Báo cáo khoa học: Glutamine stimulates the gene expression and processing of sterol regulatory element binding proteins, thereby increasing the expression of their target genes docx

When HepG2 cells were cultured with both glutamine and sterols, the increased SREBP-1a mRNA levels were not reduced by sterols, which suppressed SREBP-2 expression robustly Fig.. HepG2 c

of sterol regulatory element binding proteins, thereby

increasing the expression of their target genes

Jun Inoue, Yuka Ito, Satoko Shimada, Shin-ich Satoh, Takashi Sasaki, Tsutomu Hashidume,

Yuki Kamoshida, Makoto Shimizu and Ryuichiro Sato

Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, University of Tokyo, Japan

Introduction

Because amino acids are indispensable nutrients for cell

growth, cell culture media usually contain large

amounts of them In addition to their role as substrates

for protein synthesis, glucogenic substrates and

nitro-gen carriers, amino acids often act as critical regulators

of the transcription of certain genes For example,

amino acid supplementation with tryptophan and

glutamine induce the gene expression of collagenase and argininosuccinate synthetase, respectively [1,2] Amino acid starvation also regulates the transcription

of several genes, including fatty acid synthase, aspara-gine synthetase and C⁄ EBP homologous protein [3–5] Amino acid metabolism is strongly linked to both glucose and fatty acid metabolism Under certain

Keywords

glutamine; processing; Sp1; SREBP;

transcriptional regulation

Correspondence

R Sato, 1-1-1 Yayoi, Bunkyo-ku, Tokyo

113-8657, Japan

Fax: 81 5841 8029

Tel: 81 5841 5136

E-mail: aroysato@mail.ecc.u-tokyo.ac.jp

(Received 17 March 2011, revised 23 May

2011, accepted 1 June 2011)

doi:10.1111/j.1742-4658.2011.08204.x

Here we show that the larger the amount of glutamine added to the med-ium, the more the expression of genes related to lipid homeostasis is pro-moted by the activation of sterol regulatory element binding proteins (SREBPs) at the transcriptional and post-translational levels in human hep-atoma HepG2 cells Glutamine increases the mRNA levels of several SREBP targets, including SREBP-2 The gene expression of SREBP-1a,

a predominant form of SREBP-1 in most cultured cells and a target of the general transcription factor Sp1, is significantly augmented by glutamine via an increased binding of Sp1 to the SREBP-1a promoter In contrast, the increased expression of SREBP targets including SREBP-2 is due to stimulation of the processing of SREBP proteins by glutamine It is also shown that glutamine accelerates SREBP processing through increased transport of the SREBP⁄ SREBP cleavage-activating protein complex from the endoplasmic reticulum to the Golgi apparatus The processing of acti-vating transcription factor 6 is activated by the same proteases as SREBPs

in the Golgi in response to endoplasmic reticulum stress and is not induced

by glutamine Taken together, these results clearly demonstrate that gluta-mine brings about not only the induction of SREBP-1a transcription but also the stimulation of SREBP processing, thereby facilitating the gene expression of SREBP targets in cultured cells

Abbreviations

ATF6, activating transcription factor 6; CPT1A, carnitine palmitoyltransferase 1A; ER, endoplasmic reticulum; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFAT, glutamine:fructose-6-glyceraldehyde-3-phosphate amidotransferase; HMG, 3-hydroxy-3-methylglutaryl; Insig, insulin-inducing gene; LDL, low density lipoprotein; MTP, microsomal triglyceride transfer protein; PI3K, phosphatidylinositol 3-kinase; PLAP, placental alkaline phosphatase; p70S6K, p70 ribosomal protein S6 kinase; SCAP, SREBP cleavage-activating protein; SQS, squalene synthase; SREBP, sterol regulatory element-binding protein; S1P, site-1 protease; S2P, site-2 protease.

physiological conditions amino acids are metabolized

to either glucose precursors or acetoacetyl-CoA

Acetoacetyl-CoA is then converted to acetyl-CoA and

it subsequently enters into the tricarboxylic acid cycle,

or is used as a precursor of fatty acids in response to

their demands Under fasting conditions, the

acetyl-CoA provided by the oxidation of free fatty acids

increases the consumption of amino acids as

precur-sors of the oxaloacetate required for condensation

with acetyl-CoA [6] Although the acetyl-CoA

provided as a metabolite of amino acids can be a

sub-strate for cholesterol synthesis, the interplay between

amino acid and cholesterol metabolism remains largely

unknown

Cholesterol and fatty acid synthesis are strictly

regulated at the transcriptional level SREBPs are a

family of transcription factors which consists of the

SREBP-1a, SREBP-1c and SREBP-2 proteins that

control the transcription of genes related to

cholesterol and fatty acid metabolism [7] SREBPs are

synthesized as membrane proteins located on the

endoplasmic reticulum (ER) and are processed to

lib-erate the N-terminal halves that function as

transcrip-tion factors in the nucleus The proteolytic processing

of SREBPs is tightly regulated by the interaction

between two ER membrane proteins, SREBP

cleav-age-activating protein (SCAP) and the

insulin-induc-ing gene (Insig) When cells are depleted of sterols,

SCAP escorts SREBPs from the ER to the Golgi

Thereafter, SREBPs are processed by two proteases,

site-1 protease (S1P) and site-2 protease (S2P) within

the Golgi Once the ER membrane cholesterol content

increases, SCAP binds to cholesterol, induces

confor-mational change and becomes attached to Insig,

thereby remaining on the ER membrane

There-fore, cholesterol is a critical determinant of SREBP

activation

In the present study, we report that glutamine

treat-ment results in an increase in the promoter activities of

a number of SREBP targets, such as the low density

lipoprotein (LDL) receptor,

3-hydroxy-3-methylgluta-ryl (HMG) CoA synthase and squalene synthase (SQS)

