Integrin-FAK signaling rapidly and potently promotes mitochondrial function through STAT3
Visavadiya et al. Cell Communication and Signaling
Integrin-FAK signaling rapidly and potently promotes mitochondrial function through STAT3
Nishant P. Visavadiya 0
Matthew P. Keasey 0
Gary L. Wright
0 Equal contributors Department of Biomedical Sciences, Quillen College of Medicine, East Tennessee State University , Building 178, Maple Ave, PO Box 70582, Johnson City, TN37614 , USA
Background: STAT3 is increasingly becoming known for its non-transcriptional regulation of mitochondrial bioenergetic function upon activation of its S727 residue (S727-STAT3). Lengthy mitochondrial dysfunction can lead to cell death. We tested whether an integrin-FAK-STAT3 signaling pathway we recently discovered regulates mitochondrial function and cell survival, and treatments thereof. Methods: Cultured mouse brain bEnd5 endothelial cells were treated with integrin, FAK or STAT3 inhibitors, FAK siRNA, as well as integrin and STAT3 activators. STAT3 null cells were transfected with mutant STAT3 plasmids. Outcome measures included oxygen consumption rate for mitochondrial bioenergetics, Western blotting for protein phosphorylation, mitochondrial membrane potential for mitochondrial integrity, ROS production, and cell counts. Results: Vitronectin-dependent mitochondrial basal respiration, ATP production, and maximum reserve and respiratory capacities were suppressed within 4 h by RGD and ?v?3 integrin antagonist peptides. Conversely, integrin ligands vitronectin, laminin and fibronectin stimulated mitochondrial function. Pharmacological inhibition of FAK completely abolished mitochondrial function within 4 h while FAK siRNA treatments confirmed the specificity of FAK signaling. WT, but not S727A functionally dead mutant STAT3, rescued bioenergetics in cells made null for STAT3 using CRISPR-Cas9. STAT3 inhibition with stattic in whole cells rapidly reduced mitochondrial function and mitochondrial pS727-STAT3. Stattic treatment of isolated mitochondria did not reduce pS727 whereas more was detected upon phosphatase inhibition. This suggests that S727-STAT3 is activated in the cytoplasm and is short-lived upon translocation to the mitochondria. FAK inhibition reduced pS727-STAT3 within mitochondria and reduced mitochondrial function in a non-transcriptional manner, as shown by co-treatment with actinomycin. Treatment with the small molecule bryostatin-1 or hepatocyte growth factor (HGF), which indirectly activate S727-STAT3, preserved mitochondrial function during FAK inhibition, but failed in the presence of the STAT3 inhibitor. FAK inhibition induced loss of mitochondrial membrane potential, which was counteracted by bryostatin, and increased superoxide and hydrogen peroxide production. Bryostatin and HGF reduced the substantial cell death caused by FAK inhibition over a 24 h period. Conclusion: These data suggest that extracellular matrix molecules promote STAT3-dependent mitochondrial function and cell survival through integrin-FAK signaling. We furthermore show a new treatment strategy for cell survival using S727-STAT3 activators.
Bioenergetics; Cell death; CRISPR; ECM; Endothelial cell; Focal adhesion kinase; Integrin; Mitochondria; Vitronectin; STAT3
Integrins are heterodimer transmembrane receptors which
bind ECM molecules to promote cell adhesion and initiate
intracellular signaling that can lead to cell survival [1, 2].
Disruption of integrin binding can cause cell death,
especially for cells attached to basement membranes , e.g.,
endothelial cells in the central nervous system (CNS).
Among others, endothelial cells express ?v?3 integrins
which contribute to their survival [4, 5]. Integrin signaling
is important for normal endothelial cell function in
maintaining the blood-brain-barrier (BBB) [6, 7], whose
disruption by neural injury and stroke leads to disease progression
. FAK is one of the major integrin signaling mediators
and is activated via autophosphorylation on Y397  which
can suppress apoptosis in endothelial cells .
Mitochondria not only play a vital role in energy
production, particularly in the CNS , but also have
emerged as a key stress-signaling hub within the cell
. CNS endothelial cells have a very high
mitochondrial mass compared to those of other organs , and
mitochondrial function is important for maintaining the
BBB and ATP-dependent trans-endothelial transport
[13, 14]. Mitochondrial dysfunction after neurological
insults plays a role in BBB breakdown and tissue
degeneration [7, 15, 16]. Lengthy mitochondrial bioenergetic
dysfunction leads to depletion of ATP, increased
production of reactive oxygen/nitrogen species, calcium
dysregulation, and release of pro-apoptotic proteins, leading to
cell death [17, 18].
Integrins can prevent apoptosis through FAK-AKT
signaling [10, 19, 20], and inhibiting mitochondria-associated
bit1 [20, 21], but have not been implicated in bioenergetic
function. We recently discovered an integrin signaling
pathway that inhibits CNTF expression, involving FAK,
JNK and the S727 residue of the transcription factor
STAT3 . Depending on phosphorylation of residues
S727 or Y705, STAT3 can inhibit or promote nuclear gene
expression . Recent seminal studies identified a
nontranscriptional role of pS727-STAT3 in stimulating
mitochondrial bioenergetic function through electron
transport chain (ETC) complex I, II and V activity [24?
26], probably not by binding directly , but by binding
to prohibitin 1 . STAT3 can also reduce formation of
the mitochondrial permeability transition pore, possibly
by interacting with cyclophilin D , thus maintaining
membrane potential necessary for bioenergetic function,
as well as preventing release of cytochrome-c, which leads
to apoptosis . STAT3 can also reduce ROS formation,
possibly by improving complex I coupling, improving cell
survival under ischemic conditions [30, 31]. It is unknown
whether integrin or FAK signaling interacts with
pS727STAT3 to regulate mitochondrial function.
A few studies have identified mechanisms that activate
S727-STAT3, including cytoplasmic PKC isoforms [32, 33]
and the c-Met receptor acting via ERK and AKT [34, 35].
The small molecule bryostatin 1 activates PKC 
whereas HGF activates c-Met . They are being
developed for treating Alzheimer?s disease  and spinal cord
injury , respectively. Stimulating S727-STAT3 more
directly may be a treatment strategy that can circumvent
pathologically disrupted integrin or FAK signaling.
The present study determined whether
ECM-integrinFAK-STAT3 signaling promotes mitochondrial function
in cultured endothelial cells and whether treatments that
activate STAT3 could rescue cells against the loss of
The immortalized mouse brain endothelioma cell line
(bEnd5) was created and characterized as a good model
for CNS endothelial cells [39?41]. Culture medium
components were from Gibco. The cells were grown in in
T75 Flask (Cat # CC7682-4175, CytoOne, Ocala, FL) in
DMEM supplemented with 10% fetal calf serum, 3 mM
glutamine, 100 units/ml penicillin and 100 ?g/ml
streptomycin, 1 mM sodium pyruvate, 1% non-essential
amino acids. Cells were passaged every seven days with
cells up to 35 passages being used for experiments.
Before use, replated cells were maintained for 24 h except
where noted specifically. The cells were counted in high
resolution images obtained with a 10? objective, five
fields per well, using ImageJ software (NIH), with 3?5
wells per condition.
Mitochondrial bioenergetics measurements
Mitochondrial bioenergetics measurements were made
with an XF24 Extracellular Flux Analyzer (Seahorse
Bioscience, North Billerica, MA). Cell density was
optimized for recommended basal OCR ranges of 50?400
pMole/Min. For vitronectin (VTN) experiments, XF-24
culture plates were incubated with poly-d-lysine for 1 h
(50 ?g/ml; Sigma), washed 3? with autoclaved H2O,
dried and coated with recombinant human VTN
(0.5 ?g/cm2, Cat#SRP3186, Sigma) for 1 h. BEnd5 cells
were seeded at 100,000/well with or without RGDS
(Arg-Gly-Asp-Ser peptide, 100 ?g/ml, Cat#3498, Tocris,
Bristol, UK) or P11 (His-Ser-Asp-Val-His-Lys-NH2
peptide, 10 ?g/ml, Cat#4744, Tocris) peptides in serum-free
medium supplemented with 2% B27 (Gibco cat#
10889038) for 4 h. We also tested the effects of VTN,
EHS mouse laminin-111 (Cat#L2020, Sigma) and human
plasma fibronectin (Cat#F3879, Sigma) as 50 ?g/ml
substrate with 10 ?g/ml added to serum-free medium for
4 h. For the other experiments, 50,000 bEnd5 cells/well
were seeded without substrate and grown for 44 h
before 4 h incubation with FAK14 (5 or 10 ?M, Cat#3414,
Tocris) or stattic (5 or 10 ?M, Cat#2798, Tocris) with or
without bryostatin (50 nM, Cat#2383, Tocris) or HGF
(500 ng/ml, Cat#315-23, PeproTech, Rocky Hill, NJ) or
actinomycin D (0.3 ?g/ml, Cat#1229, Tocris) treatment
with or without 5 ?M FAK14. After treatments, cells
were washed once with XF base media supplemented
with 2.5 mM glucose and 1 mM sodium pyruvate, and
incubated for 1 h in a non-CO2 incubator. Before use,
sensor cartridges were hydrated, loaded with oligomycin
(Cat# 495455, Millipore), carbonyl cyanide
4-(trifluoromethoxy) phenylhydrazone (FCCP, Cat# 0453, Tocris,
1 ?M as used by others [42, 43], antimycin A (Cat#
A8654, Sigma), and calibrated according to
manufacturer?s instructions. XF24 plates with cells were then
loaded and mitochondrial respiration measurements
performed according to a standard software protocol. The
basal OCR, ATP production, maximum reserve and
respiratory capacity were calculated as described , with
averages calculated from five wells per condition in each
individual experiment. Bioenergetics data are presented
as a representative original trace (OCR raw data) while
bar graphs were normalized to the average of controls
for a set of experiments.
