Desnutrin, an Adipocyte Gene Encoding a Novel Patatin Domain-containing Protein, Is Induced by Fasting and Glucocorticoids: Ectopic expression of desnutrin increases triglyceride hydrolysis

Abstract

We have used rat cDNA microarrays to identify adipocyte-specific genes that could play an important role in adipocyte differentiation or function. Here, we report the cloning and identification of a 2.0-kb mRNA coding for a putative protein that we have designated as desnutrin. The novel gene is expressed predominantly in adipose tissue, and its expression is induced early during 3T3-L1 adipocyte differentiation. Desnutrin mRNA levels were regulated by the nutritional status of animals, being transiently induced during fasting. In vitro desnutrin gene expression was up-regulated by dexamethasone in a dose-dependent manner but not by cAMP, suggesting that glucocorticoids could mediate the increase in desnutrin mRNA levels observed during fasting. Desnutrin mRNA codes for a 486-amino acid putative protein containing a patatin-like domain, characteristic of many plant acyl hydrolases belonging to the patatin family. Confocal microscopy of enhanced green fluorescent protein-tagged desnutrin protein-transfected cells showed that the fusion protein localized in the cytoplasm. Moreover, cells overexpressing desnutrin by transfection showed an increase in triglyceride hydrolysis. Interestingly, we also found that the desnutrin gene expression level was lower in ob/ob and db/db obese mouse models. Overall, our data suggest that the newly identified desnutrin gene codes for an adipocyte protein that may function as a lipase and play a role in the adaptive response to a low energy state, such as fasting, by providing fatty acids to other tissues for oxidation. In addition, decreased expression of desnutrin in obesity models suggests its possible contribution to the pathophysiology of obesity.

Full text

Desnutrin, an Adipocyte Gene Encoding a Novel Patatin Domain-
containing Protein, Is Induced by Fasting and Glucocorticoids
ECTOPIC EXPRESSION OF DESNUTRIN INCREASES TRIGLYCERIDE HYDROLYSIS*
Received for publication, April 7, 2004, and in revised form, August 12, 2004
Published, JBC Papers in Press, August 27, 2004, DOI 10.1074/jbc.M403855200
Josep A. Villena, Suheeta Roy, Eszter Sarkadi-Nagy, Kee-Hong Kim, and Hei Sook Sul‡
From the Department of Nutritional Sciences and Toxicology, University of California, Berkeley, California 94720
We have used rat cDNA microarrays to identify adipo-
cyte-specific genes that could play an important role in
adipocyte differentiation or function. Here, we report
the cloning and identification of a 2.0-kb mRNA coding
for a putative protein that we have designated as des-
nutrin. The novel gene is expressed predominantly in
adipose tissue, and its expression is induced early dur-
ing 3T3-L1 adipocyte differentiation. Desnutrin mRNA
levels were regulated by the nutritional status of ani-
mals, being transiently induced during fasting. In vitro
desnutrin gene expression was up-regulated by dexa-
methasone in a dose-dependent manner but not by
cAMP, suggesting that glucocorticoids could mediate
the increase in desnutrin mRNA levels observed during
fasting. Desnutrin mRNA codes for a 486-amino acid
putative protein containing a patatin-like domain, char-
acteristic of many plant acyl hydrolases belonging to the
patatin family. Confocal microscopy of enhanced green
fluorescent protein-tagged desnutrin protein-trans-
fected cells showed that the fusion protein localized in
the cytoplasm. Moreover, cells overexpressing desnu-
trin by transfection showed an increase in triglyceride
hydrolysis. Interestingly, we also found that the desnu-
trin gene expression level was lower in ob/ob and db/db
obese mouse models. Overall, our data suggest that the
newly identified desnutrin gene codes for an adipocyte
protein that may function as a lipase and play a role in
the adaptive response to a low energy state, such as
fasting, by providing fatty acids to other tissues for ox-
idation. In addition, decreased expression of desnutrin
in obesity models suggests its possible contribution to
the pathophysiology of obesity.
Triglycerides serve as the most efficient form of energy stor-
age in times of caloric excess in many organisms. During peri-
ods of energy demand, triglycerides can be rapidly mobilized by
the hydrolytic action of lipases, releasing free fatty acids that
are oxidized to meet the energy requirement of the organism.
In mammals, adipose tissue serves as the major lipid storage
site. Whereas white adipose tissue (WAT)
1
stores triglycerides
that can be mobilized, producing fatty acids to be used by
peripheral tissues, brown adipose tissue (BAT) itself uses the
accumulated lipids to generate heat, a process known as adap-
tative thermogenesis (1). In addition, WAT plays an important
role as an endocrine organ, secreting a wide variety of factors
that are involved in various aspects of physiology, including
appetite control, peripheral metabolism, immune response,
and vascular function (2–4). The function of adipose tissue is
dependent on the energy requirements and is tightly controlled
by nutrient, neural, and hormonal signals. The crucial contri-
bution of adipose tissue to the energy metabolism and function
of an organism is manifested by pathological conditions in
which dysregulation of adipose function leads to the develop-
ment of severe diseases such as insulin resistance, diabetes
mellitus, and cardiovascular disease (5).
The adipocyte is the main cell type present in adipose tissue.
During adipose tissue development, precursor cells differenti-
ate to generate adipocytes fully equipped with the enzymatic
machinery and regulatory proteins that are needed to carry out
their function in controlling fat metabolism and energy home-
ostasis. The adipogenic process entails dramatic morphological
and biochemical changes and alteration in the expression of
hundreds of genes (6 – 8). This includes increases in the expres-
sion of genes involved in the specialized role of adipose tissue in
lipid metabolism, hormone responsiveness, the extracellular
matrix, and secretion of endocrine and regulatory factors (2).
Moreover, some of the genes are expressed only in adipose
tissue and constitute a unique functional trait of this tissue.
Identification of such genes or gene products is of crucial im-
portance in unraveling the regulatory mechanisms of adipocyte
differentiation and function. It is also a requirement for fully
understanding the etiology of the pathologies associated with
adipose tissue malfunction.
To identify novel genes that have a role in adipocyte differ-
entiation or function, we have used several approaches based
on differential gene expression. By differential screening and
differential display techniques using the adipogenic 3T3-L1 cell
line and primary preadipocytes, we previously cloned two reg-
ulatory proteins involved in the control of adipocyte differenti-
ation, Pref-1 (9) and ENC-1 (10), respectively (10, 11). We
recently employed the microarray technique to clone ADSF/
resistin (12), an adipocyte-specific secreted factor that inhibits
adipocyte differentiation in vitro and in vivo (12, 13) and that
may be a contributing factor to insulin resistance (14).
Here, we report the cloning and identification of the full-
length cDNA coding for a putative protein that we named
desnutrin. This novel gene is induced early during 3T3-L1
* This work was supported by National Institutes of Health Grants
DK050828 and DK068439 (to H. S. S.). The costs of publication of this
article were defrayed in part by the payment of page charges. This
article must therefore be hereby marked “advertisement” in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted
to the GenBank
TM
/EBI Data Bank with accession number(s) AY731699.
‡ To whom correspondence should be addressed. Tel.: 510-642-3978;
Fax: 510-642-0535; E-mail: hsul@nature.berkeley.edu.