genes in human hepatoma HepG2 cells We further

investigated the molecular mechanism by which

glutamine affects the expression of the genes involved

in cholesterol homeostasis The results clearly

demon-strate that glutamine stimulates SREBP processing and

the gene expression of SREBP targets The same

con-centrations of alanine, proline and glutamate treatment

did not influence SREBP processing Moreover,

gluta-mine treatment also causes an increase in hexosagluta-mine

biosynthesis as a substrate, thereby enhancing the

SREBP-1a mRNA levels To our knowledge, this is

the first report showing that glutamine promotes SREBP activities and stimulates the gene expression of SREBP targets

Results and Discussion

Glutamine stimulates the promoter activities of SREBP targets

More than 50 years ago, Eagle et al [8] reported the importance of glutamine in a culture medium for cell proliferation Because cholesterol is essential for the constituent of membrane, we examined whether gluta-mine regulates the expression of genes involved in cho-lesterol homeostasis To investigate the effect of glutamine on the transcription of genes related to cho-lesterol metabolism, a variety of reporter assays were preformed in HepG2 cells The promoter activities of the LDL receptor, HMG CoA synthase and SQS were increased by the addition of a 4- or 10-fold excess of glutamine in DMEM, while the promoter activity of microsomal triglyceride transfer protein (MTP) was attenuated in a dose-dependent manner (Fig 1A) In contrast, carnitine palmitoyltransferase 1A (CPT1A) promoter activity was not affected by glutamine (Fig 1A) Since these observed glutamine effects mim-icked SREBP actions, we next determined whether treatment with sterols, which almost completely inhibit SREBP processing, affects the glutamine-induced pro-moter activities The glutamine-induced promoter activities of both HMG CoA synthase and SQS were abolished by sterols (Fig 1B) When one or two of the sterol regulatory elements (SREs) in the HMG CoA synthase or SQS promoter, respectively, was mutated (SREKO), the glutamine stimulatory effects were greatly reduced (Fig 1B) Moreover, the glutamine effect on the HMG CoA synthase promoter activity was also observed in human intestinal epithelial Caco2 cells (Fig 1C) From these results, it seems likely that the glutamine actions are mediated by the activation

of SREBP

Next, we sought to confirm whether glutamine causes an elevation of the endogenous mRNA levels

of SREBP targets HepG2 cells were treated with a 10-fold excess of glutamine for 12 h Quantitative real-time PCR analyses revealed that glutamine treatment caused a significant increase in the mRNA levels of both HMG CoA synthase and the LDL receptor, and the glutamine effects were completely abolished by the addition of sterols (Fig 2) Taken together, glutamine

is evidently capable of stimulating the SREBP-mediated expression of genes related to cholesterol metabolism

Glutamine enhances the mRNA levels of SREBP

family members

There are a couple of possible explanations for the

glu-tamine-mediated promotion of SREBP functions The

most likely are, first, an increase in the gene expression

of SREBPs, and second, the enhancement of SREBP processing by glutamine We therefore investigated the effect of glutamine on the gene expression of SREBP family members, SREBP-1 and SREBP-2 SREBP-1 exists in two forms, designated 1a and 1c SREBP-1a

is the predominant isoform in most cultured cells, including HepG2 cells [9], and is a more potent transcription factor than SREBP-1c [10] Accordingly,

we examined the effect of glutamine on the gene expression of SREBP-1a and SREBP-2 in HepG2 cells

in the following experiments The mRNA levels of both SREBP-1a and SREBP-2 were significantly elevated by treatment with a 10-fold excess of gluta-mine for 12 h in HepG2 cells (Fig 3A) It has been demonstrated that SREBP-2 is an SREBP target gene, and that the transcription of SREBP-1a gene is predominantly regulated by the general transcription factor Sp1 [11,12] We next compared the glutamine-induced gene expression of SREBPs under various conditions When HepG2 cells were cultured with both glutamine and sterols, the increased SREBP-1a mRNA levels were not reduced by sterols, which suppressed SREBP-2 expression robustly (Fig 3A), implying that the gene expression of SREBP-2 is induced by the acti-vation of the SREBP in response to the higher gluta-mine concentration In contrast, the elevation of the SREBP-1a levels by glutamine was completely abolished by azaserine, an inhibitor of glutamine:fruc-tose-6-phosphate amidotransferase (GFAT), whereas the SREBP-2 mRNA level was not affected These

0.0

0.5

1.0

1.5

2.0

2.5

3.0

pHMG S

3.5

4.0

B

0.0

0.5

1.0

1.5

2.0

2.5

3.0

**

**

**

**

**

*

0.0

0.5

1.0

1.5

2.0

2.5

pCPT1A pHMG S

*

**

3.5

No addition

4 × Gln

10 × Gln

**

**

0 1 2 3

pSQS

4 5

HepG2

**

10 × Gln

10 × Gln + Sterols

No addition

4 × Gln

10 × Gln

Fig 1 Glutamine stimulates the promoter activities of SREBP

tar-gets HepG2 cells (A and B) and Caco-2 cells (C) were transfected

with 200 ng of the reporter constructs consisting of the indicated

gene promoters and 200 ng of pEF-b-Gal The cells were cultured

with medium A (A and B) or medium B (C) for 36 h and then re-fed

with the low amino acid medium supplemented with the indicated

concentration of glutamine (4 · Gln, 16 mM; 10 · Gln, 40 mM) for

12 h in the presence or absence of sterols (10 lgÆmL)1of

choles-terol plus 1 lgÆmL)1 of 25-hydroxycholesterol) Luciferase assays

were performed as described in Materials and methods The

pro-moter activities without glutamine addition are represented as 1.

