THE

JOURNAL • RESEARCH • www.fasebj.org

The regulation of mitochondria dynamics in neurite

outgrowth by retinoic acid receptor b signaling
Diogo Trigo, Maria B. Goncalves, and Jonathan P. T. Corcoran1
The Wolfson Centre for Age-Related Diseases, King’s College London, London, United Kingdom

ABSTRACT: Neuronal regeneration is a highly energy-demanding process that greatly relies on axonal mitochondrial
transport to meet the enhanced metabolic requirements. Mature neurons typically fail to regenerate after injury,
partly because of mitochondrial motility and energy deficits in injured axons. Retinoic acid receptor (RAR)-b
signaling is involved in axonal and neurite regeneration. Here we investigate the effect of RAR-b signaling on
mitochondria trafficking during neurite outgrowth and find that it enhances their proliferation, speed, and
movement toward the growing end of the neuron via hypoxia-inducible factor 1a signaling. We also show that RAR-
b signaling promotes the binding of the mitochondria to the anchoring protein, glucose-related protein 75, at the
growing tip of neurite, thus allowing them to provide energy and metabolic roles required for neurite
outgrowth.—Trigo, D., Goncalves, M. B., Corcoran, J. P. T. The regulation of mitochondria dynamics in neurite
outgrowth by retinoic acid receptor b signaling. FASEB J. 33, 000–000 (2019). www.fasebj.org
KEY WORDS: neuron • retinoid • growth cone • energy metabolism

Mitochondria are cellular organelles whose main func- movement toward the distal part of neurites and syn-
tions are to generate ATP, buffer cytosolic calcium, and apses (7) by interacting with the transport enzymes
generate reactive oxygen species. Mitochondria also play mitochondrial Rho GTPase (MIRO) 1 and 2 (8, 9). Ad-
an important role in the mechanism of axon extension, ditionally, the mitochondrial stress chaperone glucose-
regeneration, and axon branching (1, 2). The elongating related protein 75 (GRP75), a mediator of endoplasmic
axon requires a continuous supply of energy in areas distal reticulum (ER)–mitochondria coupling, has recently
from the cell body (1); proper mitochondria distribution is been associated with HIF-1a–mediated mitochondria
required for axonal regeneration (3), and indeed mito- response to hypoxia (10–12).
chondria movement is increased following axonal injury It has previously been shown, in refs. 13–15, that
(4). retinoic acid induces axonal and neurite outgrowth by
This transport is regulated by various signaling activating the retinoic acid receptor (RAR)-b, but the
pathways to efficiently respond to cellular needs and role of retinoids in mitochondria mobilization or re-
external factors. One of these pathways is hypoxia- cruitment is yet to be described. Furthermore, consid-
inducible factor 1a (HIF-1a). Induced by decreased ering the essential role of mitochondria in neurite
oxygen tension, the HIF-1a signaling pathway is tra- elongation, and taking into account the role of hypoxia
ditionally associated with a first-level cell response signaling in peripheral neurite regeneration, it is perti-
to hypoxia, namely, by regulating mitochondria re- nent to investigate whether retinoic acid promotes
cruitment, activity, and oxygen consumption (5, 6). neurite elongation in the CNS by acting via this signal-
Besides regulating mitochondria activity, HIF-1a in- ing pathway.
duces the expression of the modulator of mitochondria Here we show that in cultured cortical neurons,
movement, hypoxia up-regulated mitochondrial move- RAR-b, in parallel to promoting neurite outgrowth,
ment regulator (HUMMR), which promotes mitochondria promotes mitochondria membrane depolarization in
the soma and its anterograde transport and pro-
ABBREVIATIONS: ER, endoplasmic reticulum; GRP75, glucose-related
liferation along the neurite by activating the hypoxia
protein 75; HIF-1a, hypoxia-inducible factor 1a; HUMMR, hypoxia up- signaling pathway. We describe that HIF-1a is required
regulated mitochondrial movement regulator; MIRO, mitochondrial Rho for both retinoid-induced neurite elongation and mi-
GTPase; PBS-T, PBS with 0.02% Tween; RAR, retinoic acid receptor;
siRNA, small interfering RNA; TMRM, tetramethylrhodamine, methyl
tochondria regulation, and we additionally show that
ester RAR-b activation promotes the accumulation of mi-
1
Correspondence: The Wolfson Centre For Age-Related Diseases, King’s tochondria in the growing neurite. This is accom-
College London, Guy’s Campus, London SE1 1UL, United Kingdom. plished by facilitating the interaction of mitochondria
E-mail: [email protected] with the chaperone GRP75, possibly by mediating
doi: 10.1096/fj.201802097R mitochondria-ER interaction.

0892-6638/19/0033-0001 © FASEB 1
ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 MATERIALS AND METHODS Confocal microscopy