1
The abbreviations used are: WAT, white adipose tissue; BAT, brown
adipose tissue; RACE, rapid amplification of cDNA ends; EST, ex-
pressed sequence tag; contig, group of overlapping clones; HA, hemag-
glutinin; EGFP, enhanced green fluorescent protein; PBS, phosphate-
buffered saline; DMEM, Dulbecco’s modified Eagle’s medium; MIX,
methylisobutylxanthine; C/EBP, CAAT/enhancer-binding protein.
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 279, No. 45, Issue of November 5, pp. 47066 –47075, 2004
© 2004 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
This paper is available on line at http://www.jbc.org47066
by guest on November 5, 2015http://www.jbc.org/Downloaded from

adipocyte differentiation and is expressed predominantly in
adipose tissue. Desnutrin mRNA levels are under the control of
the nutritional status of animals, being transiently induced
during fasting. In vitro, desnutrin gene expression was regu-
lated by dexamethasone, but not by cAMP, suggesting that
glucocorticoids could mediate the increase in desnutrin mRNA
during fasting. Desnutrin mRNA codes for a 486-amino acid
protein containing a patatin-like domain, characteristic of
many plant acyl hydrolases that belong to the patatin family.
We found that ectopic overexpression of desnutrin increased
triglyceride hydrolysis in the cell, suggesting the function of
desnutrin as a lipase and its role in the adaptive response of
adipose tissue to the low energy state of fasting, releasing
substrates for oxidation to meet the energy requirement for
other tissues.
EXPERIMENTAL PROCEDURES
Genefilter Microarray Analysis—Identification of genes expressed
exclusively in adipose tissue was achieved by comparing the gene ex-
pression patterns of different mouse tissues using rat Genefilter mem-
branes (Research Genetics) as described previously (12). Briefly, filters
were hybridized with
33
P-labeled cDNAs synthesized by reverse tran
-
scription using 5
␮
g of total RNA from WAT, brain, muscle, and liver.
Only those spots found exclusively in filters hybridized with WAT
cDNAs were further analyzed.
Cloning of Desnutrin cDNA and 3⬘-Rapid Amplification of cDNA
Ends (RACE)—A BLASTn search (15) conducted using the identified
rat expressed sequence tag (EST) clone sequence (GenBank
TM
/EBI
accession number AI059513) as a query in the mouse genome data base
of NCBI revealed a single match located on the mouse chromosome 7
contig. The gene was identified as the 0610039C21Riken gene, and from
now on, we will refer to it as desnutrin. The corresponding annotated
mRNA sequence of the 0610039C21Riken gene (accession number
NM_025802) was used to design PCR primers for cloning the cDNA
fragment containing the entire open reading frame. Briefly, 5
␮
g of total
RNA from WAT was reverse-transcribed with Superscript II reverse
transcriptase (Invitrogen) and oligo(dT) in a total volume of 20
␮
l. Two
␮
l of the reaction was used as the template, and desnutrin partial cDNA
was amplified by PCR using primers 5⬘-GACAGCGTCTCCGCCTCCG-
C-3⬘ (forward) and 5⬘-GACAGGATCTTGTTCCACCCC-3⬘ (reverse). The
1634-bp PCR product was cloned into the pGEM-T-Easy vector (Pro-
mega), and the insert fragment was fully sequenced.
We performed 3⬘-RACE to obtain the 3⬘-end of desnutrin cDNA using
the desnutrin-specific primer 5⬘-TGTCCTTCACCATCCGCTTGTTG-3⬘,
oligo(dT), and WAT cDNA as the template. PCR was carried out for 35
cycles of denaturation (at 95 °C for 45 s), annealing (at 42 °C for 45 s),
and extension (at 72 °C for 1 min). The resulting fragment was cloned
into the pGEM-T-Easy vector and sequenced.
Construction of Plasmids—To obtain a hemagglutinin (HA)-tagged
desnutrin expression vector, PCR was carried out using the cDNA con-
taining the entire coding region (described above) as the template and
primers 5⬘-CGCGGAGACCCCAAGGTATC-3⬘ (forward) and 5⬘-TTACAA-
GCTGTAATCTGGAACATCGTATGGGTAGCAAGGCGGGAGGCCAGG-
TGGATCCTGTGGGGT-3⬘ (reverse). The PCR fragment was cloned into
the pGEM-T-Easy vector and then subcloned into the NotI site of the
pcDNA3.1 vector (Invitrogen) for mammalian expression.
A fusion protein construct bearing the coding region of desnutrin was
prepared by subcloning a reverse transcription-PCR product of desnutrin
into the pEGFP-N1 vector (BD Biosciences) in-frame with enhanced green
fluorescent protein (EGFP). Briefly, PCR was performed using 3T3-L1
adipocyte cDNA as the template and primers 5⬘-TCCGGACTCAGATCT-
ATGTTCCCGAGGGAGACCAAGTGG-3⬘ (forward) and 5⬘-CGTACCGTC-
GACTGCAAGCTGTAATCTGGAACATC-3⬘ (reverse). The PCR product
was purified on agarose gel, digested with BglII and SalI, and cloned into
the BglII/SalI sites of the pEGFP-N1 vector.
Animals—For tissue distribution of desnutrin mRNA and its regu-
lation by fasting and refeeding, male C57BL/6 mice were used. In the
fasting/refeeding experiments, mice were fasted for a period of up to
48 h, after which they were killed, and tissues were extracted for RNA
analysis. Mice were refed a high carbohydrate diet for 12 h after 48 h of
fasting before being killed. Desnutrin mRNA levels was also deter-
mined by reverse transcription-PCR and Northern blot analysis in
gonadal adipose tissue obtained from ob/ob and db/db male mice (Jack-
son Laboratory, Bar Harbor, ME) fasted for 12 h.
Separation of the Stromal Vascular Fraction and Adipocytes of
WAT—Inguinal WAT from mice was isolated, washed with phosphate-
buffered saline (PBS), and minced prior to digestion with 2 mg/ml
collagenase and 2% bovine serum albumin in Dulbecco’s modified Ea-
gle’s medium (DMEM). Digestion was carried out for 20 –30 min at
37 °C under constant agitation until tissue integrity was disrupted. The
cell suspension was filtered through a 70-
␮
m mesh filter to remove
undigested tissue fragments, and mature adipocytes were separated
from the stromal vascular fraction by flotation. The stromal vascular
fraction was collected with a syringe, and cells were pelleted by centrif-
ugation at 500 ⫻ g for 10 min. Cells were immediately processed for
RNA extraction.
Cell Culture and Transfection—3T3-L1 cells (American Type Culture
Collection) were cultured in DMEM containing 10% calf serum. To
induce differentiation of 3T3-L1 cells into adipocytes, 2-day post-con-
fluent preadipocytes (day 0) were treated with 1
␮
M dexamethasone
(DEX) and 0.5
␮
M methylisobutylxanthine (MIX) for 48 h. After the
induction period, cells were switched to differentiation medium (DMEM
supplemented with 10% fetal calf serum) and maintained for 5–7 days,
at which point 90% of the cells exhibited the typical adipocyte morphol-
ogy. For the experiments examining the effect of glucocorticoids on
desnutrin expression, confluent preadipocytes were treated with dexa-
methasone, MIX, dibutyryl cAMP, or dexamethasone/MIX at the con-
centrations indicated.
