cell signaling, differentiation, and survival. There is clear evidence of
compromised mitochondrial function following TBI, however, the
underlying mechanisms and consequences are not clear. MicroRNAs
(miRNAs) are small non-coding RNA molecules that regulate post-
transcriptional gene expression, and function as important mediators
of neuronal development, synaptic plasticity, and neurodegeneration.
Several miRNAs are altered following TBI, however, the involvement
of mitochondria in modulating miRNA activity is unknown. Here, we
present evidence supporting the presence of miRNA associated with
hippocampal mitochondria, as well as changes in miRNA expression
levels following a cortical contusion injury (CCI) in rats. Specifically,
we found that the miRNA processing proteins Argonaute (AGO) and
Dicer are present in mitochondria from uninjured rat hippocampus,
and immunoprecipitation of AGO associated miRNA from mito-
chondria suggests the presence of functional RNA-induced silencing
complexes. Interestingly, RT-qPCR miRNA array studies revealed
that a subset of miRNA is enriched in mitochondrial relative to cy-
toplasm. At 12 hr following CCI, several miRNA are significantly
altered in hippocampal mitochondria and cytoplasm. In addition,
levels of miR-155 and miR-223, both of which play a role in in-
flammatory processes, are significantly elevated in both cytoplasm
and mitochondria. We propose that mitochondria serve as a platform
for miRNA function and play an important role in regulating miRNA
activities in response to cellular demand and stressors such as that
observed following TBI. Supported by the Morton Cure Paralysis
Fund ( JES), PHS grants AG028383, NS085830 and NS061933
(PTN), and NS062993 (PGS).
Key words
cortical contusion injury, hippocampus, microrna, mitochondria
OC2-01
BLAST INJURY EXACERBATES NEUROPATHOLOGY AND
IMPAIRS VISUAL FUNCTION IN A TRANSGENIC APPSWE-
PSEN1DE9 MOUSE MODEL OF AD
Harper, M.M.
1,2
, Herlein, J.
1
, Hedburg-Buenz, A.
2
, Anderson,
M.G.
1,2
, Paljug, W.R.
3
, Abrahamson, E.E.
3
, Ikonomovic, M.D.
2,3
1
Veterans Affairs, Iowa City and Pittsburgh, USA
2
University of Iowa, Iowa City, USA
3
University of Pittsburgh, Pittsburgh, USA
The purpose of this study was to determine if blast-mediated TBI
accelerates neurodegeneration in APPswePSENd19e mice (APP-
PSEN), which have a genetic predisposition to develop an Alzhei-
mer’s disease (AD) phenotype characterized by deposition of Amy-
loid-beta (A
b
). We also examined if increased A
b
accumulation was
associated with exacerbated functional and structural
in vivo
retinal
outcomes which could serve as non-invasive metrics for evaluation of
brain pathology after injury.
Mice were exposed to a single blastwave: 150 kPa, 10 ms positive
peak duration. Retinal function and structure were assessed two
months following injury using the pattern electroretinogram (pERG)
and optical coherence tomography (OCT), respectively. Mice were
then euthanized and A
b
concentration and plaque load were assessed
in the hippocampus (HPC) and cortex (CTX).
Compared to sham blast-exposed littermates, blast-injured APP-
PSEN mice had significant reductions in pERG amplitudes
(19.92
–
1.91
l
V and 11.13
–
0.97
l
V, respectively;
p
<
0.01) and ret-
inal ganglion cell (RGC) complex thickness (76.71
–
2.50
l
m and
64.65
–
3.31
l
m, respectively,
p
<
0.01) two months after blast expo-
sure. Blast-exposed APP-PSEN mice also had fewer axons in the optic
nerve as well as substantial increases in soluble and insoluble A
b
40
and A
b
42 concentrations and a 2–6 fold increase in A
b
plaque load in
the HPC and CTX compared to sham-blast-exposed littermates.
The extent of both visual impairments and increases in brain A
b
levels in APP-PS mice were exacerbated two months after blast
TBI. Visual deficits in APP-PSEN mice manifested twice as fast as
we have reported using the same injury in wild-type mice. These
data suggest that analyses of retinal structure and function could
serve as
in vivo
biomarkers to evaluate the extent of blast-mediated
neuronal damage either alone or in relation to A
b
-induced neuro-
degeneration.
Funding
VA RR&D MERIT Award 1I01RX00095.
Key words
Alzheimer’s disease, vision
OC2-02
EFFICIENT ESTIMATION OF BRAIN STRAIN RESPONSES
IN CONTACT SPORTS USING A PRE-COMPUTED MODEL
RESPONSE ATLAS
Ji, S.
, Zhao, W.
Dartmouth College, Hanover, USA
Finite element models of the human head are playing an important
role in investigating the mechanisms of traumatic brain injury, in-
cluding sports concussion. A critical limitation, however, is that they
incur a substantial computational cost (typically hours on a modern
multi-core computer or even a super computer) to simulate a single
impact. Consequently, model-based brain injury studies have been
focused on single head impacts to date and current simulation
schemes are impractical to investigate the significance of repetitive
head blows especially on a large scale. In this study, we evaluated the
feasibility of a pre-computed model response atlas (pcMRA) to sig-
nificantly increase the simulation efficiency using isolated rotational
acceleration impulses parameterized with four independent variables
(peak magnitude (a
rot_p
) and duration, and rotational axis azimuth and
elevation angles). The parametric space was sampled by combining
each variable with values determined from on-field measurements to
serve as the training dataset (a
rot_p
was limited to 1.5–4.5 krad/s
2
,
approximately corresponding to the 50
th
and 95
th
percentile magni-
tudes in ice-hockey). Using randomly generated testing datasets with
a
rot_p
range 0.5–7.5 krad/s
2
(approximately from 25
th
percentile sub-
concussive to 95
th
percentile concussive a
rot_p
in college football, or
up to
*
99
th
percentile a
rot_p
in ice-hockey), the pcMRA interpolation
(a
rot_p
range 1.5–4.5 krad/s
2
) achieved a 100% success rate based on
element-wise differences in accumulated peak strain (e
p
; relative to
directly simulated ground-truths or injury-causing thresholds according
to a ‘‘double-10%’’ criterion, i.e., volume fraction of large element-
wise differences (
>
10%) for the whole-brain was
<
10%) or average
regional e
p
in generic regions (difference
<
10%). Further, (nearly)
excellent performance was maintained in extrapolation for out-of-range
a
rot_p
impulses. The computational cost to estimate element-wise
whole-brain or regional e
p
using the pcMRA was 6 sec and
<
0.01 sec,
respectively, while it was
*
30min to directly simulate a 25ms im-
pulse. These findings suggest the feasibility for pcMRA to substantially
increase the throughput in head impact simulation without significant
loss of accuracy and, therefore, its potential to accelerate the explora-
tion of the mechanisms of sports concussion.
Key words
contact sports, Dartmouth head injury model, finite element modeling,
pre-computation, repetative head impacts
A-11
1...,2,3,4,5,6,7,8,9,10 12,13,14,15,16,17,18,19,20,21,...136