All data are presented as means ± SD of three independent

experi-ments performed in triplicate *P < 0.05; **P < 0.01.

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

0.0 0.5 1.0 1.5 2.0 2.5

3.0

**

**

No addition

10 × Gln

10 × Gln + Sterols

Fig 2 Glutamine enhances the gene expression of endogenous SREBP targets in HepG2 cells HepG2 cells were cultured with the low amino acid medium containing a 10-fold excess of glutamine (10 · Gln, 40 mM) for 12 h in the presence or absence of sterols (10 lgÆmL)1of cholesterol plus 1 lgÆmL)1of 25-hydroxycholesterol) and total RNA was isolated Real-time PCR analysis was per-formed, and relative mRNA levels were obtained after normalization

to GAPDH mRNA The mRNA levels without glutamine addition are represented as 1 All data are presented as means ± SD of three independent experiments performed in triplicate **P < 0.01.

results imply the involvement of the O-glycosylation of

Sp1 in the induction of SREBP-1a transcription by

glutamine The gene expression of SREBP-1c, which

represented less than 20% of the SREBP-1a gene

expression in our experiments, was regulated in a

similar manner to SREBP-2 (data not shown) It has

been reported that glutamine treatment stimulated

O-glycosylation of Sp1 in Caco2 cells, in turn causing

an increase in Sp1 activity through induced nuclear

localization [13] To examine whether glutamine

treat-ment promotes O-glycosylation of Sp1 in HepG2 cells,

we performed immunoblotting analyses using the RL2 antibody, which recognizes N-acetylglucosamine attached to a serine or threonine residue Glutamine elevated the O-glycosylated Sp1 level, whereas azaser-ine completely abolished this effect and further reduced the basal O-glycosylated Sp1 (Fig 3B) More-over, we examined whether glutamine induces the translocation of Sp1 from the cytosol to the nucleus

As shown in Fig 3C, the amount of nuclear Sp1 was

**

**

0.0 0.5 1.0 1.5 2.0

SREBP-2 0.0

0.5

1.0

1.5

2.0

2.5

SREBP-1a

*

**

Azaserine

+ + – – – IP: anti-Sp1 IB: anti-O-GlcNAc IB: anti-Sp1

1.0 1.2 0.7

Cytosol Nuclear

Time after Gln (h) IB: anti-Sp1

1.0 0.7 0.8 1.0 1.7 1.9

No addition

10 × Gln

10 × Gln + Sterols

10 × Gln + Azaserine

D

0.0 0.1 0.2

SREBP-1a promoter 0.4

anti-Sp1

0.3

0.0 0.1 0.2

SREBP-1a-distal 0.4

IgG anti-Sp1

0.3

IP: IgG

No addition 10 × Gln

E

0.0 0.5 1.0 1.5 2.0

SREBP-1a 2.5

siLuc siSp1 0.0

0.2 0.4 0.8 1.0

Sp1 1.2

siLuc siSp1

0.6

**

No addition 10 × Gln

Fig 3 Glutamine increases the mRNA levels of SREBPs and O-glycosylated Sp1 (A) HepG2 cells were cultured with the low amino acid medium supplemented with a 10-fold excess of glutamine (10 · Gln, 40 mM) for 12 h in the presence or absence of sterols (10 lgÆmL)1of cholesterol plus 1 lgÆmL)1of 25-hydroxycholesterol) or 5 lM azaserine Real-time PCR analysis was performed, and relative mRNA levels were obtained after normalization to GAPDH mRNA The mRNA levels without the glutamine addition are represented as 1 All data are pre-sented as means ± SD of three independent experiments performed in triplicate *P < 0.05; **P < 0.01 (B) HepG2 cells were cultured with the low amino acid medium supplemented with or without a 10-fold excess of glutamine (10 · Gln, 40 mM) in the presence or absence of

5 lM azaserine for 12 h The whole cell extracts were subjected to immunoprecipitation (IP) with anti-Sp1 antibody Aliquots of immunopre-cipitates were subjected to SDS ⁄ PAGE and immunoblot (IB) analysis with anti-O-GlcNAc or anti-Sp1 antibodies, and the signals were quanti-fied with a Fujifilm LAS-3000 Luminoimager Fold change was calculated by the ratio of the intensity between the O-glycosylated Sp1 and the whole Sp1 signals The ratio in the absence of both glutamine and azaserine was set as 1 The same results were obtained in more than three separate experiments (C) HepG2 cells were cultured with the low amino acid medium for 4 h and then re-fed with the medium sup-plemented with a 10-fold excess of glutamine (10 · Gln, 40 mM) for the indicated periods The nuclear and cytosol fractions were prepared