CRISPR-Cas9 mediated knockdown of STAT3
CRISPR-Cas9 pre-cloned guide RNA cassettes targeting
mouse STAT3 were purchased from Origene (Cat#
KN316845) and knockdown performed according to 
with some alterations. Briefly, 2 ? 105 bEnd5 cells were
seeded into 6-well plates and maintained for 24 h.
Lipofectamine 3000 (Cat# L3000, Invitrogen) was used at
0.5% complexed with 5 ?g plasmid (2.5 ?g Cas9-gRNA
with 2.5 ?g Donor) in Opti-MEM. Transfections were
performed according to standard procedures, with cells
maintained for 24 h after transfection. Transfection
medium was removed from cells the following day and
replaced with fresh medium containing 2 ?g/ml
puromycin (Cat# A1113803) which caused 100% cell death in
non-transfected control cells after 4 days. Cells resistant
to puromycin treatment were grown in fresh medium
and maintained until confluent (~7?14 days). PCR
primers were designed against regions of the mouse
STAT3 gene flanking the guide-RNA target sequence
and were: Forward 5?-GGCCTTGACCTGTCTGTCAT -3?
and reverse 5?-TGTGCAGAGATCTCACCAAGT -3? to
generate an amplicon of 849 bp in a reaction with
100 ng of genomic DNA extracted from cells with the
DNeasy Blood and Tissue kit (Qiagenssa, USA, Cat#
69506). PCR products were run on an agarose gel for
purification and sequencing performed by the
Molecular Core Facility at ETSU using the following primer
5?-GGCCTTGACCTGTCTGTCAT -3?. A point
mutation was identified at the predicted Cas9-guide RNA
cut site, resulting in a frame shift. Knockout of protein
was confirmed by western blots.
Mitochondria were isolated as previously described .
Briefly, cells were homogenized in ice-cold isolation
buffer (215 mM mannitol, 75 mM sucrose, 0.1% bovine
serum albumin, 20 mM HEPES, 1 mM EGTA; pH
adjusted to 7.2 with KOH), followed by centrifugation,
twice at 1,300 g for 3 min to obtain nuclear pellets. The
supernatant was centrifuged at 13,000 g for 10 min to
obtain mitochondrial pellets. The last supernatant was
collected as the cytosolic fraction. The cold and the
EGTA in the isolation buffer ensure that proteins remain
phosphorylated and intact. Moreover, we have found
that addition of protease and phosphatase inhibitors
(Cat# P8340, Sigma), supplemented with 1 mM sodium
orthovanadate, does not reveal any differences. The
purity of sub-cellular fractions was confirmed using
Western blotting for fraction-specific markers. For another
experiment, mitochondria were treated with stattic
(10 ?M) or okadaic acid (1 ?M; Cat# 5934, Cell
Signaling) for 2 h in KCl-based respiration buffer (125 mM
KCl, 2 mM MgCl2, 2.5 mM KH2PO4, 20 mM HEPES
and 0.1% bovine serum albumin, pH 7.2) containing
oxidative substrates pyruvate (5 mM) and malate (2.5 mM)
before protein extraction.
Western blot analysis
Protein from whole cells was extracted using 1% RIPA
lysis buffer (cat# R0278, Sigma) with standard protease
and phosphatase inhibitors (Cat# P8340, Sigma).
Proteins from subcellular fractions and whole cell lysate
were separated by SDS?PAGE using Criterion 4?20%
Tris?HCl (10?250 kD) gels (Bio-Rad, Hercules, CA).
After transfer to PVDF membranes, and after washing in
TBST buffer containing 0.1% Tween 20, and blocking in
5% milk in TBST, the primary antibodies were diluted in
5% milk in TBST and incubated at 4 ?C for 18 h. Primary
antibodies against pyruvate dehydrogenase (PDH,
1:2,500, Cat#3205), pS727-STAT3 (1:400, Cat#9134),
STAT3 (1:1,000, Cat#12640), pY397-FAK (1:500,
Cat#3283), FAK (1:1,000, Cat#3285), ?-tubulin (1:2,000,
Cat#2125) all from Cell Signaling Technology (Danvers,
MA) and histone 3H3 (1:15,000, ab1791) and ATPase
(1:5,000, ab14730) from Abcam (Cambridge, MA). After
washing in TBST, the membranes were incubated in
species- and isotype-specific HRP-conjugated secondary
antibodies in TBST (1:2000, Cat#7074, Cat#7076, Cell
Signaling). Chemiluminescence (ECL, Cat# 34080,
ThermoFisher) was used to reveal immunoreactive protein
bands, detected by X-ray film or imaged on a Licor
Odyssey FC (Lincoln, NE) for 10 min.
Quantitative capillary ProteinSimple westerns
For one of the FAK inhibition experiments, quantitative
analysis of protein expression was performed according
to the ProteinSimple protocol guide with reagents of a
kit (Cat#, SM-W004 and DM-001, ProteinSimple),
except where noted. Briefly, cell lysates were diluted to
0.2 ?g/?l with 0.1? sample buffer supplemented with 1?
fluorescent molecular weight markers and 40 mM DTT
for a 5 ?l reaction (1 ?g protein/reaction). Samples were
heated at 95 ?C for 5 min before loading into 24 single
designated wells of a pre-filled plate along with blocking
reagent, primary antibodies (1:25, pS727-STAT3, 1:50,
tSTAT3, or 1:500, ?-tubulin and ATPase) in antibody
diluent, with pSTAT3 and tubulin/ATPase or STAT3 and
tubulin/ATPase mixed together for detection within the
same capillary), anti-rabbit HRP conjugated secondary
antibody, luminol-peroxide mix to generate
chemiluminescence and washing buffer. Plates were loaded into the
automated ProteinSimple ?Wes? for electrophoresis and
fluorescence imaged in real time by a CCD camera for
immunodetection in the capillary system at default
settings: Electrophoresis, 375 volts, 25 min; blocking, 5 min;
primary antibody, 30 min; secondary antibody, 30 min.
Data was analyzed using Compass software
(ProteinSimple) and expressed as peak intensity or synthetic bands.
Quantification was performed by normalizing areas under
protein peaks to ?-tubulin loading control.
siRNA and STAT3 plasmid transfections
SiRNAs against mouse FAK (Cat# L-041099,
Dharmacon) or a non-targeting negative control (Cat#
D001810-10, Dharmacon) were transfected into bEnd5
cells using lipofectamine-2000 (Cat #11668, Invitrogen)
24 h after plating cells in Seahorse plates at 12,500 cells/
well, with 50 nM siRNA and 0.5% Lipofectamine
according to manufacturer?s protocol. A second transfection
was performed 24 h later (48 h after plating) due to
FAK?s long protein half-life  and bioenergetics
assessed 5 days later. Transfection efficiency was >80%
as visualized by fluorescent microscopy using siGLO
RNA (Cat #D-001630-02, Dharmacon). Mutant S727A
STAT3 expression plasmid was obtained from Addgene
(Cat# 8708) while wild type mouse STAT3 was PCR
amplified from cDNA and directionally cloned into a
pcDNA 3.1(+) expression vector using BamHI and
EcoRI restriction sites. bEnd5 cells were transfected with
500 ng plasmid per well using lipofectamine 3000 (Cat#
L3000, Invitrogen) and maintained for 24 h prior to Sea
For confocal imaging, bEnd5 cells were seeded at 105
cells/35 mm clear bottom culture dishes
(P35G-1.5?14c, MarTek Corp, Ashland, MA). After 24 h, cells were
treated with FAK14 (5 or 10 ?M) for 4 h, followed by
30 min incubation with 20 nM tetramethylrhodamine
methyl and ethylesters, which accumulate within active
mitochondria in a potentiometric fashion (TMRM,
Cat#T668, Invitrogen) and 20 ?M Hoechst 33342 (Cat #
62249, ThermoFisher Scientific) dyes. The cells were
imaged for TMRM red fluorescence (544/590 nm) and
nuclear Hoechst blue fluorescence (460/490 nm) using a
Leica SP8 confocal microscope.
