ASC-1 transporter-dependent amino acid uptake is

required for the efficient thermogenic response of human

adipocytes to adrenergic stimulation
Rini Arianti1,2 , Bogla


rka Agnes
ta B. To
Vinnai1, Bea
Cso

th1, Abhirup Shaw1,2 , Eva } sz1 ,
Attila Va
mos , Ferenc Gyo
1,2
} ry , Pamela Fischer-Posovszky , Martin Wabitsch , Endre Kristo
3 4 4

f1
and La
szlo

Fe

su
€s 1

1 Department of Biochemistry and Molecular Biology, Faculty of Medicine, University of Debrecen, Hungary
2 Doctoral School of Molecular Cell and Immune Biology, University of Debrecen, Hungary
3 Department of Surgery, Faculty of Medicine, University of Debrecen, Hungary
4 Division of Pediatric Endocrinology and Diabetes, University Medical Center Ulm, Germany

Correspondence Brown and beige adipocytes dissipate energy by uncoupling protein 1

f and L. Fe
E. Kristo
su
€s, Department of (UCP1)-dependent and UCP1-independent thermogenesis, which may be uti-
Biochemistry and Molecular Biology, Faculty
lized to develop treatments against obesity. We have found that mRNA and
of Medicine, University of Debrecen,
protein expression of the alanine/serine/cysteine transporter-1 (ASC-1) was
H-4032 Debrecen, Hungary
Tel: +36 52 416 432 induced during adipocyte differentiation of human brown-prone deep neck
E-mail: kristof.endre@med.unideb.hu (E.K.); and beige-competent subcutaneous neck progenitors, and SGBS preadipo-
fesus@med.unideb.hu (L.F.) cytes. cAMP stimulation of differentiated adipocytes led to elevated uptake
of serine, cysteine, and glycine, in parallel with increased oxygen consump-
Endre Kristo
f and L

Fe
aszlo
su
€s contributed tion, augmented UCP1-dependent proton leak, increased creatine-driven sub-
equally to this work.
strate cycle-coupled respiration, and upregulation of thermogenesis marker
genes and several respiratory complex subunits; these outcomes were impeded
(Received 20 April 2021, revised 16 June
2021, accepted 28 June 2021) in the presence of the specific ASC-1 inhibitor, BMS-466442. Our data sug-
gest that ASC-1-dependent consumption of serine, cysteine, and glycine is
doi:10.1002/1873-3468.14155 required for efficient thermogenic stimulation of human adipocytes.

Edited by Peter Brzezinski Keywords: adipocytes; ASC-1 inhibition; gene expression; obesity; proton
leak respiration; thermogenesis; uncoupling protein 1

The activation of nonshivering thermogenesis of brown multilocular lipid droplets, high expression of UCP1,
and beige adipocytes located in either brown or white and large amounts of mitochondria [2]. Recently,
adipose tissue, respectively, dissipates energy sources UCP1-independent thermogenesis in beige adipocytes
to heat. Uncoupling protein 1 (UCP1) plays a major has been described, including a creatine-driven sub-
role in their heat generation by creating a proton leak strate cycle [3,4].
in the inner membrane of mitochondria uncoupling In mice, brown and beige cells have different pro-
the mitochondrial respiratory chain from ATP synthe- genitors and locations exhibiting distinct gene expres-
sis [1]. Both thermogenic adipocytes possess sion and functional signatures [5–8]. It has been

Abbreviations
ASC-1, alanine/serine/cysteine transporter-1; CKB, creatine kinase B; DN, deep neck; ECAR, extracellular acidification rate; FTO, fat mass
and obesity-associated; GSH, glutathione; hASCs, human adipose-derived stromal cells; NADPH, nicotinamide adenine dinucleotide phos-
phate; OCR, oxygen consumption; SC, subcutaneous; SGBS, Simpson-Golabi-Behmel syndrome; TCA, tricarboxylic acid; UCP1, Uncoupling
protein 1.

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 1
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and
distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

reported that brown adipose tissue adipocytes are small neutral amino acids, including alanine, serine,
derived from myogenic precursors, while beige adipo- cysteine, and glycine. ASC-1 is the main regulator of
cytes have a common origin with white ones develop- extracellular D-serine levels in synaptic systems [25].
ing from mesenchymal progenitors of white adipose The ASC-1 transporter was previously listed as a white
tissue and activated in response to cold and adipocyte-specific cell surface marker [26]. It has been
b-adrenergic stimuli, physical exercise, and PPAR-c recently reported that mRNA expression of ASC-1 is
stimulation [9,10], which is often referred to as “adipo- high and induced in adipocytes freshly isolated from
cyte browning.” In humans, brown and variably abdominal subcutaneous and omental white adipose
“brownable” adipose tissue depots are interspersed in tissue and shows strong inverse correlation with vis-
several anatomical regions, including cervical deep ceral obesity and insulin resistance in humans [27], and
neck (DN), supraclavicular, axillary, paraspinal, and it promotes mitochondrial respiration and insulin-
mediastinal depots [11–13]. Unlike in rodents, there is stimulated glucose uptake.
no clear distinction regarding the origin and molecular Here, we report the induction of ASC-1 during
signature of human brown adipocytes and adipose tis- white and brown differentiation of DN and SC-
sue. Several studies have reported that human brown derived neck adipocytes and Simpson-Golabi-Behmel
adipocytes isolated from DN or supraclavicular area syndrome (SGBS) cells that model beige adipogenic
resemble murine beige adipocytes [14–17] while others differentiation [28–30]. The presence of a specific ASC-
proposed that they closely resemble classic murine 1 inhibitor resulted in decreased serine, cysteine, and
brown fat but that some beige adipocytes might also glycine consumption during cAMP stimulation of DN-
be present in DN [18,19]. Using stromal vascular frac- derived white and brown adipocytes with the conse-
tions from brown and white human adipose tissue quence of decreased UCP1-dependent and UCP1-
depots and ex vivo brown and white differentiation independent respiration as well as reduced thermogenic
protocols, adipocyte cultures with high and low pro- gene inductions. cAMP-induced respiratory and gene
portion of thermogenesis competent cells can be expression changes could be also reduced by the ASC-
obtained [20–23]. 1 inhibitor in differentiated adipocytes of SC and
Emerging evidence suggests that beige adipocytes in SGBS cell origin. The results demonstrate an impor-
white adipose tissue persist in a masked form possess- tant role of ASC-1 in regulation of human adipocyte
ing the morphology of white adipocytes [2]. Their acti- thermogenesis.
vation can quickly occur, usually through adrenergic
stimulation, involving inhibition of ongoing mitophagy
[23]. Genetic predisposition contributes to the size of Materials and methods
the beige cell population; an intronic single nucleotide
polymorphism of fat mass and obesity-associated Materials
(FTO) gene shifts the balance between white and beige
All chemicals were from Sigma-Aldrich (Munich, Germany)
progenitors toward the former [24]. We have recently
unless stated otherwise.
reported that the FTO status strongly influences the
thermogenic potential of neck area adipocytes regard-
less of type of depot and differentiation protocol [22]. Ethics statement and obtained tissue samples
The enhancement of adipocyte thermogenesis can be
Tissue collection was approved by the Medical Research
a promising approach in treating obesity. Therefore, it Council of Hungary (20571-2/2017/EKU) followed by
is important to reveal all molecular elements of the EU Member States’ Directive 2004/23/EC on pre-
thermogenic regulation. Recently, we used high- sumed consent practice for tissue collection. All experi-
throughput RNA sequencing technology to analyze ments were carried out in accordance with the
global gene expression patterns of ex vivo differenti- guidelines of the Helsinki Declaration. Written informed
ated human DN and subcutaneous (SC) adipocytes consent was obtained from all participants before the
that had equal differentiation capacity [22] and found surgical procedure. During thyroid surgeries, a pair of
a set of differentially expressed genes when they were DN and SC adipose tissue samples was obtained to
compared. One of these genes was SLC7A10 which rule out interindividual variations. Patients with known
showed higher expression in DN-derived adipocytes, diabetes, malignant tumor, or with abnormal thyroid
particularly when they were differentiated according to hormone levels at the time of surgery were excluded.
a brown protocol. SLC7A10 encodes the alanine/ Human adipose-derived stromal cells (hASCs) were iso-
serine/cysteine transporter-1 (ASC-1) that facilitates lated from SC and DN fat biopsies as described previ-
the sodium-independent, bidirectional transport of ously [22,31].

