96
In addition to osteoclasts, cells of the wound epidermis are also responsive toHBO treatments. The completion of epidermal closure following amputation is
slowed by HBO treatment [ 47 ], and this process can be accelerated by treatment
with a commercially available cyanoacrylic wound dressing , Dermabond [ 48 ].
Cyanoacrylics applied to a wound rapidly polymerize to form a fl exible skin adhe-
sive, which have been shown to enhance the rate of wound closure [ 52 ]. When used
as a wound dressing for digit tip amputations, application of Dermabond creates a
sustained hypoxic microenvironment that is restricted to the stump wound epider-
mis and doubles the rate of wound closure (Fig. 5.6f, g ). The Dermabond treated
epidermis does not retract but is able to migrate directly over the amputated stump
bone, and this effect is ameliorated by HBO treatment. These data implicate hypoxia
as a positive regulator of epidermal migration during wound healing. Once amputa-
tion wound closure is complete blastema formation occurs precociously (Fig. 5.6h )
and the forming blastema becomes hypoxic. Immunohistochemical studies show
that large multinucleated cathepsin K positive osteoclasts are absent, however
cathepsin K positive pre-osteoclasts are observed (Fig. 5.6i, j ), and microCT analy-
ses show that stump bone degradation is signifi cantly reduced (Fig. 5.6k ). The pre-
cociously formed blastema re-differentiates to form the distal digit tip, but
remarkably the overshoot in regenerated bone that characterizes the endogenous
response is not observed, and the regenerated bone has a structure that is lamellar
rather than woven. In other words, the resulting regenerate is structurally identical
to the amputated digit tip when the wound epidermis is induced to close rapidly. It
is interesting to note that limb regeneration in salamanders is characterized by rapid
wound closure that is driven by a rapid cell migration response coupled with a gen-
eral injury induced epidermal swelling response [ 53 , 54 ].
Data from HBO and Dermabond studies are consistent with a model in whichoxygen availability plays a key role in regulating the histolytic phase of regenera-
tion, particularly hypoxia induced termination of osteoclast activity. These studies
also demonstrate that regulating osteoclast activity correlates with blastema size:
reduced osteoclast activity induced by Dermabond results in small blastemas
whereas enhancing osteoclast activity with HBO treatment results in large blaste-
mas. These observations suggest that osteoclast activity is linked to the number or
proliferation of stump cells that participate in the regenerative response. One pos-
sibility is that histolysis of mature stump tissues is required to release progenitor
cells so they can participate in blastema formation; enhancing histolysis results in a
larger regeneration competent progenitor cell population while reducing histolysis
results in a smaller population of progenitors. An alternative, but not mutually
exclusive, possibility is that proteolytic activities associated with histolysis degrades
extracellular matrix and releases chemotactic and/or mitogenic signals known to be
present in mature tissues. For example, bone tissue is known to store extracellular
BMPs [ 55 ], which have been shown to be essential for a regenerative response [ 10 ,
11 ]. Similarly, matrix degradation products (e.g. cryptic peptides ) and metallopro-
teinase activity have been implicated in cell recruitment and regeneration after
amputation injury [ 20 – 22 , 56 ]. Regardless of mechanism, the evidence supports the
conclusion that histolysis of mature stump tissue is a critical early phase of a
L.A. Dawson et al.