as described previously [29] and the extracts were subjected to IB with Sp1 antibody; the signals were quantified with a Fujifilm LAS-3000 Luminoimager The intensity at time 0 was set as 1 The same results were obtained in more than three separate experiments (D) HepG2 cells were cultured with the low amino acid medium supplemented with a 10-fold excess of glutamine (10 · Gln, 40 mM) for 6 h and pro-cessed for chromatin immunoprecipitation analyses as described in Materials and methods After IP with anti-Sp1 IgG, real-time PCR analy-sis was performed with a primer set covering the Sp1-binding region or distal region in the human SREBP-1a promoter The same results were obtained in two separate experiments (E) HepG2 cells were transfected with either control (siLuc) or Sp1 siRNA oligonucleotides (siSp1), cultured with medium A for 96 h and re-fed with the medium containing a 10-fold excess of glutamine (10 · Gln, 40 mM) for 24 h before harvest Real-time PCR analysis was performed, and the relative mRNA levels were obtained after normalization to GAPDH mRNA The mRNA levels transfected with siLuc without any glutamine addition are represented as 1 All data are presented as means ± SD of three independent experiments performed in triplicate **P < 0.01.

increased by 2 h after glutamine treatment,

accompa-nied by a reduction in the amount of cytosolic Sp1

In order to determine whether glutamine increases the

binding of Sp1 to the SREBP-1a promoter region

containing the Sp1-binding elements, we performed a

chromatin immunoprecipitation assay As shown in

Fig 3D, glutamine increased the Sp1 binding to the

SREBP-1a promoter region but not the distal region

of the SREBP-1a gene, indicating a

glutamine-depen-dent recruitment of Sp1 to the SREBP-1a promoter

We next examined whether the activation of Sp1 by

glutamine treatment is involved in the induction of

SREBP-1a gene expression When endogenous Sp1

expression was reduced to  20% of normal with

gene-specific small interfering RNA (siRNA), the gene

expression of SREBP-1a was significantly decreased in

HepG2 cells (Fig 3E), indicating that the basal gene

expression of SREBP-1a is under the control of Sp1

Moreover, the elevation in the mRNA level of

SREBP-1a by glutamine was abolished by the

knock-down of Sp1 expression (Fig 3E) These results

sug-gest that glutamine treatment facilitates Sp1 function

in HepG2 cells via its increased nuclear localization,

thereby stimulating SREBP-1a transcription

Induction of SREBP gene expression is not the

initial trigger for the glutamine effects

If induction of the SREBP gene expression serves as

the initial trigger for the glutamine effects, the gene

expression of SREBPs should be induced prior to the

SREBP target gene To test this prediction,

time-course experiments in the presence of glutamine were

performed in HepG2 cells At 4 h after the addition of

glutamine, the gene expression of the SREBP targets,

such as HMG CoA synthase and the LDL receptor,

was slightly increased, but then the mRNA levels of

both SREBPs and their target genes became

signifi-cantly elevated at 8 and 12 h (Fig 4) Based on the

fact that the increase in the mRNA of these genes was

nearly simultaneously observed at 8 h or later, it is

unlikely that the increased SREBP gene expression

served as the initial step for the glutamine effects

Glutamine stimulates SREBP processing

In the above experiments, HepG2 cells were cultured

with a low amino acid medium in order to detect the

glutamine effects with a high sensitivity Later it

turned out that the glutamine effects were reproduced

in cells incubated with DMEM which contained 4 mm

glutamine Indeed, when HepG2 cells were cultured in

DMEM supplemented with a 10-fold excess of

gluta-mine (40 mm), the mRNA levels of SREBP family members were significantly increased, as had occurred

in the cells cultured in the low amino acid medium (Fig S1) To investigate the effect of glutamine on SREBP processing, HepG2 cells cultured with DMEM were re-fed with a glutamine-supplemented medium and incubated for the indicated time (Fig 5A) The whole cell extracts were subjected to immunoblotting using anti-SREBP-1 and anti-SREBP-2 antibodies SREBP-1 was detected by an antibody recognizing its N-terminus [SREBP-1(N)], and SREBP-2 was detected

by antibodies recognizing its N-terminus [SREBP-2(N)] and C-terminus [SREBP-2(C)] Two antibodies recognizing the N-terminus of SREBP-1 and SREBP-2 detect their precursor and mature (nuclear) forms In contrast, SREBP-2(C) detects the precursor and the cleaved form that remains in the Golgi after release of the N-terminal mature form As shown in Fig 5A, SREBP-1 and SREBP-2 processing was induced by glutamine, as judged by the increase in the mature and cleaved forms (‘mature’ or ‘cleaved’ in Fig 5A) The fact that the glutamine-induced SREBP processing started as early as 0.5 h after glutamine treatment indi-cates that the glutamine-induced post-translational activation of SREBPs occurs prior to the stimulation

of the gene expression of SREBPs, which required 8 h

or longer (Fig 4) Similar glutamine-stimulated SREBP processing was observed when HepG2 cells

0.0

0 1.0 2.0 3.0 4.0 6.0

HMG S LDLR

Time (h)

8.0

0.0 1.0 2.0 3.0 4.0 6.0 8.0

SREBP-1a SREBP-2

Fig 4 The gene expression of SREBPs is not induced prior to the SREBP target gene by glutamine HepG2 cells were cultured with the low amino acid medium for 4 h and then re-fed with the med-ium supplemented with a 10-fold excess of glutamine (40 mM) for

2, 4, 8 or 12 h, and then total RNA was isolated Real-time PCR analysis was performed, and the relative mRNA levels were obtained after normalization to GAPDH mRNA The mRNA levels at time 0 are represented as 1 All data are presented as means ± SD

of three independent experiments performed in triplicate.