COS-7 cells (American Type Culture Collection) were cultured in
DMEM containing 10% fetal calf serum. Transient transfection of
COS-7 cells with desnutrin-EGFP or HA-desnutrin expression vectors
was carried out by the DEAE-dextran/chloroquine method as described
previously (16).
RNA Isolation and Northern Blot Analysis—Total RNA from mouse
tissues or cells was isolated using TRIzol reagent (Invitrogen) according
to the manufacturer’s instructions. For Northern blot analysis, 5–15
␮
g
of total RNA was subjected to electrophoresis on formaldehyde-contain-
ing 1.2% agarose gel and transferred onto Hybond N
⫹
nylon mem
-
branes (Amersham Biosciences). Blot hybridization was carried out in
ExpressHyb solution (Clontech) using
32
P-labeled cDNA-specific probes
for desnutrin, adiponutrin, fatty-acid synthase, adipocyte fatty acid-
binding protein (aFABP/aP2), CAAT/enhancer-binding protein-
␣
(C/
EBP
␣
), and peroxisome proliferator-activated receptor-
␥
.
Confocal Microscopy—COS-7 cells were transiently transfected with
desnutrin-EGFP (0.7
␮
g of DNA) as described above and grown on glass
coverslips. Thirty-six h post-transfection, cells were fixed with 4%
paraformaldehyde for 10 min. The samples were briefly equilibrated with
PBS, and nuclei were counterstained with 0.3
␮
M 4⬘,6-diamidino-2-phe-
nylindole (Molecular Probes) for 3– 4 min. The samples were rinsed sev-
eral times with PBS and mounted on glass microscope slides using Anti-
fade and Prolong mounting media (Molecular Probes) according to the
manufacturer’s instructions. As a positive control, the empty pEGFP-N1
vector expressing only EGFP was used. Images were captured using a
Zeiss 510 UV-visible laser scanning confocal microscope.
Subcellular Fractionation by Differential Centrifugation and Western
Blot Analysis—COS-7 cells transfected with an expression vector for
HA-tagged desnutrin were washed three times with PBS and twice with
0.25
M sucrose, 10 mM triethanolamine, and 10 mM acetic acid (pH 7.8).
Cells were harvested in ice-cold 0.25 M sucrose, 10 mM triethanolamine,
10 mM acetic acid, and 1 mM EDTA (pH 7.8) and homogenized using a
Dounce homogenizer. The crude nuclear fraction was obtained by cen-
trifugation at 1000 ⫻ g for 10 min. The supernatant was centrifuged at
18,000 ⫻ g for 10 min to obtain the mitochondrial fraction, and the
resulting supernatant was then centrifuged at 100,000 ⫻ g for1hto
separate the microsomal fraction from the cytosolic fraction. Equal
amounts of protein from each fraction were subjected to 10% SDS-
PAGE and transferred to a polyvinylidene difluoride membrane (Milli-
pore Corp.) for immunodetection using anti-HA antibody (Covance).
Separation of Lipids from Transfected COS-7 Cells—COS-7 cells
were transfected with the HA-desnutrin expression vector or the control
pcDNA3.1 vector as described above. Thirty-six h post-transfection,
cells were incubated for4hinserum-free medium (DMEM) containing
fatty acid-free bovine serum albumin (2 mg/ml) and [U-
14
C]palmitic
acid (final volume of 0.1
␮
Ci/ml; Amersham Biosciences). Cells were
washed with PBS and harvested for lipid extraction (time 0) or further
incubated with serum-containing medium for an additional 4, 8, and
16 h. Lipids from cells and media were extracted by the method of Bligh
and Dyer (17) and separated by TLC using solvent system A (hexane/
diethyl ether/acetic acid (80:20:2)) to resolve neutral lipids or solvent
system B (chloroform/methanol/ammonium hydroxide/water (65:35:5:
1)) to separate polar lipids. Radioactive lipids were detected by autora-
diography. To measure the radiolabeled lipids released to the media by
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis 47067
by guest on November 5, 2015http://www.jbc.org/Downloaded from

cells, 1 ml of cell culture medium was collected at the end of each time
point and mixed with 20 ml of CytoScint (Fisher), and the radioactivity
was measured by scintillation counting.
RESULTS
Cloning of Full-length Mouse Desnutrin cDNA—We per-
formed microarray analysis using EST cDNA filter microarrays
to identify those genes expressed exclusively in adipose tissue.
We compared the gene expression patterns in brain, muscle,
liver, and adipose tissue using rat EST cDNA arrays and iden-
tified a set of genes that showed signal only on the arrays that
were hybridized with WAT cDNAs. Candidate EST clones were
sequenced, and adipose tissue-specific expression was verified
by Northern blot analysis using RNA from liver, brain, muscle,
and WAT (data not shown). One of these clones showed exclu-
sive expression in WAT. Using the 434-bp sequence of the
selected EST clone (GenBank
TM
/EBI accession number
AI059513) as a query, the BLASTn search of the mouse genome
data base of NCBI identified a single match on the 3⬘-region of
the 0610039C21Riken gene, located in chromosome 7. We used
the sequence of the annotated 0610039C21Riken gene mRNA
(accession number NM_025802) for the design of a set of prim-
ers to amplify the coding region of the gene using mouse cDNA
reverse-transcribed with RNA prepared from WAT. In addi-
tion, the 3⬘-end was further characterized by carrying out 3⬘-
RACE. A total of nine amplification products were sequenced,
and the resulting consensus 3⬘-end sequence was used for PCR
to generate the full-length desnutrin cDNA of 1965 bp (Fig. 1).
Desnutrin encodes a protein with a length of 486 amino acids
and an estimated molecular mass of 53.6 kDa. The cDNA
contains a short 5⬘-untranslated region of 80 bases and a long
3⬘-untranslated region of 426 bases with a single polyadenyl-
ation signal (Fig. 1).
The homology search revealed the presence of related pro-
teins in a wide range of organisms, including animals and
plants. In humans, two proteins with a high degree of homology
(88%) to desnutrin have been identified. These proteins, named
TTS2.2 (GenBank
TM
/EBI accession number CAC01132) and
TTS2.1 (accession number CAC01131), appear to be distinct
proteins and have been implicated in the vesicular transport
and secretion of the ICAM-3 (intercellular adhesion mole-
cule-3) cell-surface receptor. In addition, two other puta-
tive human proteins sharing high homology with desnutrin
have been identified: GS2-like protein (accession number
NP_620169) and the hypothetical chromosome 22 open reading
frame (accession number NP_079501). In mouse, two homolo-
gous proteins can be found: a hypothetical protein (accession
number XP_128189) that appears to be the mouse homolog of
the human GS2-like protein and adiponutrin. Adiponutrin has
been recently cloned (18) as an adipose tissue-specific protein
whose expression is up-regulated during adipocyte differentia-
tion, but down-regulated during fasting in mice. Two homolo-
gous proteins are found in Drosophila melanogaster and four in
Caenorhabditis elegans. Interestingly, additional putative pro-
teins sharing some degree of homology are present in Arabi-
dopsis thaliana, such as the protein At1g33270 (accession num-
ber NP_977474), and in prokaryotes, including Bacteroides
thetaiotaomicron (accession number NP_809687) and Bacillus
subtilis (accession number NP_389387). The presence of des-
nutrin homologs and other related proteins in such a variety of
organisms suggests that desnutrin belongs to a family of pro-
teins whose members are widely expressed in all organisms,
and this points toward a basic function of these proteins that is
not exclusive of higher organisms.
