OBJECTIVES/SPECIFIC AIMS: Gliomas are the most lethal and common
primary tumor type in the central nervous system across all age groups; affected
adults have a life expectancy of just 14 months. As glioma cells invade the
surrounding normal parenchyma they remodel the composition and ultrastructure of
the surrounding extracellular matrix (ECM), suggesting that the native (i.e.,
“normal”) microenvironment is not ideal for their survival
and proliferation. Recent reports describe suppressive and/or lethal
effects of mammalian ECM hydrogels derived from normal (nonneoplastic) sources
upon various cancer types. ECM-based bioscaffolds placed at sites of neoplastic
tissue resection in humans have never been reported to facilitate cancer
recurrence. The objective of the present research is to evaluate mammalian ECM
as a novel approach to glioma therapy. METHODS/STUDY POPULATION: ECM
hydrogels from porcine dermis, small intestine, and urinary bladder were
produced as described previously. Primary glioma cells were graciously supplied
by Drs. Nduka Amankulor and Johnathan Engh, and U-87 MG were ordered through
ATCC. Cells were plated onto tissue culture plastic at
~60% confluence and allowed to attach for 24 hours before
treatment. The saline-soluble fraction (SSF) of ECM was obtained by mixing
lyophilized, comminuted ECM with 0.9% saline for 24 hours then
filtering the resulting mixture through a 10 kDa molecular weight cutoff column.
All assays and kits were followed according to the manufacturer’s
instructions. Cell viability was measured via MTT assay
(Vybrant® MTT Cell Proliferation Assay, Invitrogen)
and by live/dead staining
(LIVE/DEAD® Cell Imaging Kit, Invitrogen). Time
lapse videos were created by taking images every 20 minutes for 18 hours
(phase-contrast) or every 10 minutes for 12 hours (darkfield). NucView reagent
was ordered from Biotium. Temozolomide was ordered through Abmole. All in vivo
work was conducted according to protocols approved by the University of
Pittsburgh’s IACUC office. RESULTS/ANTICIPATED RESULTS:
ECM hydrogels derived from porcine dermis, small intestine, or urinary bladder
all decreased the viability of primary glioma cells in vitro, with urinary
bladder extracellular matrix (UBM) having the most dramatic effects. The SSF of
UBM (UBM-SSF), devoid of the fibrillar, macromolecular components of ECM, was
sufficient to recapitulate this detrimental effect upon neoplastic cells in
vitro and was used for the remainder of the experiments described herein. In a
cell viability assay normalized to the media treatment, non-neoplastic CHME5 and
N1E-115 cells scored 103% and 114% after 48 hours when
treated with UBM-SSF and 2 primary high-grade glioma cell types scored
17% and 30.5% with UBM-SSF (n=2).
Phase-contrast time-lapse video showed CHME5 and HFF thriving in the presence of
UBM-SSF for 18 hours while most primary glioma cells shriveled and died within
this time. Darkfield time-lapse video of wells containing Nucview dye,
fluorescent upon cleavage by active caspase-3, confirmed that within 12 hours
most primary glioma cells underwent apoptosis while CHME5 and HFF did not. In
culture with primary astrocytes, high grade primary glioma cells, and U-87 MG
glioma cells for 24 hours, UBM-SSF was found to significantly increase the
population of primary astrocytes compared with media
(p<0.05) while decreasing the 2 glioma cell types to
approximately one-third as many cells as the media control
(p<0.0001). A dose-response of temozolomide from 0
to 10,000 μM showed that when treating 2 non-neoplastic cell types
(CHME5 and HFF) and 2 types of primary glioma cell there was no difference in
survivability at any concentration. Contrasted to this, a dose-response of
UBM-SSF from 350 to 7000 μg/mL showed that the
non-neoplastic cells survived significantly better than the glioma cells at
concentrations of 875 μg/mL and upward
(p<0.05). In preliminary animal experiments, large
primary glioma tumors in the flanks of athymic nude mice were resected and
replaced with either UBM SSF or Matrigel (an ECM product of neoplastic cell
origin). After 7 days the resection sites with UBM-SSF had little tumor regrowth
if any compared with the dramatic recurrence seen in the Matrigel injection
sites (n=2). In a separate survival study comparing PBS to UBM-SSF
injections in the flank-resection model, all animals given PBS had to be
sacrificed at 9, 11, and 11 days (n=3) whereas animals given UBM-SSF
were sacrificed at 15, 24, and 39 days (n=3), indicating a moderate
increase in survival due to the UBM-SSF. DISCUSSION/SIGNIFICANCE OF
IMPACT: Since the introduction of the pan-cytotoxic chemotherapeutic agent TMZ
in 2005, the standard of care for patients with glioblastoma multiforme has not
improved. These findings indicate that non-neoplastic ECM contains potent
bioactive regulators capable of abrogating malignancy. Our in vitro data suggest
these molecules appear to have no deleterious effect on non-neoplastic cells
while specifically inducing apoptosis in glioma cells. Our in vivo data suggest
that these molecules may be useful in delaying glioma recurrence, thus resulting
in extended lifespan. Delivering soluble fractions of ECM to a tumor site may
represent a novel approach to glioma therapy, sidestepping traditional cytotoxic
therapies in favor of utilizing putative endogenous anti-tumor pathways.
Robust Optical Flow Algorithm for General, Label-free Cell Segmentation