were cultured with the low amino acid medium, which

contained 0.25 mm glutamine (Fig S2A,B) In order to

examine how high the glutamine concentration must

be to induce SREBP processing, HepG2 cells were

incubated with various concentrations of glutamine

(4, 10, 20, 30 or 40 mm) for 4 h Figure 5B

demon-strates that SREBP processing is upregulated by

gluta-mine in a dose-dependent manner in HepG2 cells

Glutamine accelerates the ER-to-Golgi transport

of the SREBP⁄ SCAP complex When SREBP precursors are processed to liberate N-terminal active forms, the SREBP⁄ SCAP complex must be translocated to the Golgi, where two proteases responsible for SREBP cleavage reside Therefore, the glutamine-induced SREBP processing is assumed to be caused by an acceleration of the ER-to-Golgi translo-cation of the SREBP⁄ SCAP To determine whether the SREBP translocation is promoted by glutamine,

we adopted two in vitro assays First, we performed the reporter assay devised by Sakai et al [14], which monitors SREBP processing by determining the secreted alkaline phosphatase activity HEK293 cells were cotransfected with the expression plasmids for SCAP and the C-terminal half of SREBP-2 that was fused to the secreted form of placental alkaline phosphatase (PLAP-BP2) In agreement with previous studies [14,15], the secreted PLAP activity was increased when the cells were cultured in cholesterol-depleted conditions only in the presence of SCAP (data not shown) As shown in Fig 6A, the PLAP secretion was remarkably induced in the presence of SCAP, and a 10-fold excess of glutamine significantly enhanced the secretion Since the PLAP cleavage is mediated by S1P in this assay, it is possible that glutamine stimulated the S1P activity Alternatively, glutamine might simply accelerate the ER-to-Golgi transport of the PLAP-BP2⁄ SCAP complex Second,

we used the SCAP-null CHO cells expressing GFP-SCAP established by Nohturfft et al [16] to determine whether glutamine influences the movement of SCAP When the cells were cultured in medium C containing 2.5 mm glutamine, GFP-SCAP was diffusively distrib-uted within the cells (Fig 6B), which partially over-lapped with the Golgi marker GM130 (Fig 6C, D, H) Treatment with a 10-fold excess of glutamine for 6 h brought about a remarkable degree of overlap between GFP-SCAP and GM130 (Fig 6E–H), indicating that GFP-SCAP had moved to the Golgi in response to glutamine treatment These data indicate that gluta-mine treatment promotes the ER-to-Golgi transport of the SREBP⁄ SCAP complex, thereby stimulating SREBP processing Next, we examined whether activating transcription factor 6 (ATF6) processing is stimulated by glutamine, because ATF6 is also pro-cessed by S1P and S2P after its translocation from ER

to the Golgi in response to ER stress Unexpectedly, glutamine suppressed the gene expression of ATF6 (Fig 7A) However, the gene expression of BiP, which

is known to be an ATF6 target gene, was not altered

by glutamine treatment (Fig 7A) Therefore, we next

Mature IB: SREBP-1 (N)

A

Precursor

Precursor

Cleaved

IB: actin

B

Time after Gln (h)

4 10 20 30 40 Gln (mM)

Mature Precursor

Mature IB: SREBP-1 (N)

Precursor

Precursor

Cleaved IB: actin

Mature Precursor 8

IB: SREBP-2 (N)

IB: SREBP-2 (C)

IB: SREBP-2 (N)

IB: SREBP-2 (C)

Fig 5 Glutamine stimulates the processing of both SREBP-1 and

SREBP-2 (A) HepG2 cells were cultured with medium A containing

4 mM glutamine for 48 h and re-fed with the medium containing a

10-fold excess of glutamine (40 mM) for the indicated period of

time before harvest The whole cell extracts were subjected to

SDS ⁄ PAGE and immunoblotting (IB) with anti-SREBP-1(N) (2A4),

SREBP-2(N) (Rs004), SREBP-2(C) (1C6) or b-actin

anti-bodies (B) HepG2 cells were cultured with medium A containing

4 mM glutamine for 40 h and then re-fed with the medium

contain-ing the indicated concentration of glutamine for 4 h The whole cell

extracts were subjected to SDS ⁄ PAGE and IB with the antibodies

as described in (A) The same results were obtained in more than

three separate experiments.

determined whether glutamine stimulates ATF6 processing using exogenously expressed flag-tagged ATF6a HepG2 cells were transfected with the expres-sion plasmid for Flag-ATF6a, and then treated with either tunicamycin, an inhibitor of protein N-glycosyl-ation, or glutamine While treatment with tunicamycin, which causes ER stress, increased the amount of the processed form of Flag-ATF6a (denoted as N in Fig 7B, lanes 1 and 2), glutamine treatment had no effect (Fig 7B, lanes 4 and 5) These results indicate that glutamine stimulates the ER-to-Golgi transport of the SREBP⁄ SCAP complex without any significant influence on the S1P and S2P protease activities

Cell swelling is not involved in the glutamine-stimulated SREBP processing The uptake of glutamine into HepG2 cells is mediated

by a sodium-dependent transporter [17] When a large amount of glutamine is taken up by cells, osmotic cell

10 μm

0

5

10

15

20

A

PLAP-BP2 cleavage (relative light units)

**

25

1 × Gln

10 × Gln

S1P SCAP

ER

Golgi

PLAP

PLAP-BP2

Secretion into medium

0

10

20

Gln (m M )