Desnutrin mRNA Is Expressed Predominantly in Adipose
Tissue—Northern blot analysis was performed to determine
the expression pattern of desnutrin in various types of mouse
tissues. Hybridization of mouse RNAs with a desnutrin probe
spanning nucleotides 20 –1632 revealed a single mRNA of ⬃2.0
kb in size (Fig. 2A). The appearance of a single band is in
agreement with the presence of a single polyadenylation signal
in the desnutrin cDNA sequence. Considering the average
poly(A) length, we conclude that our desnutrin cDNA sequence
represents full-length cDNA. As shown in Fig. 2A, desnutrin
mRNA was highly expressed in various depots of WAT as well
as BAT. Low but detectable desnutrin levels were also found in
other tissues such as heart and testis, whereas desnutrin
mRNA was found at a very low, barely detectable level in liver,
spleen, thymus, kidney, brain, skeletal muscle, and lung.
Therefore, although ubiquitously expressed, desnutrin is
highly expressed in adipose tissues only. It is interesting to
note that the desnutrin expression levels differed depending on
the type or location of adipose tissue. Overall, BAT showed
higher desnutrin mRNA levels than WAT. Among the white
adipose depots, gonadal fat showed the highest level of expres-
sion compared with inguinal and renal WAT. Adiponutrin, a
closely related gene known to be expressed exclusively in adi-
pose tissue, showed a pattern of expression similar to that of
desnutrin in the different adipose depots, with a higher level of
expression in BAT and gonadal WAT (Fig. 2A). We next exam-
ined desnutrin gene expression in gonadal fat from the genet-
ically obese mouse models ob/ob and db/db by Northern blot
analysis (Fig. 2B) as well as by reverse transcription-PCR (data
not shown). We observed ⬃50% lower desnutrin mRNA levels
in db/db mice and 80% lower levels in ob/ob mice compared
with wild-type C57BL/6 mice, suggesting a possible signifi-
cance of desnutrin function in obesity.
To identify the cell type that is responsible for the high
expression level of desnutrin mRNA in adipose tissue, cellular
components of inguinal WAT were fractionated into adipocytes
and a stromal vascular fraction containing preadipocytes as
well as endothelial cells and resident macrophages. As shown
in Fig. 2C (left panels), the desnutrin transcript was not de-
tected in the stromal vascular fraction, but was found exclu-
sively in the adipocyte fraction, along with adipocyte markers
such as fatty-acid synthase, adipocyte fatty acid-binding pro-
tein/aP2, C/EBP
␣
, and adiponutrin. The adipocyte-specific ex-
pression of desnutrin mRNA was corroborated in 3T3-L1 cells,
an established cell line that can be induced to differentiate into
adipocytes upon appropriate hormone treatment and used as a
model system for adipogenesis. Desnutrin mRNA was found
only in mature adipocytes, but not in proliferating or confluent
3T3-L1 preadipocytes (Fig. 2C, right panels). We next exam-
ined desnutrin gene expression during the course of conversion
of 3T3-L1 preadipocytes to adipocytes. As predicted and shown
in Fig. 3, desnutrin mRNA was not detected in preadipocytes
(day 0). However, its expression was rapidly increased when
cells were induced to differentiate. The maximal level of des-
nutrin mRNA was reached after 6 days, when cells were fully
differentiated into adipocytes, as judged by accumulation of
lipids and expression of adipocyte markers such adipocyte fatty
acid-binding protein/aP2, fatty-acid synthase, peroxisome pro-
liferator-activated receptor-
␥
, C/EBP
␣
, and adiponutrin. Inter-
estingly, a detectable level of desnutrin mRNA could be ob-
served in 3T3-L1 cells as early as 24 h after induction of
differentiation, preceding the expression of the various late
adipocyte markers. Thus, desnutrin is an adipocyte gene in-
duced early during 3T3-L1 adipocyte differentiation.
Induction of Desnutrin mRNA by Fasting and by Glucocor-
ticoids—Expression of the adiponutrin gene, which is closely
related to desnutrin, is dramatically down-regulated by fast-
ing, and its levels are restored upon refeeding (18). Because of
the homology between the two putative proteins, we decided to
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis47068
by guest on November 5, 2015http://www.jbc.org/Downloaded from

FIG.1. Nucleotide and deduced amino acid sequences of the full-length mouse desnutrin cDNA. The full-length cDNA encoding
desnutrin was cloned from WAT. The predicted amino acid sequence corresponds to the longest open reading frame (486 amino acids) and is shown
under the nucleotide sequence. The boldface nucleotides indicate the polyadenylation signal, and the asterisk shows the stop codon. Amino acids
8 –180 (in boldface) indicate the conserved patatin domain.
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis 47069
by guest on November 5, 2015http://www.jbc.org/Downloaded from

investigate whether desnutrin gene expression could also be
regulated by the nutritional status of the organism. WAT from
animals subjected to fasting/refeeding was analyzed for desnu-
trin expression. As shown in Fig. 4A, the levels of desnutrin
mRNA were higher in 24-h fasted mice than in mice that
underwent food deprivation for the same time period and then
were refed for 12 h. These results show evidence of nutritional
regulation of desnutrin gene expression. To determine whether
desnutrin expression is down-regulated during refeeding or
up-regulated by food deprivation, a set of time course experi-
ments were conducted in which mice were fasted for 12, 24, and
48 h or fed for 12 additional h after 48 h of fasting. As shown in
Fig. 4B, compared with fed mice, desnutrin mRNA levels in-
creased rapidly after 12 h of fasting, reaching a maximal level
and then gradually decreasing to basal levels 48 h after fasting
was initiated. Interestingly the expression pattern of the des-
nutrin gene was inversely correlated with that of the adiponu-
FIG.4. Nutritional regulation of desnutrin mRNA levels. A,
expression of desnutrin mRNA in WAT from mice fasted for 24 h (F)or
fasted and then refed for 12 h (R) as assessed by Northern blot analysis;
B, time course analysis of desnutrin and adiponutrin mRNA expression
levels in WAT during fasting. Mice were fasted for 12, 24, or 48 h, and
total RNA was extracted for examination of desnutrin mRNA levels by
Northern blot analysis. Desnutrin mRNA levels in WAT were compared
with those in mice that were fasted for 48 h and subsequently fed for
12 h. FAS, fatty-acid synthase.
FIG.2.Desnutrin mRNA levels in various adult mouse tissues
and cells. A,10
␮
g of total RNA from various mouse tissues was
analyzed by Northern blotting and hybridized with radiolabeled desnu-
trin and adiponutrin cDNA probes. Sk, skeletal; Ing, inguinal; Gon,
gonadal; Ren, renal. B, shown is desnutrin mRNA expression in gonadal
WAT from 12-h fasted wild-type (WT), db/db, and ob/ob mice. C,5
␮
gof
total RNA from cells of the stromal vascular fraction (SVF) or adipo-
cytes isolated from mouse inguinal adipose tissue (left panels)or10
␮
g
of total RNA from 3T3-L1 cells at the indicated days of differentiation
(right panels) was examined by Northern blot analysis for the expres-
sion of desnutrin and various adipocyte markers. Prol. Pre., proliferat-
ing preadipocytes; Conf. pre., confluent preadipocytes; FAS, fatty-acid
synthase; aFABP, adipocyte fatty acid-binding protein.