2 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies
R. Arianti et al. ASC-1 transporter is needed for adipocyte thermogenesis

Differentiation and treatment of hASCs and Oxygen consumption and extracellular

SGBS preadipocytes acidification rate measurement
White and brown adipocytes were differentiated from stro- Oxygen consumption (OCR) and extracellular acidification
mal vascular fraction of adipose tissue containing hASCs rate (ECAR) were measured using an XF96 oximeter (Sea-
or SGBS preadipocytes according to described protocols horse Biosciences, North Billerica, MA, USA) as described
[20,21]. Differentiated adipocytes were maintained in previously [33]. After recording the baseline OCR, 500 µM
DMEM-F12-HAM medium and treated with a single bolus dibutyryl-cAMP, 100 nM BMS-466442, or combination of
of 500 µM dibutyryl-cAMP (cat#D0627) for 10 h to mimic the two compounds was injected to the cells. Then, stimu-
in vivo cold-induced thermogenesis [23]. BMS-466442 lated OCR was recorded every 30 min. The adipocytes
(Aobious INC, Gloucester, MA, USA cat#AOB6567) was were treated with the creatine analogue 2 mM b-
administered in 100 nM to inhibit ASC-1 transporter activ- guanidinopropionic acid (b-GPA) which interferes with
ity [32]. creatine-driven substrate cycle [3,34]. Proton leak respira-
tion was determined after injecting oligomycin at 2 lM con-
centration. Cells received a single bolus of antimycin A at
Quantification of amino acids and calculation of 10 lM concentration for baseline correction (measuring
their uptake by cells nonmitochondrial respiration). The OCR was normalized
to protein content.
Frozen cell culture supernatants were filtered using 3 kDa
filters (Pall Corporation, Port Washington, NY, USA), and
then, 10 lL of filtrate was derivatized with AccQ
Tag Ultra Statistical analysis
Derivatization Kit (Waters, Milford, MA, USA). Chro-
The results are expressed as mean  SD. Normality of the
matographic separation was carried out on H-class UPLC
data was tested by Kolmogorov–Smirnov test. For multiple
(Waters) using AccQTag Ultra Column (2.1 x 100 mm),
comparisons of groups, statistical significance was deter-
AccQTag Eluent A and B, and gradient provided in the
mined by one-way analysis of variance followed by Tukey
AccQTag Ultra Chemistry Kit (Waters). Detection of
post hoc test. In comparison of two groups, two-tailed
amino acid derivatives was performed at 260 nm in the
paired Student’s t-test was used. The data were visualized
PDA detector of the UPLC. Concentration of the amino
and analyzed by using GRAPHPAD PRISM 8 (GraphPad Soft-
acids was calculated with the Empower software (Waters)
ware, San Diego, CA, USA).
using a 7-point calibration curve.
Flux of amino acids into or from adipocytes was calcu-
lated by comparing concentration differences measured at Results
starting and end point of 10 h of dibutyryl-cAMP treat-
ment with or without the presence of ASC-1 inhibitor. The Alanine/serine/cysteine transporter-1 is induced
number of cells in wells was calculated using KOVA glass- in human white and brown adipocytes
tic slide 10 with grids (Kova International Inc, Garden differentiated from deep neck and subcutaneous
Grove, CA, USA, cat#K304680) as described (https://www. progenitors and SGBS cells
kovaintl.com/downloads/DI-91064-17).
In an attempt to extend our knowledge about regula-
tory mechanism of thermogenesis, we have started to
RNA isolation, RNA-seq analysis, and study adipocyte populations differentiated to white
quantitative real-time PCR (RT-qPCR) and brown cells at the same extent from stromal vas-
Cells were collected, total RNA was isolated, and RT- cular fraction of paired DN and SC adipose tissue
qPCR was performed as described previously [28,33]. Gene sites of nine donors and compared their global gene
primers and probes were designed and supplied by Thermo expression patterns by global RNA sequencing [22].
Fisher Scientific (Waltham, MA, USA) as listed in We found that according to mRNA expression-based
Table S1. Global transcriptome analysis by high- ProFAT and BATLAS scores [35,36], DN-derived
throughput mRNA sequencing was performed on Illumina brown and even white adipocytes had high browning
sequencing platform as detailed in our recent paper [22]. potential probability compared with SC cells and pos-
sessed elevated expression of brown-marker genes
(e.g., UCP1, CKMT1A/B, CPT1A/B, CIDEA,
Immunoblotting and densitometry
PM20D1, DIO2, LEPR, FABP3); the expression pro-
Immunoblotting and densitometry were carried out as file of these genes showed a very similar pattern in the
described previously [23]. Antibodies and working dilutions samples according to Pearson’s correlation analysis
are listed in Table S2. (Fig. S1). Looking for potential thermogenesis

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 3
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