Primary neuronal cell cultures Multichannel fluorescence images were captured using a
Zeiss LSM 700 laser-scanning confocal microscope (Carl
Mouse primary cortical neurons were prepared as pre- Zeiss, Oberkochen, Germany) with a 363 oil-immersion
viously described in ref. 16. Cells were plated onto 5 mg/ml Aprochromat objective with an image size of 512 3 512
poly-D-lysine-coated 24-well cell culture plates, 75-cm2 flasks, pixels, with a pinhole aperture of 1 Airy unit. Settings for
or 35-mm glass-bottom culture dishes (MatTek, Ashland, MA, gain, contrast, and brightness were optimized initially and
USA), depending on the experiment, at a density of 15 3 104 held constant throughout each study so that all sections were
cells per ml. Cells were cultured in neurobasal medium digitized under the same conditions.
(Thermo Fisher Scientific, Waltham, MA, USA) containing 2% For colocalization studies, z stacks of the whole imaged neu-
B27 serum-free supplement, 2 mM L-glutamine, 1.5% glucose, ron were taken (serial scans at different focal planes with a sep-
penicillin (100 U/ml), and streptomycin (100 g/ml), incubated aration optimized by the software) as previously described in ref.
at 37°C in a humidified atmosphere of 5% CO2 and 95% air. 18. For time-lapse imaging analysis, TMRM loaded cells were
Cultures were 98% neurons, judged by bIII-tubulin staining. imaged in a time series of 100 images 10 s apart. The Fiji distri-
Unless mentioned otherwise, mouse primary cortical cultures bution (https://fiji.sc/) of ImageJ software (National Institutes of
were treated with 100 nM CD2019 (synthesized by Sygnature Health, Bethesda, MD, USA) was used to measure pixel signal
Chemical Services, Nottingham, United Kingdom) or vehicle intensity and colocalization studies.
(DMSO 0.1%, v/v) for 72 h. CD2019 is a RAR-b agonist capable For neurite outgrowth assays, treated primary cortical
of inducing axonal outgrowth in central primary cultures with neuron cultures were fixed in 4% paraformaldehyde, per-
a 5-fold and 12-fold selectivities for RAR-b over RAR-a and meabilized with 0.1% Triton X-100 for 4 min, and stained with
RAR-g, respectively (16). The dose and treatment duration Alexa Fluor 488 phalloidin, or alternatively with the neuron-
were based on our previous in vivo studies on activation of specific anti–bIII-tubulin. These were then imaged by confo-
RAR-b signaling in the adult rat brain. HIF-1a was inhibited cal microscopy, and ImageJ-Fiji software was used to measure
with CAY10585 (20 mm; Cayman Chemical, Ann Arbor, MI, the length of the longest neurite in each neuron.
USA), an (aryloxyacetylamino)benzoic acid analog that was
determined by a reporter assay to inhibit HIF-1a protein ac-
cumulation and its target gene expression under hypoxic con- Transfection of cortical neurons
ditions, without altering HIF-1b levels (17).
GRP75 small interfering RNA (siRNA) (sc-35521; Santa Cruz
Biotechnology, Dallas, TX, USA) and Lipofectamine 2000 trans-
Immunocytochemistry fection reagent (Thermo Fisher Scientific) in OptiMem (2% v/v)
were prepared and allowed to stand for 5 min. They were then
Immunocytochemistry was performed as previously de- mixed and incubated at room temperature for 20 min before
scribed in ref. 16. Cortical neuron cultures were washed with addition to cortical neurons plated in glass bottom dishes or
PBS for 1 min, fixed in 4% paraformaldehyde for 20 min, 6-well plates, in a total volume of 1 ml, or in 24-well plates, in a
washed 3 times for 5 min each in PBS, and permealized with total volume of 250 ml, with a final siRNA concentration of 10 nM.
0.1% Triton X-100 in PBS for 4 min prior to being incubated in Cells were incubated for 7 h before adding an equal volume of
primary antibody in PBS with 0.02% Tween (PBS-T) over- supplemented neurobasal medium and incubated for additional
night. Primary antibody was removed by washing 3 times 24 h, following which the medium was replaced with fresh
for 5 min each in PBS-T; cultures were then incubated in the supplemented neurobasal medium. Cells were incubated for an
secondary antibody for 1 h at room temperature in PBS-T. additional period of 24 h.
The coverslips were then mounted using FluorSave reagent
(Merck, Darmdstadt, Germany). Antibodies used were as
follows: mouse monoclonal anti–bIII-tubulin (1:1000 for Western blotting
immunocytochemistry, against peptide EAQGPK; Promega,
Madison, WI, USA), mouse monoclonal anti–HIF-1a (1:500, Proteins were separated by SDS-PAGE on 10% (w/v) poly-
H1a67, aa 400–550; Abcam, Cambridge, MA, USA), mouse acrylamide gels and then transferred to a 0.45-mm pore size ni-
monoclonal anti-GRP75 (1:100, ab2799; Abcam), mouse trocellulose membrane (BA85; GE Healthcare, Waukesha, WI,
monoclonal anti-actin (1:5000, AC-15; MilliporeSigma, Bur- USA) using a Trans-Blot SD Semi-Dry Transfer Cell (Bio-Rad,
lington, MA, USA), and Alexa Fluor 488 phalloidin (1:40, Hercules, CA, USA). Nitrocellulose membranes were incubated
A12379; Thermo Fisher Scientific). Secondary antibodies for in blocking solution consisting of 5% (w/v) skimmed milk
immunohistochemistry were Alexa Fluor 594 and Alexa powder in PBS-T (0.1%, w/v) for 1 h at room temperature, fol-
Fluor 488 (1: 1000; Thermo Fisher Scientific). Hoechst stain lowed by incubation with the appropriate primary antibody
was used to stain nuclei (1:10000; Thermo Fisher Scientific). diluted in blocking solution overnight at 4°C. Membranes were
Secondary antibodies for Western blotting were Alexa Fluor washed 3 times in PBS-T and incubated with species-specific
680 and Alexa Fluor 800 (1:5000; Thermo Fisher Scientific). secondary antibodies in blocking solution for 2 h at room tem-
ER was stained with Cytopainter ER Staining Kit, Green perature in the dark, after which the membranes were washed as
Fluorescence (1:1000, ab139481; Abcam), according to the above. Protein levels were corrected for loading differences by
manufacturer’s instructions. normalizing against b-actin levels. Proteins were detected by
Imaging of mitochondria was performed by dyeing cells scanning at 700 and 800 nm using the Odyssey Detection System
according to the manufacturer’s instructions with mitotracker (Li-Cor Biosciences, Lincoln, NE, USA).
red (500 nM; Thermo Fisher Scientific) for 30 min prior to
fixation. Alternatively, cells were loaded with 20 nM tetra-
methylrhodamine, methyl ester (TMRM, T668; Thermo Quantitative analysis
Fisher Scientific) for 45 min, prior to being placed in an
incubator attached to a confocal microscope. TMRM is Signal intensity was analyzed by defining regions of interest with
a cell-permeant fluorescent dye, sequestered by active the actin or phalloidin staining and measuring the signal mean
mitochondria. or integrated density with ImageJ-Fiji (19). Kymographs were

2 Vol. 33 June 2019 The FASEB Journal x www.fasebj.org TRIGO ET AL.

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 plotted and analyzed using the Velocity Measurement Tool of RESULTS
ImageJ-Fiji. Colocalization was assessed by the Pearson product-
moment correlation coefficient (Pearson’s r value) (20), obtained RAR-b recruits mitochondria to the
from the Coloc 2 plugin for ImageJ-Fiji. Fusion and fission events growing neurite
were counted individually in each time series, using the 3D
Project Tool to obtain 3-dimensional reconstructions of z stacks
of the neurite to distinguish fusion or fission events from over- Given the high energy requirements of neurite outgrowth,
lapping mitochondria or mitochondria traveling together. we first asked whether activating RAR-b, required for
Quantified anti–HIF-1a signals were normalized to a back- neurite outgrowth in the CNS, was reflected in mito-
ground area of 25 3 25 mm from each imaging field. chondrial behavior in the neurite. Treatment of neuronal
cultures with the RAR-b agonist CD2019 (100 nM for 72 h)
results in neurite elongation (54% 6 32, P = 0.0018) (Fig.
Statistical analysis 1A, B). We next assessed mitochondrial distribution along
those neurites. Imaging of neurons stained with the mi-
Statistical studies were performed using Prism 5 for Mac OS X tochondria dye mitotracker red (Fig. 1C) showed no sta-
and Prism 6 for Windows (GraphPad Software, La Jolla, CA,
USA). Unless stated otherwise, data were analyzed using
tistically significant differences in mitotracker red signal
nondirectional, 2-tailed, 2-sample, equal variance Student’s on the cell body or the whole neurite between the RAR-
t test. Comparisons were made between appropriate groups, b agonist– and vehicle-treated neurons. However, com-
with a = 0.05. paring the intensity of mitotracker red signal in the distal