H

**

30

40

2.5 40

Fig 6 Glutamine stimulates the ER-to-Golgi transport of the

SREBP⁄ SCAP complex (A) HEK293 cells were transfected with

pCMV-PLAP-BP2(513-1141), pCMV-b-gal and either pCMV-SCAP or

its empty vector After transfection, the cells were cultured with

the medium containing either 4 mM (1 · Gln) or 40 mM (10 · Gln)

glutamine for 16 h Then, aliquots of the medium were removed

and assayed for the PLAP activity The data were normalized to the

cellular b-galactosidase activity as described in Materials and

meth-ods All data are presented as means ± SD of three independent

experiments performed in triplicate **P < 0.01 (B–G) Stably

trans-fected CHO ⁄ pGFP-SCAP cells were set up as described in

Materi-als and methods On day 1, the cells were cultured with medium

C, which contained 2.5 mM (B, C and D) or 40 mM (E, F and G)

glu-tamine, for 6 h Then, the cells were fixed and incubated with the

primary antibody against GM130, and subsequently incubated with

the Cy3-conjugated secondary antibody The cells were imaged for

GFP-SCAP (B and E) or GM130 (C and F) Panels D and G are

merged images of GFP-SCAP and the Golgi marker GM130 Scale

bar, 10 lm (H) Quantification of the percentage of Golgi-localized

GFP-SCAP in (B)–(G) The signals were quantified with an ImageJ.

All data are presented as means ± SD of three independent

experi-ments.

0

0.8 0.6 1.0 1.2

ATF6

A

1.4

IB: anti-Flag

Flag-ATF6 α (P)

IB: anti- β-actin

+ +

Tunicamycin – +

– – – –

– – – –

B

Flag-ATF6 α (P*)

Flag-ATF6 α (N)

**

0.4 0.2

0

0.8 0.6 1.0 1.2

BiP

1.4

0.4 0.2

1 × Gln

10 × Gln

Fig 7 Glutamine does not stimulate ATF6 processing (A) HepG2 cells were cultured with medium A containing 4 mM glutamine for

48 h and then re-fed with the medium containing either 4 mM (1 · Gln) or 40 mM (10 · Gln) glutamine for 24 h Real-time PCR analysis was performed, and the relative mRNA levels were obtained after normalization to GAPDH mRNA All data are pre-sented as means ± SD of three independent experiments per-formed in triplicate **P < 0.01 (B) HepG2 cells transfected with pCMV-Flag-ATF6 were treated with 3 lgÆmL)1tunicamycin for 4 h

or 40 mM glutamine for 7 h The whole cell extracts were sub-jected to SDS ⁄ PAGE and immunoblotting (IB) with Flag or anti-b-actin antibodies Flag-ATF6a(P*) denotes the non-glycosylated form of pATF6a(P).

swelling often occurs A previous report has

demon-strated that culture of CHO-7 cells with a hypotonic

medium caused osmotic cell swelling and the ER stress

that inhibits general protein synthesis, thereby

stimu-lating SREBP processing though a reduction in the

Insig-1 protein due to its rapid rate of turnover [18]

This raises the possibility that glutamine-mediated cell

swelling, if it occurs, could result in the stimulation of

SREBP processing though various ER stress responses

Since both alanine and proline also induce cell swelling

[19], we examined the effect of these amino acids on

SREBP processing While glutamine stimulated

SREBP processing, the same concentration of alanine

or proline did not influence SREBP processing

(Fig 8), implying that an osmotic change, if it did

occur in the cells, was not involved in the stimulation

of SREBP processing by glutamine In addition, the

effect of glutamate on SREBP processing was also

examined because glutamine is capable of being

con-verted to glutamate [20] As shown in Fig 8, treatment

with glutamate did not influence SREBP processing,

indicating that the SREBP processing induced by

glu-tamine is not mediated by glutamate function

Furthermore, it has been reported that the

acceler-ated SREBP processing driven by osmotic cell swelling

is not inhibited by treatment with sterols because of

the reduction of Insig-1 protein level by cell swelling

[18] In order to examine the effect of sterols on the

glutamine-induced SREBP processing, HepG2 cells

were treated with glutamine in the presence or absence

of sterols for 4 h, and then immunoblotting analyses

were performed The stimulation of SREBP processing

by glutamine was completely attenuated by sterols

(Fig 9) These results suggest that the stimulatory

effect of glutamine on SREBP processing is not

medi-ated by the reduction of Insig-1 protein level caused

by osmotic cell swelling in HepG2 cells

PI3K-Akt pathway is involved in SREBP-1

processing but not in SREBP-2 processing

Next, we examined how glutamine stimulates SREBP

processing in HepG2 cells One possible mechanism

could be that glutamine modulates certain protein

kinase signaling pathways, which in turn activates

SREBP processing It has been reported that glutamine

stimulates Akt phosphorylation in HepG2 cells [21]

and both phosphatidylinositol 3-kinase (PI3K) and

p70 ribosomal protein S6 kinase (p70S6K) in rat

primary hepatocytes [22] When HepG2 cells were

trea-ted with 40 mm glutamine for 1 h, the levels of active

phosphorylated Akt, which is a substrate of PI3K,

were increased (Fig 10A) We were unable to detect

phosphorylated p70S6K despite the presence of gluta-mine (data not shown) It has been shown that SREBP processing is stimulated by the activation of the PI3K-Akt [15,23,24] and mTORC-p70S6K1 pathways [25] Therefore, we next treated cells with the PI3K inhibi-tor LY294002 and the mTORC1 inhibiinhibi-tor rapamycin

to assess whether these kinase pathways are involved

in glutamine-stimulated SREBP processing Interest-ingly, LY294002 inhibited glutamine-stimulated SREBP-1 processing, whereas it did not affect the stimu-lation of SREBP-2 processing by glutamine (Fig 10B)