For spectrofluorometry, bEnd5 cells were grown for
24 h at 10,000/well in 96 wells (Cat#3997, Corning) and
treated with FAK14 for 4 h. Afterwards, mitochondrial
membrane potential was estimated by incubation with 25
nM TMRM and 300 nM DAPI (Dilactate, Invitrogen), to
normalize for the number of cells, for 30 min at 37 ?C.
Fluorescence (TMRM: 544/590, DAPI 360/450 nm
excitation/emission) was measured with a plate reader. O?2?
formation was measured by incubation in 10 ?M
2-,7dichlorodihydrofluorescein diacetate (DCFH2-DA, Cat#
D6883, Sigma) and 5 U/ml horseradish peroxidase (HRP,
CatP8375, Sigma) as described before . Oxidized DCF
fluorescence (485/520 nm) was measured after 10 min at
37 ?C. H2O2 production was measured using 1 ?M
Amplex Red (Cat#A36006, Invitrogen) and 0.25 U/ml
HRP at 30 ?C as described . Formation of fluorescent
resorufin from Amplex Red was measured (530/590 nm).
Statistical analyses were performed with Students t-test
for two samples or with a One Way Analysis of Variance
(ANOVA) followed by a Bonferroni post-hoc test using
GraphPad Software (La Jolla, CA). The number of
experiments indicated in the text are independent
Integrin antagonists inhibit mitochondrial function
BEnd5 endothelial cells were cultured for 4 h in
serumfree conditions on VTN substrate to provide a ligand for
integrins, especially ?v?3 integrin present on endothelial
cells. The tetrapeptide RGDS, which blocks RGD
binding ligands , as well as the P11 peptide which blocks
?v?3 integrin , substantially inhibited mitochondrial
function (Fig. 1a), with significantly reduced ATP
production, maximum reserve capacity and maximum
respiratory capacity (Fig. 1b). The 4 h incubation with
RGDS and P11 did not cause any obvious changes in cell
number (Fig. 1c, d), or protein content (Fig. 1e),
indicating that the effects were not due to cell detachment.
Conversely, maximum reserve and respiratory capacity
were stimulated when integrin agonists VTN, laminin or
fibronectin were added directly to bEnd5 cell culture
medium for 4 h under serum-free conditions (Fig. 1f ).
FAK inhibition causes mitochondrial dysfunction
FAK inhibitor 14 (FAK14, also named Y15:
1,2,4,5-Benzenetetramine tetrahydrochloride, MW = 284) is a
waterFig. 1 RGD and ?v?3 integrin blockade suppresses mitochondrial bioenergetics. a Traces from a representative experiment show mitochondrial
bioenergetic deficits in mouse brain bEnd5 cells plated on vitronectin (VTN), in the presence of RGD integrin (RGDS) or ?v?3 integrin (P11)
blocking peptides, in serum free medium for 4 h, as measured by oxygen consumption rate (OCR) in an XF24 Seahorse Flux Analyzer. Data are
means of 5 wells per condition. b Integrin blockade reduced all four measures of cellular respiration relative to control (no VTN) and VTN treated
cells. c Cells cultured in XF24 microplates for 4 h on no substrate or on VTN, with or without integrin blockade, seemed similar in number but
cells without substrate and in the presence of RGDS were more rounded. The cell number (d) and protein content (e) in such cells, determined
without subsequent Seahorse analyses, were not significantly different across all conditions. Cell counts were independently verified. f Cells
cultured for 4 h with VTN, laminin (LAM) or fibronectin (FN) as substrate and added to serum-free media showed increased mitochondrial
bioenergetics. All data are means ? SEM of 3?4 independent experiments. *p < 0.05, **p < 0.01, and ***p < 0.001
soluble small-molecule FAK inhibitor that directly
inhibits the essential Y397 autophosphorylation and thus,
subsequent total phosphorylation and activation of FAK,
with an IC50 of 1 ?M . FAK14 has high specificity
because it did not inhibit nine other recombinant
kinases, importantly the Pyk2 homologue of FAK, as well
as c-RAF, c-Src, EGFR, VEGFR-3, IGF-1, Met, PDGFR-?,
and PI3K, in an in vitro kinase assay . FAK
phosphorylation is greatly decreased already at 1 ?M in
cultured cells . FAK14 dose-dependently reduced
mitochondrial function, affecting ATP production,
maximum reserve capacity and maximum respiratory capacity
of bEnd5 cells within 4 h (Fig. 2a, b). At 10 ?M, FAK14
completely abolished mitochondrial bioenergetic function.
FAK14 did not affect the number of cells or protein
content when cells were incubated for 4 h in Seahorse
microwells (without performing subsequent bioenergetics
measurements; Fig. 2c, d). SiRNA knockdown of FAK
(siFAK) in bEnd5 cells over a 5 day period (to account for
the >20 h half-life of FAK , and confirmed by westerns
for FAK, not shown) also led to mitochondrial dysfunction
(Fig. 2e). The effect was not as large as seen with FAK14,
perhaps because siRNA mediated knockdown of FAK
reduced protein expression by ~80% over the course of
Fig. 2 FAK inhibition causes mitochondrial dysfunction. a Pharmacological inhibition of the integrin signaling molecule, FAK, with FAK14 treatment of
bEnd5 cells for 4 h produced a robust and dose-dependent suppression of mitochondrial bioenergetic function as shown in a representative OCR
trace. Cells were grown for 44 h without substrate before treatment. b FAK14 causes reductions in basal respiration, ATP production, and maximum
reserve and respiratory capacity. The cell number (c) and protein content (d) were not different across conditions. Cell counts were independently
verified. e SiRNA-mediated knockdown of FAK (siFAK) decreased ATP production, reserve capacity and respiratory capacity relative to a non-targeting
siRNA control (siControl). Data are mean ? SEM from 3 independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ns = not significant versus control
5 days, likely giving cells time to compensate to some
extent as well as retaining a small amount of functional
FAK. FAK14, as a small molecule inhibitor, would block
all molecules with more or less immediate effect.
S727-STAT3 inhibition causes mitochondrial dysfunction
STAT3 was deleted in bEnd5 cells using the
CRISPRCas9 method. Transfection of these cells with WT
plasmids increased bioenergetic function whereas
transfection with S727A mutants was without any effect
(Fig. 3a). STAT3 protein deletion was confirmed by
western blot (Fig. 3b) and the point mutation by
sequencing (Fig. 3c). This shows that S727-STAT3 regulates
mitochondrial function and is consistent with findings of
Stattic (6-nitrobenzo[b]thiophene 1,1-dioxide, MW =
211) is a small molecule inhibitor of STAT3, selectively
blocking STAT3 SH2 domains at an IC50 of 5 ?M,
inhibiting STAT3 activation, dimerization and nuclear
translocation . It has high specificity against STAT3
compared to STAT1, and other transcription factors
(cMyc, Max, Jun) and leads to reduced Y705 and S727
STAT3 (but not JAK1, JAK2, c-Src, AKT, JNK or ERK)
phosphorylation [29, 51, 52]. Stattic caused
mitochondrial dysfunction in bEnd5 cells as early as 2 h (Fig. 3d).
Most of the pS727-STAT3 was present in the nucleus
and mitochondria, as shown by sub-cellular fractionation
(Fig. 3e), consistent with other studies . Stattic
treatment of bEnd5 cells caused a reduction in mitochondrial
pS727-STAT3 within 4 h (Fig. 3f ).
To identify where S727-STAT3 is phosphorylated and
dephosphorylated, freshly isolated bEND5 mitochondria
were treated for 2 h with stattic or the serine/threonine
phosphatase inhibitor, okadaic acid. Stattic did not have
an effect, suggesting that STAT3 is phosphorylated in
the cytoplasm before being imported, as others have
found . Okadaic acid increased pS727-STAT3
(Fig. 3g). A higher molecular weight pS727 band was
visible after treatment with okadaic acid in a position
where a faint band of total STAT3 could be seen after
longer exposure (not shown).
FAK inhibition reduces mitochondrial S727-STAT3
FAK inhibitor FAK14 (Fig. 4a) caused a reduction in
mitochondrial pS727-STAT3 after 4 h. The higher
molecular weight pS727-STAT3 could be detected in the
freshly isolated mitochondria (Fig. 4a) and disappeared
upon FAK inhibition with FAK14. The results were
confirmed by a quantitative capillary western method, as
shown in representative traces (Fig. 4b), which were
converted with software to synthetic bands for presentation
purposes (Fig. 4c), and quantified (Fig. 4d). Another FAK
inhibitor, PF573228, also reduced pS727-STAT3, in
concert with pFAK, as shown in whole cell lysates (Fig. 4e).