regulators, we found that UCP1 and the brown- inhibitor treatment. Alanine release may indicate that
marker genes were clustered with SLC7A10 which brown adipocytes degrade amino acids for fueling the
encodes the amino acid transporter ASC-1. SLC7A10 tricarboxylic acid (TCA) cycle and convert nitrogen to
mRNA was elevated in DN as compared to SC adipo- pyruvate forming alanine as part of a possible glucose/
cytes with higher levels in brown than in white cells alanine cycle in vivo [37]. Since ASC-1 operates prefer-
(Fig. 1A). The same expression pattern of ASC-1 and entially, although not exclusively, in an exchange mode
UCP1 was observed when we validated the RNA-seq [38], alanine release may allow uptake of more serine,
data by RT-qPCR analysis in adipocytes obtained cysteine, and glycine. In this context, we also observed
from a separate set of donors (Fig. 1B). We noticed significant release of glutamate which can be trans-
that the expression of ASC-1 was low in both DN and ported by ASC-1.
SC preadipocytes and was induced during the brown
and white differentiation process in parallel with vari-
ASC-1 inhibition hinders UCP1-dependent and
ous levels of UCP1 induction. The transporter protein
UCP1-independent oxygen consumption upon
was present in the adipocytes after differentiation and
cAMP stimulation
neither the differentiation protocol nor the anatomical
origin of the cells resulted in consistently different Next, we investigated the importance of ASC-1 in ther-
ASC-1 protein levels (Fig. 1C) while UCP1 levels were mogenic activation of adipocytes by monitoring oxygen
higher in DN and brown cells compared with SC and consumption rate and blocking ASC-1 transporter
white ones. activity by adding the inhibitor, BMS-466442 during
We have also investigated the expression of ASC-1 short-term adrenergic stimulation. Of note, the inhibitor
transporter in SGBS cells which can model beige dif- did not influence mRNA and protein expression of
ferentiation [28]. It was induced in differentiating ASC-1 in either the primary or SGBS adipocyte
SGBS adipocytes (Fig. 1D) and both white and brown (Fig. S2). As expected, OCR was elevated immediately
cells contained ASC-1 protein with a higher level in upon cAMP addition to white and brown DN adipo-
the former (Fig. 1E). In accordance with published cytes (Fig. 3A) which preserve their thermogenesis
results, SGBS beige adipocytes, obtained with brown potential during differentiation. The respiratory
differentiation protocol, showed higher level of UCP1 response of the brown adipocytes compared with white
mRNA as well as protein expression [28]. ones was higher in accordance with higher expression of
UCP1 and other thermogenesis-related genes [22]. The
ASC-1 inhibitor significantly reduced cAMP-stimulated
Facilitated serine, cysteine, and glycine uptake by
oxygen consumption in both types of adipocytes
adrenergic stimulation of DN adipocytes is
(Fig. 3A). Basal respiration was not affected by ASC-1
decreased in the presence of ASC-1 inhibitor
inhibition in spite of the decreased uptake of serine, cys-
To learn the functional significance of ASC-1 in adipo- teine, and glycine in resting adipocytes (Fig. 2), showing
cytes, we calculated the consumption of alanine, serine, that the transporter plays a respiration facilitating role
cysteine, and glycine by the differentiated brown and during thermogenic activation of adipocytes. To deter-
white DN adipocytes after their adrenergic stimulation mine the effect of the inhibitor on UCP1-dependent
by the cell-permeable dibutyryl-cAMP for 10 h. We portion of cellular respiration, we injected oligomycin
found that serine, cysteine, glycine, and also the ASC- that blocks ATP-synthase activity and makes estimation
1 substrate threonine were transported at a high basal of UCP1-dependent proton leak possible. This was
rate into unstimulated brown and white adipocytes reduced by the ASC-1 inhibitor in both white and
(significantly more serine and cysteine into white than brown cAMP-stimulated DN adipocytes (Fig. 3A)
into brown cells), which could be inhibited by the pointing to a significant influence of ASC-1 transported
addition of the specific noncompetitive ASC-1 inhibi- amino acids on mitochondrial proton gradient genera-
tor, BMS-466442 (Fig. 2), in a nanomolar concentra- tion and its uncoupling. Contribution of a UCP1-
tion which was effective in previous cellular independent creatine-driven substrate cycle to the stim-
experiments [32]. cAMP treatment led to significant ulated respiration was estimated by applying the cre-
facilitation of the uptake of these amino acids, and atine analogue, b-GPA, and calculating the consequent
this was suppressed in the presence of the ASC-1 inhi- reduction in oxygen consumption [3,33]; b-GPA-
bitor. Interestingly, alanine uptake was low into white inhibited portion of respiration was elevated following
adipocytes and brown cells even transported alanine cAMP treatment, and it could be blunted by the ASC-1
out both in resting and cAMP-stimulated conditions; inhibitor in both white and brown DN adipocytes. Non-
none of the latter were significantly influenced by the mitochondrial respiration (Fig. 3A), basal, and cAMP-

4 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies
R. Arianti et al. ASC-1 transporter is needed for adipocyte thermogenesis

Primary Adipocytes
UCP1 (qPCR)
ASC-1 (qPCR)
(A) ASC-1 (RNA-Seq) (B)
*

Normalized mRNA expression

200 0.05
0.03

Normalized mRNA expression

* * *
Normalized mRNA count

* * 0.04
150 *
0.02 0.03
100 *
** 0.02
* 0.01 **
50 0.01

0 0.00
0.00 PA W B PA W B
PA W B PA W B PA W B PA W B
SC DN SC DN
SC DN

(C) ASC-1 (Densitometry) UCP1 (Densitometry)

1.5 1.5
SC DN

Normalized optical density

*
*
PA W B PA W B 1.0 1.0

32 kDa UCP1
*
0.5 0.5
57 kDa ASC-1

55 kDa TUB 0.0

PA W B PA W B
0.0
PA W B PA W B
SC DN SC DN

SGBS Adipocytes

(D) ASC-1 (qPCR) UCP1 (qPCR) (E) SGBS

* ASC-1 (Densitometry) UCP1 (Densitometry)
0.020 0.40

*
0.35 PA W B Normalized optical density 2.5 1.5
Normalized mRNA
Normalized mRNA

0.30
0.015 0.25 2.0 *
expression
expression

0.20
32 kDa UCP1 1.0
1.5
0.010 0.15
0.10 57 kDa ASC-1 1.0
0.05 0.5
0.005
0.0005 55 kDa TUB 0.5

0.000 0.0000 0.0 0.0

PA W B PA W B PA W B PA W B

Fig. 1. mRNA and protein expression of alanine/serine/cysteine transporter-1 (ASC-1) in human neck area and Simpson-Golabi-Behmel
syndrome (SGBS) adipocytes. (A) ASC-1 mRNA expression determined by RNA-seq. n = 9 for all groups. Statistical analysis was performed
by one-way ANOVA with the option "each row represents matched, or repeated measures, data." (B) Validation of ASC-1 mRNA expression
by RT-qPCR. n = 6 for all groups. (C) Detection and quantification of ASC-1 and uncoupling protein 1 (UCP1) by immunoblotting and
densitometry. n = 7 for all groups. (D) mRNA and (E) protein expression of ASC-1 and UCP1; n = 5 for all groups. PA, preadipocytes;
W, white; B, brown differentiation. Unless indicated, statistical analysis was performed by paired t-test, *P < 0.05, **P < 0.01.

stimulated ECAR (Fig. 3B) were not influenced by inhi- The basal and cAMP-stimulated ECARs of SC and
bition of ASC-1. SGBS adipocytes were not affected by ASC-1 inhibitor
As we observed UCP1 protein expression in white and (Fig. S3C). These data suggest that SC neck and SGBS
brown SC and SGBS adipocytes, we also investigated the adipocytes also possess thermogenic potential and ASC-
respiratory response in these cells. The basal respiration 1-mediated amino acid transport is needed for its efficient
of brown SC and SGBS adipocytes was higher than of activation.
white ones (Fig. S3A-B). cAMP increased oxygen con-
sumption in both white and brown SC and SGBS adipo-
Inhibition of ASC-1 abrogates the upregulation of
cytes. ASC-1 inhibition significantly decreased cAMP-
thermogenic markers and mitochondrial
stimulated UCP1-dependent and UCP1-independent
complexes upon cAMP stimulation
(b-GPA-inhibited) oxygen consumption in both white
and brown SC and SGBS adipocytes, but did not affect Based on the observed effect of ASC-1 transporter inhi-
the basal and nonmitochondrial respiration (Fig. S3A-B). bition on cAMP-stimulated respiration of adipocytes,