Figure 1. RAR-b activation

promotes mitochondria trans-
port and accumulation toward
the neurite terminal. A, B)
Imaging (nuclear stain in
blue; scale bar, 20 mm) (A)
and quantification (B) of neu-
rite outgrowth in neurons
treated with either vehicle or
CD2019 for 3 d. At least 5
neurons were quantified from
each of at least 6 slides per
condition. C–E) CD2019 com-
pared with vehicle treatment
resulted in an increase in the
length of the longest neurite
(scale bar, 10 mm) (C ), ac-
companied by an increase of
more than 3-fold in signal of
mitotracker red in the neu-
rites most distal 20 mm2
(terminal) (D), with no differ-
ences in total mitochondria
presence in the whole neurite
(E ). At least 5 neurons were
quantified from each of at
least 6 slides per condition.
F ) This increase in mito-
tracker red signal is also sup-
ported by an increase in
mitochondria presence in the
neurites most distal 20 mm2.
G) Examples of kymographs
recorded from neurons to de-
termine the direction and
speed of moving mitochon-
dria of live neurons treated
with either vehicle or CD2019,
at least 30 min prior to imag-
ing. At least 5 neurons were
quantified from 5 (vehicle)
and 8 (CD2019) slides. H, I )
Mitochondria in neurons
treated with agonist show an
increase in the speed of an-
terograde (antero) and retro-
grade (retro) movement (H ), and although the percentage of mitochondria moving retrogradely is not altered, the
percentage of those moving anterogradely is increased by the agonist (I ). N.s., not significant. *P , 0.05, **P , 0.005.

MITOCHONDRIA IN RETINOID-MEDIATED NEURITE GROWTH 3

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 20 mm2 of the neurite revealed a statistically significant RAR-b signaling regulates HIF-1a expression
increase in mitotracker red signal (0.75 6 0.4 to 3.03 6 1.6,
P = 0.038). This increase was more significant when the Taking into account that activation of the hypoxia sig-
number of mitochondria in the neurite terminal was tallied naling pathway promotes axonal regeneration, we next
individually (7.8 6 0.4 to 9.4 6 0.6, P = 0.0044), indicating asked if this pathway was regulated by RAR-b signaling.
that this increase in mitotracker red signal results from an Given that HIF-1a can be regulated by retinoids by non-
elevated mitochondrial presence in the growing neurite transcriptional mechanisms (24) and that we see an effect
following RAR-b activation (Fig. 1D–F). of RAR-b on mitochondrial function after 30 min, we
This population increment could result either from in- looked at both acute (30 min) and chronic (72 h) CD2019
creased recruitment and transport toward the neurite or treatment on HIF-1a expression. At both time points,
from greater mitochondria proliferation in this region. RAR-b activation results in an increase in HIF-1a expres-
Concerning the former, imaging of live neurons loaded sion by immunochemistry compared with control cul-
with TMRM (20 nM) revealed that acute activation of tures: 65% increase at 30 min (P = 0.036) and 33% increase
RAR-b (treated for 30 min prior to imaging with agonist, at 72 h (P = 0.049) (Fig. 3A, B).
100 nM) increases the speed of movement of healthy mi- These observations, together with the data showing
tochondria, both in anterograde (P = 0.019) and retrograde mitochondria recruitment to the neurite terminal, suggests
directions (P = 0.048). However, even though the speed of that RAR-b promotes mitochondria recruitment toward
moving mitochondria appears to also be increased in- the growing neurite, where energy production is pro-
dependently of direction of transport, RAR-b activation moted via increased oxidative phosphorylation, in a HIF-
increases the percentage of mitochondria moving ante- 1a–mediated process. Having established the importance
rogradely from 8.9 6 3.5% to 15.2 6 3.7% (P = 0.009), not of HIF-1a signaling in neurite outgrowth, we further
altering the percentage of mitochondria moving toward studied its effect by inhibiting HIF-1a in cultured neurons
the soma (P = 0.43) (Fig. 1G–I). with CAY10585 (10 mg/ml), which inhibits HIF-1a accu-
As for the latter, even though comparing the mean mulation in a concentration and time-dependent manner
length of mitochondria in the neurite revealed no differ- and significantly suppresses transcriptional activity of
ences, irrespective of their movement status (P = 0.99 HIF-1a target genes, obtaining complete inhibition at
for anterograde-moving, P = 0.54 for retrograde-moving, similar concentrations (17). Neurons were treated for 72 h,
and P = 0.74 for immobile mitochondria), analysis of mi- which decreased the presence of HIF-1a in lysates to 37%
tochondria size distribution showed a significant in- of that in control-treated lysates (P = 0.018), in a manner
crease in smaller mitochondria, particularly ,1 mm, from not rescuable by CD2019, which had a similar decrease in
20.6 6 1.0% to 29.3 6 5.0% (P = 0.0026). Moreover, neu- HIF-1a levels to 24% of control (P = 0.0002) (Fig. 3C, D). In
ronal activation of RAR-b affected mitochondrial dy- the CD2019 plus CAY10585–treated cultures, the RAR-
namics, substantiated by the increase in number of b–mediated neurite outgrowth was also inhibited, with
mitochondria fission events, from 0.0077 6 0.006 per mi- the mean length of the longer neurite being comparable to
crometer in vehicle-treated neurons to 0.0204 6 0.012 per the length seen in vehicle-treated cultures (1-way ANOVA
micrometer in agonist-treated neurons (P = 0.035) (Fig. followed by Tukey’s test, P , 0.005).
2A–E). Increased fission is a behavior associated with mi- Considering that RAR-b activation has an effect in
tochondria biogenesis (21) and previously observed in mitostasis regulation, as early as after 30 min of treatment,
regenerating axons (22, 23). we next studied the role of HIF-1a in this regulation. In-
Having identified that RAR-b activation recruits mito- hibition of HIF-1a does not alter the total percentage of
chondria to the growing neurite tip, we then analyzed its mitochondria moving in the neurite in live neurons (P =
effect in mitochondria activity. This was accomplished by 0.48) but prevents the increase in speed of moving mito-
measuring mitochondria membrane potential following chondria observed with acute activation of RAR-b, de-
treatment with the RAR-b agonist, registering the intensity creasing the mean speed of moving mitochondria to that of
of TMRM signal in the neurite of live cortical neurons in vehicle-treated neurons (1-way ANOVA followed by
culture. Acute activation of RAR-b (100 nM treatment Tukey’s test, P , 0.005). While not altering mitochondria
30 min prior to imaging) suggests a decrease in mito- presence in the neurite or the percentage of mitochondria
chondrial membrane potential in the neurite (P = 0.13 for moving in either direction, when compared with control,
anterograde-moving, P = 0.16 for retrograde-moving, and the increase in anterograde-moving mitochondria ob-
P = 0.29 for immobile mitochondria). This decrease in served with activation of RAR-b was prevented with
mitochondrial membrane potential was confirmed to be a HIF-1a inhibition, not altering the percentage of retrograde-
statistically significant 20% decrease in TMRM integrated moving mitochondria. Additionally, the decrease in TMRM
density signal when all the mitochondria in the neuron signal intensity observed following activation of RAR-b
were taken into account (P = 0.012), suggesting that this is also absent when HIF-1a is inhibited (Fig. 3E–K).
decrease in membrane potential is more focused on the
cell body. This was indeed confirmed when quantifica- Mitochondria interaction with GRP75 is
tion of mean TMRM signal discriminated between cell required for neurite elongation
body, where a 22% decrease in mean TMRM signal in-
tensity was observed (P = 0.019), and the neurite, with The mitochondria-associated stress chaperone GRP75
mean TMRM signal intensity remaining unaltered (P = has recently been described to play a role in neuronal
0.819) (Fig. 2F–K). ER–mitochondrial coupling and sensitivity to oxidative