Fig. 3 S727-STAT3 inhibition causes mitochondrial dysfunction. a Expression plasmids for wildtype (WT), but not functionally dead S727A mutant,
STAT3 increased bioenergetics function in bEnd5 cells rendered STAT3?/? by a CRISPR-Cas9 method. N = 3. b Deletion of STAT3 protein was
confirmed by western blot most clearly in cells transfected with the G1A guide RNA (G1A and G1B were replicate conditions of the same
Cas9guideRNA plasmid) which was used for the bioenergetics experiment in (A). c DNA sequence data confirmed a point mutation and frame shift in
exon 1. d Pharmacological inhibition of STAT3 with stattic for 2 h in bEnd5 cells reduced mitochondrial bioenergetic measures. n = 3. e
pS727STAT3 is predominantly present in the mitochondria and nucleus as shown in Western blots of protein extracts from sub-cellular fractions.
Markers = H3 protein, nucleus (N), ?-tubulin, cytoplasm (C), PDH, mitochondria (M). W = whole cell lysate. tSTAT3 = total STAT3. f Cells treated for
4 h with the STAT3 inhibitor, stattic (10 ?M), had less pS727-STAT3 in their isolated mitochondria. The blot is representative of three experiments.
g pS727-STAT3 was increased in isolated bEnd5 mitochondria treated with okadaic acid (OA: 1 ?M) in respiration buffer, as detected by western
blotting (3 independent experiments are shown). Stattic did not have any effect. A higher molecular weight band appeared after OA treatment in
the pS727-STAT3 blot. A faint band could be seen in the same position after longer exposure of the total STAT3 blot (not shown)
Treatment with the transcription inhibitor, actinomycin,
for 4 h did not significantly affect mitochondrial
bioenergetics, nor the effects of FAK inhibition (Fig. 4f ),
consistent with the known non-transcriptional role of
pS727-STAT3 . The results so far suggest that
integrin signaling through FAK promotes mitochondrial
function by activating STAT3 on S727.
S727-STAT3 activators rescue mitochondrial function and
cells against FAK inhibition
Bryostatin-1 and HGF increase phosphorylation of
S727STAT3 via PKC, and ERK activation, respectively [34, 54].
Here, both mitigated the reduced mitochondrial
maximum reserve and respiratory capacity caused by a 4 h
FAK14 treatment of bEnd5 cells (Fig. 5a). Further, FAK14
dose-dependently decreased pS727-STAT3, which was
prevented by bryostatin and HGF (Fig. 5b). The
substantial changes in pS727-STAT3 are likely not
resulting from changes in cell survival (Fig. 6) because the total
STAT3 and actin loading controls were similar. Bryostatin
could not reduce mitochondrial dysfunction caused by
stattic (Fig. 5c), evidence that its protective effects against
FAK inhibition (Fig. 5a, and below) were mediated by
BEnd5 cell survival was assessed in regular tissue
culture plates. FAK14 dose- and time-dependently caused
loss of bEnd5 cells (Fig. 6a-e), up to more than 50%
(Fig. 6d, e), and may have reduced proliferation between
4 and 24 h (controls in Fig. 6b vs. c and d vs. e). No cell
loss was observed in the bioenergetics experiments
(Fig. 2), probably due to the different culture conditions.
Bryostatin and HGF promoted bEnd5 cell survival
caused by FAK14 treatment at both 4 h (Fig. 6a, b, d)
and 24 h (Fig. 6c, e).
Fig. 4 FAK inhibits mitochondrial S727-STAT3 phosphorylation. a A 4 h FAK14 treatment of bEnd5 cells reduced pS727-STAT3 in both the
mitochondrial and cytoplasmic fractions. Blots are representative for 5 experiments. b This reduction was confirmed by quantitative capillary
western blotting with representative chemiluminescent spectrograms and synthetic bands (c). d Quantitation was performed of spectrograms
confirmed a clear and significant decrease in pS727-STAT3 following 4 h FAK14 treatment in the mitochondrial fractions (n = 3). e Treatment with
another more lipophilic FAK antagonist (PF573228: PF at 10 or 20 ?M) for 4 or 8 h showed decreases in pS727-STAT3 in conjunction with
decreased pFAK in whole cell lysates. f Incubation with the global transcriptional inhibitor actinomycin D (0.3 ?g/ml, 4 h) did not significantly
change mitochondrial bioenergetics under control or FAK14 conditions
FAK inhibition causes mitochondrial depolarization and
A normal mitochondrial membrane potential (??m) is
essential for ATP generation . FAK14 caused a
dosedependent reduction in TMRM fluorescent signal, a
measure of ??m, in bEnd5 cells after 4 h as shown by
confocal images (Fig. 7a) and spectrofluorometry of cell
extracts (Fig. 7b). At 10 ?M, FAK14 caused a complete
loss of membrane potential to levels seen with the FCCP
uncoupler of mitochondrial oxidative phosphorylation.
Bryostatin increased mitochondrial polarization, as
assessed with TMRM, in a dose-dependent manner in
the presence of FAK14, but did not have an effect by
itself (Fig. 7c). Mitochondrial failure can be induced by
excessive ROS formation . A 4 h treatment with
FAK14 dose-dependently increased superoxide (Fig. 7d)
and H2O2 (Fig. 7e) formation. Bryostatin did not affect
this FAK14-induced ROS formation (data not shown).
The present study demonstrates that
ECM?integrinFAK signaling regulates mitochondrial bioenergetics via
pS727-STAT3. This adds an important component to
what was known about integrin signaling functions.
ECM-integrin binding is important for cell adhesion and
regulates actin organization, cell movement and cell
cycle control . Integrins can play an important role
in cell survival, especially of cells that are attached to
Fig. 5 pS727-STAT3 stimulation preserves mitochondrial bioenergetics against FAK inhibition. a Bryostatin and HGF preserved reserve and
respiratory capacity in bEnd5 cells incubated with FAK14 (5 ?M, F5) for 4 h. Data are mean ? SEM from 3 independent experiments. *p < 0.05, **p
< 0.01, **p < 0.001. b Western blots of whole cell lysates showing that bryostatin (BST) and HGF preserve pS727-STAT3 in the presence of FAK14
(representative for 3 independent experiments). c Bryostatin does not affect the mitochondrial bioenergetics dysfunction caused by STAT3
inhibition with stattic. N = 3
basement membranes, such as endothelial cells which
undergo anoikis when attachment is disrupted .
Disrupted integrin signaling can induce apoptosis by activating
the pro-apoptotic Bad and Bax or reducing anti-apoptotic
Bcl-2 [3, 58]. RGDS peptide can trigger pro-apoptotic
induction and activation of caspase 8 and 9 in human
endothelial cells . The P11 antagonist of ?v?3 integrin can
cause apoptosis of human umbilical vein endothelial cells
by up-regulating p53 expression resulting in caspase
activation . Moreover, lengthy low bioenergetics leads
to pathological mitochondrial pore formation and
cytochrome-C release thus initiating apoptotic death [17,
18, 61]. Our data suggest that the integrin-FAK-STAT3
pathway is involved in suppressing such cell death
Fig. 6 Bryostatin and HGF reduce bEND5 cell death due to FAK inhibition. a Images are representative of bEnd5 cells grown for 4 h with 5 ?M
FAK14 +/?bryostatin (BST) in regular tissue culture wells. b to e) Cell counts show that bryostatin and HGF can promote survival despite FAK
inhibition with 5 or 10 ?M FAK14 for 4 or 24 h. Data are mean ? SEM from 3 independent experiments. *p < 0.05, **p < 0.01 and ***p < 0.001
Fig. 7 FAK inhibition depresses mitochondrial membrane potential and increases ROS production. a FAK inhibition with FAK14 for 4 h caused a
dose-dependent reduction in mitochondrial membrane potential, as shown by confocal images of red TMRM fluorescence in bEnd5 cells. Cells
with complete mitochondrial failure are indicated by arrowheads and detectable only by blue nuclear Hoechst staining. b Spectrometry of TMRM
in extracts was used to quantify the effects, and shows that 10 ?M FAK14 (FAK10) reduced the membrane potential to the same level as FCCP.
Values were normalized to nuclear DAPI staining to account for differences in cell numbers. Data are mean ? SEM from 3 independent
experiments. *p < 0.05, **p < 0.01, ***p < 0.001. c Bryostatin (BST) dose-dependently preserved mitochondrial membrane potential in the presence
of FAK14. N = 3. FAK14 dose-dependently induced formation of d) superoxide (O.2-) and e) H2O2. N = 3
mechanisms by maintaining mitochondrial function and
integrity. This is consistent with the finding that
fibronectin knockdown induces apoptosis in rat mesangial cells in
a mitochondria-dependent manner, mainly as a result of
cytochrome c release and downstream caspase-3 and ?9
activation . Conversely, laminin and fibronectin protect
pancreatic cancer cells from death by mechanisms
involving the inhibition of both mitochondrial depolarization
and caspase activity . Previously, we have shown that
an ?v?3 agonist peptide can rescue endothelial cells after
contusive spinal cord injury in adult mice .