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 5
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

Serine Cysteine Glycine

**
* *
100 ** 200 *
W
80 ** W
W
* B
** B B **
**
Consumption (nmol/105 cells) **
80 60
150
*
**
* 40
60 * *
*
* 100 **
** 20
40

50
20 0

0 0 –20
– + – + – + – + – + – + – + – + – + – +
cAMP (500 µM) – + – +
– – + + – – + + – – + + – – + + – – + + – – + +
BMS-466442 (100 nM)

Alanine Threonine Glutamate

20 W 250 W 0
B B
*
Consumption (nmol/105 cells)

* **
10 200

–50
0 150

–10 100
–100

–20 50
W
B
–30 0 –150
cAMP (500 µM) – + – + – + – + – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – – + + – – + + – – + + – – + + – – + + – – + +

Fig. 2. Effect of 10 h of thermogenic induction and ASC-1 inhibitor (BMS-466442) treatment on serine, cysteine, glycine, alanine, threonine,
and glutamate consumption by human deep neck area-derived adipocytes. W, white; B, brown differentiation; n = 3 for all groups. Statistical
analysis was performed by paired t-test, *P < 0.05, **P < 0.01.

we presumed that induction of thermogenic genes was We also investigated the effect of ASC-1 inhibitor
also affected. DN adipocytes had the tendency to on other brown/beige adipocyte markers in primary
express PGC1a, UCP1, and CKMT2 browning markers adipocytes. cAMP increased the mRNA expression of
at a higher extent than the SC ones (Fig. 4A) as PM20D1 [39], CKMT1A/B [3], ELOVL3 [40], CITED1
reported previously [22]. As anticipated, cAMP elevated [15], DIO2 [41], and TBX1 [16] which was blunted by
the mRNA and protein expression of these genes in ASC-1 inhibition (Fig. S4).
both white and brown SC and DN adipocytes, with In addition to thermogenic markers, we also investi-
some exceptions in white SC cells (Fig. 4B). Inhibition gated the effect of ASC-1 transporter inhibitor on pro-
of ASC-1 resulted in reduced cAMP-dependent upregu- tein expression of mitochondrial complex subunits.
lation of PGC1a, CKMT2, and UCP1 mRNA and pro- cAMP treatment elevated the expression of complex I
tein expression. cAMP could also increase the mRNA and II subunits in DN (except in some brown cells
and protein expression of the three thermogenic genes with already high levels) and partially in SC adipo-
in white and brown SGBS adipocytes in most cases and cytes. The elevated levels could be reduced by the
this was also prevented in the presence of the ASC-1 addition of the ASC-1 inhibitor (Fig. 5A-C,E). The
inhibitor (Fig. 4C,D). amount of mitochondrial complex III and V subunits

6 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies
R. Arianti et al. ASC-1 transporter is needed for adipocyte thermogenesis

DN W 60
(A) cAMP

OCR (pmol O2/min* µg protein)

BMS-466442
cAMP+BMS-466442 β-GPA Oligomycin
30
OCR (pmoles O2/min* µg protein)

Antimycin A
40

*
20 *
Untreated
cAMP
BMS-466442 *
20 *
cAMP+BMS-466442 *
10 *

0 0
Basal
0 100 200 300 400 500 600 660 Maximal Stimulation β-GPA Inhibted Proton Leak Non-mitochondrial
Respiration
Time (min) cAMP (500 μM) – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – + + – – + + – – + + – – + + –

DN B 60
cAMP

OCR (pmol O2/min* µg protein)

BMS-466442
cAMP+BMS-466442 β-GPA Oligomycin
50 *
OCR (pmoles O2/min* µg protein)

Antimycin A *
40
40

30 Untreated
cAMP
BMS-466442 * *
20 *
20 cAMP+BMS-466442 *

10

0 0
0 100 200 300 400 500 600 660 Basal
Maximal Stimulation β-GPA Inhibted Proton Leak Non-mitochondrial
Respiration
Time (min) cAMP (500 μM) – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – + + – – + + – – + + – – + + –

(B) 25 25

*
20 20
ECAR (mpH/min)

15 15
*

10 10

5 5

0 0 Basal
Basal
Maximal Stimulation Maximal Stimulation
cAMP (500 μM) – + – + – + – +
BMS-466442 (100 nM) – + + – – + + –

Fig. 3. Effect of ASC-1 inhibitor (BMS-466442) on cAMP-stimulated oxygen consumption rate (OCR) and extracellular acidification rate
(ECAR) in human deep neck (DN) white and brown adipocytes. (A) OCR of adipocytes was detected for 10 h following cAMP stimulation
without and with ASC-1 inhibition; representative curves of four measurements (left panels). OCR at basal, maximal stimulation, after b-GPA
inhibition, oligomycin, and antimycin A addition (right panels) was quantified in adipocytes derived from four independent donors. (B) ECARs
of DN white and brown at basal and maximal stimulation; n = 4 for all groups. W, white; B, brown differentiation. Statistical analysis was
performed by paired t-test, *P < 0.05, **P < 0.01.

was not affected by either cAMP or the inhibitor while Discussion

complex IV levels in DN cells were decreased by the
SLC transporters act as metabolic gates for the cells
latter (Fig. 5D,F). cAMP stimulation did not lead to and facilitate the transport of important biomole-
significant increase in mitochondrial complex subunits cules such as glucose, amino acids, fatty acids, vita-
in either white or brown SGBS adipocyte and ASC-1 mins, and ions [42]. Thermogenic brown and beige
inhibition did not have pronounced effects (Fig. S5). adipocytes utilize glucose and fatty acids in a

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 7
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

Primary Adipocytes
(A) PGC1α (qPCR) CKMT2 (qPCR) UCP1 (qPCR)
0.10 0.15 W
0.15 W
W *
B * B B

Normalized mRNA expression

Normalized mRNA expression

Normalized mRNA expression

0.08 *
*
*
* 0.10 * 0.10
0.06
* *
** * *
0.04 * *
0.05 0.05 *

0.02 **
*
0.00 0.00 0.00
SC DN SC DN SC DN
cAMP (500 µM) – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + +

PGC1α (Densitometry) CKMT2 (Densitometry) UCP1 (Densitometry)

1.5 0.8 2.0 *
(B) W
B
W
B
W
B
*

Normalized optical density

SC DN * 0.6 * 1.5
*
1.0 * *
W B W B *
* *
0.4 1.0
91 kDa PGC1a *
* *
0.5 * *
47 kDa CKMT2 0.2 0.5 *
*
32 kDa UCP1

55 kDa 0.0 0.0 0.0

TUB
SC DN SC DN SC DN
cAMP (500 µM) – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – +
– – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + +
BMS-466442 (100 nM)

SGBS Adipocytes
(C)
PGC1α (qPCR) CKMT2 (qPCR) UCP1 (qPCR)
0.15 W
0.08 W
0.15 W
B B B
*
Normalized mRNA expression

