Content area
Full Text
NEWS AND VIEWS.com/naturemedicineinto the lung, so keeping cells at the site of
infection. From here, the cells again move
to the local nodes where they re-encounter
antigen (Fig. 1).At first sight this roundabout pathway
seems overly complex, but it has two very
useful features: first, the site of T-cell proliferation is anatomically separated from the
site of infection; second, there is a built-in
delay between the onset of infection and the
delivery of primed T cells back to the lung.
These characteristics may serve to limit
damage to the infected airway by T cells
that might otherwise cause overexuberant
responses, leading to immunopathology and
respiratory failure if activated cells are present
at the start of infection6.Part of the reason that cells are maintained
in this loop seems to be that processed antigen is retained in the draining lymph node
for at least 2 months. This finding is reminiscent of the classic experiments by Julia
et al., showing that peptide-loaded DC can
remain in the airways for as long as 8 weeks
and seem responsible for keeping CD4+ T
cells in place7.Simple viral persistence is unlikely to
explain the findings of Zammit et al., because
the authors were unable to detect viral RNA
by reverse transcriptase-PCR 30 days after
infection, in either the lung or local node,
even when looking for two separate viral
genes. However, they did not check for viral
DNA transcripts, which have been found
in mice infected with LCMV. This anomaly
may be due to endogenous retroviral reverse
transcriptase, which is present in mice and
hamsters but not in human cells8. This aside,
it does seem likely that antigen displayed on
the class I major histocompatibility complex
(MHC) of cells within the lymph node is
responsible for maintaining T-cell activation. Using T cells to detect the antigen may
seem a rather circular method, but perhaps
there is no alternative: as few as three MHC
class I molecules complexed with peptide
are sufficient to trigger a CD8+ T cell9, a
sensitivity unimaginable with other current
techniques.This study raises several questions. Is it
actually DCs that hold and present antigen, and in what form? Is there a limit to
the number of different antigens that can
be presented at one time, and do different