FIG.3. Desnutrin mRNA levels during adipocyte differentia-
tion of 3T3-L1 cells. Two-day post-confluent 3T3-L1 preadipocytes
(day 0) were induced to differentiate by treatment with 1
␮
M dexa-
methasone and 0.5 m
M MIX for 2 days and then maintained in differ-
entiation medium for an additional 5 days. Ten
␮
g of total RNA pre-
pared from cells collected at the indicated time points was examined for
the expression of desnutrin and other adipocyte markers by Northern
blot analysis. aFABP, adipocyte fatty acid-binding protein; FAS, fatty-
acid synthase; PPAR
␥
, peroxisome proliferator-activated receptor-
␥
.
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis47070
by guest on November 5, 2015http://www.jbc.org/Downloaded from

trin gene, whose expression practically disappeared when mice
were fasted, but was induced upon refeeding. The fatty-acid
synthase gene, known to be tightly regulated by fasting and
refeeding (19), was used as a control. As predicted, fatty-acid
synthase mRNA levels were not detectable during fasting, but
increased drastically during the refeeding period. Together,
these results indicate that desnutrin gene expression is rapidly
and transiently induced in WAT by fasting.
During fasting, circulating glucagon and glucocorticoids lev-
els are elevated, and these hormones serve as mediators of the
adaptive response to starvation by inducing changes in gene
expression to face the new metabolic conditions. Because des-
nutrin mRNA is induced during fasting, we investigated
whether cAMP, the mediator of glucagon action, or glucocorti-
coids can regulate desnutrin expression. Confluent 3T3-L1
preadipocytes, which do not have detectable levels of desnutrin
mRNA, were treated for 48 h with dexamethasone, MIX, dibu-
tyryl cAMP, or dexamethasone/MIX. As shown in Fig. 5A,
neither MIX (an inhibitor of phosphodiesterase) nor the cAMP
analog dibutyryl cAMP induced expression of the desnutrin
gene, indicating that cAMP and glucagon are not involved in
the induction of desnutrin mRNA during fasting. However,
dexamethasone, a synthetic glucocorticoid, was able to signifi-
cantly increase desnutrin mRNA levels in preadipocytes. The
combination of dexamethasone and MIX seemed to have a more
exacerbated effect on desnutrin gene expression, probably be-
cause of the initiation of adipocyte differentiation by dexa-
methasone/MIX treatment. Overall, these results indicate that
glucocorticoids (but not cAMP) induce desnutrin gene expres-
sion in 3T3-L1 cells. An increase in desnutrin mRNA was also
observed when fully differentiated adipocytes were treated
with dexamethasone (Fig. 5B). The increase did not appear to
be as significant as that shown in preadipocytes, probably
because of the high basal levels of desnutrin mRNA in adipo-
cytes. The induction of desnutrin mRNA by dexamethasone
was time- and dose-dependent (Fig. 5, C and D). Confluent
preadipocytes treated for 48 h with increasing amounts of
dexamethasone ranging from 1 n
M to 10
␮
M showed detectable
levels of desnutrin mRNA at concentrations as low as 1 n
M and
a gradual increase in parallel with the dose of dexamethasone.
The half-maximal response occurred in the nanomolar range,
in agreement with the reported K
d
for dexamethasone with the
glucocorticoid receptor of 3T3-L1 cells (20), suggesting that the
effect of dexamethasone on desnutrin mRNA levels was prob-
ably mediated by glucocorticoid binding to its receptor. More-
over, desnutrin expression was time-dependent, first detecta-
ble 24 h after dexamethasone treatment (Fig. 5D).
Intracellular Localization of EGFP-tagged Desnutrin Pro-
tein—PSORTII analysis of the deduced desnutrin amino acid
sequence predicted a cytoplasmic localization of the protein. To
determine whether the desnutrin protein is, in fact, located in
the cytoplasm, we transiently transfected an expression vector
for a desnutrin open reading frame-EGFP fusion protein into
COS-7 cells and analyzed its localization by confocal micros-
copy. A C-terminal EGFP fusion protein was generated rather
than an N-terminal EGFP fusion protein in an attempt to avoid
potential masking of putative signal sequence. As shown in Fig.
6, desnutrin-EGFP showed a homogeneous distribution in the
cells excluding the nucleus, suggesting a cytosolic localization
of the desnutrin gene-encoded protein. On the other hand, in
agreement with a previous report (18), examination of adipo-
nutrin-EGFP expression showed a granular appearance con-
sisting of punctate structures in the cytoplasm (data not
shown).
To confirm the cytoplasmic localization of the desnutrin pro-
tein, COS-7 cells were transfected with an HA-tagged desnu-
trin expression vector, and crude subcellular fractionation fol-
lowed by Western blot analysis was performed. HA-desnutrin
was not detected in the mitochondrial or microsomal fraction,
but a strong signal was detected in the cytosolic fraction (Fig.
6B). A strong signal was detected also in the nuclear fraction,
but this probably was due to the presence of unbroken cells, as
corroborated by optic microscopy (data not shown). Thus, we
conclude that putative desnutrin is a cytoplasmic protein.
Effect of Ectopic Expression of Desnutrin on Triglyceride
Hydrolysis—So far, no function has been attributed to any of
FIG.5.Regulation of desnutrin mRNA levels by glucocorticoids. A, confluent preadipocytes treated for 48 h with 1
␮
M dexamethasone
(DEX), 0.5 m
M MIX, 0.5 mM dibutyryl cAMP (Bt2AMP), or 1
␮
M dexamethasone and 0.5 mM MIX for 48 h (DEX/MIX). After the incubation period,
cells were harvested, and desnutrin mRNA levels were analyzed by Northern blotting. B, desnutrin mRNA levels in fully differentiated adipocytes
untreated (C) or treated for 48 h with 1
␮
M dexamethasone. C, dose-dependent induction of desnutrin mRNA by dexamethasone in 3T3-L1
preadipocytes. D, time course analysis of desnutrin mRNA expression during dexamethasone treatment in 3T3-L1 preadipocytes. Cells were
treated with 1
␮
M dexamethasone for the indicated times.
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis 47071
by guest on November 5, 2015http://www.jbc.org/Downloaded from

the desnutrin protein homologs in mammals. In an attempt to
identify a potential functional domain in desnutrin, we com-
pared its amino acid sequence with the Pfam protein family
data base. A so-called patatin domain present in the region
spanning amino acids 8 –179 (Fig. 1) of desnutrin was identi-
fied. The domain owes its name to patatin, a storage protein
found in plants that possesses lipid acyl hydrolase and broad
esterase activity. Interestingly, ClustalW alignment of desnu-
trin and the related proteins described above showed that the
highest degree of homology concentrates, precisely, in this
patatin domain. Several regions conserved among the desnu-
trin-related proteins in metazoans could also be found in pro-
teins of organisms as phylogenetically distant as plants, fungi,
and prokaryotes (Fig. 7) (data not shown). Three highly con-
served regions could be identified (Fig. 7): a glycine-rich
GXGXXG nucleotide binding motif, a GXSXG serine hydrolase
motif, and a DX(G/A) motif containing a conserved aspartate
residue. The serine and aspartate residues constitute a cata-
lytic dyad that is required for the lipase activity of patatin (21).
