W. Bi et al. / Bioorg. Med. Chem. Lett. 26 (2016) 4042–4051
4045
mitochondrial morphology was classified into three categories:
tubular (normal), intermediate (tubular with swollen regions),
and fragmented (small and globular) (Fig. 2A). The method for
quantification involved determining the percentage of cells with
abnormal mitochondrial morphologies as a surrogate measure of
the proportion of cells with fragmented mitochondria. When cells
with intermediate or fragmented mitochondria were expressed as
a percentage of the total cells counted (100 cells were counted per
experiment and the data was averaged over four independent
experiments per treatment), the non-treated cells contained pre-
dominantly long and evenly distributed tubular mitochondria
throughout the cell.
During 2 h of simulated reperfusion, the number of cells with
intermediate and fragmented mitochondria increased significantly.
The simulated ischemia was found to significantly increase the
percentage of cells with intermediate or fragmented mitochondria
when compared to the control cells, whereas reperfusion resulted
in the majority of cells displaying fragmented mitochondria. The
percentage of control cells exhibiting long tubular mitochondria
decreased from 85.3 6.1% (sham control) to 15.8 4.3% (s-I/R
alone), and the mitochondria with the intermediate as well as
the fragmented morphology were predominantly aggregated in
the perinuclear region. Interestingly, the addition of compound
11a,b to the perfusate of control cells undergoing ischemia
resulted in significant protection from s-I/R induced cell death,
and attenuated the fragmented mitochondrial appearance. The
cells treated with 11a,b + s-I/R exhibited an increase in the per-
centage of cells with tubular mitochondria (56.4 6.2% (11a + s-I/
R); 50.3 5.7% (11b + s-I/R) versus 15.8 4.3% (s-I/R alone)
(Fig. 2B).
To determine whether or not the mitochondrial protection pro-
vided by compounds 11a,b was associated with mitigation of mito-
chondrial ROS formation, the mitochondrial ROS levels were
measured following s-I/R using confocal imaging microscopy cou-
pled with MitoProbe.28–30 By measuring the changes in fluores-
cence intensity of MitoProbe at different time points, the total
ROS generation in the mitochondria was estimated. An ꢀ2.3 fold
increase was observed for mitochondrial ROS levels during simu-
lated ischemia. Further, mitochondrial ROS rapidly increased
within the first 10 min following reperfusion and was followed
by mitochondrial swelling. The increase in mitochondrial ROS pre-
ceded mitochondrial swelling in some areas of the mitochondria,
indicating that opening of the mitochondrial permeability transi-
tion pore (MPT) was occurring after mitochondrial ROS formation.
Within 120 min of reperfusion, an ꢀ6.4 fold elevation of ROS in the
mitochondria was associated with simulated reperfusion; how-
ever, significant inhibition of mitochondrial ROS was achieved via
administration of compounds 11a,b as shown by the changes in
fluorescence intensity of MitoProbe (Fig. 2C).
Enhanced mitochondrial ROS production, membrane potential
depolarization, and mitochondrial swelling have been shown to
trigger the opening of the MPT and translocation of the mitochon-
drial apoptotic protein cytochrome c from membrane to cytosol;
therefore, we examined cytochrome c release in the presence/
absence of compounds 11a,b as a function of simulated I/R. Confo-
cal images revealed that cytochrome c was localized to the mito-
chondrial area in the sham-control cells. After the cells were
subjected to reperfusion, cytochrome c translocated to the cytoso-
lic and nuclear regions, with simultaneous mitochondrial swelling
(Fig. 3A). Pretreatment with compounds 11a,b significantly inhibit
cytochrome c release and mitochondrial swelling during s-I/R from
78.5% 4.5% (s-I/R alone) to 23.5 5.1% (11a + s-I/R), or
33.7% 5.3% (11b + s-I/R) (Fig. 3B), which was also associated with
a low production of mitochondrial ROS. Our data indicate that
mitochondrial ROS overproduction during s-I/R contributes to
mitochondrial fragmentation and cytochrome c release that is pre-
vented by compounds 11a,b.
Tourniquet application is broadly employed in orthopedic, vas-
cular, and reconstructive surgery to provide a bloodless surgical
field, but associates with enhanced production of free radicals
within the ischemic tissue prior to, and during, reperfusion. This
can be exacerbated through enhanced production of inflammatory
adhesion molecules and cytokines. We induced complete limb I/R
injury in a rat model by occluding collateral blood flow around
the femoral and iliac arteries using a rubber band tourniquet
(RBT). The rat model underwent 3 h of unilateral lower limb ische-
mia, and then 4 h of reperfusion post tourniquet release.18 Com-
pounds 11a,b were then further examined to ascertain their
capacity for protection against local and remote organ injury using
this tourniquet-induced limb I/R model.
In our model, routine histological evaluation (hematoxylin–
eosin staining) of frozen cross-sections of skeletal muscle derived
from sham group rats revealed myofibers of uniform size and
shape that were tightly arranged and in essentially direct contact
with separation by a thin level of endomysium (image not shown).
Conversely, 3 h of ischemia followed by reperfusion in our tourni-
quet model led to extensive interstitial edema and damage charac-
terized by swelling and inflammatory cell infiltration (Fig. 4).
Muscle tissue damage post reperfusion was accompanied by an
increase in the lipid peroxidation byproduct malondialdehyde
(MDA) (Table 4), suggesting a correlation between enhanced lipid
peroxidation and oxygen free radicals. Intervention with 11a pre-
vented interstitial edema and left the cross-sectional area of the
microvessel essentially unchanged (Fig. 5). The histological results
indicated that treatment with 11a resulted in only mild edema and
subtle inflammatory cell infiltration in the skeletal muscle, with no
muscle fibers from this group showing obvious histological
changes. We further performed modified tri-chrome staining in
order to quantify intramuscular collagen deposition, as increased
deposition of connective tissue has negative consequences on con-
tractile muscle function via decrease of myofiber occupancy. As
shown in Figure 5B, there was approximately a 2-fold increase
(17–20% vs ꢀ9%) in percent area occupied by collagen in saline
+ I/R groups, relative to 11a + I/R treated group. Overall, 11a treat-
ment improved muscle morphology and reduced fibrosis.
We also noted an irregular distribution of oxidative metabolism
enzymes that was seen in the NADH-stained skeletal muscle tis-
sues from the I/R rat group treated with tourniquet blood flow
restriction. Typically, lightly stained fibers were categorized as
Type II (fast-twitch glycolytic, mitochondrial-poor), while fibers
with intense staining were categorized as Type I (slow-twitch
oxidative, mitochondrial-rich). Type II fibers, possessing decreased
antioxidant enzyme activities in comparison to Type I fibers, are
predicted to be more susceptible to I/R induced damage. However,
our results indicated that mitochondria-rich Type I fibers suffered
greater damage compared to Type II fibers.
11a intervention significantly mitigated most of the preceding
damage, although occasional ‘ragged red fibers’ were still observed
(lower row in Fig. 4B) in addition to occasional vacuolated fibers
seen in H/E sections (lower row in Fig. 4A). The majority of muscle
fibers were intact, with well-defined edges, consistent texture, and
homogeneous morphology. Additionally, NADH reactions were
weak in most fibers (lower row in Fig. 4C), reflecting substantial
tissue protection associated with 11a intervention following I/R
treatment. Along these lines, we found that mitochondrial-rich
Type I fibers could withstand I/R intervention with essentially no
damage, suggesting that compound 11a intervention protects
oxidative muscles to some extent. These muscles, rich in mito-
chondria, would otherwise be expected to suffer extensive freed
radical production and damage associated with I/R.
Bi, Wei