Normalized mRNA expression

Normalized mRNA expression

**
0.06
0.10 0.10
*
0.04 *
*
*
*
0.05 0.05
*
0.02

0.00 0.00 0.00

cAMP (500 µM) – + – + – + – + – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – – + + – – + + – – + + – – + + – – + + – – + +

PGC1α (Densitometry) CKMT2 (Densitometry) UCP1 (Densitometry)

(D) 1.5 3 W
1.5 W
W *
B * B ** B *
*
Normalized optical density

1.0 2 1.0 *
*
SGBS Adipocytes *

W B
0.5 1 0.5
91 kDa PGC1a
47 kDa CKMT2
32 kDa UCP1
55 kDa TUB 0.0 0 0.0
cAMP (500 µM) – + – – – + – + – + – – + – + – + – + – + – + – + – + – +
+ + +
BMS-466442 (100 nM) – – + – – – – + + – – + + – – + + – – + + – – + + – – + +
+ + +

Fig. 4. Effect of ASC-1 inhibitor on the expression of thermogenic marker genes in human primary SC, DN, and SGBS adipocytes. (A-B)
mRNA and protein expression of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PCG1a), creatine kinase,
mitochondrial 2 (CKMT2), and uncoupling protein 1 (UCP1) detected by RT-qPCR and immunoblotting in human neck adipocytes. n = 3 for
all groups. (C–D) mRNA and protein expression of PCG1a, CKMT2, and UCP1 detected by RT-qPCR and immunoblotting in SGBS
adipocytes. n = 3 for all groups. W, white; B, brown differentiation. Statistical analysis was performed by paired t-test, *P < 0.05,
**P < 0.01.

significant proportion as fuels to generate heat and transporter ASC-1 (encoded by SLC7A10), induced
SLC transporters mediate their uptake via the glu- during adipocyte differentiation, in stimulated ther-
cose transporter-4, encoded by SLC2A4, and the mogenesis of human adipocytes. Inhibition of ASC-1
fatty acid transporter-1, encoded by SLC27A1 during cAMP treatment prevented efficient response
[43,44]. Brown adipose tissue can actively oxidize of highly thermogenic DN-derived brown and white
branched-chain amino acids during cold-induced adipocytes (with a mixed population of brown and
thermogenesis [45]; however, the role of amino acid beige cells) as well as lower thermogenesis competent
influx in the metabolism and regulation of thermo- adipocyte populations of subcutaneous origin which
genic adipocytes has not been fully investigated so contain beige cells.
far. The data presented here have revealed the ASC-1 mRNA was induced and expressed at a
importance of the alanine, serine, cysteine, glycine greater extent in DN and SC brown adipocytes as

8 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies
R. Arianti et al. ASC-1 transporter is needed for adipocyte thermogenesis

Complex I
(Densitometry)
(A) (B) *
SC DN 1.5 W

Normalized optical density

*
B

W B W B
*
18 kDa 1.0 *
I–NDUFB8 *
29 kDa II–SDHB *

48 kDa III–UQCRC2
0.5
22 kDa IV–COX II *
*
54 kDa V–ATP5A
55 kDa TUB 0.0
SC DN
cAMP (500 µM) – + – + – + – + – + – + – + – + cAMP (500 µM) – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – – + + – – + + – – + + – – + + BMS-466442 (100 nM) – – + + – – + + – – + + – – + +

Complex II Complex III

(C) (Densitometry) (D) (Densitometry)
0.6 W
1.5 W
Normalized optical density

*
B * B
*

*
0.4 1.0
*
*
*

0.2 0.5

0.0 0.0
SC DN SC DN
cAMP (500 µM) – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + +

Complex IV Complex V
(E) (Densitometry) (F ) (Densitometry)
1.5 W
2.0 W
Normalized optical density

*
B B
*
1.5
* *
1.0
*
* 1.0
*
0.5
0.5

0.0 0.0
SC DN SC DN
cAMP (500 µM) – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – +
BMS-466442 (100 nM) – – + + – – + + – – + + – – + + – – + + – – + + – – + + – – + +

Fig. 5. Effect of ASC-1 inhibitor on the protein expression of mitochondrial complex subunits in primary SC and DN adipocytes. (A)
Expression of mitochondrial complex subunits detected by immunoblotting. (B-F) Quantification of complex I-V immunoblotting by
densitometry; n = 3. W, white; B, brown differentiation. Statistical analysis was performed by paired t-test, *P < 0.05

compared to white ones, similarly to UCP1 (Fig. 1A, [26,46]. In mice, its presence in a subpopulation of
B). At the protein level, ASC-1 expression was either subcutaneous preadipocytes of adolescent adipose tis-
not different or moderately higher in white as com- sue and in differentiating preadipocyte cell lines could
pared to brown adipocytes of DN and SC origin even inhibit beige differentiation [47]. In a human
(Fig. 1C). ASC-1 was described earlier as a cell surface study, slightly more ASC-1 mRNA was detected in SC
marker of white adipocytes in humans and mice white tissue samples obtained by needle biopsies from

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 9
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