The high homology between desnutrin, adiponutrin, TTS2, and
GS2 indicates that these proteins belong to the same family,
and the similarity in structure suggests a similar function.
The homology of the protein encoded by the desnutrin gene to
the patatin-like protein family led us to investigate whether
ectopic overexpression of desnutrin could affect hydrolysis of
triglycerides due to its potential lipase activity. We labeled
cellular lipid with [U-
14
C]palmitic acid for4hinCOS-7 cells
expressing HA-desnutrin or control cells transfected with
empty vector (Fig. 8A), and the lipid composition was analyzed
by thin layer chromatography. As shown in Fig. 8B, after the
labeling period (time 0), cells expressing desnutrin or cells
transfected with a control empty vector showed the same lipid
profile, and no quantitative difference among the different
types of lipids was observed between control and desnutrin-
expressing cells. This indicates that synthesis of lipids was not
affected by desnutrin overexpression. The labeled triglyceride
levels did not change significantly when the control cells were
maintained in normal medium for an additional 8 h. On the
other hand, in desnutrin-overexpressing cells, the labeled tri-
glyceride levels decreased significantly by ⬃40% (Fig. 8, B, left
panel; and D, lower panel). There were no differences in the
phospholipid and other lipid profiles (Fig. 8B, right panel),
indicating that ectopic expression of HA-desnutrin affected
only triglyceride hydrolysis. The reduction in labeled triglycer-
ides observed in HA-desnutrin-transfected cells was time-de-
pendent; and after 16 h of incubation, their levels in the cells
were reduced further by 80% (Fig. 8C), whereas in control cells,
there was only a 30% reduction in labeled triglycerides. These
results clearly indicate that the triglycerides in desnutrin-
overexpressing cells were hydrolyzed at a faster rate than in
control cells.
The intracellular labeled free fatty acid levels decreased
significantly during the 8-h period of incubation in both des-
nutrin-transfected and control cells. Therefore, despite the
lower levels of labeled triglycerides detected in the desnutrin-
transfected cells, we could not detect higher levels of intracel-
lular free fatty acids, suggesting that the fatty acids resulting
from triglyceride hydrolysis were either released to the me-
dium or rapidly metabolized in the cell. We therefore examined
the radiolabeled lipids in the culture medium of the COS cells
(Fig. 8D, upper panels). TLC analysis of lipids extracted from
the medium showed that the labeled lipids found in the me-
dium were mainly free fatty acids (Fig. 8D, upper left panel). As
shown in Fig. 8D (upper right panel), the labeled free fatty
acids in the medium gradually increased when the cells were
maintained in normal medium up to 8 h. Furthermore, desnu-
trin-expressing cells showed higher levels of radiolabeled free
fatty acids released to the medium compared with control
empty vector-transfected cells. Therefore, the decrease in la-
beled cellular triglyceride levels was accounted for in part by an
increase in free fatty acids release to the medium, providing
further evidence that the desnutrin-overexpressing cells have a
higher rate of lipolysis.
DISCUSSION
In our search for genes that could play a crucial role in
adipocyte differentiation or function, we previously have suc-
cessfully used EST cDNA microarrays to identify genes that
are expressed exclusively in adipose tissue (12). Using the
same approach, we now have identified and cloned from adi-
pose tissue a 1965-bp cDNA encoding a 486-amino acid puta-
tive protein that we have named desnutrin. The predicted
amino acid sequence shows that desnutrin belongs to a distinct
new family of proteins that have common structural features
and that are regulated by the nutritional condition of the ani-
mals. So far, the genes encoding the two members of this
family, desnutrin and adiponutrin, have been cloned and iden-
tified to be predominantly adipocyte-specific.
Whereas adiponutrin mRNA expression is restricted to adi-
pose tissue, desnutrin mRNA, although expressed predomi-
nantly at a high level in adipose tissue, is also found at a low
level in a variety of tissues. This favored expression in adipose
tissue clearly suggests the involvement of the desnutrin gene in
a function preferential to but not exclusive of adipose tissue.
The increase in triglyceride hydrolysis associated with desnu-
trin overexpression and the induction of desnutrin gene expres-
sion by fasting are consistent with this hypothesis. Indeed,
although most tissues use their triglycerides for hydrolysis and
oxidation during energy depletion, the energy reserve is not
sufficient, and these cells need to rely on the substrates pro-
FIG.6.Subcellular localization of desnutrin-EGFP fusion pro-
tein. A, COS-7 cells were transfected with the desnutrin-EGFP expres-
sion vector, and localization of the fusion protein was assessed by
confocal microscopy. B, COS-7 cells were transfected with an HA-tagged
desnutrin expression vector, and nuclear (Nuc.), mitochondrial (Mit.),
microsomal (Mic.), and cytosolic (Cyt.) fractions were prepared as de-
scribed under “Experimental Procedures.” Five
␮
g of protein from each
fraction was subjected to SDS-PAGE, transferred to a polyvinylidene
difluoride membrane, and analyzed for the presence of HA-desnutrin
using anti-HA antibody. As a positive control for the immunodetection,
10
␮
g of whole cell lysate was used.
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis47072
by guest on November 5, 2015http://www.jbc.org/Downloaded from

vided by adipose tissue, an organ specialized in energy storage
in the form of triglycerides. It is worth noting that desnutrin
mRNA levels vary depending on the adipose tissue type and
depot. The differential regional expression of the desnutrin
gene could be a factor contributing to the metabolic heteroge-
neity observed among different adipose depots (22, 23). Inter-
estingly, desnutrin gene expression is lower in genetically ob/ob
and db/db obese mouse models, suggesting a potential role of
desnutrin in the pathophysiology of obesity.
The homology data from a BLASTp search did not provide
clues to the potential role of desnutrin. On the other hand, a
Pfam protein family data base search identified the presence of
a patatin-like domain in the N-terminal region of desnutrin.
Patatin is a member of a multigene family of proteins found in
potato and other solanaceous plants (24). It accounts for 40% of
the total soluble potato tuber protein and is considered to be a
storage protein. In addition, patatin displays broad lipid acyl
hydrolase activity (25, 26). This enzymatic activity relies on a
catalytic dyad formed by a conserved serine residue in the
GXSXG motif, characteristic of esterases, and a conserved as-
partate residue belonging to the DX(G/A) motif (21). The struc-
tural resemblance of desnutrin to patatins, including the pres-
ence of the conserved residues that constitute the catalytic
dyad, and the increase in triglyceride hydrolysis observed in
cells overexpressing the desnutrin open reading frame support
our hypothesis that the desnutrin protein may be a lipase.
When ectopically overexpressed, desnutrin seems to have
effects specifically on triglycerides since no change in choles-
terol or phospholipids was observed in the lipid labeling exper-
iments. The 85-kDa calcium-independent phospholipase A
2
,a
mammalian protein with eight ankyrin motifs and a patatin
domain with the serine-aspartate catalytic dyad, selectively
hydrolyzes phospholipids at the sn-2 position (27). On the other
hand, plant patatins act on a broad range of substrates, includ-
ing phospholipids, glycolipids, sulfolipids, and mono- and dia-
cylglycerols, but not triacylglycerols (25). Interestingly, ExoU,
the Pseudomonas aeruginosa encoded type III cytotoxin, a re-
cently described lipase containing a patatin domain, has tri-
glycerides and other neutral lipids among its substrates (28).