4 Vol. 33 June 2019 The FASEB Journal x www.fasebj.org TRIGO ET AL.

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 Figure 2. Mitochondria trans-
port and dynamics on the neu-
rite is regulated by RAR-b. A–C)
Acute treatment with CD2019
compared with vehicle (at least
30 min prior to imaging) alters
mitochondrial dynamics, specif-
ically increasing the number of
fission events—imaging (scale
bar, 5 mm) (A), quantification
(B)—but the mean size of
moving or stationary mitochon-
dria in the neurite remains
unaltered (C ). D, E ) However,
analysis of the distribution of
mitochondria size frequency
showed that treatment with
CD2019 increased the number
of mitochondria shorter than 1
mm long: imaging (D), quanti-
fication (E ). F–I) Measurement
of TMRM signal intensity in the
neurite of live neurons shows
that treatment with the agonist
decreases the intensity of
TMRM signal—imaging (scale
bar, 5 mm) (F ), quantification
(G)—which is confirmed when
mitochondrial TMRM signal in
the whole neuron is analyzed,
with treatment with CD2019 for
30 min causing a decrease of
20%: imaging (scale bar, 20
mm) (H ), quantification (I ). J,
K ) This reflects an increase in
mean TMRM intensity signal in
the cell body (red-delimited re-
gion), because the mean TMRM
signal in the neurite (yellow-
delimited region) remained unal-
tered following RAR-b activation:
imaging (scale bar, 20 mm) (J),
quantification (K). At least 5 neu-
rons were quantified from each of
at least 6 slides per condition. *P ,
0.05, **P , 0.005.

stress and hypoxia, in a process involving HIF-1a (10, 25). GRP75 and mitotracker red increased from 0.35 6 0.03 in
Because GR75 is highly linked with mitochondrial re- vehicle-treated neurons to 0.47 6 0.04 (Pearson’s r value,
sponse to hypoxia and oxygen limitation (11, 12), and P = 0.0091) (Fig. 4A–D).
considering that it has been shown to bind RARs (26), we Having established that GRP75 and its interaction with
next looked at its expression levels to further clarify its role mitochondria are up-regulated following RAR-b activa-
in mediating the process of mitochondria recruitment to tion, we next silenced GRP75 with siRNA to explain its role
the growing neurite. in mitochondria recruitment by RAR-b. Treatment with
Immunostaining of cultured cortical neurons that had GRP75 siRNA was sufficient to ablate the effect of neurite
been treated with CD2019 (100 nM) or vehicle for 72 h elongation in primary cultures of cortical neurons fol-
shows that in the presence of the RAR-b agonist there is an lowing treatment with RAR-b agonist (100 nM for 72 h,
increase in colocalization of GRP75 and the mitochondria 1-way ANOVA followed by Tukey’s test, P , 0.05). Live
dye mitotracker red, suggesting that not only is GRP75 up- imaging of GRP75-silenced neurons additionally revealed
regulated in RAR-b activated neurites but its binding to that acute treatment with RAR-b agonist (30 min prior to
mitochondria is also elevated, established by colocaliza- imaging) causes a significant increase of 102% in the
tion analysis. Indeed, RAR-b activation caused a 40% in- number of mitochondria moving in the retrograde di-
crease in GRP75 signal coincident with mitotracker red rection (P = 0.039) (Fig. 4E–H). Having established that
(P = 0.038), and colocalization between immunostained activation of RAR-b promotes GRP75 association with

MITOCHONDRIA IN RETINOID-MEDIATED NEURITE GROWTH 5

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 Figure 3. HIF-1a is involved in regulation of mitochondria dynamics by RAR-b. A, B) Neuronal cultures treated with CD2019 for
30 min show an increase in the levels of HIF-1a, which was maintained when treatment was extended for 3 d compared with
vehicle-treated cultures: imaging (nuclear stain in blue; green signal is phalloidin; scale bar, 10 mm) (A), quantification (B). Ten
neurons were quantified from each of at least 4 slides per condition. C, D) Western blot of cell lysates of cortical primary cultures
treated with vehicle, CAY15085, or CD2019 for 3 d shows that CAY15085 alone decreases the levels of HIF-1a which is not rescued
by CD2019: imaging (C ), quantification, (D). Western blots were repeated 3 times. E, F ) Imaging (green signal is bIII-tubulin;
(continued on next page)

6 Vol. 33 June 2019 The FASEB Journal x www.fasebj.org TRIGO ET AL.

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 mitochondria and that this association is required to ac- axon requires a continuous supply of energy in areas
cumulate mitochondria in the growing neurite, and given distal from the cell body (31), proper mitochondria dis-
that GRP75 mediates ER-mitochondria interaction, we tribution is required for axonal regeneration (3), and
examined whether the association between the ER axonal injury increases mitochondria movement (4).
and mitochondria was similarly up-regulated by RAR-b Therefore, for a pathway to promote axonal growth, it
activation. must also be capable of actively recruiting mitochondria
Colocalization analysis of cortical primary cultures to the growing axon.
stained with Cytopainter ER and the mitochondria dye We describe that RAR-b signaling–dependent growth
mitotracker red shows a solid Pearson’s r 5 0.6 between is associated with anterograde mitochondria movement,
ER and mitochondria. The levels of coincident signal resulting in an increased presence of mitochondria in the
between ER and mitotracker red in the whole neuron neurite terminal. In large cells such as axons, the main
were unaltered following RAR-b activation, but when reason for mitochondria anterograde movement is typi-
this analysis was restricted to the distal 20 mm2 of the cally to distribute new, healthy mitochondria (32) to sites
neurite, an increase of 40% (P = 0.025) was observed in of high energy demand, such as the node of Ranvier or the
agonist-treated neurons (Fig. 4I–K). These results sug- growing tip (33, 34).
gest that, following RAR-b activation, mitochondria We observe that activation of RAR-b in primary
recruited to the growing terminal are coupled to the ER cultures of cortical neurons increases the number of
in an interaction regulated by GRP75. submicrometer-long mitochondria in the neurite, result-
ing from an increase in mitochondria fission, pinching off
segments that result in daughter mitochondria. This
DISCUSSION process is associated with mitochondrial redistribution to
regions of increased metabolic demand (30, 35), and no
In this study, we have observed that in primary cultures of overall changes to the mean length of mitochondria were
cortical neurons, activation of RAR-b not only induces observed, independent of mitochondria movement sta-
neurite outgrowth but also mobilizes mitochondria to the tus. This modulation of mitochondria dynamics occurring
growing neurite (Fig. 5). These processes are mediated by following RAR-b activation appears, then, to be related
HIF-1a and are accomplished by up-regulating mitochon- not to mitochondria degradation but to increased mito-
drial presence in the tip of the growing neurite by increasing chondria proliferation (21) required for axonal growth
the speed and number of mitochondria moving ante- (36).
rogradely and by promoting mitochondria proliferation.
Recruited mitochondria accumulate in the growing termi-
nal, where they closely associate with the chaperone GRP75 Recruitment and regulation of mitochondria
and the ER. In parallel, mitochondria membrane potential is dynamics by the hypoxia signaling pathway
not statistically significantly decreased in the neurite but is
The neuronal process of mitochondria transport occurs
evidently decreased when the cell body is considered. This
through different mechanisms, depending on the di-
suggests that mitochondria membrane potential, and con-
rection of movement (37), occurring through dynein
sequentially oxidative phosphorylation and energy pro-
motors in the case of retrograde transport (38), and
duction, could be favored in the growing neurite at the
kinesin motors in the case of anterograde transport (39).
expense of the cell body. We observed no correlation be-
Even though the regulation mechanism of these motor
tween mitochondria membrane potential and direction of
proteins is not yet clear (40), the MIRO protein family is
transport in the neurites. Even though retrogradely moving
associated with mitochondrial axonal stopping (41, 42).
mitochondria have been associated with depolarized
membranes (27, 28), other studies found no distinction be- This mechanism is particularly interesting, considering
that MIRO acts by interacting with the hypoxia medi-
tween anterograde and retrograde populations (29, 30), as
ator HUMMR, a neuronal protein up-regulated in mi-
was the case of our observations.
tochondria via HIF-1a, which promotes anterograde
transport (7).
Regulation of mitochondria dynamics Additionally, activation of RAR-b in cortical neuro-
by RAR-b nal cultures decreases somal mitochondria membrane
potential, similar to what has been described in isolated
Previous work has shown that activation of RAR-b mitochondria (43). In parallel to promoting neurite
stimulates neurite growth (13–15). Because the process of outgrowth, we have presented evidence that activation
constant cytoskeletal rearrangement in the elongating of RAR-b starts a signaling cascade that fulfills the

scale bar, 20 mm) (E ) and analysis (F ) of neurons treated with CAY10585, CD2019, or both for 3 d. CAY10585 prevents the
neurite outgrowth induced by CD2019. 10 neurons were quantified from each of at least 3 slides per condition. G–J ) Acute
treatment with CD2019 30 min prior to imaging (G) showed that inhibition of HIF-1a by CAY15085 does not alter mitochondria
presence in the neurite (H ) or retrograde movement but does prevent the increases in anterograde-moving mitochondria (I )
and speed of mitochondria (J ) observed in the presence of CD2019. At least 5 neurons were quantified from each of at least 6
slides per condition. K ) Moreover, the decrease in mitochondria membrane potential, measured by TMRM following RAR-b
agonist treatment, does not occur in the presence of the HIF-1a inhibitor CAY15085. At least 5 neurons were quantified from
each of at least 6 slides per condition. CAY, CAY15085; n.s., not significant. *P , 0.05; **P , 0.005; ***P , 0.001.

MITOCHONDRIA IN RETINOID-MEDIATED NEURITE GROWTH 7

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 Figure 4. In the growing neurite, mitochondria interact with GRP75 and the ER following RAR-b activation. A–D) Neurons
treated with CD2019 for 3 d [imaging of cell body and neuron (scale bar, 10 mm) (A); zoomed-in imaging of the neurite in a
series of z-stack slices (scale bar, 20 mm) (B)] show an increase in colocalization between GRP75 and mitotracker red signals
compared with vehicle-treated neurons (C, D). E, F ). Treatment of neurons with siRNA for GRP75 negated the effect of neurite
elongation of treatment with RAR-b agonist: imaging (green signal is bIII-tubulin; scale bar, 20 mm) (E ), quantification (F ). G,
H ) Live imaging of neurons treated with GRP75 siRNA showed that acute treatment with RAR-b agonist (at least 30 min prior to
imaging) doubled the number of mitochondria moving in the retrograde direction, whereas the number of anterograde-moving
mitochondria is unaltered compared with vehicle-treated neurons: imaging (G), quantification (H ). I, J ) In similarly treated
preparations, ER staining and mitotracker red staining (scale bar, 5 mm) (I ) show a solid Pearson’s r 5 0.5 (J ), which is not
altered significantly with activation of RAR-b. K ) The neurite’s more distal 20 mm2 reveals a significant increase in the overlap
between ER and mitochondria. At least 5 neurons were quantified from each of 3 slides per condition. N.s., not significant. *P ,
0.05, **P , 0.005.

8 Vol. 33 June 2019 The FASEB Journal x www.fasebj.org TRIGO ET AL.

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 Figure 5. Model for mitochon-
dria recruitment by RAR-b in
the growing neurite. Activation
of RAR-b in cortical neuronal
cultures promotes neurite elon-
gation. This process involves, in
part, up-regulation of HIF-1a
signaling, which promotes mi-
tochondria membrane depolar-
ization, decreasing oxidative
phosphorylation in the cell
body, but not neurite tip. Acti-
vation of RAR-b also causes a
HIF-1a–dependent increase in
mitochondria proliferation and
increases the speed of mito-
chondria movement along the
neurite. Following activation of
RAR-b, GRP75 mediates mito-
chondria binding to the ER in
the terminal, anchoring the
recruited mitochondria to this
region, decreasing the num-
ber of mitochondria being
transported retrogradely, and
effectively increasing the mito-
chondrial presence in the grow-
ing neurite terminal.

heightened energy need in the neurite terminal. As a interacts with mediators of mitochondrial transport (9),
result, RAR-b activation appears to decrease oxidative promoting anterograde mitochondria movement (7). The
phosphorylation in the cell body, reflected in a decrease HIF-1a–mediated increase in anterograde mitochondria
in membrane potential, minimizing substrate usage transport following RAR-b activation could occur using
and oxygen consumption away from the growing the same machinery. We show that HIF-1a modulates
neurite, where energy requirement is higher. mitochondria activity, promoting its motility and mini-
Our data show that HIF-1a is up-regulated in cortical mizing oxidative phosphorylation at the cell body. At the
neuronal culture following RAR-b activation. We have growing neurite, where membrane depolarization signals
shown that this can be via a nongenomic effect of RAR-b are absent, mitochondria membrane repolarizes, pro-
signaling, because it occurs as early as 30 min after agonist moting oxidative phosphorylation, by which mitochon-
treatment. Nongenomic events of retinoid signaling are dria can generate the energy required for the process of
now emerging as important players in the maintenance neurite elongation.
and regeneration of the nervous system, and include ac-
tivation of AKT (44), synapse transmission (45), and pro-
liferation of stem cells, which is also due to increased Mitochondria accumulation in the growing
stability of HIF-1a by retinoid signaling (46). In addition, neurite is dependent on the interaction
chronic treatment with the RARb agonist maintains HIF- with GRP75
1a expression, which suggests that RAR-b signaling may
controls HIF-1a at the translational level as well as regu- The increase in mitochondria movement throughout the
lating its stability. HIF-1a expression is necessary for neuron is stopped at the growing neurite terminal, where
retinoid-mediated neurite outgrowth to occur, because mitochondria are halted. If this process were only caused
inhibition of HIF-1a with the HIF-1a inhibitor CAY10585 by the absence of promovement signals, a residual sto-
prevents the increase in neurite length observed following chastic increase in retrograde transport should still
RAR-b activation. Inhibition of HIF-1a also prevents the presumably occur because of increased mitochondria
recruitment of mitochondria to the growing neurite, ne- presence. Because such an increase was not observed, it
gating the increase in speed of mitochondria movement follows that mitochondria recruited to the neurite must
observed following treatment with RAR-b agonist in live actively be maintained at this location.
neurons. Moreover, inhibition of HIF-1a also prevents the The chaperone GRP75 is an essential mitochondria-
decrease in mitochondria membrane polarization ob- targeted chaperone (48, 49), up-regulated in stress situa-
served in the cell body in response to RAR-b activation. tions such as starvation, oxidative stress or ischemia (50,
Besides promoting peripheral axonal regeneration, 51), and, importantly, is involved in mitochondria distri-
hypoxia-pathway signals are crucial in regulating mito- bution (52).
chondria dynamics: HIF-1a regulates mitochondrial oxy- We show here that RAR-b–mediated recruitment of
gen consumption (47), and the HIF-1a–induced HUMMR mitochondria to the growing neurite is accompanied by an

MITOCHONDRIA IN RETINOID-MEDIATED NEURITE GROWTH 9

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 up-regulation in GRP75 and an increase in mitochondria 2. Cartoni, R., Norsworthy, M. W., Bei, F., Wang, C., Li, S., Zhang, Y.,
colocalization with GRP75. Although found in other sub- Gabel, C. V., Schwarz, T. L., and He, Z. (2017) The mammalian-
specific protein Armcx1 regulates mitochondrial transport during
cellular locations, GRP75 was originally identified as a axon regeneration. Neuron 94, 689
mitochondrial matrix-resident chaperone protein (53), and 3. Han, S. M., Baig, H. S., and Hammarlund, M. (2016) Mitochondria
data suggest that its preferential location in the neurite is localize to injured axons to support regeneration. Neuron 92,
1308–1323
indeed associated with the mitochondria. 4. Mar, F. M., Simões, A. R., Leite, S., Morgado, M. M., Santos, T. E.,
We further show that GRP75 is required for neurite Rodrigo, I. S., Teixeira, C. A., Misgeld, T., and Sousa, M. M. (2014)
outgrowth, because its silencing with siRNA ablates the CNS axons globally increase axonal transport after peripheral
effect of neurite elongation. Interestingly, silencing of conditioning. J. Neurosci. 34, 5965–5970
5. Iyer, N. V., Kotch, L. E., Agani, F., Leung, S. W., Laughner, E., Wenger,
GRP75 also resulted in a drastic increase in mitochondria R. H., Gassmann, M., Gearhart, J. D., Lawler, A. M., Yu, A. Y., and
retrograde transport (i.e., in the direction of the cell body) Semenza, G. L. (1998) Cellular and developmental control of O2
following activation of RAR-b. These results indicate that homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev. 12,
GRP75 is required to maintain recruited mitochondria at 149–162
6. Lai, D., Tan, C. L., Gunaratne, J., Quek, L. S., Nei, W., Thierry, F., and
the neurite terminal. Bellanger, S. (2013) Localization of HPV-18 E2 at mitochondrial
In addition to its role in energy homeostasis, mitochon- membranes induces ROS release and modulates host cell metabo-
dria coordinate with the ER, being physically tethered lism. PLoS One 8, e75625
through the mitochondria-associated ER membrane com- 7. Li, Y., Lim, S., Hoffman, D., Aspenstrom, P., Federoff, H. J., and
Rempe, D. A. (2009) HUMMR, a hypoxia- and HIF-1alpha-inducible
plex to regulate cellular levels of intracellular calcium (54), protein, alters mitochondrial distribution and transport. J. Cell Biol.
and the importance of ER in the axon has recently been 185, 1065–1081
characterized (55). Considering that the mitochondria-ER 8. Stowers, R. S., Megeath, L. J., Górska-Andrzejak, J., Meinertzhagen,
interface is mediated by GRP75, which colocalizes with I. A., and Schwarz, T. L. (2002) Axonal transport of mitochondria to
synapses depends on milton, a novel Drosophila protein. Neuron 36,
mitochondria and mitochondria-associated ER membranes 1063–1077
(56), the increase in mitochondria-GRP75 colocalization 9. Reis, K., Fransson, A., and Aspenström, P. (2009) The Miro GTPases:
following RAR-b activation raises the hypothesis that it at the heart of the mitochondrial transport machinery. FEBS Lett. 583,
1391–1398
could be up-regulating mitochondria-ER interaction. 10. Mylonis, I., Kourti, M., Samiotaki, M., Panayotou, G., and Simos, G.
Indeed, RAR-b activation is accompanied by increased (2017) Mortalin-mediated and ERK-controlled targeting of HIF-1a to
association between the ER and mitochondria in the distal mitochondria confers resistance to apoptosis under hypoxia. J. Cell Sci.
end of the neurite, where mitochondria-ER interaction is 130, 466–479
11. Williamson, C. L., Dabkowski, E. R., Dillmann, W. H., and Hollander,
exacerbated. This interaction could be increased because of J. M. (2008) Mitochondria protection from hypoxia/reoxygenation
structural purposes (i.e., anchoring mitochondria in specific injury with mitochondria heat shock protein 70 overexpression. Am. J.
neuronal locations) or functional reasons (such as calcium Physiol. Heart Circ. Physiol. 294, H249–H256
regulation), and GRP75 has long been described to de- 12. Liu, Y., Liu, W., Song, X. D., and Zuo, J. (2005) Effect of GRP75/
mthsp70/PBP74/mortalin overexpression on intracellular ATP level,
termine ER-mitochondria coupling, mediating the in- mitochondrial membrane potential and ROS accumulation following
teraction of the mitochondrial voltage-dependent anion glucose deprivation in PC12 cells. Mol. Cell. Biochem. 268, 45–51
channel with the reticular inositol 1,4,5-triphosphate re- 13. Corcoran, J., and Maden, M. (1999) Nerve growth factor acts via
ceptor (56).Therefore, mitochondria could not only be retinoic acid synthesis to stimulate neurite outgrowth. Nat. Neurosci. 2,
307–308
playing a crucial role generating chemical energy required 14. Dmetrichuk, J. M., Carlone, R. L., and Spencer, G. E. (2006) Retinoic
for axonal and neurite outgrowth, but additionally helping acid induces neurite outgrowth and growth cone turning in
control the calcium presence, an important regulator of invertebrate neurons. Dev. Biol. 294, 39–49
15. Hoecker, J., Liffert, R., Burch, P., Wehlauch, R., and Gademann, K.
axon outgrowth and guidance (57). (2013) Caged retinoids as photoinducible activators: implications for
cell differentiation and neurite outgrowth. Org. Biomol. Chem. 11,
3314–3321
ACKNOWLEDGMENTS 16. Goncalves, M. B., Clarke, E., Hobbs, C., Malmqvist, T., Deacon, R.,
Jack, J., and Corcoran, J. P. (2013) Amyloid b inhibits retinoic acid
This work was supported by Grant WT092718 from the synthesis exacerbating Alzheimer disease pathology which can be
Wellcome Trust (London, United Kingdom). The authors attenuated by an retinoic acid receptor a agonist. Eur. J. Neurosci. 37,
declare no conflicts of interest. 1182–1192
17. Lee, K., Lee, J. H., Boovanahalli, S. K., Jin, Y., Lee, M., Jin, X., Kim, J. H.,
Hong, Y. S., and Lee, J. J. (2007) (Aryloxyacetylamino)benzoic acid
analogues: a new class of hypoxia-inducible factor-1 inhibitors. J. Med.
AUTHOR CONTRIBUTIONS Chem. 50, 1675–1684
18. Trigo, D., and Smith, K. J. (2015) Axonal morphological changes
D. Trigo and M. B. Goncalves designed the experiments, following impulse activity in mouse peripheral nerve in vivo: the
return pathway for sodium ions. J. Physiol. 593, 987–1002
collected, analyzed, and interpreted the data, and drafted 19. Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M.,
and revised the manuscript; and J. P. T. Corcoran Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B.,
conceived and designed the experiments, interpreted Tinevez, J. Y., White, D. J., Hartenstein, V., Eliceiri, K., Tomancak, P.,
the data, and drafted and revised the manuscript. and Cardona, A. (2012) Fiji: an open-source platform for biological-
image analysis. Nat. Methods 9, 676–682
20. Costes, S. V., Daelemans, D., Cho, E. H., Dobbin, Z., Pavlakis, G., and
Lockett, S. (2004) Automatic and quantitative measurement of
REFERENCES protein-protein colocalization in live cells. Biophys. J. 86, 3993–4003
21. Garnier, A., Fortin, D., Zoll, J., N’Guessan, B., Mettauer, B., Lampert,
1. Morris, R. L., and Hollenbeck, P. J. (1993) The regulation of E., Veksler, V., and Ventura-Clapier, R. (2005) Coordinated changes
bidirectional mitochondrial transport is coordinated with axonal in mitochondrial function and biogenesis in healthy and diseased
outgrowth. J. Cell Sci. 104, 917–927 human skeletal muscle. FASEB J. 19, 43–52

