Discussion
The results from the present study demonstrate that wounds inflicted on
well-differentiated respiratory epithelium, grown from primary NECs in
ALI culture, heal at different rates depending on the atopic and
asthmatic status of the individual donors. Healing is slowest in atopic
asthmatics. However, atopic status appears to have more influence than
asthmatic status. In addition, our data strongly suggest that the
regular use of inhaled corticosteroids prior to nasal cells being
harvested improves wound healing in our ex vivo model. Finally,
wounded epithelium takes longer to heal when the wound occurs at the
peak of an RSV infection, regardless of health status, and that prior
regular use of inhaled corticosteroids improves wound healing under
these circumstances as well.
Before discussing the implications of our findings, some technical
issues need to be acknowledged. Wound healing involves three processes:
spreading, migration and proliferation 4,6. Our model
only addresses spreading and migration as our end-point is when the
wound is physically closed. Thus we are not able to comment on
proliferative events. Our ex vivo model of wound closure uses a
fully-differentiated epithelium grown from primary NECs in ALI culture.
Many previous studies have wounded monolayer cultures of tracheal
epithelial cells 15-18 . A major advantage of ALI
cultures is they contain the full variety of epithelial cells6,19; this is not the case with monolayer cultures,
which consist of basal cells only. Thus the results from the present
study may differ from monolayer wounding studies. However, the general
impairment of wound repair we report following RSV infection in ALI
culture is similar to previous reports of delayed wound repair following
rhinovirus infection of monolayer cultures 16.
Finally, ALI cultures are not well suited to measuring cytokine
secretion into the surface lining fluid. The surface lining fluid layerin vitro is thin and the maximum volume that can be removed is
around 1-2µL without the addition of a wash solution and even this
stimulates the cells 20 . Thus lack of cytokine
responses in the present study should be treated with caution and is
unlikely to reflect in vivo conditions.
Previous monolayer culture studies, reported incomplete wound healing in
asthmatic children 15,17,18 . In the present study,
delayed wound repair was related more close to atopy than asthma;
cultures from non-atopic asthmatics healed at similar rates to those
from healthy subjects and delayed healing was seen in atopic
non-asthmatics. Atopy is associated with increased levels of the
T-helper(Th)-2 cytokine, IL-13 12. IL-13 induces
epithelial proliferation by inducing TGF α, which in turn binds to the
EGF receptor (EGFR) on epithelial cells 12. In the
present study, we confirm that signalling through EGFR is necessary for
epithelial wound repair; repair rate was markedly reduced by blocking
EGFR. However, adding additional EGF did not hasten wound repair. The
role of IL-13 is more difficult to understand. Exogenous IL-13 did
reduce wound healing, however, the amount added was at least six orders
of magnitude (ng/ml) greater than endogenous production by epithelial
cells (pg/ml). Blocking the effects of endogenous IL-13 had no effect on
the rate of wound healing. Taken together, these data suggest that
excessive production of IL-13 could influence wound healing but not in
concentrations made by epithelial cells. One caveat to that premise is
that epithelial cells grown in monolayer culture have been shown to
secrete high enough levels of IL-13 to facilitate wound healing21. However, that study used 1HAEo cells, a SV-40
transformed cell line. How these results relate to normal epithelial
cells is uncertain. Monocyte derived macrophages secrete IL-13 in ng/ml
quantities, especially when polarized to the alternatively activated
phenotype 22 . IL-13 is also secreted by Th-2 T cells.
As such, IL-13 secreted into the airway in vivo could be involved
in an exaggerated delay in wound healing in atopic non-asthmatics as
well as in atopic asthmatics and warrants further investigation.
The data from the present study confirm previous reports, using
different respiratory viruses 16,23, that infection of
the respiratory epithelium impairs wound repair in all subjects, but
especially in asthmatics. In the context of the present study, these
results serve to demonstrate the utility of the model rather than
advancing knowledge about epithelial response to external stressors.
Perhaps the most surprising result from the present study was the
improved wound healing in subjects taking regular inhaled
corticosteroids prior to cell harvesting. There is an implied, if not
written, “belief” that harvested cells are unlikely to be influenced
by in vivo exposures after several culture passages in
vitro . Our data definitely challenge this premise. In our case, these
cells can only have been exposed via the systemic circulation, either by
the extremely low levels of corticosteroid absorbed into the circulation
or by some downstream effect of corticosteroids in the lung. Despite
this, the effects of prior use of inhaled corticosteroids in improving
the rate of wound healing, including after RSV infection, are clear and
indisputable (Table 2). The real challenge is in understanding what
these findings mean! By definition, innate immune responses, such as
those from epithelial cells, do not have a response memory. However,
this concept has recently been challenged by elegant studies showing
that skin wounds in mice heal more quickly when the mouse had previously
been wounded in the same area 24-26. Whether similar
mechanisms operate in human respiratory epithelium remains to be
demonstrated.
In summary, data from the present study have demonstrated that atopic
subjects per se are more likely to have delayed healing of a
wounded respiratory epithelium, however the mechanisms involved remain
unclear. Exposure to inhaled corticosteroids prior to harvest improve
wound healing in vitro , including following RSV infection.