This disparity in substrate specificity, despite the high degree
of conservation in the residues that form the catalytic domain,
suggests that the domain(s) responsible for substrate recogni-
tion may rely on regions that are less conserved among the
different lipases containing a patatin domain. This differential
specificity can also have important functional significance. The
delivery of ExoU by P. aeruginosa is associated with lung
injury and sepsis in animal models (29). These pathologies
seem to be caused by alterations in the membrane permeability
of the host cells due to the lipase activity of the toxin (28). In
patatins, the broad lipase activity has been associated with a
defense function, inhibiting the growth of some insect larvae by
disrupting their mid-gut membranes (30). On the other hand,
phospholipases A
2
specifically cleave the sn-2 ester bond of
substrate phospholipids, and the released fatty acids can func-
tion as second messengers or precursors of eicosanoids that
mediate signal transduction (31). The apparent specificity of
the desnutrin gene-encoded protein for triglycerides, its tissue
distribution, and the induction of expression by fasting suggest
that the desnutrin gene could play a role in the response of the
organism to starvation, enhancing hydrolysis of triglycerides
and providing free fatty acids to other tissues to be oxidized in
situations of energy depletion. In this regard, hormone-sensi-
tive lipase is known to be a key enzyme in the mobilization of
fatty acids from acylglycerols in adipocytes as well as, albeit
low, in non-adipocytes (32). In hormone-sensitive lipase null
mice, catecholamine-induced glycerol and fatty acid release is
significantly blunted. However, significant levels of lipolysis
occur in hormone-sensitive lipase null cells, and diglycerides
accumulate in adipocytes, indicating the presence of additional
lipase(s) that are not catecholamine-sensitive and that are
mainly triglyceride lipases. Interestingly, as discussed below,
desnutrin gene expression is regulated by glucocorticoids, but
not by cAMP. We propose that desnutrin may function as a
triglyceride lipase. However, because our present study was
performed by transfection of the open reading frame of the
desnutrin gene, examination of the endogenous protein using
specific antibodies as well as RNA interference approaches in
adipocytes will further clarify this conclusion.
The nutritional regulation of the murine patatin-like pro-
tein-encoding genes seems to be specific to each of the family
members. Indeed, adiponutrin gene expression is abolished by
fasting and up-regulated by feeding. The desnutrin gene, how-
FIG.7.Multiple sequence alignment of amino acid sequences corresponding to the patatin domain of desnutrin and desnutrin-
related proteins. Alignment of desnutrin and desnutrin-related proteins revealed the presence of a highly conserved 180-amino acid N-terminal
region identified as a patatin domain in the Pfam protein family data base. Alignment of the following sequences is shown: human TTS2.1, TTS2.2,
and GS2-like protein; mouse desnutrin and adiponutrin; D. melanogaster CG5295-PA; C. elegans C05D11.7; and A. thaliana Atlg33270. Identical
residues in all aligned sequences are indicated by dark gray boxes. Residues conserved in 80% of the aligned sequences are indicated by light gray
boxes. The glycine-rich motif (GXGXXG), the active serine hydrolase motif (GXSXG), and the aspartate active site (DX(G/A)) are shown.
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis 47073
by guest on November 5, 2015http://www.jbc.org/Downloaded from

ever, shows an opposing pattern of expression, being tran-
siently up-regulated by fasting. The opposite expression pat-
tern of desnutrin and adiponutrin genes may reflect a
differential hormonal regulation during fasting/feeding, and
this suggests different functions for these proteins. We have
demonstrated that desnutrin gene expression is induced by
glucocorticoids, one of the hormones involved in the response of
an organism to fasting, but not by cAMP. On the other hand,
induction of adiponutrin gene expression by glucose has been
reported to be counteracted by isoproterenol and forskolin,
which mimic catecholamine action by raising the intracellular
cAMP levels. The mode of action of glucocorticoids on desnutrin
gene expression is still unclear. Undoubtedly, the fact that low
concentrations of dexamethasone, close to the K
d
for the glu
-
cocorticoid receptor, can induce desnutrin expression suggests
that its effect is mediated by the glucocorticoid receptor. How-
ever, we have not found glucocorticoid response elements in the
5⬘-promoter region of the desnutrin gene. It is known that the
glucocorticoid receptor can exert its effects not only by binding
directly to the glucocorticoid response element, but also by
interacting with other transcription factors such as activator
protein-1 (33) and C/EBP
␤
(34) by a mechanism termed cross-
coupling. Interestingly, several consensus binding sites for
C/EBP that could mediate the glucocorticoid effect are present
in the desnutrin promoter region (data not shown). Another
possibility is an indirect action of glucocorticoids on desnutrin
gene expression through an intermediary factor that is induced
by glucocorticoids. This may explain the somewhat slow induc-
tion of desnutrin gene expression by dexamethasone. However,
we did not observe any difference in desnutrin mRNA induction
by dexamethasone in the presence of cycloheximide (data not
shown).
In conclusion, with the cloning of desnutrin cDNA, we define
a new family of mammalian genes coding for proteins charac-
terized by the presence of a patatin-like domain as a structural
feature. Expression of the two members of the family identified
and characterized so far appears to be under strict hormonal
and nutritional control. We suggest naming this gene family of
desnutrin homologs and the protein they encode “nutrins,”
given the control nutritional status exerts on its expression in
adipose tissue. Both desnutrin and adiponutrin genes are
highly expressed in adipose tissue, and desnutrin protein ac-
tivity seems to be involved in triglyceride hydrolysis and, by
extension, in energy homeostasis. In this regard, genetic obe-
sity models show lower levels of desnutrin expression, suggest-
ing its potential contribution to the pathophysiology of obesity.
However, in vitro biochemical studies, including characteriza-
tion of the endogenous protein, are needed for definitive evi-
dence for the function of desnutrin and adiponutrin. It would be
interesting to see whether desnutrin, which we found localized
in the cytoplasm, could be translocated to lipid droplets, which
could be regulated by perilipin, as in the case of hormone-
sensitive lipase (35). Furthermore, in vivo experiments on the
TLC autora-diograms using solvent system A to separate neutral lipids
and free fatty acids (FFA)(left panel) and solvent system B to separate
phospholipids and free fatty acids (right panel). TG, triglycerides; Chol.,
cholesterol; DAG, diacylglycerol; PC, phosphatidylcholine; PI, phos-
phatidylinositol. C, shown are the results from time course analysis of
labeled triglyceride content in control and HA-desnutrin-transfected
COS-7 cells. Labeled lipids were isolated, resolved by TLC, and detected
by autoradiography. Bands corresponding to triglycerides are shown. D,
labeled lipids in the cell culture medium corresponded mostly to free
fatty acids as determined by TLC (upper left panel). Labeled lipids in
the cell culture medium, mostly FFA, were quantified in a scintillation
counter (upper right panel). Densitometric scanning of intracellular
triglycerides from two independent experiments was also performed
(lower panel). Results are expressed as the means ⫾ S.E. *, p ⬍ 0.05; **,
p ⬍ 0.01.
FIG.8.Effect of ectopic expression of desnutrin on hydrolysis
of triglycerides in COS-7 cells. COS-7 cells transfected with
pcDNA3.1 (control (Cont)) or the HA-desnutrin expression vector (Desn)
were labeled with [U-
14
C]palmitic acid for 4 h. After washing, cells were
maintained in serum-containing medium. At this time, cells were col-
lected (time 0) or incubated for an additional 8 h before harvesting.