Integrin-FAK signaling also can regulate mitochondrial
biogenesis and morphology. FAK inactivation in
cardiomyocytes causes structural abnormalities in mitochondria
. Moreover, FAK interacts with mitochondrial
transcriptional cascades and enhances mechanical
stressinduced mitochondrial biogenesis . It remains to be
determined how the FAK-STAT3 pathway affects
biogenesis and morphology.
Most of our experiments were conducted with VTN as
an integrin ligand. Its effects on mitochondrial function
were mediated at least by ?v?3 integrin as shown by the
inhibitory effects of the P11 peptide. Fibronectin and
laminin also had a stimulating effect, consistent with their
role in promoting endothelial cell survival [3, 67, 68],
whereas RGD inhibition reduced bioenergetic function.
Other ECM molecules also utilize RGD integrins and it
remains to be determined whether additional basement
membrane or ECM molecules  are involved in
regulating mitochondrial function. Vitronectin is present in the
basement membrane but also at high concentrations in
the blood [70, 71]. It will be interesting to determine the
role of plasma VTN on mitochondrial function or whether
integrin activation is predominantly ab-luminal, luminal,
or both in the vasculature. FAK may be predominantly
clustered with the VTN receptor on the ab-luminal side of
endothelial cells, where they would make contact with the
basement membrane . If so, it is possible that
pathological VTN leakage from the blood would promote
survival of endothelial cells.
Our data suggest that FAK activates S727-STAT3. It
remains to be determined which pathways downstream
of FAK might be involved, but could include the serine
kinases JNK1  or ERK . The latter is also
activated by HGF . The bryostatin results suggest that
one or more of the PKC kinases can activate STAT3
even under reduced FAK signaling conditions. FAK can
also be activated by growth factor receptors and
Gprotein-linked receptors [74, 75] raising the possibility
that they too could promote to mitochondrial function.
The ECM may be involved in orchestrating these
miscellaneous extracellular regulators. Our acute
pharmacological approach with FAK14 and stattic showed a
remarkably complete cessation of mitochondrial
bioenergetic function within 4 h. A mitochondrial role for
STAT3 has been convincingly demonstrated with STAT3
S727A mutants suppressing bioenergetic function by ~
50-70% . Our results, validate these findings by
confirming that expression of STAT3 S727A mutants
cannot increase bioenergetic function in STAT3
knockout cells using CRISPR-Cas9, whereas wild type STAT3
controls can. Other pathways that regulate STAT3 pS727
activation might also be involved under more
physiological conditions. For example, CNTF can promote
mitochondrial function through NFKB expression and activity
in dorsal root ganglion neurons . It remains to be
determined whether this effect is fully
transcriptiondependent or might also involve S727-STAT3 activation.
The mechanisms and roles of STAT3 within the
mitochondria are not well-understood. Our results with
actinomycin are consistent with the known
nontranscriptional nature of the role of STAT3 in
mitochondrial function [24, 26], and support a non-transcriptional
effect of FAK inhibition on mitochondria. In an apparent
contradiction, in keratinocytes, S727-STAT3 can repress
transcription of mitochondrial ETC genes, but no
functional measures were reported . Our finding that
pS727-STAT3 is mainly found in the mitochondrial and
nuclear fractions, but less in the cytoplasm, is consistent
with the findings that STAT3 translocates upon S727
phosphorylation in the cytoplasm . Moreover,
treatment of isolated mitochondria with stattic, known to
reduce phosphorylation and translocation , did not
reduce mitochondrial pS727-STAT3. This suggests that
the effects of integrin or FAK inhibition are mediated by
a reduced translocation to the mitochondria, which may
be dependent on chaperones such as HSPs [78, 79]. We
also find for the first time that dephosphorylation of
S727-STAT3 occurs within the mitochondria, as isolated
mitochondria treated for 2 h with the serine/threonine
phosphatase inhibitor, okadaic acid, had more
pS727STAT3. The rapidity suggests that mitochondrial
pS727STAT3 is short-lived and its constant import dependent
on external stimuli which activate FAK signaling. This
would represent a mechanism for rapid adjustments to
cellular energy demands. Okadaic acid is known to
inhibit PP2A (and PP1) which has been associated with
pS727-STAT3 dephosphorylation in the cytoplasm .
PP2A has been found in mitochondria  and has been
linked to mitochondrial bioenergetics by promoting
mitochondrial division through Drp1 dephosphorylation .
We detected an ~100 kDa pS727-STAT3 in okadaic
acid-treated or freshly isolated mitochondria. This form
may also have been found in avian neurons upon CNTF
stimulation, showing a similar short-lived nature and
dependency on persistent signaling . A higher
molecular weight STAT3 can also be detected in Cos1 cells
(Fig. 4 in  and splenocytes (Fig. 3 in ). Our
results suggest that only a very minor fraction of
mitochondrial STAT3 exists in this form and might be newly
imported STAT3 which is post-translationally modified.
Mitochondrial STAT3 is thought to maintain
optimal oxidative phosphorylation, by interacting with
ETC complexes I, II, and V, and inhibiting formation
of the mitochondrial permeability transition pore
which would lead to loss of mitochondrial membrane
potential [26, 29]. Our results suggest that
liganddependent integrin-FAK activation of S727-STAT3 is
an important mechanism which maintains a
physiological mitochondrial function. Together, our data
support the idea that the integrin-FAK signaling
pathway, through STAT3, prevents the loss of
mitochondrial membrane potential and decreased ATP
synthesis, which are key mediators of mitochondrial
dysfunction and subsequent triggers for release of
pro-apoptotic factors and cell death .
Our results also suggest that this pathway promotes
the potential antioxidant role of STAT3. In this scenario,
STAT3 binds to complex I iron sulfur clusters with its
cysteines being oxidized to serve as electron donors,
thus reducing ROS generation [30, 31]. Indeed, STAT3
can reduce ROS formation in cardiomyocytes ,
preventing excessive ROS levels which can lead to cell death
. Integrins and ECM molecules have been found to
reduce [86, 87] or promote [88, 89] ROS production. It
will be interesting to test whether pS727-STAT3 is
differentially regulated under those conditions or in those
cells. Mitochondrial ROS are also involved in
physiological processes, including proliferation, differentiation
and migration . It remains to be determined whether
the integrin-STAT3 pathway plays a role in these
physiological functions. Our finding that bryostatin did not
reduce superoxide and H2O2 formation caused by FAK
inhibition may explain why it was not fully effective in
rescuing mitochondrial function or cell survival, despite
restoring normal pS727-STAT3 levels. This raises the
possibility that FAK can regulate mitochondrial function
through other pathways.
Our study also explored a potential therapeutic
strategy for endothelial cells with dysfunctional integrin
signaling such as is seen after traumatic detachment, by
circumventing integrin-FAK signaling through direct
STAT3 activation. Our results are proof of principle that
this can be achieved, also because we have seen similar
decreases in bioenergetics, pS727-STAT3, and
mitochondrial membrane potential after FAK inhibitor
treatment in primary endothelial cells derived from adult
mouse brain (Visavadiya et al., unpublished results).
However, although it remains to be seen whether this
therapeutic approach can be optimized and applied to
animal models of disease. This is a reasonable
expectation because bryostatin is being explored as a treatment
for Alzheimer?s disease  and HGF is in clinical trials
for spinal cord injury . This new therapeutic strategy
could have implications for disorders where endothelial
cells are disturbed, e.g., to maintain proper BBB integrity
during neurological or other insults. FAK inhibitors are
in clinical trials for solid tumors [91, 92] due to their
inhibition of angiogenesis and cell adhesion [49, 93]. Our
results suggest that mitochondrial dysfunction caused by
such agents could also explain some of the effects. In
apparent contrast to our proposed effects of FAK
inhibition on bioenergetics function and ROS, mitochondrial
ROS could trigger gastric cancer progression by
stimulating cancer cell migration via the ?5-integrin induction
. Also, mitochondrial bioenergetic dysfunction
caused by mitochondrial DNA mutations leads to
increased motility and migration in cytoplasmic hybrid
cells ?cybrid cells? which harbor A3243T mutation in the
leucine transfer RNA gene through a mechanism
involving ECM molecules . It remains to be determined
whether the integrin-FAK-STAT3 pathway plays different
roles in different cell types, and whether it is absent in
cancer cells which defeat anoikis to metastasize.
