Content area
Full Text
NEWS AND VIEWSOne of the more interesting features of
this study comes from how the structure
was derived. Most mAbs to CD28 (conventional mAbs) bind an epitope located near
the CD80 or CD86 binding site(s) on CD28
(ref. 6; Fig. 1). These mAbs have signaling
properties very similar to those of the natural ligands of CD28, in that their binding to
CD28 strongly costimulates TCR signals but
only weakly signals in the absence of TCR
stimulation. In contrast, the mAb that Evans
et al.2 used to induce CD28 crystallization
is representative of another class of mAbs,
called mitogenic mAbs, which fully activate
T cells even in the absence of TCR stimulation (ref. 6; Fig. 1). Fortunately, this situation
does not occur naturally, because T cells activated in this way would lack any antigenic
specificity and could indiscriminately attack
normal tissues. So, the promiscuous T cell
activation triggered by mitogenic mAbs is
nonphysiological, but it may shed light on
how T cells normally become activated.How do these two types of mAbs (conventional and mitogenic) produce such different
outcomes? Previous studies have shown that
these mAbs bind distinct epitopes on CD28,
with conventional mAbs binding close to the
ligand-binding site on CD28 and mitogenic
mAbs, on the laterally exposed CD region6.
Other studies2,6,7 have discussed ways that differential binding might trigger distinct signals,
including differential crosslinking, binding
to different populations of CD28, proximity
effects and antibody-induced conformational
effects. The available evidence does not favor
(or exclude) any of these (or other) possibilities and they are not necessarily mutually
exclusive. What is now needed are experimental data directly addressing how these mAbs
signal differently.Evans et al. directly address this issue by
comparing structures of complexes of mitogenic and conventional mAbs with CD28.
They have asked whether these mAbs look different from each other when they are bound to
CD28. What they have found suggests that the
mitogenic mAb used in these studies bound to
the side of a CD28 monomer as it projected
from the cell surface, whereas a conventional
mAb bound to the top portion of the receptor (Fig. 1). In addition, both mAbs formed
crosslinked complexes with CD28 that from
a distance seemed very similar. These studies suggest that the ability or inability to form
crosslinked...