In contrast, rapamycin did not influence glutamine-induced SREBP processing (Fig 10B) Taken together, these results suggest that the PI3K signaling pathway plays a key role in SREBP-1 processing in the presence

of glutamine but not in glutamine-stimulated SREBP-2 processing Recent reports indicate that insulin enhances SREBP-1 processing because of the phosphorylation of precursor SREBP-1 that is induced by Akt, which in turn enhances the affinity of the SREBP-1⁄ SCAP complex for the Sec23⁄ 24 proteins of the COPII vesicles and its transport to the Golgi apparatus [26] Thus, it is conceivable that the glutamine-stimulated SREBP-1 processing is mediated by the direct phosphorylation of the precursor SREBP-1 induced by Akt Further studies are required to determine how the glutamine-activated PI3K pathway is involved in SREBP-1 processing

In conclusion, the present study shows that treatment with glutamine causes the stimulation of

Gln – Ala Pro Glu Treated amino acid (20 mM)

Mature IB: anti-SREBP-1 (N)

Precursor

Precursor

Cleaved IB: anti-SREBP-2( C)

IB: anti- β-actin

Fig 8 Treatment with alanine, proline or glutamate does not stim-ulate SREBP processing HepG2 cells were cultured with medium

A containing 4 mM glutamine for 48 h and re-fed with the medium containing the indicated amino acid at a concentration of 20 mM for

4 h before harvest The whole cell extracts were subjected to SDS ⁄ PAGE and immunoblotting (IB) with SREBP-1(N), anti-SREBP-2(C) or anti-b-actin antibodies The same results were obtained in more than three separate experiments.

SREBP-1a gene expression in HepG2 cells due to

the activation of hexosamine biosynthesis pathway

Furthermore, the post-translational processing of

SREBPs is stimulated by treatment with glutamine

Since it is well known that the activation of SREBPs

stimulates the synthesis of lipids such as fatty acids

and cholesterol, the new concept presented here is that

glutamine acts as a stimulator of SREBP activities in

addition to serving as a nutrient and⁄ or signaling

mol-ecule These multiple and combined effects of

gluta-mine may fulfill the cellular demand for lipids that is

physiologically necessary for rapid cellular growth

Given that serum glutamine concentration is

approxi-mately 1 mm and rarely reaches 10 mm, glutamine has

a low potential for affecting SREBP processing in the

liver In contrast, because it is assumed that the

gluta-mine concentration in the cells of the small intestine

could transiently and sufficiently increase just after the

intake of glutamine-rich foods, amino acids may

strongly activate SREBPs in vivo, as was observed with

the cultured cells in this study Further studies are

required to determine whether glutamine-stimulated

SREBP activities contribute to the rapid growth of

cultured cells and whether dietary glutamine activates

SREBP in the liver and intestine in vivo

Materials and methods

Materials

serum, dialyzed fetal bovine serum, azaserine, tunicamycin, LY294002 and rapamycin were purchased from Sigma

medium were from Wako (Osaka, Japan)

Low amino acid medium

The low amino acid medium [8] was kindly provided by Ajinomoto (Kawasaki, Japan) It contains comparatively low concentrations of 20 amino acids compared with DMEM The amino acid concentrations are as follows:

5 lm Trp; 12.5 lm Cys, Met, His; 25 lm Gly, Ala, Ser, Asn, Glu, Asp, Phe, Typ, Arg, Pro; 50 lm Thr, Val, Leu, Ile, Lys; 250 lm Gln

Precursor

10 × Gln – + –

– – Sterols

+ + + IB: SREBP-1 (N)

Mature Precursor

Mature

IB: actin

Precursor

Cleaved

IB: SREBP-2 (N)

IB: SREBP-2 (C)

Fig 9 Stimulation of SREBP processing by glutamine is

attenu-ated by sterols HepG2 cells were cultured with medium A

contain-ing 4 mM glutamine for 48 h and re-fed with a 10-fold excess of

glutamine (10 · Gln, 40 mM) containing medium in the presence

or absence of sterols (10 lgÆmL)1of cholesterol plus 1 lgÆmL)1of

25-hydroxycholesterol) for 4 h before harvest The whole cell

extracts were subjected to SDS ⁄ PAGE and immunoblotting (IB)

with anti-SREBP-1(N), anti-SREBP-2(N), anti-SREBP-2(C) or

anti-b-actin antibodies The same results were obtained in more than

three separate experiments.

Time after Gln (h) IB: p-Akt IB: Akt

IB: SREBP-1 (N)

Mature Precursor

IB: actin

Precursor

Cleaved

Time after Gln (h) 0 4 0 4

IB: SREBP-2 (C)

B A

+

Rapamycin

Fig 10 Glutamine activates the PI3K-Akt pathway, and the inhibi-tor of this pathway differentially affects the processing of SREBP-1 and SREBP-2 (A) HepG2 cells were cultured with medium A con-taining 4 mM glutamine for 48 h and re-fed with a 10-fold excess of glutamine (40 mM) containing medium for the indicated period of time before harvest Whole cell extracts were subjected to SDS ⁄ PAGE and immunoblotting (IB) with anti-phospho-Akt (Ser473) and anti-Akt antibodies (B) HepG2 cells were cultured with med-ium A containing 4 mM glutamine for 48 h After pre-incubation with or without 10 lM LY294002 or 25 nM rapamycin for 30 min, the cells were re-fed with a 10-fold excess of glutamine (40 mM) containing medium in the presence or absence of either 10 lM LY294002 or 25 nM rapamycin for 4 h before harvest The whole cell extracts were subjected to SDS ⁄ PAGE and IB with anti-SREBP-1(N), anti-SREBP-2(C) or anti-b-actin antibodies The same results were obtained in triplicate experiments.