In conclusion, our data suggest that integrin-FAK
signaling maintains mitochondrial bioenergetic function by
maintaining membrane potential and ETC function, thus
reducing ROS formation, via pS727-STAT3. Further, we
propose a treatment strategy that can circumvent
dysfunction of integrin signaling which may help to
develop new neuroprotective treatments. Our finding
also provides a platform to investigate fundamental
mechanisms of how ligand binding to integrins
modulate physiological mitochondrial bioenergetic function in
ACT: Actinomycin D; AKT: v-Akt murine thymoma viral oncogene/protein
kinase B (PKB); ANOVA: One way analysis of variance; ATP: Adenosine
triphosphate; ATPase: Adenosine triphosphate synthase; BBB: Blood?brain
barrier; Bcl-2: B-cell lymphoma 2; bEnd5: Mouse brain endothelioma cell line;
bit1: Bcl-2 inhibitor of transcription; BST: Bryostatin; Cas9: CRISPR associated
protein 9; CNS: Central nervous system; CNTF: Ciliary neurotrophic factor;
CRISPR: Clustered regularly interspaced short palindromic repeats; DAPI:
4?,6diamidino-2-phenylindole; DCFH2-DA: 2-,7-dichlorodihydrofluorescein
diacetate; DMEM: Dulbecco?s modified eagle medium;
DNA: Deoxyribonucleic acid; ECL: Enhanced chemiluminescence;
ECM: Extracellular matrix; EGTA: Ethylene glycol-bis(?-aminoethyl
ether)N,N,N?,N?-tetraacetic acid; ERK: Extracellular signal-regulated kinases;
ETC: Electron transport chain; F10: FAK14 (10 ?M); F5: FAK14 (5 ?M);
FAK: Focal adhesion kinase; FAK14: 1,2,4,5-Benzenetetramine
tetrahydrochloride; FCCP: Carbonyl cyanide 4-(trifluoromethoxy)
phenylhydrazone; FN: Fibronectin; H2O2: Hydrogen peroxide; H3: Histone H3;
HEPES: 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; HGF: Hepatocyte
growth factor; HRP: Horseradish peroxidase; HSPs: Heat shock proteins;
JNK: c-Jun N-terminal kinases; KCl: Potassium chloride; KH2PO4: Potassium
phosphate monobasic; KOH: Potassium hydroxide; LAM: Laminin;
MgCl2: Magnesium chloride; NFKB: Nuclear factor kappa B; O.2-: Superoxide;
OA: Okadaic acid; OCR: Oxygen consumption rate; PDH: Pyruvate
dehydrogenase; PF10: PF573228 (10 ?M); PF20: PF573228 (20 ?M);
3,4-Dihydro-6-[[4-[[[3-(methylsulfonyl)phenyl]methyl]amino]-5(trifluoromethyl)-2-pyrimidinyl]amino]-2(1H)-quinolinon; PKC: Protein kinase C;
pS727-STAT3: Phosphorylated serine 727 residue of STAT3;
PVDF: Polyvinylidene difluoride; pY397-FAK: Phosphorylated tyrosine 397
residue of FAK; RGD: Arg-Gly-Asp; RGDS: Arg-Gly-Asp-Ser;
RIPA: Radioimmunoprecipitation assay; RNA: Ribonucleic acid; ROS: Reactive
oxygen species; S5: Stattic (5 ?M); S727-STAT3: Serine 727 residue of STAT3;
SDS-PAGE: Sodium dodecyl sulfate-polyacrylamide gel electrophoresis;
siFAK: Small interfering RNA for FAK; siRNA: Small interfering RNA;
STAT3: Signal transducer and activator of transcription 3; TBST: Tris-buffered
saline tween 20; TMRM: Tetramethylrhodamine, methyl ester; tSTAT3: Total
signal transducer and activator of transcription 3; VTN: Vitronectin;
?v?3: alpha v beta 3; ??m: Membrane potential
NPV, MPK, VR, KB, CL, CJ and TH performed experiments and analyzed and
interpreted data. MPK and TH conceived the overall idea. NPV, MPK and TH
wrote the paper. GLW provided intellectual input as well as technical
guidance for the Seahorse experiments. All authors read and approved the
Received: 2 September 2016 Accepted: 6 December 2016
1. Giancotti FG , Ruoslahti E. Integrin signaling. Science . 1999 ; 285 : 1028 - 32 .
2. Hynes RO . The extracellular matrix: not just pretty fibrils . Science . 2009 ; 326 : 1216 - 9 .
3. Meredith JEJ , Schwartz MA. Integrins, adhesion and apoptosis . Trends Cell Biol . 1997 ; 7 : 146 - 50 .
4. Brooks PC , Montgomery AM , Rosenfeld M , Reisfeld RA , Hu T , Klier G , et al. Integrin alpha v beta 3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels . Cell . 1994 ; 79 : 1157 - 64 .
5. Wang J , Milner R. Fibronectin promotes brain capillary endothelial cell survival and proliferation through alpha5beta1 and alphavbeta3 integrins via MAP kinase signalling . J Neurochem . 2006 ; 96 : 148 - 59 .
6. Del Zoppo GJ , Milner R. Integrin-matrix interactions in the cerebral microvasculature . Arterioscler Thromb Vasc Biol . 2006 ; 26 : 1966 - 75 .
7. Doll DN , Hu H , Sun J , Lewis SE , Simpkins JW , Ren X. Mitochondrial crisis in cerebrovascular endothelial cells opens the blood-brain barrier . Stroke . 2015 ; 46 : 1681 - 9 .
8. Baeten KM , Akassoglou K. Extracellular matrix and matrix receptors in bloodbrain barrier formation and stroke . Dev Neurobiol . 2011 ; 71 : 1018 - 39 .
9. Toutant M , Costa A , Studler J-M , Kadar? G , Carnaud M , Girault J-A. Alternative splicing controls the mechanisms of FAK autophosphorylation . Mol Cell Biol . 2002 ; 22 : 7731 - 43 .
10. Frisch SM , Vuori K , Ruoslahti E , Chan-Hui PY. Control of adhesion-dependent cell survival by focal adhesion kinase . J Cell Biol . 1996 ; 134 : 793 - 9 .
11. de Castro IP , Martins LM , Tufi R. Mitochondrial quality control and neurological disease: an emerging connection . Expert Rev Mol Med . 2010 ; 12 :e12.
12. Harbauer AB , Zahedi RP , Sickmann A , Pfanner N , Meisinger C. The protein import machinery of mitochondria-a regulatory hub in metabolism, stress, and disease . Cell Metab . 2014 ; 19 : 357 - 72 .
13. Oldendorf WH , Cornford ME , Brown WJ . The large apparent work capability of the blood-brain barrier: a study of the mitochondrial content of capillary endothelial cells in brain and other tissues of the rat . Ann Neurol . 1977 ; 1 : 409 - 17 .
14. Grammas P , Martinez J , Miller B. Cerebral microvascular endothelium and the pathogenesis of neurodegenerative diseases . Expert Rev Mol Med . 2011 ; 13 :e19.
15. Sullivan PG , Rabchevsky AG , Waldmeier PC , Springer JE. Mitochondrial permeability transition in CNS trauma: cause or effect of neuronal cell death ? J Neurosci Res . 2005 ; 79 : 231 - 9 .
16. Hall ED , Wang JA , Bosken JM , Singh IN. Lipid peroxidation in brain or spinal cord mitochondria after injury . J Bioenerg Biomembr . 2016 ; 48 : 169 - 74 .
17. Murphy AN , Fiskum G , Beal MF . Mitochondria in neurodegeneration: bioenergetic function in cell life and death . J Cereb Blood Flow Metab . 1999 ; 19 : 231 - 45 .
18. Mattson MP , Kroemer G . Mitochondria in cell death: novel targets for neuroprotection and cardioprotection . Trends Mol Med . 2003 ; 9 : 196 - 205 .
19. Ruoslahti E , Reed JC. Anchorage dependence, integrins, and apoptosis. Cell . 1994 ; 77 : 477 - 8 .
20. Lu Q , Rounds S. Focal adhesion kinase and endothelial cell apoptosis . Microvasc Res . 2012 ; 83 : 56 - 63 .
21. Jan Y , Matter M , Pai J-T , Chen Y-L , Pilch J , Komatsu M , et al. A mitochondrial protein, Bit1, mediates apoptosis regulated by integrins and Groucho/TLE corepressors. Cell . 2004 ; 116 : 751 - 62 .
22. Keasey MP , Kang SS , Lovins C , Hagg T. Inhibition of a novel specific neuroglial integrin signaling pathway increases STAT3-mediated CNTF expression . Cell Commun Signal . 2013 ; 11 : 35 .
23. Lim CP , Cao X. Serine phosphorylation and negative regulation of Stat3 by JNK . J Biol Chem . 1999 ; 274 : 31055 - 61 .
24. Wegrzyn J , Potla R , Chwae Y-J , Sepuri NBV , Zhang Q , Koeck T , et al. Function of mitochondrial Stat3 in cellular respiration . Science . 2009 ; 323 : 793 - 7 .
25. Boengler K , Ungefug E , Heusch G , Schulz R. The STAT3 inhibitor stattic impairs cardiomyocyte mitochondrial function through increased reactive oxygen species formation . Curr Pharm Des . 2013 ; 19 : 6890 - 5 .
26. Gough DJ , Corlett A , Schlessinger K , Wegrzyn J , Larner AC , Levy DE . Mitochondrial STAT3 supports Ras-dependent oncogenic transformation . Science . 2009 ; 324 : 1713 - 6 .
27. Phillips D , Reilley MJ , Aponte AM , Wang G , Boja E , Gucek M , et al. Stoichiometry of STAT3 and mitochondrial proteins: Implications for the regulation of oxidative phosphorylation by protein-protein interactions . J Biol Chem . 2010 ; 285 : 23532 - 6 .
28. Han J , Yu C , Souza RF , Theiss AL. Prohibitin 1 modulates mitochondrial function of Stat3 . Cell Signal . 2014 ; 26 : 2086 - 95 .
29. Boengler K , Hilfiker-Kleiner D , Heusch G , Schulz R. Inhibition of permeability transition pore opening by mitochondrial STAT3 and its role in myocardial ischemia/reperfusion . Basic Res Cardiol . 2010 ; 105 : 771 - 85 .
30. Szczepanek K , Chen Q , Derecka M , Salloum FN , Zhang Q , Szelag M , et al. Mitochondrial-targeted Signal transducer and activator of transcription 3 (STAT3) protects against ischemia-induced changes in the electron transport chain and the generation of reactive oxygen species . J Biol Chem . 2011 ; 286 : 29610 - 20 .
31. Zhang Q , Raje V , Yakovlev VA , Yacoub A , Szczepanek K , Meier J , et al. Mitochondrial localized Stat3 promotes breast cancer growth via phosphorylation of serine 727 . J Biol Chem . 2013 ; 288 : 31280 - 8 .
32. Xuan Y-T , Guo Y , Zhu Y , Wang O-L , Rokosh G , Messing RO , et al. Role of the protein kinase C-epsilon-Raf-1-MEK-1/2-p44/42 MAPK signaling cascade in the activation of signal transducers and activators of transcription 1 and 3 and induction of cyclooxygenase-2 after ischemic preconditioning . Circulation . 2005 ; 112 : 1971 - 8 .
33. Aziz MH , Manoharan HT , Sand JM , Verma AK . Protein kinase cepsilon interacts with Stat3 and regulates its activation that is essential for the development of skin cancer . Mol Carcinog . 2007 ; 46 : 646 - 53 .
34. Nakagami H , Morishita R , Yamamoto K , Taniyama Y , Aoki M , Matsumoto K , et al. Mitogenic and antiapoptotic actions of hepatocyte growth factor through ERK, STAT3, and AKT in endothelial cells . Hypertension . 2001 ; 37 : 581 - 6 .
35. Borowiak M , Garratt AN , Wustefeld T , Strehle M , Trautwein C , Birchmeier C. Met provides essential signals for liver regeneration . Proc Natl Acad Sci U S A . 2004 ; 101 : 10608 - 13 .
36. Kraft AS , Smith JB , Berkow RL . Bryostatin, an activator of the calcium phospholipid-dependent protein kinase, blocks phorbol ester-induced differentiation of human promyelocytic leukemia cells HL-60 . Proc Natl Acad Sci U S A . 1986 ; 83 : 1334 - 8 .
37. Sun M-K , Nelson TJ , Alkon DL . Towards universal therapeutics for memory disorders . Trends Pharmacol Sci . 2015 ; 36 : 384 - 94 .
38. Sakai K , Aoki S , Matsumoto K. Hepatocyte growth factor and Met in drug discovery . J Biochem . 2015 ; 157 : 271 - 84 .
39. Steiner O , Coisne C , Engelhardt B , Lyck R. Comparison of immortalized bEnd5 and primary mouse brain microvascular endothelial cells as in vitro blood-brain barrier models for the study of T cell extravasation . J Cereb Blood Flow Metab . 2011 ; 31 : 315 - 27 .
40. Yang T , Roder KE , Abbruscato TJ . Evaluation of bEnd5 cell line as an in vitro model for the blood-brain barrier under normal and hypoxic/aglycemic conditions . J Pharm Sci . 2007 ; 96 : 3196 - 213 .
41. Wagner EF , Risau W. Oncogenes in the study of endothelial cell growth and differentiation . Semin Cancer Biol . 1994 ; 5 : 137 - 45 .
42. Shabalina IG , Jacobsson A , Cannon B , Nedergaard J. Native UCP1 displays simple competitive kinetics between the regulators purine nucleotides and fatty acids . J Biol Chem . 2004 ; 279 : 38236 - 48 .
43. Dranka BP , Benavides GA , Diers AR , Giordano S , Zelickson BR , Reily C , et al. Assessing bioenergetic function in response to oxidative stress by metabolic profiling . Free Radic Biol Med . 2011 ; 51 : 1621 - 35 .
44. Keasey MP , Lemos RR , Hagg T , Oliveira JRM . Vitamin-D receptor agonist calcitriol reduces calcification in vitro through selective upregulation of SLC20A2 but not SLC20A1 or XPR1 . Sci Rep . 2016 ; 6 : 25802 . Nature Publishing Group.
45. Visavadiya NP , Patel SP , VanRooyen JL , Sullivan PG , Rabchevsky AG . Cellular and subcellular oxidative stress parameters following severe spinal cord injury . Redox Biol . 2015 ; 8 : 59 - 67 .
46. Ochel HJ , Schulte TW , Nguyen P , Trepel J , Neckers L. The benzoquinone ansamycin geldanamycin stimulates proteolytic degradation of focal adhesion kinase . Mol Genet Metab . 1999 ; 66 : 24 - 30 .
47. Plow EF , Pierschbacher MD , Ruoslahti E , Marguerie GA , Ginsberg MH . The effect of Arg-Gly-Asp-containing peptides on fibrinogen and von Willebrand factor binding to platelets . Proc Natl Acad Sci U S A . 1985 ; 82 : 8057 - 61 .
48. Choi Y , Kim E , Lee Y , Han MH , Kang I-C. Site -specific inhibition of integrin alpha v beta 3-vitronectin association by a ser-asp-val sequence through an Arg-Gly-Asp-binding site of the integrin . Proteomics . 2010 ; 10 : 72 - 80 .
49. Golubovskaya VM , Nyberg C , Zheng M , Kweh F , Magis A , Ostrov D , et al. A small molecule inhibitor , 1 , 2 , 4 , 5 - benzenetetraamine tetrahydrochloride, targeting the y397 site of focal adhesion kinase decreases tumor growth . J Med Chem . 2008 ; 51 : 7405 - 16 .
50. Hochwald SN , Nyberg C , Zheng M , Zheng D , Wood C , Massoll NA , et al. A novel small molecule inhibitor of FAK decreases growth of human pancreatic cancer . Cell Cycle . 2009 ; 8 : 2435 - 43 .
51. Schust J , Sperl B , Hollis A , Mayer TU , Berg T. Stattic : a small-molecule inhibitor of STAT3 activation and dimerization . Chem Biol . 2006 ; 13 : 1235 - 42 .
52. Li C , Iness A , Yoon J , Grider JR , Murthy KS , Kellum JM , et al. Noncanonical STAT3 activation regulates excess TGF-beta1 and collagen I expression in muscle of stricturing Crohn's disease . J Immunol . 2015 ; 194 : 3422 - 31 .
53. Tammineni P , Anugula C , Mohammed F , Anjaneyulu M , Larner AC , Sepuri NBV . The import of the transcription factor STAT3 into mitochondria depends on GRIM-19, a component of the electron transport chain . J Biol Chem . 2013 ; 288 : 4723 - 32 .
54. Sud N , Kumar S , Wedgwood S , Black SM . Modulation of PKCdelta signaling alters the shear stress-mediated increases in endothelial nitric oxide synthase transcription: role of STAT3 . Am J Physiol Lung Cell Mol Physiol . 2009 ; 296 : L519 - 26 .
55. Dimroth P , Kaim G , Matthey U. Crucial role of the membrane potential for ATP synthesis by F(1)F(o) ATP synthases . J Exp Biol . 2000 ; 203 : 51 - 9 .
56. Orrenius S , Gogvadze V , Zhivotovsky B. Mitochondrial oxidative stress: implications for cell death . Annu Rev Pharmacol Toxicol . 2007 ; 47 : 143 - 83 .
57. Mahabeleshwar GH , Byzova TV . Vascular integrin signaling . Meth Enzymol . 2008 ; 443 : 199 - 226 .
58. Stupack DG , Cheresh DA . Get a ligand, get a life: integrins, signaling and cell survival . J Cell Sci . 2002 ; 115 : 3729 - 38 .
59. Aguzzi MS , Giampietri C , De Marchis F , Padula F , Gaeta R , Ragone G , et al. RGDS peptide induces caspase 8 and caspase 9 activation in human endothelial cells . Blood . 2004 ; 103 : 4180 - 7 .
60. Bang J-Y , Kim E-Y , Kang D-K , Chang S-I , Han MH , Baek K-H , et al. Pharmacoproteomic analysis of a novel cell-permeable peptide inhibitor of tumor-induced angiogenesis . Mol Cell Proteomics . 2011 ; 10 :M110.005264.
61. Gogvadze V , Orrenius S , Zhivotovsky B. Multiple pathways of cytochrome c release from mitochondria in apoptosis . Biochim Biophys Acta . 2006 ; 1757 : 639 - 47 .