the neck than in DN brown samples [36] and this was treatment in mice diminishes oxidative stress by
reflected in the ASC-1 protein levels we found in SC enhancing GSH synthesis and lowers the production
and DN tissue lysates (Fig. S6). On the other hand, of hepatic reactive oxygen species [51]. Serine can also
ASC-1 mRNA was expressed at the same level in be a main source of nicotinamide adenine dinucleotide
white and brown adipocytes differentiated from human phosphate (NADPH) which is required for the reduc-
SC- and DN-derived immortalized preadipocyte clones tion of oxidized GSH [52,53].
[26]. Regarding our observations, SC, DN, and SGBS Short-term adrenergic stimulation of differentiated
brown differentiated adipocytes had higher ASC-1 DN adipocytes leads to the upregulation of thermo-
mRNA and either the same or moderately less ASC-1 genic genes, such as UCP1 and PGC1a. This response
protein levels compared to white ones and the presence was also reduced by ASC-1 inhibitor in both white
of ASC-1 did not suppress thermogenic activity. Suffi- and brown human SC, DN, and SGBS adipocytes
cient transporter capacity was available in both adipo- (Fig. 4). Furthermore, we found that the cAMP-
cyte types for cAMP-stimulated increase in serine, stimulated mRNA expression of additional browning
cysteine, and glycine influx during elevated UCP1- genes, such as ELOVL3, DIO2, and CITED1, was also
dependent OCR (Figs 2 and 3). attenuated by the ASC-1 inhibitor. ELOVL3 is more
The extra need of adipocytes for serine, cysteine, expressed in brown adipose tissue upon cold exposure
and glycine became apparent when we monitored their [54] and defects in this gene lead to the decreased level
cAMP-stimulated OCR in the presence of the selective of long-chain fatty acids which activate UCP1. DIO2
ASC-1 inhibitor, BMS-455442. While the basal OCR is highly expressed in brown adipocytes, and its deple-
was not affected by the inhibitor, it strongly reduced tion leads to the downregulation of lipolysis and adap-
maximal OCR elicited by cAMP in all adipocyte types tive thermogenesis [55]. CITED1, a beige-selective
studied. We found that proton leak respiration, which transcription factor, is highly expressed in human
reflects the abundance and activity of UCP1 in brown browning cells [15].
or masked beige adipocytes, and activation of the Several studies have reported UCP1-independent
creatine-driven substrate cycle were mainly responsible thermogenesis in browning cells utilizing a creatine
for the observed OCR burst upon cAMP stimulation futile cycle [3], calcium cycling [4], and uncoupling by
of adipocytes. Both were inhibited in the presence of N-acyl amino acids [38]. The inhibition of ASC-1
the compound BMS-455442, clearly demonstrating reduced the UCP1-independent thermogenesis medi-
that ASC-1-mediated amino acid uptake is required ated by the creatine-driven substrate cycle (Figs 3A
for the thermogenic response of adipocytes (Figs 3A and S4A,B). In addition, cAMP-induced upregulation
and S3A,B). Although we have originally assumed that of CKMT2 and CKMT1a/b was also reduced by the
cAMP stimulates serine, cysteine, and glycine uptake ASC-1 inhibitor. CKMT1a/b and CKMT2, the mito-
through ASC-1, the amino acid consumption data sug- chondrial creatine kinases, phosphorylate creatine gen-
gest that cAMP activates other transporters for these erating phosphocreatine [3] to which creatine kinase B
amino acids. However, ASC-1-mediated basic trans- (CKB) can also contribute [56]. Three types of amino
port has a critical role in the respiration and gene acids methionine, glycine, and arginine are required
induction response to thermogenic stimulation since for creatine synthesis [57]. The interruption of glycine
the presence of the ASC-1 inhibitor led to suppression influx by ASC-1 inhibitor may prevent increased cre-
of both. atine synthesis. In addition, ASC-1 inhibition also
Serine is an important metabolic source to generate reduced cAMP-stimulated mRNA expression of
one-carbon units in mammalian cells [48] via its break- PM20D1 in the adipocytes; PM20D1 regulates the syn-
down by serine hydroxymethyltransferases. The meta- thesis of the endogenous uncoupler N-lipidated/N-
bolism of one-carbon units has a functional interaction fatty-acyl amino acids, and it was found enriched in
with the mitochondrial oxidative phosphorylation brown adipose tissue where it contributes to energy
(OXPHOS) system that is crucial for ATP and heat expenditure [38].
generation in mammalian cells [49]. We found that ASC-1 has been recently identified as a novel regula-
mitochondrial OXPHOS system was affected by the tor of energy metabolism in white adipocytes enhanc-
inhibition of ASC-1 activity as cAMP-induced eleva- ing their mitochondrial respiration and preventing
tion of mitochondrial complex I and II subunit protein development of adipocyte hypertrophy and insulin
expression was blunted by the ASC-1 inhibitor resistance [27]; a major part of supporting evidence
(Fig. 5). One-carbon units are also required for the was obtained by applying the same ASC-1 inhibitor,
transsulfuration pathway during glutathione (GSH) BMS-466442, that we have used in the work presented
synthesis as a response to oxidative stress [50]. Serine here. The significant role of ASC-1 in controlling

10 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies
R. Arianti et al. ASC-1 transporter is needed for adipocyte thermogenesis

obesity has been also underlined by the observation in Data accessibility

multiple cohorts that its adipose tissue expression is
RNA-seq data were deposited in the Sequence Read
inversely correlated with visceral adiposity and insulin
Archive (SRA) database [https://www.ncbi.nlm.nih.gov/
resistance [27]. The capacity to waste energy through
sra] under accession number PRJNA607438. Other data
thermogenesis by brown and beige adipocytes is low in
that support the findings of this study are available
obese individuals [58,59], particularly in those who
from the corresponding authors [fesus@med.unideb.hu,
carry the FTO rs1421085 obesity-risk allele [24]. Based
kristof.endre@med.unideb.hu] upon reasonable request.
on our presented data, it can be assumed that the
decreased expression of ASC-1 in hypertrophic white
adipose tissue of obese individuals may also result in a References
reduced thermogenesis response of masked beige cells
1 Chouchani ET, Kazak L and Spiegelman BM (2019)
which contribute to excess energy storage, lipid accu-
New advances in adaptive thermogenesis: UCP1 and
mulation, and obesity. This might, at least partially,
beyond. Cell Metab 29, 27–37.
explain the fact that obese individuals possess high 2 Kajimura S, Spiegelman BM and Seale P (2015) Brown
amounts of brownable but thermogenically not active and beige fat: physiological roles beyond heat
fat [13]. Consequently, stimulation of ASC-1 expres- generation. Cell Metab 22, 546–559.
sion in adipose tissue of obese patients and, thereby, 3 Kazak L, Chouchani ET, Jedrychowski MP, Erickson
facilitation of its beneficial effect on overall energy bal- BK, Shinoda K, Cohen P, Vetrivelan R, Lu GZ,
ance [27] and thermogenic responsiveness may have Laznik-Bogoslavski D, Hasenfuss SC et al. (2015) A
therapeutic value. creatine-driven substrate cycle enhances energy
expenditure and thermogenesis in beige fat. Cell 163,
643–655.
Acknowledgements
4 Ikeda K, Kang Q, Yoneshiro T, Camporez JP, Maki H,
We thank Jennifer Nagy for technical assistance and Homma M, Shinoda K, Chen Y, Lu X, Maretich P
Dr. K
aroly Jambrovics for help in cell analysis. This et al. (2017) UCP1-independent signaling involving
research was funded by the European Union and the SERCA2b-mediated calcium cycling regulates beige fat
European Regional Development Fund (GINOP-2.3.2- thermogenesis and systemic glucose homeostasis. Nat
15-2016-00006), the European Union and the Euro- Med 23, 1454–1465.
pean Social Fund (EFOP-3.6.3-VEKOP-16-2017- 5 Gesta S, Tseng YH and Kahn CR (2007)
00009), and the National Research, Development and Developmental origin of fat: tracking obesity to its
Innovation Office (NKFIH-FK131424 and K129139) source. Cell 131, 242–256.
6 Berry DC, Stenesen D, Zeve D and Graff JM (2013)
of Hungary. EK was supported by the J
anos Bolyai
The developmental origins of adipose tissue.
Fellowship of the Hungarian Academy of Sciences and

Development 140, 3939–3949.
the UNKP-20-5 New National Excellence Program of
7 Sanchez-Gurmaches J and Guertin DA (2014)
the Ministry for Innovation and Technology from the
Adipocytes arise from multiple lineages that are
source of the National Research, Development and
heterogeneously and dynamically distributed. Nat
Innovation Fund. BAV
was supported by the UNKP-


Commun 5, 4099.
20-2-I New National Excellence Program of the Min- 8 Ikeda K, Maretich P and Kajimura S (2018) The
istry for Innovation and Technology from the source Common and Distinct Features of Brown and Beige
of the National Research, Development and Innova- Adipocytes. Trends Endocrinol Metab 29, 191–200.
tion Fund. PF-P was funded by the German Research 9 Lee YH, Petkova AP, Mottillo EP and Granneman JG
Association (Fi1700/7-1). (2012) In vivo identification of bipotential adipocyte
progenitors recruited by b3-adrenoceptor activation and
high-fat feeding. Cell Metab 15, 480–491.
Author contributions
10 Wang QA, Tao C, Gupta RK and Scherer PE (2013)
LF and RA conceived and designed the experiments Tracking adipogenesis during white adipose tissue
with contributions from EK. RA, BAV,
and EK per- development, expansion and regeneration. Nat Med 19,
formed the experiments. BBT analyzed the RNA-seq 1338–1344.
data. FGy provided tissue samples, PF-P and MW 11 Cypess AM, Lehman S, Williams G, Tal I, Rodman D,
provided SGBS cells. ECs carried out measurement of Goldfine AB, Kuo FC, Palmer EL, Tseng YH, Doria A
amino acid concentrations. AS and AV generated pri- et al. (2009) Identification and importance of brown
mary cell cultures for the experiments. RA and LF adipose tissue in adult humans. N Engl J Med 360,
wrote the manuscript with inputs from EK and BBT. 1509–1517.