Cell culture

HEK293T and HepG2 cells were maintained in medium A

Caco-2 cells were maintained in medium B (DMEM

strepto-mycin, nonessential amino acids and 10% fetal bovine

Plasmid constructs

Reporter plasmids, pHMG S containing the human HMG

CoA synthase promoter, pHMG S SREKO containing a

mutated sequence of the SRE, pSQS containing the human

SQS promoter, pSQS SREKO containing two mutated

SREs, pLDLR containing the human LDL receptor

pro-moter, and pMTP containing the human MTP promoter

were described previously [27,28] A reporter plasmid,

pCPT1A, was constructed by inserting a 2.8-kb PCR

fragment coding the 5¢-promoter region and intron 1

(–286⁄ +2540) of the human CPT1A gene into a pGL4

basic vector (Promega, Madison, WI, USA) An expression

amino acid fusion protein consisting of an initiator

methio-nine followed by the secreted form of human PLAP (amino

acids 2–506), one novel amino acid (Y) generated by blunt

ligation, and the COOH-terminal half of human SREBP-2

(amino acids 513–1141) pCMV-PLAP-BP2(513-1141) was

constructed according to the method reported in a previous

paper [14] An expression plasmid for pCMV-SCAP was

constructed by inserting a 4.3-kb BglII-ClaI PCR fragment

coding full length human SCAP into the same restriction

sites of the pCMV vector (Stratagene, Santa Clara, CA,

USA) An expression plasmid for Flag-ATF6 was

con-structed by inserting a 1.7-kb BglII-XbaI PCR fragment

coding full length human ATF6 into the same restriction

Antibodies

Monoclonal anti-SREBP-1 (2A4), anti-SREBP-2 (1C6) and

polyclonal anti-Sp1 (PEP2) were obtained from Santa Cruz

Monoclonal anti-b-actin (AC-15) was from Sigma

Mono-clonal anti-GM130 (35) was from BD Biosciences (Franklin

Lakes, NJ, USA) Monoclonal anti-O-GlcNAC (RL2) was

from Thermo Scientific (Waltham, MA, USA) Polyclonal

anti-Sp1 used for chromatin immunoprecipitation assays

was from Millipore (Billerica, MA, USA) Polyclonal

anti-Akt and anti-phospho-anti-Akt were from Cell Signaling

Tech-nology (Beverly, MA, USA) Peroxidase-conjugated affinity

affinity purified goat anti-mouse IgG and Cy3-conjugated affinity purified donkey anti-mouse IgG were from Jackson Immunoresearch Laboratories (West Grove, PA, USA) The anti-SREBP-2 polyclonal serum (Rs004) has been described previously [11]

Luciferase assays

HepG2 cells and Caco-2 cells were plated in 12-well plates at

20 h, and then transfected with 200 ng of one of the reporter plasmids and 200 ng pEF-b-Gal, an expression plasmid for b-galactosidase, by the calcium phosphate method Twenty-four hours later, the medium was replaced with the low amino acid medium supplemented with 5% dialyzed fetal bovine serum and the indicated concentration of glutamine After incubation for another 12 h, the luciferase and b-galac-tosidase activities were measured as described previously [28] Normalized luciferase values were determined by dividing the luciferase activity by the b-galactosidase activity

Real-time PCR

Total RNA was extracted from HepG2 cells using an RNeasy mini kit (Qiagen, Valencia, CA, USA) according

to the manufacturer’s instructions cDNA was synthesized and amplified from 2 lg total RNA using a high capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA, USA) Quantitative real-time PCR (Taqman probe and SYBR green) analysis was performed on an

Expression was normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) control The TaqMan ID

Hs00231882_m1; LDL receptor, Hs00181192_m1; HMG-CoA synthase, Hs00266810_m1; CPT1A, Hs00157079_m1; Sp1, Hs00916521_m1; GAPDH, 4352934 The sequences of the primer sets used were as follows: SREBP-1a, 5¢-TCAG CGAGGCGGCTTTGGAGCAG-3¢ and 5¢-CATGTCTTC GATGTCGGTCAG-3¢ [9]; ATF6, 5¢-ATGTCTCCCCTTT CCTTATATGGT-3¢ and 5¢-AAGGCTTGGGCTGAATT GAA-3¢; BiP, 5¢-GACCTGGGGACCACCTACTC-3¢ and 5¢-TTCAGGAGTGAAGGCGACAT-3¢

Chromatin immunoprecipitation assays

Chromatin immunoprecipitation assays were performed as described previously [14] Real-time PCR was performed with the following primers: SREBP-1a promoter region

BP-1a promoter region reverse (5¢-GGTCTGCGCCACAA ATCTC-3¢), SREBP-1a distal region forward (5¢-AAAGTA CATAAAAGACAATGACCATCAC-3¢) and SREBP-1a distal region reverse (5¢-CTTGAGTTGTTTCTCTGCAGCTT-3¢)