Total lipids were extracted, and lipid composition was analyzed by TLC
using two solvent systems. A, shown is desnutrin expression in COS-7
cells after transient transfection with the HA-tagged desnutrin expres-
sion vector or the empty pcDNA3.1 vector. B, shown are representative
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis47074
by guest on November 5, 2015http://www.jbc.org/Downloaded from

gain or loss of function using transgenic technology will cer-
tainly clarify the role of desnutrin in the response of the orga-
nism to caloric shortage as well as its possible implication in
pathologies associated with altered adipose function or lipid
metabolism such as obesity, lipodystrophy, and diabetes.
Acknowledgment—We thank Dr. L. Rubio for invaluable help with
sequence-editing software.
REFERENCES
1. Rosen, E. D., and Spiegelman, B. M. (2000) Annu. Rev. Cell Dev. Biol. 16,
145–171
2. Gregoire, F. M., Smas, C. M., and Sul, H. S. (1998) Physiol. Rev. 78, 783– 809
3. MacDougald, O. A., and Mandrup, S. (2002) Trends Endocrinol. Metab. 13,
5–11
4. Fruhbeck, G., Gomez-Ambrosi, J., Muruzabal, F. J., and Burrell, M. A. (2001)
Am. J. Physiol. 280, E827–E847
5. Unger, R. H., and Orci, L. (2001) FASEB J. 15, 312–321
6. Burton, G. R., Nagarajan, R., Peterson, C. A., and McGehee, R. E., Jr. (2004)
Gene (Amst.) 329, 167–185
7. Guo, X., and Liao, K. (2000) Gene (Amst.) 251, 45–53
8. Sadowski, H. B., Wheeler, T. T., and Young, D. A. (1992) J. Biol. Chem. 267,
4722– 4731
9. Smas, C. M., and Sul, H. S. (1993) Cell 73, 725–734
10. Zhao, L., Gregoire, F., and Sul, H. S. (2000) J. Biol. Chem. 275, 16845–16850
11. Villena, J. A., Kim, K.-H., and Sul, H. S. (2002) Horm. Metab. Res. 34, 664 – 670
12. Kim, K.-H., Lee, K., Moon, Y. S., and Sul, H. S. (2001) J. Biol. Chem. 276,
11252–11256
13. Kim, K.-H., Zhao, L., Moon, Y. S., Kang, C., and Sul, H. S. (2004) Proc. Natl.
Sci. U. S. A. 101, 6780 – 6785
14. Banerjee, R. R., Rangwala, S. M., Shapiro, J. S., Rich, A. S., Rhoades, B., Qi,
Y., Wang, J., Rajala, M. W., Pocai, A., Scherer, P. E., Steppan, C. M., Ahima,
R. S., Obici, S., Rossetti, L., and Lazar, M. A. (2004) Science 303, 1195–1198
15. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W.,
and Lipman, D. J. (1997) Nucleic Acids Res. 25, 3389–3402
16. Gulick, T. (1996) in Current Protocols in Molecular Biology (Ausubel, F. M.,
Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and
Struhl, K., eds) John Wiley & Sons, Inc., New York
17. Bligh, E. G., and Dyer, W. J. (1959) Can. J. Med. Sci. 37, 911–917
18. Baulande, S., Lasnier, F., Lucas, M., and Pairault, J. (2001) J. Biol. Chem. 276,
33336 –33344
19. Sul, H. S., and Wang, D. (1998) Annu. Rev. Nutr. 18, 331–351
20. Nakada, M. T., Stadel, J. M., Poksay, K. S., and Crooke, S. T. (1987) Mol.
Pharmacol. 31, 377–384
21. Rydel, T. J., Williams, J. M., Krieger, E., Moshiri, F., Stallings, W. C., Brown,
S. M., Pershing, J. C., Purcell, J. P., and Alibhai, M. F. (2003) Biochemistry
42, 6696 – 6708
22. Martin, M. L., and Jensen, M. D. (1991) J. Clin. Investig. 88, 609 –613
23. Wahrenberg, H., Lonnqvist, F., and Arner, P. (1989) J. Clin. Investig. 84,
458 – 467
24. Ganal, M. W., Bonierbale, M. W., Roeder, M. S., Park, W. D., and Tanksley,
S. D. (1991) Mol. Gen. Genet. 225, 501–509
25. Hirschberg, H. J., Simons, J. W., Dekker, N., and Egmond, M. R. (2001) Eur.
J. Biochem. 268, 5037–5044
26. Andrews, D. L., Beames, B., Summers, M. D., and Park, W. D. (1988) Biochem.
J. 252, 199 –206
27. Tang, J., Kriz, R. W., Wolfman, N., Shaffer, M., Seehra, J., and Jones, S. S.
(1997) J. Biol. Chem. 272, 8567– 8575
28. Sato, H., Frank, D. W., Hillard, C. J., Feix, J. B., Pankhaniya, R. R., Moriyama,
K., Finck-Barbancon, V., Buchaklian, A., Lei, M., Long, R. M., Wiener-
Kronish, J., and Sawa, T. (2003) EMBO J. 22, 2959–2969
29. Allewelt, M., Coleman, F. T., Grout, M., Priebe, G. P., and Pier, G. B. (2000)
Infect. Immun. 68, 3998 – 4004
30. Strickland, J. A., Orr, G. L., and Walsh, T. A. (1995) Plant Physiol. 109,
667– 674
31. Six, D. A., and Dennis, E. A. (2000) Biochim. Biophys. Acta 1488, 1–19
32. Holm, C. (2003) Biochem. Soc. Trans. 31, 1120 –1124
33. Schule, R., Rangarajan, P., Kliewer, S., Ransone, L. J., Bolado, J., Yang, N.,
Verma, I. M., and Evans, R. M. (1990) Cell 62, 1217–1226
34. Nishio, Y., Isshiki, H., Kishimoto, T., and Akira, S. (1993) Mol. Cell. Biol. 13,
1854 –1862
35. Sztalryd, C., Xu, G., Dorward, H., Tansey, J. T., Contreras, J. A., Kimmel,
A. R., and Londos, C. (2003) J. Cell Biol. 161, 1093–1103
Adipose Desnutrin, a Patatin-like Protein, Increases Lipolysis 47075
by guest on November 5, 2015http://www.jbc.org/Downloaded from

Sul
Sarkadi-Nagy, Kee-Hong Kim and Hei Sook
Josep A. Villena, Suheeta Roy, Eszter
TRIGLYCERIDE HYDROLYSIS
DESNUTRIN INCREASES
ECTOPIC EXPRESSION OF
Is Induced by Fasting and Glucocorticoids:
Novel Patatin Domain-containing Protein,
Desnutrin, an Adipocyte Gene Encoding a
Lipids and Lipoproteins:
doi: 10.1074/jbc.M403855200 originally published online August 27, 2004
2004, 279:47066-47075.J. Biol. Chem.
10.1074/jbc.M403855200Access the most updated version of this article at doi:
.JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the
Alerts:
When a correction for this article is posted•
When this article is cited•
to choose from all of JBC's e-mail alertsClick here
http://www.jbc.org/content/279/45/47066.full.html#ref-list-1
This article cites 34 references, 17 of which can be accessed free at
by guest on November 5, 2015http://www.jbc.org/Downloaded from