10 Vol. 33 June 2019 The FASEB Journal x www.fasebj.org TRIGO ET AL.

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber
 22. Lathrop, K. L., and Steketee, M. B. (2013) Mitochondrial dynamics in (2009) Miro1 is a calcium sensor for glutamate receptor-dependent
retinal ganglion cell axon regeneration and growth cone guidance. localization of mitochondria at synapses. Neuron 61, 541–555
J. Ocul. Biol. 1, 9 42. Wang, X., and Schwarz, T. L. (2009) The mechanism of Ca2+
23. Kiryu-Seo, S., Tamada, H., Kato, Y., Yasuda, K., Ishihara, N., Nomura, -dependent regulation of kinesin-mediated mitochondrial motility.
M., Mihara, K., and Kiyama, H. (2016) Mitochondrial fission is an Cell 136, 163–174
acute and adaptive response in injured motor neurons. Sci. Rep. 6, 43. Notario, B., Zamora, M., Viñas, O., and Mampel, T. (2003) All-trans-
28331 retinoic acid binds to and inhibits adenine nucleotide translocase and
24. Zhang, J., Song, L. P., Huang, Y., Zhao, Q., Zhao, K. W., and Chen, induces mitochondrial permeability transition. Mol. Pharmacol. 63,
G. Q. (2008) Accumulation of hypoxia-inducible factor-1 alpha pro- 224–231
tein and its role in the differentiation of myeloid leukemic cells in- 44. Agudo, M., Yip, P., Davies, M., Bradbury, E., Doherty, P., McMahon, S.,
duced by all-trans retinoic acid. Haematologica 93, 1480–1487 Maden, M., and Corcoran, J. P. (2010) A retinoic acid receptor beta
25. Honrath, B., Metz, I., Bendridi, N., Rieusset, J., Culmsee, C., and agonist (CD2019) overcomes inhibition of axonal outgrowth via
Dolga, A. M. (2017) Glucose-regulated protein 75 determines ER- phosphoinositide 3-kinase signalling in the injured adult spinal cord.
mitochondrial coupling and sensitivity to oxidative stress in neuronal Neurobiol. Dis. 37, 147–155
cells. Cell Death Discov. 3, 17076 45. Liou, J. C., Ho, S. Y., Shen, M. R., Liao, Y. P., Chiu, W. T., and Kang,
26. Shih, Y. Y., Lee, H., Nakagawara, A., Juan, H. F., Jeng, Y. M., Tsay, Y. G., K. H. (2005) A rapid, nongenomic pathway facilitates the synaptic
Lin, D. T., Hsieh, F. J., Pan, C. Y., Hsu, W. M., and Liao, Y. F. (2011) transmission induced by retinoic acid at the developing synapse. J. Cell
Nuclear GRP75 binds retinoic acid receptors to promote neuronal Sci. 118, 4721–4730
differentiation of neuroblastoma. PLoS One 6, e26236 46. Mishra, S., Kelly, K. K., Rumian, N. L., and Siegenthaler, J. A. (2018)
27. Hüttemann, M., Lee, I., Pecinova, A., Pecina, P., Przyklenk, K., and Retinoic acid is required for neural stem and progenitor cell
Doan, J. W. (2008) Regulation of oxidative phosphorylation, the proliferation in the adult hippocampus. Stem Cell Reports 10,
mitochondrial membrane potential, and their role in human disease. 1705–1720
J. Bioenerg. Biomembr. 40, 445–456 47. Papandreou, I., Cairns, R. A., Fontana, L., Lim, A. L., and Denko, N. C.
28. Cho, Y., Shin, J. E., Ewan, E. E., Oh, Y. M., Pita-Thomas, W., and Cavalli, (2006) HIF-1 mediates adaptation to hypoxia by actively down-
V. (2015) Activating injury-responsive genes with hypoxia enhances regulating mitochondrial oxygen consumption. Cell Metab. 3, 187–197
axon regeneration through neuronal HIF-1a. Neuron 88, 720–734 48. Voos, W., Martin, H., Krimmer, T., and Pfanner, N. (1999)
29. Miller, K. E., and Sheetz, M. P. (2004) Axonal mitochondrial transport
Mechanisms of protein translocation into mitochondria. Biochim.
and potential are correlated. J. Cell Sci. 117, 2791–2804
Biophys. Acta 1422, 235–254
30. Lin, M. Y., Cheng, X. T., Tammineni, P., Xie, Y., Zhou, B., Cai, Q., and
49. Jin, J., Hulette, C., Wang, Y., Zhang, T., Pan, C., Wadhwa, R., and
Sheng, Z. H. (2017) Releasing syntaphilin removes stressed
Zhang, J. (2006) Proteomic identification of a stress protein,
mitochondria from axons independent of mitophagy under
mortalin/mthsp70/GRP75: relevance to Parkinson disease. Mol.
pathophysiological conditions. Neuron 94, 595–610.e6
Cell. Proteomics 5, 1193–1204
31. Zhou, B., Yu, P., Lin, M. Y., Sun, T., Chen, Y., and Sheng, Z. H. (2016)
50. Massa, S. M., Longo, F. M., Zuo, J., Wang, S., Chen, J., and Sharp, F. R.
Facilitation of axon regeneration by enhancing mitochondrial
transport and rescuing energy deficits. J. Cell Biol. 214, 103–119 (1995) Cloning of rat grp75, an hsp70-family member, and its ex-
32. Saxton, W. M., and Hollenbeck, P. J. (2012) The axonal transport of pression in normal and ischemic brain. J. Neurosci. Res. 40, 807–819
mitochondria. J. Cell Sci. 125, 2095–2104 51. Lee, A. S. (2001) The glucose-regulated proteins: stress induction and
33. Verburg, J., and Hollenbeck, P. J. (2008) Mitochondrial membrane clinical applications. Trends Biochem. Sci. 26, 504–510
potential in axons increases with local nerve growth factor or 52. Kawai, A., Nishikawa, S., Hirata, A., and Endo, T. (2001) Loss of the
semaphorin signaling. J. Neurosci. 28, 8306–8315 mitochondrial Hsp70 functions causes aggregation of mitochondria
34. Sajic, M., Mastrolia, V., Lee, C. Y., Trigo, D., Sadeghian, M., Mosley, in yeast cells. J. Cell Sci. 114, 3565–3574
A. J., Gregson, N. A., Duchen, M. R., and Smith, K. J. (2013) Impulse 53. Gao, Z., Niu, X., Zhang, Q., Chen, H., Gao, A., Qi, S., Xiang, R.,
conduction increases mitochondrial transport in adult mammalian Belting, M., and Zhang, S. (2017) Mitochondria chaperone GRP75
peripheral nerves in vivo. PLoS Biol. 11, e1001754 moonlighting as a cell cycle controller to derail endocytosis provides
35. Hollenbeck, P. J. (1996) The pattern and mechanism of an opportunity for nanomicrosphere intracellular delivery. Oncotarget
mitochondrial transport in axons. Front. Biosci. 1, d91–d102 8, 58536–58552
36. Vaarmann, A., Mandel, M., Zeb, A., Wareski, P., Liiv, J., Kuum, M., 54. Rowland, A. A., and Voeltz, G. K. (2012) Endoplasmic reticulum-
Antsov, E., Liiv, M., Cagalinec, M., Choubey, V., and Kaasik, A. (2016) mitochondria contacts: function of the junction. Nat. Rev. Mol. Cell
Mitochondrial biogenesis is required for axonal growth. Development Biol. 13, 607–625
143, 1981–1992 55. Luarte, A., Cornejo, V. H., Bertin, F., Gallardo, J., and Couve, A.
37. Hollenbeck, P. J., and Saxton, W. M. (2005) The axonal transport of (2018) The axonal endoplasmic reticulum: one organelle-many
mitochondria. J. Cell Sci. 118, 5411–5419 functions in development, maintenance, and plasticity. Dev. Neurobiol.
38. Susalka, S. J., and Pfister, K. K. (2000) Cytoplasmic dynein subunit 78, 181–208
heterogeneity: implications for axonal transport. J. Neurocytol. 29, 56. Szabadkai, G., Bianchi, K., Várnai, P., De Stefani, D., Wieckowski,
819–829 M. R., Cavagna, D., Nagy, A. I., Balla, T., and Rizzuto, R. (2006)
39. Hirokawa, N., Sato-Yoshitake, R., Kobayashi, N., Pfister, K. K., Bloom, Chaperone-mediated coupling of endoplasmic reticulum and mito-
G. S., and Brady, S. T. (1991) Kinesin associates with anterogradely chondrial Ca2+ channels. J. Cell Biol. 175, 901–911
transported membranous organelles in vivo. J. Cell Biol. 114, 295–302 57. Hutchins, B. I., Li, L., and Kalil, K. (2011) Wnt/calcium signaling
40. Devine, M. J., Birsa, N., and Kittler, J. T. (2016) Miro sculpts mediates axon growth and guidance in the developing corpus cal-
mitochondrial dynamics in neuronal health and disease. Neurobiol. losum. Dev. Neurobiol. 71, 269–283
Dis. 90, 27–34
41. Macaskill, A. F., Rinholm, J. E., Twelvetrees, A. E., Arancibia-Carcamo, Received for publication October 2, 2018.
I. L., Muir, J., Fransson, A., Aspenstrom, P., Attwell, D., and Kittler, J. T. Accepted for publication February 18, 2019.

MITOCHONDRIA IN RETINOID-MEDIATED NEURITE GROWTH 11

ww.fasebj.org by Univ of Penn Libr Electronic Acq Dept (165.123.34.86) on March 16, 2019. The FASEB Journal Vol. ${article.issue.getVolume()}, No. ${article.issue.getIssueNumber