From: Mario Roederer (Roederer@Beadle.stanford.edu)
Date: Fri Oct 30 1998 - 11:44:41 EST
<c=US%a=_%p=Specialty_Labora%l=PRIMARY-981028233759Z-63810@primary.specialtylab s.com> of Wed, 28 Oct 1998 15:37:59 -0800 Content-Type: TEXT/plain; charset=US-ASCII Susan, Identifying naive T cells in the human is nontrivial and you must take care to do it properly! In any case, you must use at least 3 colors to do so: CD4 or CD8, and a pair of antibodies including a CD45 isoform (as noted below). There are several ways to uniquely identify T cell subsets in the human. Currently, the best are the combined use of CD45RA with one of CD11a, CD62L, or CD27. Naive T cells are CD45RA+ and CD62L+, CD27+, and CD11a-dull. You may use CD45RO in place of CD45RA, noting that its expression is inverse to CD45RA; thus, naive T cells are CD45RO- and CD62L+, CD27+, CD11a dull. Don't make the mistake of using CD45RO and CD45RA in the same stain; you get no additional information from using either one alone (see notes below for more info). Also note that CD45 isoform expression is very different on CD4 and CD8 T cells! Therefore, you CANNOT use isotype gates for identifying naive T cells, and you CANNOT use the same gate set for CD4+ cells as for CD8+ cells. In particular, the CD8+ memory cells that are "CD45RA-" actually express a good amount of CD45RA! You will need some experience with staining healthy adults to see where to set the gates. CD62L is probably the best antigen to use in combination with CD45RA. It is bright; the CD62L+ cells are easy to distinguish from CD62L- cells. The main problem with CD62L is that it falls off cells during freeze/thaw protocols, and cannot be reliably used on frozen blood samples. CD27 is a good choice as well, although far fewer functional studies have been done with CD27+CD45RA+ cells to really prove that these are pure naive T cells. CD11a is the most robust identifier of naive vs. memory cells in the CD45RA+ subset. However, it is the most difficult to use, because you must set a gate that distinguishes bright from dull cells, without that much separation. Again, a bit of experience looking at healthy adult samples will teach you quickly where to set this gate. CD45RA and/or CD45RO staining is insufficient without an additional marker (CD62L, CD27, or CD11a). There is a distinct set of CD45RA+CD45RO- memory T cells. In the CD8 compartment, this subset averages around 10-30% of total CD8; in CD4 it averages 5% of total CD4 (in healthy adults). In adults with acute or chronic disease such as HIV, these percentages go way up. In advanced HIV disease, CD4 T cells can be more than 50% CD45RA+, but nearly all are memory T cells! In unpublished studies using 10-color flow cytometry, I have correlated the exact expression of all of these markers (together with functional studies to verify memory/naive as well as possible). I do find that about 1% (CD4) or 5% (CD8) of the "naive" T cells, when identified using CD45RA with CD62L OR with CD27 are actually activated memory T cells! If you use CD45RA, CD62L, and CD27 altogether (a 4-color combination with CD4 or CD8), then this percentage drops by an order of magnitude. Thus, a three color combination of CD4/8 with CD45RA and CD62L or CD11a identifies naive T cells to about 95-99% purity. CD11a was much better, achieving nearly 100% purity. mr
This archive was generated by hypermail 2.1.6 : Thu Jan 01 2004 - 17:35:26 EST