62. Wu D , Chen X , Guo D , Hong Q , Fu B , Ding R , et al. Knockdown of fibronectin induces mitochondria-dependent apoptosis in rat mesangial cells . J Am Soc Nephrol . 2005 ; 16 : 646 - 57 .
63. Vaquero EC , Edderkaoui M , Nam KJ , Gukovsky I , Pandol SJ , Gukovskaya AS . Extracellular matrix proteins protect pancreatic cancer cells from death via mitochondrial and nonmitochondrial pathways . Gastroenterology . 2003 ; 125 : 1188 - 202 .
64. Han S , Arnold SA , Sithu SD , Mahoney ET , Geralds JT , Tran P , et al. Rescuing vasculature with intravenous angiopoietin-1 and alphavbeta3 integrin peptide is protective after spinal cord injury . Brain . 2010 ; 133 : 1026 - 42 .
65. Peng X , Kraus MS , Wei H , Shen T-L , Pariaut R , Alcaraz A , et al. Inactivation of focal adhesion kinase in cardiomyocytes promotes eccentric cardiac hypertrophy and fibrosis in mice . J Clin Invest . 2006 ; 116 : 217 - 27 .
66. Tornatore TF , Dalla Costa AP , Clemente CFMZ , Judice C , Rocco SA , Calegari VC , et al. A role for focal adhesion kinase in cardiac mitochondrial biogenesis induced by mechanical stress . Am J Physiol Heart Circ Physiol . 2011 ; 300 : H902 - 12 .
67. Isik FF , Gibran NS , Jang YC , Sandell L , Schwartz SM . Vitronectin decreases microvascular endothelial cell apoptosis . J Cell Physiol . 1998 ; 175 : 149 - 55 .
68. Davis GE , Senger DR . Endothelial extracellular matrix: biosynthesis, remodeling, and functions during vascular morphogenesis and neovessel stabilization . Circ Res . 2005 ; 97 : 1093 - 107 .
69. Naba A , Clauser KR , Ding H , Whittaker CA , Carr SA , Hynes RO . The extracellular matrix: tools and insights for the ?omics? era . Matrix Biol . 2016 ; 49 : 10 - 24 .
70. Seiffert D , Schleef RR . Two functionally distinct pools of vitronectin (Vn) in the blood circulation: identification of a heparin-binding competent population of Vn within platelet alpha-granules . Blood . 1996 ; 88 : 552 - 60 .
71. Seiffert D , Keeton M , Eguchi Y , Sawdey M , Loskutoff DJ . Detection of vitronectin mRNA in tissues and cells of the mouse . Proc Natl Acad Sci U S A . 1991 ; 88 : 9402 - 6 .
72. Li S , Kim M , Hu YL , Jalali S , Schlaepfer DD , Hunter T , et al. Fluid shear stress activation of focal adhesion kinase . Linking to mitogen-activated protein kinases . J Biol Chem . 1997 ; 272 : 30455 - 62 .
73. Short SM , Talbott GA , Juliano RL . Integrin-mediated signaling events in human endothelial cells . Mol Biol Cell . 1998 ; 9 : 1969 - 80 .
74. Sieg DJ , Hauck CR , Ili? D , Klingbeil CK , Schaefer E , Damsky CH , et al. FAK integrates growth-factor and integrin signals to promote cell migration . Nat Cell Biol . 2000 ; 2 : 249 - 56 .
75. Streblow DN , Vomaske J , Smith P , Melnychuk R , Hall L , Pancheva D , et al. Human cytomegalovirus chemokine receptor US28-induced smooth muscle cell migration is mediated by focal adhesion kinase and Src . J Biol Chem . 2003 ; 278 : 50456 - 65 .
76. Saleh A , Roy Chowdhury SK , Smith DR , Balakrishnan S , Tessler L , Martens C , et al. Ciliary neurotrophic factor activates NF-?B to enhance mitochondrial bioenergetics and prevent neuropathy in sensory neurons of streptozotocininduced diabetic rodents . Neuropharmacology . 2013 ; 65 : 65 - 73 .
77. Macias E , Rao D , Carbajal S , Kiguchi K , DiGiovanni J. Stat3 binds to mtDNA and regulates mitochondrial gene expression in keratinocytes . J Invest Dermatol . 2014 ; 134 : 1971 - 80 .
78. Young JC , Hoogenraad NJ , Hartl FU . Molecular chaperones Hsp90 and Hsp70 deliver preproteins to the mitochondrial import receptor Tom70 . Cell . 2003 ; 112 : 41 - 50 .
79. D'Silva P , Liu Q , Walter W , Craig EA . Regulated interactions of mtHsp70 with Tim44 at the translocon in the mitochondrial inner membrane . Nat Struct Mol Biol . 2004 ; 11 : 1084 - 91 .
80. Woetmann A , Nielsen M , Christensen ST , Brockdorff J , Kaltoft K , Engel AM , et al. Inhibition of protein phosphatase 2A induces serine/threonine phosphorylation, subcellular redistribution, and functional inhibition of STAT3 . Proc Natl Acad Sci U S A . 1999 ; 96 : 10620 - 5 .
81. Floyd BJ , Wilkerson EM , Veling MT , Minogue CE , Xia C , Beebe ET , et al. Mitochondrial protein interaction mapping identifies regulators of respiratory chain function . Mol Cell . 2016 ; 63 : 621 - 32 .
82. Dickey AS , Strack S. PKA/AKAP1 and PP2A/Bbeta2 regulate neuronal morphogenesis via Drp1 phosphorylation and mitochondrial bioenergetics . J Neurosci . 2011 ; 31 : 15716 - 26 .
83. Wishingrad MA , Koshlukova S , Halvorsen SW. Ciliary neurotrophic factor stimulates the phosphorylation of two forms of STAT3 in chick ciliary ganglion neurons . J Biol Chem . 1997 ; 272 : 19752 - 7 .
84. Haq R , Halupa A , Beattie BK , Mason JM , Zanke BW , Barber DL . Regulation of erythropoietin-induced STAT serine phosphorylation by distinct mitogenactivated protein kinases . J Biol Chem . 2002 ; 277 : 17359 - 66 .
85. Wang C , Youle RJ . The role of mitochondria in apoptosis* . Annu Rev Genet . 2009 ; 43 : 95 - 118 .
86. Chen X , Abair TD , Ibanez MR , Su Y , Frey MR , Dise RS , et al. Integrin alpha1beta1 controls reactive oxygen species synthesis by negatively regulating epidermal growth factor receptor-mediated Rac activation . Mol Cell Biol . 2007 ; 27 : 3313 - 26 .
87. Roggia MF , Ueta T. alphavbeta5 integrin/FAK/PGC-1alpha pathway confers protective effects on retinal pigment epithelium . PLoS ONE . 2015 ; 10 :e0134870.
88. Kheradmand F , Werner E , Tremble P , Symons M , Werb Z. Role of Rac1 and oxygen radicals in collagenase-1 expression induced by cell shape change . Science . 1998 ; 280 : 898 - 902 .
89. Taddei ML , Parri M , Mello T , Catalano A , Levine AD , Raugei G , et al. Integrinmediated cell adhesion and spreading engage different sources of reactive oxygen species . Antioxid Redox Signal . 2007 ; 9 : 469 - 81 .
90. Poli G , Leonarduzzi G , Biasi F , Chiarpotto E. Oxidative stress and cell signalling . Curr Med Chem . 2004 ; 11 : 1163 - 82 .
91. Infante JR , Camidge DR , Mileshkin LR , Chen EX , Hicks RJ , Rischin D , et al. Safety , pharmacokinetic, and pharmacodynamic phase I dose-escalation trial of PF-00562271, an inhibitor of focal adhesion kinase, in advanced solid tumors . J Clin Oncol . 2012 ; 30 : 1527 - 33 .
92. Jones SF , Siu LL , Bendell JC , Cleary JM , Razak ARA , Infante JR , et al. A phase I study of VS-6063, a second-generation focal adhesion kinase inhibitor, in patients with advanced solid tumors . Invest New Drugs . 2015 ; 33 : 1100 - 7 .
93. Cabrita MA , Jones LM , Quizi JL , Sabourin LA , McKay BC , Addison CL . Focal adhesion kinase inhibitors are potent anti-angiogenic agents . Mol Oncol . 2011 ; 5 : 517 - 26 .
94. Hung W-Y , Huang K-H , Wu C-W , Chi C-W , Kao H-L , Li AF-Y , et al. Mitochondrial dysfunction promotes cell migration via reactive oxygen species-enhanced beta5-integrin expression in human gastric cancer SC-M1 cells . Biochim Biophys Acta . 1820 ; 2012 : 1102 - 10 .
95. Nunes JB , Peixoto J , Soares P , Maximo V , Carvalho S , Pinho SS , et al. OXPHOS dysfunction regulates integrin-beta1 modifications and enhances cell motility and migration . Hum Mol Genet . 2015 ; 24 : 1977 - 90 .