Neurobiology of Disease
Histone Deacetylase Inhibitors Protect Against Pyruvate
Dehydrogenase Dysfunction in Huntington’s Disease
X Luana Naia,
1,2
* X Teresa Cunha-Oliveira,
1,3
* X Joana Rodrigues,
1
X Tatiana R. Rosenstock,
1
X Ana Oliveira,
1
Ma´rcio Ribeiro,
1
X Catarina Carmo,
1
Sofia I. Oliveira-Sousa,
1
XAna I. Duarte,
1,3
X Michael R. Hayden,
4
and X A. Cristina Rego
1,2
1
Center for Neuroscience and Cell Biology, University of Coimbra, 3004-504 Coimbra, Portugal,
2
Faculty of Medicine, University of Coimbra, 3000-548
Coimbra, Portugal,
3
Institute for Interdisciplinary Research, University of Coimbra, 3030-789 Coimbra, Portugal, and
4
Centre for Molecular Medicine and
Therapeutics, Child and Family Research Institute, University of British Columbia, Vancouver, British Columbia, V5Z 4H4 Canada
Transcriptional deregulation and changes in mitochondrial bioenergetics, including pyruvate dehydrogenase (PDH) dysfunction, have
been described in Huntington’s disease (HD). We showed previously that the histone deacetylase inhibitors (HDACIs) trichostatin A and
sodium butyrate (SB) ameliorate mitochondrial function in cells expressing mutant huntingtin. In thiswork, we investigated the effect of
HDACIs on the regulation of PDH activity in striatal cells derived from HD knock-in mice and YAC128 mice. Mutant cells exhibited
decreased PDH activity and increased PDH E1alpha phosphorylation/inactivation, accompanied by enhanced protein levels of PDH
kinases 1 and3 (PDK1 andPDK3). Exposure todichloroacetate, an inhibitorof PDKs, increased mitochondrial respiration and decreased
production of reactive oxygen species in mutant cells, emphasizing PDH as an interesting therapeutic target in HD. Treatment with SB
and sodium phenylbutyrate, another HDACI, recovered cell viability and overall mitochondrial metabolism in mutant cells. Exposure to
SB also suppressed hypoxia-inducible factor-1 (HIF-1
␣
) stabilization and decreased the transcription of the two most abundant PDK
isoforms, PDK2 and PDK3, culminating in increased PDH activation in mutant cells. Concordantly, PDK3 knockdown improved mito-
chondrial function, emphasizing the role of PDK3 inactivation on the positive effects achieved by SB treatment. YAC128 mouse brain
presented higher mRNA levels of PDK1–3 and PDH phosphorylation and decreased energy levels that were significantly ameliorated after
SB treatment. Furthermore, enhanced motor learning and coordination were observed in SB-treated YAC128 mice. These results suggest
that HDACIs, particularly SB, promote the activity of PDH in the HD brain, helping to counteract HD-related deficits in mitochondrial
bioenergetics and motor function.
Key words: Huntington disease; metabolism; mitochondria; PDH kinase; pyruvate dehydrogenase; sodium butyrate
Introduction
Huntington’s disease (HD) is an autosomal-dominant neurode-
generative disease characterized by the expression of mutant
huntingtin (mHTT), bearing an expanded polyglutamine tract in
its N-terminus, and progressive loss of striatal and cortical neu-
rons (
Gil and Rego, 2008). Among several targets, mHTT affects
mitochondrial homeostasis (Gil and Rego, 2008) and disturbs
Received May 17, 2014; revised Dec. 22, 2016; accepted Dec. 30, 2016.
Author contributions: L.N.,T.C.-O., and A.C.R.designed research; L.N.,T.C.-O., J.R., T.R.R.,A.O., M.R., C.C.,S.I.O.-
S., and A.I.D. performed research; M.R.H. and A.C.R. contributed unpublished reagents/analytic tools; L.N., T.C.-O.,
J.R., T.R.R., A.I.D., and A.C.R. analyzed data; L.N., T.C.-O., and A.C.R. wrote the paper.
This work was supported by projects PTDC/SAU-FCF/108056/2008, EXPL/BIM-MEC/2220/2013, PEst-C/SAU/
LA0001/2013-2014, and UID/NEU/04539/2013 funded by Fundac¸a˜o para a Cieˆncia e a Tecnologia (FCT), Portugal,
cofinanced by Programa Operacional Competitividade e Internacionalizac¸a˜o (COMPETE), Quadro de Refereˆncia
Estrate´gica Nacional (QREN), European Union (Fundo Europeu de Desenvolvimento Regional); Santa Casa da
Miserico´rdia de Lisboa (SCML), Mantero Belard Neuroscience Prize (first edition); and Fundac¸a˜o Luso-Americana
para o Desenvolvimento (FLAD) Life Science 2020 project. T.C.-O. and T.R.R. were supported by FCT postdoctoral
fellowships SFRH/BPD/34711/2007 and SFRH/BPD/44246/2008,respectively; M.R. and L.N. weresupported by the
doctoral fellowships SFRH/BD/41285/2007 and SFRH/BD/86655/2012, respectively, cofinanced by the Programa
Operacional PotencialHumano, QREN,and theEuropean Union.We thankDr. HenriqueGira˜o (Institute forBiomed-
ical Imagingand LifeSciences, Facultyof Medicine,University ofCoimbra) forkindly providingthe antibodyagainst
HIF-1
␣
.
The authors declare no competing financial interests.
*L.N. and T.C.-O. contributed equally to this work.
Significance Statement
The present work provides a better understanding of mitochondrial dysfunction in Huntington’s disease (HD) by showing that the
pyruvate dehydrogenase (PDH) complex is a promisingtherapeutic target. In particular, the histone deacetylaseinhibitor sodium
butyrate (SB) may indirectly (through reduced hypoxia-inducible factor 1 alpha stabilization) decrease the expression of the most
abundant PDH kinase isoforms (e.g., PDK3), ameliorating PDH activity and mitochondrial metabolism and further affecting
motor behavior in HD mice, thus constituting a promising agent for HD neuroprotective treatment.
2776 • The Journal of Neuroscience, March 8, 2017 • 37(10):2776 –2794
gene expression by interfering with nuclear transcription (co)factors
(Zhai et al., 2005; Benn et al., 2008), some linked to mitochondrial
biogenesis and function (Cui et al., 2006; Cunha-Oliveira et al.,
2012). mHTT also affects mitochondrial transmembrane potential
(Naia et al., 2015), calcium-buffering capacity (Damiano et al.,
2010), oxidative phosphorylation (Schatz, 1995; Milakovic and
Johnson, 2005), and the production of reactive oxygen species
(ROS) that originate in mitochondria (Ribeiro et al., 2014).
Symptomatic and presymptomatic HD patients evidence al-
tered brain glucose metabolism in the caudate, putamen, cortex,
and CSF (Kuhl et al., 1982; Kuwert et al., 1989; Jenkins et al., 1993;
Koroshetz et al., 1997). The onset of energy and mitochondrial
modifications at presymptomatic stages suggest that energy def-
icits may constitute early phenomena in HD pathogenesis (Mo-
chel and Haller, 2011). Previously, we observed reduced activity
of complex I in mitochondria from platelets of presymptomatic
and symptomatic HD carriers (Silva et al., 2013). Moreover, bio-
energetic dysfunction in HD human cybrids was associated with
decreased pyruvate dehydrogenase (PDH) activity and protein
levels and increased phosphorylation/inactivation of the E1alpha
subunit (Ferreira et al., 2011). PDH was shown to be impaired in
HD human brains (Sorbi et al., 1983; Butterworth et al., 1985)
and the PDH indirect activator dichloroacetate (DCA) showed
protective effects in HD mouse models (Andreassen et al., 2001),
revealing PDH as a central player in HD-associated mitochon-
drial dysfunction. This is in agreement with the observation of
elevated lactate levels in the striatum of presymptomatic HD pa-
tients (Jenkins et al., 1998), which suggests the occurrence of
PDH defects in vivo. The PDH complex (PDC) is a large complex
of three functional enzymes, E1, E2, and E3, and the current
understanding of PDC regulation involves inhibitory serine
phosphorylation of PDH E1alpha by specific PDH kinases
(PDKs), whereas dephosphorylation of PDH E1alpha by PDH
phosphatases (PDPs) activates the PDC (Harris et al., 2002; Patel
and Korotchkina, 2006). Importantly, the activity of PDKs may
be regulated at the transcriptional level (Kim et al., 2006; Patel
and Korotchkina, 2006), suggesting that PDC dysfunction may
be related to transcriptional deregulation in HD.
Gene transcription is generally favored by histone acetylases,
whereas histone deacetylases (HDACs) mainly repress gene ex-
pression (Struhl, 1998), similarly to mHTT. Previous studies
indicated that HDAC inhibitors (HDACIs) ameliorate the tran-
scriptional changes in HD (Zuccato et al., 2010). A phase I clinical
trial with the HDACI sodium phenylbutyrate (PB) showed recov-
ery of expression of 12 transcripts that were upregulated in un-
treated HD patients (Borovecki et al., 2005; Hogarth et al., 2007;
Brett et al., 2014). In addition, we showed previously that mHTT
induced deficits in mitochondrial calcium handling, which was
ameliorated by treatment with HDACIs, namely trichostatin A
(TSA), a pan-lysine HDACI, and sodium butyrate (SB), which
acts on class I and IIa HDACs, preferentially modulating Lys
acetylation sites in the nucleus (Oliveira et al., 2006; Scho¨lz et al.,
2015), similarly to PB. Therefore, in the present work, we ana-
lyzed the influence of TSA, SB, and PB in improving PDC func-
tion in striatal cells expressing full-length mHTT and confirmed
the effect of SB in YAC128 mouse model. Our results suggest
that the HDACI SB promotes PDH activity in both in vitro and
ex vivo models by regulating the expression of PDKs, helping
to counteract HD-related deficits in mitochondrial function.
In addition, SB improved brain mitochondrial bioenergetics
and positively influenced motor learning and coordination in
YAC128 mice.
Materials and Methods
Materials
Fetal bovine serum (FBS) and penicillin/streptomycin were from Invit-
rogen. DMEM, Opti-MEM, bovine serum albumin (BSA), trypan blue
(0.4%), nicotinamide adenine dinucleotide hydrate (NAD), coenzyme
A (CoA), carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone
(FCCP), thiamine pyrophosphate (TPP
⫹
), pyruvate, acetyl-L-carnitine
(ALCAR), DCA, TSA, SB, and PB were from Sigma Aldrich. Calcium
chloride and MgCl
2
were from Merck. PureZOL RNA isolation reagent,
iScript cDNA Synthesis kit, SsoFast EvaGreen Supermix, Protein Assay
and PVDF membranes were from Bio-Rad. The Phospho-PDH in-cell
ELISA kit and NAD/NADH kit were from Abcam. Lipofectamine 3000
was from Thermo Fisher Scientific. The XF Cell Mito Stress Test Kit and
XF24 cell culture microplates were from Seahorse Bioscience. Mini-
osmotic pumps were from Alzet. Isoflurane was from Esteve. All other
reagents were of analytical grade.
Cell culture and treatments
Immortalized striatal neurons derived from knock-in mice expressing full-
length normal (STHdh
7/7
, clone 2aA5, RRID:CVCL_M590, referred as wild-
type cells) or full-length mHtt with 111 glutamines (STHdh
111/111
, clone
109-1A, RRID:CVCL_M591, referred as mutant cells) were generously pro-
vided by Dr. Marcy MacDonald (Massachusetts General Hospital, Charles-
ton, MA) and treated according to slight modifications of established
procedures (
Trettel et al., 2000). Briefly, cells were grown adherent at 33°C in
DMEM supplemented with 10% FBS, 4 mM glutamine, 100 U/ml penicillin,
0.1 mg/ml streptomycin, and 400 g/ml geneticin in an atmosphere of 95%
air/5% CO
2
. Twenty-four hours before the experiment (and 1 d after cell
plating), the cells were incubated with TSA (10, 50, and 100 nM), SB (100,
250, and 500
M), PB (100, 250, and 500
M), ALCAR (100
M, 500
M, and
5mM), or DCA (300
M,1mM,and3mM).
For hypoxia-inducible factor 1 alpha (HIF-1
␣
) stabilization/accumu-
lation, striatal cells were maintained for 1 h before the experiment in a
controlled hypoxic atmosphere (0.1%–0.5% O
2
; Forma Anaerobic
System, Thermo Fisher Scientific) at 33°C. The HIF-1
␣
subunit is
continuously synthesized and degraded under normoxic conditions, but
accumulates rapidly after exposure to low oxygen tensions (Salceda and
Caro, 1997).
Constructs and transfection
Immortalized striatal cells were cultured in DMEM supplemented with
10% FBS and transfected on the following day (80% confluence) with the
pLKO.1/shRNA-PDK3 vector (clone ID: TRCN0000023835; Dharma-
con), designed by the RNAi Consortium, or the pLKO.1-scrambled neg-
ative control (TRC-RHS6848; Open Biosystems). STHdh cells were
transfected with 0.75
g of DNA/cm
2
of growth area in opti-MEM with
-
out FBS or antibiotics, following the Lipofectamine 3000 (ThermoFisher
Scientific) manufacturer instructions. Medium was changed 4 h after
transfection and cells were cultured for 48 h.
Total cell extracts
Adherent cells were washed 2 times in ice-cold PBS containing the fol-
lowing (in mM): 137 NaCl, 2.7 KCl, 1.8 KH
2
PO
4
,10Na
2
HPO
4
䡠 2H
2
O,
pH 7.4, and then scraped in 500
l of lysis buffer (150 mM NaCl, 50 mM
Tris, 5 mM EGTA, 1% Triton X-100; 0.5% deoxycholate, 0.1% SDS, pH
7.5) supplemented with 1 mM DTT, 1 mM PMSF, 1
g/ml protease cock-
tail inhibitor (containing chymostatin, pepstatin A, leupeptin, and anti-
pain), 1
M TSA (HDAC inhibitor), and 10 mM nicotinamide (sirtuins
inhibitor). Cellular extracts were frozen three times in liquid nitrogen
and then sonicated for 30 s and centrifuged at 20,800 ⫻ g for 10 min
(Eppendorf Centrifuge 5417R) to remove cell debris. The pellet was dis-
carded, the supernatant (total extract) collected, and the protein content
Correspondence should be addressedtoAna Cristina Rego, Ph.D.,CenterforNeuroscience and Cell Biology,
and Faculty of Medicine, Rua Larga University of Coimbra, polo I, 3004-504 Coimbra, Portugal. E-mail:
a.cristina.rego@gmail.com or acrego@cnc.uc.pt.
DOI:10.1523/JNEUROSCI.2006-14.2016
Copyright © 2017 the authors 0270-6474/17/372777-19$15.00/0
Naia, Cunha-Oliveira et al. • HDACIs Alleviate Energy Metabolism in HD J. Neurosci., March 8, 2017 • 37(10):2776 –2794 • 2777
quantified with the Bio-Rad protein assay. Extracts were stored at ⫺20°C
until being used for Western blotting experiments.
For the analysis of PDH activity, cell extracts were obtained by scrap-
ing the cells using 100
l of buffer containing 25.0 mM KH
2
PO
4
, 0.5 mM
EDTA, pH 7.25, supplemented with 0.01% Triton X-100.
NAD
⫹
/NADH quantification
The total levels of NAD
⫹
and NADH were measured using a commer
-
cially available kit (Abcam) following the manufacturer’s instructions.
Briefly, cells were plated in six-well plates and cell extracts were obtained
with NADH/NAD extraction buffer. The samples were frozen three times
in liquid nitrogen and centrifuged at 20,800 ⫻ g for 10 min (Eppendorf
Centrifuge 5417R) to remove cell debris. The extracted samples were
divided in two: one remained in ice (total NAD) and the other one was
heated to 60°C for 30 min to decompose NAD
⫹
(NADH). Both samples
were transferred to a 96-well plate and a NAD
⫹
-cycling mixture enzyme
was added and incubated at room temperature for 5 min to convert
NAD
⫹
to NADH. Finally, NADH developer was added into each well
and the reaction occurred at room temperature for 2 h. Absorbance was
monitored using a microplate reader Spectra Max Plus 384 (Molecular
Devices) at 450 nm.
PDH activity assay
PDH activity was assayed at 30°C by measuring the reduction of NAD
⫹
at 340 nm in a Spectra Max Plus 384 microplate reader (Molecular De-
vices) upon mixing 9
g of protein with 0.5 mM NAD
⫹
in the presence of
200
M TPP
⫹
,40
M CoA, and 4.0 mM pyruvate. The assay was per
-
formed in the presence of 2.5
M rotenone to prevent NADH consump-
tion by complex I (Zhou et al., 2009). PDH activity was calculated as
absorbance units per minute per milligram of protein and converted to
the percentage of wild-type control cells.
Determination of PDH E1alpha subunit protein levels
and phosphorylation
PDH expression and phosphorylation were assessed using the Phospho-
PDH In-Cell ELISA kit from Abcam and following the manufacturer’s
instructions. Briefly, cells were plated in 96-well plates, transfected with
shRNA-PDK3 or scramble shRNA control or incubated with the HDACI
compounds, and fixed with 3.7% formaldehyde in PBS. Cells were per-
meabilized, blocked, and incubated overnight with an antibody against
total PDH E1alpha, combined with another antibody against one of the
phosphorylated forms of PDH E1alpha at Ser293 (site 1), Ser300 (site 2),
or Ser232 (site 3). Separate secondary antibodies were used against PDH
E1alpha or the phosphorylated serines and developed by dual colorimet-
ric detection. Cell density was normalized after Janus green staining.
Absorbance was monitored using a microplate reader Spectra Max Plus
384 (Molecular Devices).
Cell viability assessment
At the time of cell incubation with the different compounds, the culture
medium was replaced for fresh medium. The nuclear morphology of the
striatal wild-type and mutant cells exposed to different concentrations of
the compounds (TSA, PB, SB, ALCAR, and DCA) was analyzed by fluo-
rescence microscopy using a double-staining procedure with Hoechst
33342 and propidium iodide (PI). After a washing step with saline me-
dium containing the following (in m
M): 120 NaCl, 3.5 KCl, 0.4 KH
2
PO
4
,
20 HEPES, 5 NaHCO
3
, 1.2 Na
2
SO
4
, and 15 glucose, the cells were incu
-
bated with 4
g/ml Hoechst 33342 for 15 min, washed 2 times with saline
medium, and incubated with 4
g/ml PI for 3 min in the dark at room
temperature. Cells were washed three times in saline medium to remove
extracellular dyes and further examined and scored as depicted in Figure
2 by analyzing images taken in the AxioObserver Z1 upright microscope
(Zeiss).
Analysis of oxidative stress
Cells ( plated in 96-well plates) were loaded for 30 min with 20
M
dihydro-dichlorofluorescein diacetate (DCFH
2
-DA), a stable, nonfluo
-
rescent, cell-permeable compound, at 33°C in Na
⫹
medium containing
the following (in mM): 135 NaCl, 5 KCl, 0.4 KH
2
PO
4
, 1.8 CaCl
2
,20
HEPES, and 5.5 glucose, pH 7.4. When internalized by the cell,
DCFH
2
-DA is hydrolyzed to DCFH by intracellular esterases, retained
inside the cell, and rapidly oxidized to the highly green fluorescent com-
pound 2,7-dichlorofluorescein (DCF), a redox indicator probe that re-
sponds to changes in intracellular iron signaling and peroxynitrite
formation (
Kalyanaraman et al., 2012). DCF fluorescence (480 nm exci-
tation, 550 nm emission) was monitored continuously for1hat33°C
using a Spectrofluorometer Gemini EM (Molecular Devices) microplate
reader. DCF fluorescence was corrected for variations in total protein
content in the wells after quantification of cell protein in each well by the
Bio-Rad protein assay.
Oxygen consumption rate (OCR) measurements
Oxygraph oxygen electrode. To measure the OCR, cells were plated in 75
cm
2
flasks and incubated on the following day in the absence or presence
of 500
M SB, 250
M PB, or 3 mM DCA for 24 h. Cells were detached by
trypsinization and trypsin was inactivated by adding culture medium
containing FBS. The cells were then washed in prewarmed (33°C) respi-
ration buffer containing the following (in mM): 132 NaCl, 4 KCl, 1 CaCl
2
,
1.2 NaH
2
PO
4
, 1.4 mM MgCl
2
, 6 glucose, and 10 HEPES, pH 7.4, and
resuspended in 300
l of respiration buffer. The rate of O
2
consumption
was recorded under constant magnetic stirring in an Oxygraph DW1
oxygen electrode chamber (Hansatech Instruments) previously cali-
brated for the oxygen dissolved at 33°C in oxygenated water (maximum)
and in N
2
-saturated water (minimum). After recording the basal rate of
oxygen consumption, cells were sequentially exposed to oligomycin (10
g/ml) to avoid the reversal of ATP synthase and FCCP (2.5
M) to assess
maximal respiration rates. Potassium cyanide (KCN) (700
M) was then
added to confirm mitochondrial O
2
consumption, manifested as a de
-
crease in O
2
consumption due to inhibition of mitochondrial complex
IV. At the end of the experiment, the cells were pelleted, resuspended in
1 M NaOH, and protein was quantified by the Bio-Rad protein assay.
Results are expressed in nanomoles of O
2
per minute per milligram of
protein.
Cell respirometry. OCR of transfected striatal cells was measured using
the XF24 flux analyzer (Seahorse Bioscience) following the manufactur-
er’s instructions. Striatal cells were grown on the 24-well custom plates at
30,000 cells per well and transfected 48 h before the experiment (as
described in the “Constructs and transfection” session) or incubated
with 500
M SB 24 h before the experiment. Before running the experi-
ment, cell culture medium was removed and cells were washed once with
1 ml of DMEM 5030 medium, pH 7.4, supplemented with 4.5 g/L glucose
and4m
M glutamine and then incubated in 450
l of supplemented
DMEM at 33°C in a CO
2
-free incubator for 1 h. Four baseline measure
-
ments of OCR were sampled before sequential injection of mitochon-
drial inhibitors and four metabolic determinations were sampled after
the addition of each mitochondrial inhibitor before injection of the sub-
sequent inhibitors. The mitochondrial inhibitors used were oligomycin
A(1
M), FCCP (0.9
M), and antimycin A (1
M) plus rotenone (1
M).
OCR was automatically calculated and recorded using Seahorse Biosci-
ence software. After the assays, the protein level was determined for each
well to confirm equal cell density per well. Results are expressed in pico-
moles of O
2
per minute per microgram of protein.
YAC128 mice surgery
YAC128 mice, described previously by
Slow et al. (2003), express full-
length mutant HTT with ⬃128 CAG repeats from a yeast artificial chro-
mosome (YAC) transgene (RRID: MGI_MGI:3613515). YAC128 (line
HD53) and wild-type mice were housed in the housing facility of the
Center for Neuroscience and Cell Biology and Faculty of Medicine at the
University of Coimbra (Coimbra, Portugal) under conditions of con-
trolled temperature (22–23°C) and under a 12 h light/12 h dark cycle with
lights on at 07:00 h. All studies were performed according to the Helsinki
Declaration and the Guide for the Care and Use of Laboratory Animals
from the National Institutes of Health. Food and water were available ad
libitum throughout the experiment. Animals used in this study (total of
9–13 9-month-old male mice per treatment) were maintained in accor-
dance with the guidelines of the Institutional Animal Care and Use of
Committee and were performed in accordance with the European Com-
munity directive (2010/63/EU). All efforts were made to minimize ani-
2778 • J. Neurosci., March 8, 2017 • 37(10):2776 –2794 Naia, Cunha-Oliveira et al. • HDACIs Alleviate Energy Metabolism in HD
mal suffering and to reduce the number of animals used. Animals were
subjected to small surgeries for the subcutaneous implantation (and re-
moval) of mini-osmotic pumps (28 d pumps, model 2004; Alzet). Each
pump was filled with saline (0.9% NaCl) or SB (1 mg/kg/d) diluted in
saline solution. For the subcutaneously implantation of the pumps, ani-
mals were anesthetized with isofluorane using a vaporizer apparatus (EZ
Anesthesia) and immediately afterward received a subcutaneous injec-
tion of butorphanol (1:50; 2.5 ml/kg), a morphine-type synthetic opioid
analgesic. Animals were kept with the pumps for 30 d (2 more days than
the treatment itself because, in order to start releasing the solutions, the
pump needs to be at the same temperature as the animal’s body). To
remove the pumps, mice were subjected to the same surgical procedure
as described above. The body weight was measured for all animals after
treatment. Behavior analysis was performed 2 d after the surgery (recov-
ery period).
Rotarod motor evaluation
Mice were allowed to adapt to the behavior test room for 2 h before the
behavior studies. Procedures were consistent for all subjects and tests
made at minimum noise levels. Motor coordination and motor learning
were assessed on a rotarod apparatus (Letica Scientific Instruments) us-
ing a previously described procedure (
Naia et al., 2016b). Behavior
equipment was cleaned thoroughly with 70% ethanol before each animal
trial. Separate cohorts of mice were trained at a fixed speed of 4 rotations
per minute (rpm) to a maximum time of 40 s. The training time (motor
learning) corresponds to the time it took the mice to stand on the rod
without falling for 40 consecutive seconds. Motor coordination and bal-
ance were assessed using the accelerating rotarod task after all of the
animals learned the previous task. In this test, the rotarod accelerated
from 5 to 40 rpm over 5 min and the latency to fall was recorded for each
of the four trials for each mouse. Resting time between each trial was set
for 30 min.
Total fractions of YAC128 mice cortex
Total fractions from cerebral cortex were obtained from SB or saline-
treated YAC128 and wild-type mice brains. Animals were killed by very
rapid decapitation in accordance to European Union guideline 86/609/
EEC and Annex II of Portuguese Law No. 113/2013. Mice brains were
collected and the cortex from one brain hemisphere was dissected on ice
(0– 4°C) and homogenized in lysis buffer, as described for total cell ex-
tracts, which was supplemented with 100 n
M okadaic acid, 25 mM NaF, 1
mM Na
3
VO
4
,1mM DTT, 1 mM PMSF, 1
g/ml protease inhibitor mix
-
ture (chymostatin, pepstatin A, leupeptin, and antipain), 1
M TSA
(HDACI) plus 10 mM nicotinamide (sirtuins inhibitor) at 4°C. Samples
were then incubated on ice for 30 min and centrifuged at 20,800 ⫻ g for
10 min at 4°C to discard intact organelles and dead cells. All samples were
kept at ⫺80°C until use.
Western blotting
Total extracts, obtained as described previously, were denatured with
denaturing buffer (50 mM Tris-HCl pH 6.8, 2% SDS, 5% glycerol, 600
mM DTT, and 0.01% bromophenol blue) at 95°C, for 5 min. Equivalent
amounts of protein were separated by 7.5–12% SDS-PAGE gel electro-
phoresis and electroblotted onto PVDF membranes (Bio-Rad). The
membranes were blocked for1hinTris-buffered saline (TBS) solution
containing 0.1% Tween (TBS-Tween) and 5% fat-free milk or 5% BSA,
followed by an overnight incubation with primary antibodies (
Table 1)at
4°C with gentle agitation. Membranes were then washed 3 times for 15
min each time with TBS-Tween and incubated with secondary antibodies
conjugated with alkaline phosphatase (1:20,000) for1hatroom temper-
ature with gentle agitation. Immunoreactive bands were visualized after
incubation with ECF substrate (GE Healthcare), which is converted into
a highly fluorescent product reagent by alkaline phosphatase. This prod-
uct strongly fluoresces at 540 –560 nm when the blots are illuminated
with UV light (maximum excitation at 430 nm) in a Versa Doc 3000
Imaging System (Bio-Rad). Bands were quantified using Quantity One
software (Bio-Rad).
Measurement of intracellular levels of adenine nucleotides
Striatal cells were washed and then scraped with ice-cold PBS and cortex
from one mouse brain hemisphere was homogenized with ice-cold PBS.
The same volume of 0.6 M perchloric acid supplemented with 25 mM
EDTA-Na
⫹
was added. Extracts were then centrifuged at 20,800 ⫻ g for
2 min at 4°C to remove cell debris. The resulting pellet was solubilized
with 1 M NaOH and further analyzed for total protein content by the
Bio-Rad protein assay. After neutralization with 10 M KOH/1.5 M Tris,
samples were centrifuged at 20,800 ⫻ g for 5 min at 4°C. The resulting
supernatants were assayed for ATP, ADP, and AMP determination by
separation in a reverse-phase high-performance liquid chromatography
with detection at 254 nm. The chromatographic apparatus used was a
Beckman-System Gold controlled by a computer. The detection wave-
length was 254 nm and the column used was a Lichrospher 100 RP-18 (5
m). An isocratic elution with 100 mM phosphate buffer (KH
2
PO
4
), pH
6.5, and 1% methanol was performed with a flow rate of 1 ml/min. Peak
identity was determined by following the retention time of standards.
Energy charge was used as a measurement of general energy status of
mice cortex, taking into account ATP, ADP, and AMP concentrations,
and was calculated according to the following formula: (ATP ⫹
1/2*ADP)/(ATP ⫹ ADP ⫹ AMP).
Analysis of gene expression by qRT-PCR
Total RNA was extracted with TRIzol reagent (Life Technologies) for cell
extracts or with Purezol (Bio-Rad) for tissue extracts, following the man-
ufacturers’ protocols, and quantified using a Nanodrop 2000 (Thermo
Fisher Scientific), confirming that A260/280 and A260/230 were ⬎1.9.
RNA integrity was verified with the Experion RNA StdSens kit (Bio-Rad)
for cell extracts or by 1.0% agarose gel electrophoresis for tissue extracts
and RNA was converted into cDNA using the iScript cDNA synthesis kit
(Bio-Rad) following the manufacturer’s instructions. qRT-PCR was per-
formed using the SsoFast Eva Green Supermix in a CFX96 real-time PCR
system (Bio-Rad), with the primers defined in
Table 2 at 500 nM (for cell
extracts) or 300 nM (for tissue extracts). Amplification of 12.5 ng (for cell
Table 1. List of the primary antibodies used in this work
Primary antibody
Molecular
weight
(kDa) Reference Dilution
Host
species
Anti-acetyl-histone 3 17 06-599 (Millipore, RRID: AB_2115283) 1:750 Rabbit
Anti-histone 3 17 #9715 (Cell Signaling Technology, RRID:
AB_331563)
1:2000 Rabbit
Anti-PDH kinase 1
(PDK1)
47 #3820 (Cell Signaling Technology, RRID:
AB_1904078)
1:1000 Rabbit
Anti-PDK kinase 3
(PDK3)
47 ab182574 (Abcam, RRID: AB_2631347) 1:1000 Rabbit
Anti-pSer232
PDH-E1
␣
44 AP1063 (Millipore, RRID: AB_10616070) 1:750 Rabbit
Anti-PDH-E1
␣
43 #2784 (Cell Signaling Technology, RRID:
AB_2162928)
1:1000 Rabbit
Anti-PDH
phosphatase
(PDPc)
53 sc-87354 (Santa Cruz, RRID: AB_2017921) 1:200 Goat
Anti-HIF-1
␣
⬃130 Ab1 (Abcam, RRID: AB_296474) 1:500 Mouse
Anti-tubulin 50 T6199 (Sigma-Aldrich, RRID: AB_477583) 1:1000 Mouse
Anti-

-actin 42 A5316 (Sigma-Aldrich, RRID: AB_476743) 1:5000 Mouse
Table 2. Sequence (5ⴕ–3ⴕ) of primers used for qRT-PCR experiments
Gene Accession no. Forward primer Reverse primer
PDHA1 NM_008810 TCATCACTGCCTATCGAGCAC GTTGCCTCCATAGAAGTTCTTGG
PDK1 NM_172665 GGACTTCGGGTCAGTGAATGC CGCAGAAACATAAACGAGGTCT
PDK2 NM_133667.2 CGGACTCTAAGCCAGTTCACA GTGGGCACCACGTCATTGT
PDK3 NM_145630.2 CCGTCGCCACTGTCTATCAA TGCGCAGAAACATATAGGAAGTTT
PDK4 NM_013743.2 CCGCTTAGTGAACACTCCTTC TCTACAAACTCTGACAGGGCTTT
PDP1 NM_001098231.1 CGGGCACTGCTACCTATCCTT ACAATTTGGACGCCTCCTTACT
NM_001098230.1
NM_001033453.3
PDP2 NM_001024606.1 GGCTGAGCATTGAAGAAGCATT GCCTGGATTTCTAGCGAGATGT
TBP NM_013684.3 ACCGTGAATCTTGGCTGTAAAC TCAGCATTTCTTGCACGAAGT
Naia, Cunha-Oliveira et al. • HDACIs Alleviate Energy Metabolism in HD J. Neurosci., March 8, 2017 • 37(10):2776 –2794 • 2779
extracts) or 10 ng (for tissue extracts) of template was performed with an
initial cycle of 30 s at 95.0°C, followed by 40 cycles of5sat95°C plus 5 s
at 63°C (for cell extracts) or 61°C (for tissue extracts). At the end of each
cycle, Eva green fluorescence was recorded to enable determination of
Cq. After amplification, melting temperature of the PCR products were
determined by performing melting curves, with 0.5°C increments every
5 s, from 65°C to 95°C, with fluorescence recording after each tempera-
ture increment. For each set of primers, amplification efficiency was
assessed and no template and no transcriptase controls were run. For cell
extracts, relative normalized expression was determined with CFX96
Manager software (version 3.0; Bio-Rad), using TATA-binding protein
(TBP) as the reference gene. Normalization was also performed for the
reference gene hypoxanthine guanine phosphoribosyl transferase (Hprt;
accession number NM_013556.2), with similar results. For tissue ex-
tracts, beta-actin (NM_007393.3), POLR2F (NM_027231.1), GAPDH
(NM_008084.2), and Hprt (NM_013556.2) were tested as reference
genes, but none of these genes was stable in YAC128 mice treated with SB.
Therefore, normalizations were performed in relation to wild-type mice
treated with saline.
Statistical analysis
Results are expressed as mean ⫾ SEM of the number of independent
experiments indicated in the figure legends. Comparisons between mul-
tiple groups were performed with two-way or one-way ANOVA, fol-
lowed by Bonferroni post hoc test, for comparison between experimental
groups. Comparison between two groups was performed with the Stu-
dent’s t test (GraphPad Prism Version 6.0, RRID: SCR:002798). The F
test was performed to determine the interaction term. Significance was
accepted at p ⬍ 0.05.
Results
PDH dysfunction in cells expressing mutant huntingtin
Mitochondrial bioenergetics has been described to be altered in
HD (
Cunha-Oliveira et al., 2012). PDC is a central mitochondrial
enzyme complex in cellular bioenergetics, linking glycolysis to
the TCA cycle and oxidative phosphorylation (Harris et al., 2002;
Patel and Korotchkina, 2006). PDC activity can be modulated by
phosphorylation (Kolobova et al., 2001) and deregulation of
PDC activity and phosphorylation was previously shown by
us in HD human cybrids, together with decreased mitochondrial
NADH/total NAD ratio (Ferreira et al., 2011).
Using HD mutant mouse striatal cells, we observed an in-
crease in mitochondrial NAD
⫹
/NADH levels by 2.3-fold (p ⬍
0.001;
Fig. 1A), suggesting a dysfunction in cellular dehydroge-
nases. We also assessed PDH activity by following the production
of NADH that accompanies the irreversible conversion of pyru-
vate into acetyl-CoA; a decrease in PDH activity by ⬃30% ( p ⬍
0.0001) was observed in mutant cells, compared with wild-type
striatal cells (Fig. 1B), which may explain the substantial increase
in pyruvate levels detected in mutant cells (Naia et al., 2016a).
The protein levels of the E1alpha subunit of PDH, along with its
Figure 1. Deregulation ofNAD
⫹
/NADH ratioand PDH activity,E1alpha subunit expression,and phosphorylation in cells expressing mutant huntingtin (STHdh
Q111/Q111
). NAD
⫹
/NADH ratio(A)
andPDH activity(B)were measuredin totalextracts of wild-typeandmutant cells.TotalPDH E1alphaproteinlevels (C)and phosphorylation atregulatoryserines 293(site1), 300(site2), or232 (site
3) (D) wereassessed usingthe phospho-PDH incell ELISAkit. Protein levelsof PDK1(E), PDK3 (F ), andPDP1c (G) inwild-type andmutant cells weredetermined byWestern blotting. Dataare shown
as the mean ⫾ SEM of total NAD
⫹
/NADH ratio (in pmol/mg protein), total PDH activity (in slope/mg protein), PDH E1alpha/Janus green, phosphorylated form/total PDH E1alpha. Where
mentioned, the average value obtained for wild-type was considered 100%. Absolute values for wild-type controls were 1178.00 ⫾ 247.00 AU/min/mg protein (B), 3.60 ⫾ 0.98 (C), 3.94 ⫾ 1.16
for pSer293, 3.28 ⫾ 0.62 for pSer300, and 4.32 ⫾ 1.42 for pSer232 (D). Statistical analysis was performed by Student’s t test;
t
p ⬍ 0.05,
ttt
p ⬍ 0.001, and
tttt
p ⬍ 0.0001 compared with the
respective wild-type/control cells.
2780 • J. Neurosci., March 8, 2017 • 37(10):2776 –2794 Naia, Cunha-Oliveira et al. • HDACIs Alleviate Energy Metabolism in HD