sites, at least initially. 6- Which level of expression do we have to measure? Major lymphocytes/leucocytes antigens expression levels range from few thousands (about 5,000 CD4 per monocyte) to few hundred thousands (about 400,000 CD16 on granulocyte). Some T lymphocytes express low levels of Fas (less than 2,000 sites per cell) and the maximum amount we encountered was 130,000 sites per cell (HUT 78). Thus for this particular case a specially tuned calibrator had to be designed with a working range going from 1,000 up to 200,000. Our flow cytometers have 3 to 4 log decades, but we'd rather stay well inside the working range. Quiescent platelets express very low levels of GMP140 (less than 500) and only moderate amounts after mild activation (with ADP: about 1,500 per cell). They are also rather small cells, too. We therefore had to develop a special calibrator ranging from 500 up to 50,000 molecules per platelet. Endothelial cells may express 1,000,000 ICAM-1 per cell under activation (George et al, 1996, Cytometry). Although this could be measured by extrapolation, we decided it was better to design beads covering up to 1,200,000 molecules per cell. 7- Do we know MAbs which will recognize only free receptors? This very interesting point has to deal with the feasibility of a quantitative receptor occupancy assay. We got an example with GpIIb/IIIa, fibrinogen receptor and antiaggregant drugs. We were lucky enough to find one anti GpIIIa MAb which competes with the binding of antiaggregant drugs and another one which binding is not inhibited at all. The binding of one is limited to drug-free GpIIIa whereas the other one reports for total receptor (free and occupied). This magic combination associated with a highly standardized, quantitative, no-wash, whole blood, immunoflow assay makes the basis of our recently described GpIIb/IIIa occupancy assay (Besson-Faure et al, 1997). Sorry, folks, it's now patented. 8- Receptor conformation-related epitopes? Some receptors express activation-related conformation epitopes. Do we have MAbs to tag those activated receptor neo-epitopes? PAC-1 does that for GpIIb/IIIa. However PAC-1 is an IgM. Bad boy for quantitation. LIBS and RIBS appear on receptor (GpIIb/IIIa) or ligand (fibrinogen) upon receptor-ligand interaction. However these seem to be somewhat elusive epitopes. Multi chain receptors (cytokine receptors) may also generate neo-epitopes or hidden epitopes that are initially present on the corresponding independent chains. Such a level of complexity associated with the low level of expression involved makes these receptors a good challenge for the next future. I guess that that's all I wanted to say. I hope that it answers (at least partly) the initial question. I also hope that others will comment or share their own experience on this very stimulating flow cytometry challenge. Best regards to all. Philippe Poncelet, PhD Scientific Director, BioCytex Marseilles - France Tel:+33 - 4 91 94 29 39 Fax:+33 - 4 91 47 24 71 E-mail: Still my boss's one: mcanton@wcube.fr
This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:50:10 EST