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 11
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

12 Virtanen KA, Lidell ME, Orava J, Heglind M, 24 Claussnitzer M, Dankel SN, Kim KH, Quon G,
Westergren R, Niemi T, Taittonen M, Laine J, Savisto Meuleman W, Haugen C, Glunk V, Sousa IS, Beaudry
NJ, Enerb€ack S et al. (2009) Functional brown adipose JL, Puviindran V et al. (2015) FTO Obesity variant
tissue in healthy adults. N Engl J Med 360, 1518–1525. circuitry and adipocyte browning in humans. N Engl J
13 Leitner BP, Huang S, Brychta RJ, Duckworth CJ, Med 373, 895–907.
Baskin AS, McGehee S, Tal I, Dieckmann W, Gupta 25 Rutter AR, Fradley RL, Garrett EM, Chapman KL,
G, Kolodny GM et al. (2017) Mapping of human Lawrence JM, Rosahl TW and Patel S (2007) Evidence
brown adipose tissue in lean and obese young men. from gene knockout studies implicates Asc-1 as the
Proc Natl Acad Sci USA 114, 8649–8654. primary transporter mediating d-serine reuptake in the
14 Lee P, Werner CD, Kebebew E and Celi FS. (2014). mouse CNS. Eur J Neuorsci 25, 1757–1766.
Functional thermogenic beige adipogenesis is inducible 26 Ussar S, Lee KY, Dankel SN, Boucher J, Haering MF,
in human neck fat. Int J Obes 38, 170–176. Kleinridders A, Thomou T, Xue R, Macotela Y,
15 Sharp LZ, Shinoda K, Ohno H, Scheel DW, Tomoda Cypess AM et al. (2014) ASC-1, PAT2, and P2RX5 are
E, Ruiz L, Hu H, Wang L, Pavlova Z, Gilsanz V et al. cell surface markers for white, beige, and brown
(2012) Human BAT possesses molecular signatures that adipocytes. Sci Transl Med 6, 247ra103.
resemble beige/brite cells. PLoS ONE 7, e49452.  Tallapragada D, Madsen A, Skartveit L,
27 Jersin RA,
16 Wu J, Bostr€ om P, Sparks LM, Ye L, Choi JH, Giang Fjære E, McCann A, Dyer L, Willems A, Bjune JI,
AH, Khandekar M, Virtanen KA, Nuutila P, Schaart Bjune MS et al. (2021) Role of the neutral amino acid
G et al. (2012) Beige adipocytes are a distinct type of transporter SLC7A10 in Adipocyte lipid storage,
thermogenic fat cell in mouse and human. Cell 150, obesity, and insulin resistance. Diabetes 70, 680–695.
366–376. 28 Klus

Ver
eb Z, V
amos A, Fischer-Posovszky P,
oczki A,
17 Shinoda K, Luijten IH, Hasegawa Y, Hong H, Sonne Wabitsch M, Bacso Z, F
es€ us L and Krist

of E (2019)
SB, Kim M, Xue R, Chondronikola M, Cypess AM, Differentiating SGBS adipocytes respond to PPARc
Tseng YH et al. (2015) Genetic and functional stimulation, irisin and BMP7 by functional browning
characterization of clonally derived adult human brown and beige characteristics. Sci Rep 9, 5823.
adipocytes. Nat Med 21, 389–394. 29 Wabitsch M, Brenner RE, Melzner I, Braun M, M€ oller
18 Jespersen NZ, Larsen TJ, Peijs L, Daugaard S, Homøe P, Heinze E, Debatin KM and Hauner H (2001)
P, Loft A, de Jong J, Mathur N, Cannon B, Nedergaard Characterization of a human preadipocyte cell strain
J et al. (2013) A classical brown adipose tissue mRNA with high capacity for adipose differentiation. Int J
signature partly overlaps with brite in the supraclavicular Obes Relat Metab Disord 25, 8–15.
region of adult humans. Cell Metab 17, 798–805. 30 Fischer-Posovszky P, Newell FS, Wabitsch M and
19 Sanchez-Gurmaches J, Hung CM and Guertin DA Tornqvist HE (2008) Human SGBS cells - a unique
(2016) Emerging complexities in adipocyte origins and tool for studies of human fat cell biology. Obesity facts
identity. Trends Cell Biol 26, 313–326. 1, 184–189.
20 Elabd C, Chiellini C, Carmona M, Galitzky J, Cochet 31 Krist
of E, Klus

Veress R, Shaw A, Combi ZS,
oczki A,
O, Petersen R, P
enicaud L, Kristiansen K, Bouloumi
e Varga K, Gy} ory F, Balajthy Z, Bai P, Bacso Z et al.
A, Casteilla L et al. (2009) Human multipotent adipose- (2019) Interleukin-6 released from differentiating human
derived stem cells differentiate into functional brown beige adipocytes improves browning. Exp Cell Res 377,
adipocytes. Stem cells 27, 2753–2760. 47–55.
21 Krist
of E, Doan-Xuan QM, Bai P, Bacso Z and F
es€ us 32 Brown JM, Hunihan L, Prack MM, Harden DG,
L (2015) Laser-scanning cytometry can quantify human Bronson J, Dzierba CD, Gentles RG, Hendricson A,
adipocyte browning and proves effectiveness of irisin. Krause R, Macor JE et al. (2014) In vitro
Sci Rep 5, 12540. Characterization of a small molecule inhibitor of the
22 T
oth BB, Arianti R, Shaw A, V
amos A, Ver
eb Z, alanine serine cysteine transporter -1 (SLC7A10). J
P
oliska S, Gy}ory F, Bacso Z, F
es€ us L and Krist
of E Neurochem 129, 275–283.
(2020) FTO intronic SNP strongly influences human 33 Krist
of E, Doan-Xuan QM, S
arv
ari AK, Klus


oczki A,
neck adipocyte browning determined by tissue and Fischer-Posovszky P, Wabitsch M, Bacso Z, Bai P,
PPARc specific regulation: a transcriptome analysis. Balajthy Z and F
es€ us L (2016) Clozapine modifies the
Cells 9, 987. differentiation program of human adipocytes inducing
23 Szatm
ari-T
oth M, Shaw A, Csom
os I, Mocs
ar G, browning. Transl Psychiat 6, e963.
Fischer-Posovszky P, Wabitsch M, Balajthy Z, L
anyi C, 34 Fitch CD and Chevli R (1980) Inhibition of creatine
Gy} ory F, Krist
of E et al. (2020) Thermogenic and phosphocreatine accumulation in skeletal muscle
activation downregulates high mitophagy rate in human and heart. Metab Clin Exp 29, 686–690.
masked and mature beige adipocytes. Int J Mol Sci 21, 35 Cheng Y, Jiang L, Keipert S, Zhang S, Hauser A, Graf
6640. E, Strom T, Tsch€ op M, Jastroch M and Perocchi F

12 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies
R. Arianti et al. ASC-1 transporter is needed for adipocyte thermogenesis

(2018) Prediction of adipose browning capacity by 48 de Koning TJ, Snell K, Duran M, Berger R, Poll-The
systematic integration of transcriptional profiles. Cell BT and Surtees R (2003) L-serine in disease and
Rep 23, 3112–3125. development. Biochem J 371 (Pt 3), 653–661.
36 Perdikari A, Leparc GG, Balaz M, Pires ND, Lidell 49 Lucas S, Chen G, Aras S and Wang J (2018) Serine
ME, Sun W, Fernandez-Albert F, M€ uller S, Akchiche catabolism is essential to maintain mitochondrial
N, Dong H et al. (2018) BATLAS: deconvoluting respiration in mammalian cells. Life science alliance 1,
brown adipose tissue. Cell Rep 25, 784–797. e201800036.
37 Felig P (1973) The glucose-alanine cycle. Metab Clin 50 McCarthy MJ (2014) Oxidative stress: a link between
Exp 22, 179–207. cardiovascular disease and psychiatric illness? Acta
38 Fukasawa Y, Segawa H, Kim JY, Chairoungdua A, Psychiatr Scand 130, 161–162.
Kim DK, Matsuo H, Cha SH, Endou H and Kanai 51 Zhou X, He L, Wu C, Zhang Y, Wu X and Yin Y
Y (2000) Identification and characterization of a Na (2017) Serine alleviates oxidative stress via supporting
(+)-independent neutral amino acid transporter that glutathione synthesis and methionine cycle in mice. Mol
associates with the 4F2 heavy chain and exhibits Nutr Food Res 61, 1700262.
substrate selectivity for small neutral D- and L- 52 Ducker GS and Rabinowitz JD (2017) One-carbon
amino acids. J Biol Chem 275, 9690–9698. metabolism in health and disease. Cell Metab 25, 27–42.
39 Long JZ, Svensson KJ, Bateman LA, Lin H, 53 Fan J, Ye J, Kamphorst JJ, Shlomi T, Thompson CB
Kamenecka T, Lokurkar IA, Lou J, Rao RR, and Rabinowitz JD (2014) Quantitative flux analysis
Chang MR, Jedrychowski MP et al. (2016) The reveals folate-dependent NADPH production. Nature
secreted enzyme PM20D1 Regulates lipidated amino 510, 298–302.
acid uncouplers of mitochondria. Cell 166, 54 J€
orgensen JA, Zadravec D and Jacobsson A (2007)
424–435. Norepinephrine and rosiglitazone synergistically induce
40 Tvrdik P, Asadi A, Kozak LP, Nedergaard J, Cannon Elovl3 expression in brown adipocytes. Am J Physiol
B and Jacobsson A (1997) Cig30, a mouse member of a Endocrinol Metab 293, E1159–E1168.
novel membrane protein gene family, is involved in the 55 de Jesus LA, Carvalho SD, Ribeiro MO, Schneider M,
recruitment of brown adipose tissue. J Biol Chem 272, Kim SW, Harney JW, Larsen PR and Bianco AC
31738–31746. (2001) The type 2 iodothyronine deiodinase is essential
41 Cannon B and Nedergaard J (2004) Brown adipose for adaptive thermogenesis in brown adipose tissue. J
tissue: function and physiological significance. Physiol Clin Investig 108, 1379–1385.
Rev 84, 277–359. 56 Rahbani JF, Roesler A, Hussain MF, Samborska B,
42 Zhang Y, Zhang Y, Sun K, Meng Z and Chen L (2019) Dykstra CB, Tsai L, Jedrychowski MP, Vergnes L,
The SLC transporter in nutrient and metabolic sensing, Reue K, Spiegelman BM et al. (2021) Creatine kinase B
regulation, and drug development. J Mol Cell Biol 11, 1– controls futile creatine cycling in thermogenic fat.
13. Nature 590, 480–485.
43 Wu Q, Kazantzis M, Doege H, Ortegon AM, Tsang B, 57 da Silva RP, Nissim I, Brosnan ME and Brosnan JT
Falcon A and Stahl A (2006) Fatty acid transport (2009) Creatine synthesis: hepatic metabolism of
protein 1 is required for nonshivering thermogenesis in guanidinoacetate and creatine in the rat in vitro and
brown adipose tissue. Diabetes 55, 3229–3237. in vivo. American journal of physiology. Endocrinol
44 El Bacha T, Luz M and Da Poian A (2010) Dynamic Metab 296, E256–E261.
adaptation of nutrient utilization in humans. Nat Educ 3, 58 Jung RT, Shetty PS, James WP, Barrand MA and
8. Callingham BA (1979) Reduced thermogenesis in
45 Yoneshiro T, Wang Q, Tajima K, Matsushita M, Maki obesity. Nature 279, 322–323.
H, Igarashi K, Dai Z, White PJ, McGarrah RW, 59 Rosenbaum M and Leibel RL (2010). Adaptive
Ilkayeva OR et al. (2019) BCAA catabolism in brown thermogenesis in humans. Int J Obes 34, S47–S55.
fat controls energy homeostasis through SLC25A44.
Nature 572, 614–619.
46 Garcia RA, Roemmich JN and Claycombe KJ Supporting information
(2016) Evaluation of markers of beige adipocytes in
white adipose tissue of the mouse. Nutr Metab 13, Additional supporting information may be found
24. online in the Supporting Information section at the end
47 Suwandhi L, Altun I, Karlina R, Miok V, Wiedemann of the article.
T, Fischer D, Walzthoeni T, Lindner C, B€ ottcher A, Fig. S1. The gene-expression data based heat map
Heinzmann SS et al. (2021) Asc-1 regulates white versus shows the Pearson’s correlation of 1049 genes
beige adipocyte fate in a subcutaneous stromal cell expressed differentially in deep neck and subcutaneous
population. Nat Commun 12, 1588. derived preadipocytes and differentiated adipocytes.

FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies 13
ASC-1 transporter is needed for adipocyte thermogenesis R. Arianti et al.

Fig. S2. Effect of ASC-1 inhibitor (BMS-466442) on Fig. S5. Effect of ASC-1 inhibitor on the protein
ASC-1 mRNA and protein expression in human neck expression of mitochondrial complex subunits in SGBS
area and SGBS adipocytes. adipocytes.
Fig. S3. Effect of ASC-1 inhibitor (BMS-466442) on Fig. S6. The ASC-1 mRNA and protein expression in
cAMP stimulated OCR and ECAR in human SC neck human neck tissue lysates.
and SGBS adipocytes. Table S1. Genes primers and probes.
Fig. S4. Effect of ASC-1 inhibitor (BMS-466442) on Table S2. Antibodies used in immunoblotting.
the mRNA expression of thermogenic marker genes in Supplementary Material
primary SC and DN adipocytes.

14 FEBS Letters (2021) ª 2021 The Authors. FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies