Glucose 6-phosphate produced by gluconeogenesis and by glucokinase

is equally effective in activating hepatic glycogen synthase

Roger R. Gomis‡§¶+, Cristián Favre‡§¶+, Mar García-Rocha‡§, Josep Mª Fernández -

Novell‡, Juan C. Ferrer‡ and Joan J. Guinovart ‡§**

+ Both authors have equally contributed to this work

From the ‡Departament de Bioquímica i Biologia Molecular and the § Institut de

Recerca Biomèdica de Barcelona-Parc Científic de Barcelona, Universitat de

Barcelona, E-08028 Barcelona, Spain

** To whom correspondence should be addressed:

Joan J. Guinovart

Institut de Recerca Biomèdica de Barcelona-Parc Científic de Barcelona

c/Josep Samitier, 4-5, E-08028 Barcelona, Spain

Tel: +34-934037163; Fax: +34-934037114; e-mail: guinovart@pcb.ub.es

Running title: Glycogen synthesis from gluconeogenic precursors

1The abbreviations used are: Glc-6-P, glucose 6-phosphate; DHA, dihydroxyacetone;

GK, glucokinase; HK I, hexokinase I; G6Pase, glucose 6-phosphatase; Glc-1-P, glucose

1-phosphate; UDP-Glc, UDP-glucose; GS, glycogen synthase; RLGS, rat liver GS; GP,

glycogen phosphorylase; DMEM, Dulbecco’s modified Eagle’s medium;.

Copyright 2003 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on January 7, 2003 as Manuscript M212151200 by guest on M

arch 11, 2020http://w

ww

.jbc.org/D

ownloaded from

2

SUMMARY

Glucose 6-phosphate (Glc-6-P) produced in cultured hepatocytes by direct

phosphorylation of glucose or by gluconeogenesis from dihydroxyacetone (DHA)

was equally effective in activating glycogen synthase (GS). However, glycogen

accumulation was higher in hepatocytes incubated with glucose than in those

treated with DHA. This difference was attributed to decreased futile cycling

through GS and glycogen phosphorylase (GP) in the glucose-treated hepatocytes,

owing to the partial inactivation of GP induced by glucose. Our results indicate

that the gluconeogenic pathway and the glucokinase-mediated phosphorylation of

glucose deliver their common product to the same Glc-6-P pool, which is accessible

to liver GS. As observed in the treatment with glucose, incubation of cultured

hepatocytes with DHA caused the translocation of GS from a uniform cytoplasmic

distribution to the hepatocyte periphery and a similar pattern of glycogen

deposition. We hypothesize that Glc-6-P has a major role in glycogen metabolism

not only by determining the activation state of GS but also by controlling its

subcellular distribution in the hepatocyte.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

3

INTRODUCTION

Glucose 6-phosphate (Glc-6-P)1 is a key metabolite in hepatic carbohydrate

metabolism. It is produced from glucose and ATP, by the action of glucokinase (GK),

from three carbon-atom precursors through gluconeogenesis and also as a result of

glycogen breakdown. In turn, Glc-6-P is a substrate for glycolysis, for the pentose-

phosphate pathway and for the production of glucose, via glucose 6-phosphatase

(G6Pase)-catalyzed hydrolysis, or glycogen, through a pathway that involves the

successive conversion of Glc-6-P into glucose 1-phosphate (Glc-1-P) and UDP-glucose

(UDP-Glc), the substrate of glycogen synthase (GS).

Glycogen synthase (GS) and GK are the key enzymes in the control of hepatic

glycogen synthesis from glucose (1). The activity of the former is tightly regulated by

phosphorylation and by allosteric effectors (2), mainly Glc-6-P. This metabolite is an

allosteric activator of GS, and more importantly, it also promotes the permanent,

covalent activation of GS by dephosphorylation (3). Furthermore, upon incubation with

glucose, GS is redistributed in the hepatocyte, which may be an additional mechanism

of control (4). In the absence of glucose hepatic GS is uniformly distributed throughout

the cytoplasm, but it concentrates at the hepatocyte periphery when it is present (5,6).

This affects the pattern of hepatic glycogen deposition. In hepatocytes with low initial

glycogen content, the polysaccharide is synthesized first near the plasma membrane and

then, as the synthesis progresses, in more internal locations (6,7).

Glucose also affects the subcellular distribution of GK, both in vitro (8-10) and

in vivo (11,12). In the absence of substrate, GK concentrates in the nucleus, where it

remains bound to its regulatory protein and it translocates to the cytoplasm when the

levels of glucose or fructose increase.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4

Hepatic GS, but not muscle GS, differentiates between Glc-6-P produced by GK

or hexokinase I (HK I). Only Glc-6-P produced from glucose by GK can induce the

dephosphorylation and consequent activation of liver GS and trigger the synthesis of

glycogen (13,14). These results indicate there are at least two pools of Glc-6-P inside

the hepatocyte. The pool accessible to liver GS is replenished by the action of GK,

while the Glc-6-P produced by HK I is delivered to a cellular compartment from which

GS is excluded.

Here we examine, in cultured hepatocytes, biochemical and cellular aspects of

the synthesis of glycogen from dihydroxyacetone (DHA), an efficient gluconeogenic

substrate (15). We used recombinant adenovirus technology to compare the role of Glc-

6-P produced by gluconeogenesis with that arising from the direct phosphorylation of

glucose by GK, in the activation of hepatic GS and its translocation to the cell

periphery.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

5

EXPERIMENTAL PROCEDURES

Preparation of recombinant adenovirus – The AdCMV-RLGS adenovirus has

been described in a previous report (1). AdCMV-GK (16) and AdCMV-G6Pase (17)

were generous gifts from Dr. C.B. Newgard.

Hepatocyte isolation and treatment with recombinant adenovirus – Hepatocytes

were isolated by collagenase perfusion from male Wistar rats (180-225 g) fasted for 24

h, as described (18). Cells were suspended in Dulbecco’s modified Eagle’s medium

(DMEM), supplemented with 10 mM glucose, 10 % fetal bovine serum (Biological

Industries, Israel), 100nM insulin, and 100nM dexamethasone (Sigma, St. Louis, USA)

and then seeded onto plastic plates of 60 mm diameter treated with 0.1 % gelatin

(Sigma) at a final density of 8 x 104 cells/cm2. After cell attachment (4 h at 37 ºC), the

medium was replaced by DMEM without glucose, FBS and hormones, and hepatocytes

were treated for 2 h with AdCMV-RLGS, AdCMV-GK or AdCMV-G6Pase at a

multiplicity of infection (moi) of 10 for the first adenovirus and 4 for the other two. The

medium was again replaced by DMEM without glucose, FBS and hormones, and cells

were kept at 37 ºC for an additional 16-18 h, to allow for the expression of the transgen.

Finally, hepatocytes were incubated in the same medium supplemented with glucose or

DHA at the concentrations indicated in the figure legends. At the end of each

manipulation, cell monolayers were washed in phosphate-buffered saline and frozen in

liquid N2 until analysis, or fixed for immunofluorescence studies.

Metabolite determinations – To measure glycogen content, cell monolayers were

scraped into 30 % KOH, the extract was then boiled for 15 min and centrifuged at 5,000

x g for 15 min. Glycogen was measured in the cleared supernatants as described (19).

The intracellular concentration of Glc-6-P was measured using a spectrophotometric

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

6

assay (20). Glucose concentration in the incubation medium was measured as described

(21).

Enzyme activity assays – Glucose-phosphorylating activity and GS: Frozen cell

monolayers from 60 mm diameter plates were scraped using 100 µl of homogenization

buffer, which consisted of 10 mM Tris-HCl (pH 7.0), 150 mM KF, 15 mM EDTA, 15

mM 2-mercaptoethanol, 10 µg/ml leupeptin, 1 mM benzamidine, and 1 mM

phenylmethylsulfonyl fluoride. Thawing induced cell lysis. Protein concentration was

measured using a Bio-Rad assay reagent as described (22). Total GS activity was

measured in homogenates in the presence of 6.6 mM Glc-6-P, as described (23). Active

GS (I or a form) was measured in the absence of Glc-6-P. Glucose-phosphorylating

activity was measured spectrophotometrically in the supernatant fraction of hepatocyte

extracts centrifuged at 10,000 x g for 15 min, using 1 mM or 100 mM glucose at 30 ºC,

as described (24).

G6Pase activity: Frozen cells were scraped into a buffer solution containing 50

mM Tris-HCl (pH 7.4), 1mM EDTA, 100mM KCl, 600mM sucrose and 15mM 2-

mercaptoethanol. Homogenates were completely lysed by sonication. The activity of

G6Pase was assessed spectrophotometrically in the supernatant fraction after

centrifugation at 10000 x g for 15 min at 4ºC, as in (25).

Immunofluorescence analyses of GS and glycogen distribution – Hepatocytes

seeded onto gelatin-coated glass coverslips (10 mm x 10 mm) were washed with PBS

and fixed for 30 min in PBS containing 4 % paraformaldehyde. After fixation, cells

were incubated with NaBH4 (1mg/ml) to reduce auto-fluorescence, permeabilized with

0.2 % (v/v) Triton X-100 in PBS and blocked with 3 % bovine serum albumin (w/v) in

PBS. Next, cells were incubated for 1h at room temperature with the L1 anti-liver GS

(6) at dilution 1/500 (v/v), washed in PBS, and treated for 30 min with T-6391 Texas

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

7

Red®-X goat anti-rabbit Ig G (H+L) conjugate (Molecular Probes). Alternatively, cells

were treated for 1 h at room temperature with a monoclonal IgM antibody against

glycogen, a generous gift from Dr. O. Baba (26), diluted in PBS containing 3 % bovine

serum albumin (w/v), washed with PBS and subjected to incubation for 30 min with a goat

anti-mouse IgM secondary antibody conjugated to tetrametylrhodamine (Chemicon).

For actin staining, hepatocytes were treated for 30 min with fluorescein

conjugated phalloidin (Molecular Probes). Coverslips were finally air-dried and

mounted on glass microscope slides using Immuno Floure Mounting Medium (ICN

Biomedicals, Inc.). Fluorescence images were obtained with a Leica TCS 4D (Leica

Lasertechnik, Heidelberg, Germany) confocal scanning laser microscope adapted to an

inverted Leitz DMIRBE microscope and a 63 x (NA 1.4 oil) Leitz Plan-Apo objective.

The light source was an argon/krypton laser (75 mW) and optical sections (0.1 µm)

were obtained.

Electrophoresis and Immunoblotting – Active and total liver glycogen

phosphorylase (GP) protein levels were measured in hepatocyte homogenates by

Western blot. Electrotransfer of proteins from the gel to the nitrocellulose was

performed at 200 V (constant) and room temperature for 2 h, using a Bio-Rad miniature

transfer apparatus, as described (27). The nitrocellulose blots were incubated overnight

at 4ºC in blocking buffer (1 % BSA, 0.05 % Tween-20 in phosphate-buffered saline).

Blots were then incubated for 1 h with antibodies against phosphorylated (active) or

total GP. The antibody against active GP was raised in chicken using the peptide

9EKRRQI[pSer]IRGI19, which is located near the N-terminus of rat liver GP and

contains the Ser residue that is phosphorylated in the active enzyme. The antibody

against total GP was obtained by immunizing rabbits with the peptide

829EPSDLKISLSKESSNGVNANGK850, which constitutes the C-terminal end of the

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

8

protein. Membranes were then washed and incubated for 1 h with a secondary anti-

chicken/rabbit antibody conjugated to horseradish peroxidase. Immunoreactive bands

were visualized using an ECL kit (Amersham), following the manufacturer’s

instructions.

Statistical analysis – Data are expressed as the mean ± S.E.M. Statistical

significance was determined by unpaired two-tailed Student’s t test. Statistical

significance was assumed at p < 0.05.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

9

RESULTS

Effect of DHA on glycogen metabolism – We evaluated glycogen deposition in

hepatocytes that were synthesizing glucose from DHA by gluconeogenesis. Hepatocytes

were maintained for 16 h in a medium devoid of glucose and DHA and then incubated

for 2 h with increasing concentrations of DHA. These cells showed a dose-dependent

increase in glycogen content (Fig. 1A), intracellular Glc-6-P concentration (Fig. 1B) and

active GS levels (Fig. 1C), whereas total GS activity did not change in any of the

conditions assayed (data not shown). These effects were maximal at 10 mM DHA (Fig.

1) and, therefore, this concentration of the gluconeogenic precursor was chosen to

perform subsequent experiments.

Effect of DHA on the subcellular distribution of GS and glycogen deposition –To

examine the effect of DHA on the intracellular distribution of hepatic GS, cultured

hepatocytes were incubated for 2h in the presence of 10 mM DHA, 10 mM glucose and,

25 mM glucose, or left untreated. Immunofluorescence analysis, using an antibody that

specifically recognizes liver GS (6), showed that incubation with DHA produced an

accumulation of the enzyme at the periphery of the hepatocytes, similar to that observed

after glucose treatment. In control cells, GS was distributed throughout the cytoplasm

and after incubation with glucose or DHA, the enzyme formed patches near the

hepatocyte plasma membrane (Fig 2A).

Immunodetection of glycogen particles with a monoclonal anti-glycogen

antibody (26) showed that DHA also induced the deposition of the polysaccharide at the

cell periphery, as observed with the hepatocytes incubated with glucose (Fig. 2B).

Effect of GS overexpression on hepatocytes incubated with DHA or glucose – In

control uninfected hepatocytes, a 2 h-treatment with 10 mM DHA induced the

deposition of 16.1 ± 0.6 µg of glycogen/106 cells. This value was 40 % lower than that

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

10

measured for hepatocytes incubated with 25 mM glucose (28.1 ± 0.6 µg of glycogen/106

cells) (Fig. 3A), even though the levels of Glc-6-P and active GS were somewhat higher

in the former condition (Fig. 3B,C).

When total GS was overexpressed 4-fold by treatment of cultured hepatocytes

with the AdCMV-RLGS adenovirus (Table I), the ability of these cells to accumulate

glycogen increased substantially. However, the differences in glycogen content with the

respective uninfected controls were higher in the glucose-treated hepatocytes than in

those incubated with 10 mM DHA. Thus, GS-overexpressing hepatocytes treated with 5

or 25 mM glucose accumulated approximately 160 % more glycogen than their

corresponding controls and reached a level of 73 ± 10 µg/106 cells at 25 mM glucose

(Fig. 3A). In contrast, AdCMV-RLGS-infected hepatocytes incubated with 10 mM

DHA deposited 28 ± 3 µg glycogen/106 cells, which is only 70 % higher than that

observed for their uninfected counterparts, and represents 40 % of the glycogen

accumulated by GS-overexpressing hepatocytes treated with 25 mM glucose. This large

difference was observed in spite of the fact that intracellular Glc-6-P concentration (Fig.

3B) and active GS levels (Fig. 3C) were similar in AdCMV-RLGS-infected hepatocytes

incubated with 25 mM glucose or with 10 mM DHA.

On the other hand, the increase in glycogen content observed in GS-

overexpressing hepatocytes incubated with DHA was accompanied by a 13 % decrease

in the glucose output, which changed from 215 ± 5 to 188 ± 2 µg glucose/106 cells in

uninfected and GS-overexpressing cells, respectively.

Effect of GK overexpression on hepatocytes incubated with DHA or glucose – A

3-fold overexpression of GK (Table I) caused an increase of about 120% in the

intracellular Glc-6-P concentration of cultured hepatocytes treated with 5 (0.45 ± 0.10

nmol Glc-6-P /106 cells) or 25 mM glucose (1.54 ± 0.16 nmol Glc-6-P /106 cells), when

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

11

compared with their uninfected controls (0.21 ± 0.05 and 0.66 ± 0.07 nmol Glc-6-P /106

cells, respectively) (Fig. 3B). As a result, the levels of active GS (1.74 ± 0.07 and 2.10 ±

0.16 mU/106 cells, respectively) (Fig. 3C) and glycogen content (31.7 ± 3.4 and 67.2 ±

6.9 µg/106 cells, respectively) (Fig. 3A) of AdCMV-GK infected hepatocytes treated

with 5 or 25 mM glucose also rose significantly and in a glucose-concentration

dependent manner, as described previously (1).

In contrast, GK overexpression did not affect the Glc-6-P concentration of

hepatocytes treated with 10 mM DHA (0.80 ± 0.05 nmol Glc-6-P /106 cells in

uninfected hepatocytes vs. 0.86 ± 0.14 nmol Glc-6-P /106 cells in AdCMV-GK-treated

hepatocytes) (Fig. 3B). Accordingly, the levels of active GS (1.47 ± 0.06 and 1.43 ±

0.16 mU/106 cells, respectively) (Fig. 3C) and glycogen content (16.0 ± 0.9 and 19.3 ±

1.5 µg/106 cells, respectively) (Fig. 3A) remained unchanged compared to their

respective controls.

Effect of G6Pase overexpression on hepatocytes incubated with DHA or glucose

– The ability of hepatocytes to accumulate glycogen was drastically reduced when

G6Pase was overexpressed 3-fold in these cells (Table I). Cultured hepatocytes treated

with AdCMV-G6Pase and incubated with 10 mM DHA, 5 mM or 25 mM glucose

showed a decrease of 75, 35, and 50 % in glycogen content, respectively, compared to

their uninfected controls (Fig. 3A). The larger drop in glycogen accumulation in

hepatocytes incubated with 10 mM DHA was consistent with the larger effect of the

overexpression of G6Pase on Glc-6-P concentration and active GS levels in these cells,

which were respectively 25 % and 45 % of those determined for the uninfected controls

(Fig. 3B,C). The concentration of Glc-6-P and active GS attained in G6Pase

overexpressing hepatocytes incubated with 25 mM glucose were respectively 50 % and

75 % of the corresponding values in uninfected hepatocytes (Fig. 3B,C).

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

12

Conversely, the glucose output in G6Pase overexpressing hepatocytes incubated

with 10 mM DHA (253 ± 3 µg glucose/106) was almost 20 % higher than in the

uninfected cells (215 ± 5 µg glucose/106 cells)

Correlation of glycogen content and endogenous active GS with Glc-6-P

concentration in hepatocytes incubated with DHA or glucose – We analyzed the relative

effectiveness of Glc-6-P, produced by direct phosphorylation of glucose or by

gluconeogenesis from DHA, in activating GS and promoting glycogen deposition in

cultured hepatocytes. The values of glycogen content (Fig. 4A) and active GS (Fig. 4B),

which were determined in previous experiments, were plotted against the corresponding

Glc-6-P concentration values. The points derived from hepatocytes infected with the

AdCMV-RLGS adenovirus were not included in the plots, since in these experiments

the total amount of GS was artificially augmented and, therefore, they are not

comparable with the rest of the series, as we previously pointed out (1).

There was a strong correlation between the levels of active GS and Glc-6-P

concentration, irrespective of the origin of this metabolite. This linear correlation held

throughout the range of Glc-6-P concentrations, from the lowest values, attained by

means of G6Pase overexpression, to the highest, obtained in GK overexpressing

hepatocytes (Fig. 4B).

However, when glycogen content was plotted against Glc-6-P concentration, two

sets of points were clearly distinguishable (Fig. 4A). The points corresponding to

hepatocytes incubated with glucose or with DHA fell in two separate straight lines with

distinct slopes. Thus, the sensitivity of glycogen deposition to intracellular Glc-6-P

concentration, measured as the slope of the plot, is about two-fold higher in cultured

hepatocytes incubated with glucose than in those treated with DHA (Fig. 4A).

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

13

Activation state of glycogen phosphorylase in hepatocytes incubated with DHA

or glucose – To determine whether incubation with glucose or DHA affected the levels

of total or active glycogen phosphorylase (GP), we performed Western blot analysis of

hepatocyte homogenates, using two antibodies directed respectively against two

unrelated peptides contained in the sequence of rat liver GP. The first antibody, which

was raised against a peptide that consists of the 22 C-terminal amino acids of the

protein, was used to provide a measure of total GP. In Western blots the immune, but

not the pre-immune serum, recognized a band of the expected molecular mass in rat

liver extracts, but not in muscle homogenates (data not shown). Muscle expresses a

different GP isoform, which does not contain in its sequence the peptide used for the

immunizations of the rabbits. The second antibody was raised in chicken against an 11

amino acid peptide, which includes the phosphorylated Ser residue at position 15 of the

rat liver GP, and it also recognized a band of the expected molecular mass in liver

extracts. This band was more intense in Western blots of homogenates from cultured

hepatocytes treated with 100 µM forskolin, a drug that is known to trigger the

phosphorylation and, thus the activation, of GP through an increase in the intracellular

levels of cAMP (Fig. 5A). The binding of this antibody to active GP was blocked by the

presence of the phosphorylated peptide used for the immunizations, but not by the

analogous peptide in which the critical Ser residue (Ser15) is not phosphorylated (Fig.

5A). These observations indicate that the phospho-specific antibody only recognizes the

active phosphorylated form of liver GP.

Western blot analysis of cultured hepatocytes showed that the levels of total GP

remained essentially unchanged in the incubations with 10 mM DHA, 5 mM or 25 mM

glucose, compared to control cells (Fig. 5B). However, while the levels of active GP in

control cells and those incubated for 2 h with 10 mM DHA did not significantly change,

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

14

incubation with 5 or 25 mM glucose led to a marked decrease in the amount of active

GP. The final concentration of glucose in the medium of the hepatocytes incubated with

the gluconeogenic precursor was 0.7 ± 0.1 mM.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

15

DISCUSSION

In previous reports we have shown that Glc-6-P produced by GK in cultured

hepatocytes (13,28) and in FTO2B cells (14) is more effective in promoting the covalent

activation of hepatic GS and the deposition glycogen than Glc-6-P derived from the

catalytic action of HK I. This result suggested that Glc-6-P is compartmentalized in at

least two pools in these cells. Liver GS is excluded from the compartment where the

Glc-6-P produced by HK I is directed, while it has access to the Glc-6-P pool

replenished by the GK-mediated phosphorylation of glucose. This second pool is also

accessible to other enzymes and provides substrate for several metabolic routes. Thus,

overexpression of GK enhances not only glycogen deposition, but also glycolysis, both

in cultured hepatocytes (28) and in FTO2B cells (29). Glc-6-P from this pool can also

be directed to hydrolysis by G6Pase, since overexpression of the catalytic subunit of this

system reduces glycogen deposition and lactate production, while it increases hydrolysis

of Glc-6-P (21).

The main conclusion of this study is that the gluconeogenic pathway and the

direct phosphorylation of glucose by GK deliver their common product to the same

"general" Glc-6-P pool, which feeds the above mentioned metabolic processes. The

activation state of GS strongly correlates with the intracellular levels of Glc-6-P in

hepatocytes incubated with glucose (3,30,31). As shown in the present study, this

positive correlation also holds when cultured hepatocytes are incubated with

gluconeogenic precursors, such that Glc-6-P arising from gluconeogenesis is as

effective in activating hepatic GS as Glc-6-P produced by GK (Fig. 4B). The conversion

of Glc-6-P into glucose by G6Pase and later re-phosphorylation to Glc-6-P prior to

triggering the covalent activation of GS can be ruled out since, in contrast to what

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

16

occurs in hepatocytes incubated with glucose, GK overexpression has negligible effects

on GS activation and glycogen deposition in hepatocytes incubated with DHA (Fig. 3).

However, although liver GS activation state is equally sensitive to Glc-6-P

produced by GK or by gluconeogenesis, glycogen accumulation in cultured hepatocytes

is more efficient when Glc-6-P originates from direct phosphorylation of glucose by GK

(Fig. 4A). Similar levels of Glc-6-P attained by treating hepatocytes with glucose or

with DHA have distinct quantitative effects on glycogen deposition. An explanation for

this could reside in the observation that while incubation with DHA has the main effect

of increasing intracellular Glc-6-P concentration, treatment with glucose simultaneously

triggers the partial dephosphorylation and inactivation of GP (Fig. 5) (2,32), thus

causing the arrest of glycogenolysis and allowing higher rates of glycogen

accumulation. This is possible because GLUT-2, the main glucose transporter in

hepatocytes, essentially maintains intra- and extra-cellular glucose concentration in

equilibrium. Hepatocytes can synthesize glucose from gluconeogenic precursors, but

after 2 h incubation with 10 mM DHA, the medium in which the cultured cells were

kept only contained 0.7 mM glucose. This concentration is too low to significantly

inactivate GP.

Through the use of GP inhibitors which cause its inactivation by

dephosphorylation, it has been shown (15,33,34) that active GP plays a role in

controlling the phosphorylation state of GS. The covalent inactivation of GP relieves

glycogen synthase phosphatase from the allosteric inhibition caused by active GP, thus

leading to the dephosphorylation and activation of GS (2). However, in our

experimental conditions, the levels of active GS strongly correlate with the intracellular

Glc-6-P concentration, regardless of the GP activation state, indicating that GP

inactivation is not a prerequisite for the covalent activation of GS. An identical

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

17

conclusion was obtained from two previous studies in which the treatment with fructose

(35) or lithium chloride (36) of hepatocytes isolated from fasted rats led to the

simultaneous activation of GS and GP. These two apparently contradictory conclusions

may be reconciled by assuming that the levels of active GP do have an effect in

determining the GS activation state, as long as the intracellular concentration of Glc-6-P

remains constant. However, an increase in the Glc-6-P concentration would override the

allosteric inhibition of glycogen synthase phosphatase by active GP, thus causing a

corresponding increase in the levels of active GS in cultured hepatocytes.

Our results also suggest that the lower efficiency shown by Glc-6-P produced by

gluconeogenesis in stimulating glycogen deposition could be attributed to a futile cycle

through GS and GP, since both enzymes co-exist as their respective active forms in

hepatocytes incubated with DHA. Glycogen cycling has been shown to occur in vivo

and in vitro by several techniques that include the utilization of stable or radioactive

isotopes (37-40) or selective GP inhibitors (41). Meijer and coworkers (15,40) have

analyzed how the degree of glucose cycling, through GK and G6Pase, and glycogen

cycling affect glycogen accumulation in isolated hepatocytes. By means of acute

inhibition of the hepatic G6Pase system with S4048, a chlorogenic acid derivative that

inhibits translocation of Glc-6-P across the endoplasmic-reticulum membrane, these

authors have altered the partitioning of Glc-6-P produced by gluconeogenesis into

glucose production and storage as glycogen. This partitioning can also be modulated by

the overexpression of the enzymes involved in this metabolism. Thus, GS

overexpression in cultured hepatocytes re-directs part of the Glc-6-P produced by

gluconeogenesis from DHA to glycogen deposition, with the consequent reduction of

glucose output. Conversely, the overexpression of the catalytic subunit of G6Pase has

opposite effects.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

18

Another point that we have addressed in this paper is the study of the subcellular

distribution of hepatic GS. We have previously shown that incubation of isolated (5) or

cultured hepatocytes (6) with glucose causes the translocation of GS from the cytoplasm

to the cell periphery, where initial glycogen synthesis takes place. When added to

cultured hepatocytes, DHA has the same effect as glucose on the localization of GS and

the pattern of glycogen deposition. We propose that Glc-6-P is the key metabolite that

controls the subcellular distribution of hepatic GS, such that an increase in the

intracellular concentration of Glc-6-P triggers the accumulation of GS at the hepatocyte

periphery. This is based in the following reasoning.

First, the mere activation of GS induced by treatment with lithium chloride, a

known inhibitor of glycogen synthase kinase-3 (42, 43), is not sufficient to cause GS

translocation to the hepatocyte periphery (6). Second, the common metabolites in the

biosynthetic pathway of glycogen from glucose and from gluconeogenic precursors are

Glc-6-P, which is in equilibrium with Glc-1-P, and UDP-Glc, the substrate of GS and

immediate precursor in glycogen synthesis . Of these three intermediates, UDP-Glc can

be ruled out as the signal molecule, since its intracellular concentration appears to be

buffered and remains unchanged upon wide variations of Glc-6-P. Thus, GK (1) and

G6Pase (21) overexpression in cultured hepatocytes led, respectively, to a large increase

or decrease in the Glc-6-P levels, which in turn translated into increased or decreased

glycogen accumulation. However, the intracellular concentration of UDP-Glc did not

significantly change in any of these situations. Third, Glc-6-P is involved in the control

of the aggregation state of hepatic GS and its translocation from a soluble form to a

form that sediments at low centrifugal forces, both in isolated hepatocytes (44) and in

vivo (31).

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

19

Therefore Glc-6-P, not only constitutes the cross-road of several metabolic

pathways, but it also has a key role as a signal molecule in the regulation of hepatic

glycogen metabolism. It determines the activation state of GS and controls the

aggregation state and the subcellular distribution of GS in hepatocytes.

ACKNOWLEDGMENTS – We thank Dr. C. B. Newgard for his generous gift of the

AdCMV-GK and AdCMV-G6Pase adenoviruses and Dr. Otto Baba for the monoclonal

anti-glycogen antibody. We also thank Mrs. Anna Adrover for her excellent technical

assistance and Mr. Robin Rycroft for assistance in preparing the English manuscript.

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

20

REFERENCES

1. Gomis, R. R., Ferrer, J. C., and Guinovart, J. J. (2000) Biochem. J. 351, 811-816

2. Bollen, M., Keppens, S., and Stalmans, W. (1998) Biochem. J. 336, 19-31

3. Villar-Palasí, C., and Guinovart, J. J. (1997) FASEB J. 11, 544-558

4. Ferrer, J. C., Baqué, S., Fernández-Novell, J. M., García-Rocha, M., and

Guinovart, J.J. (1999) Technological and Medical Implications of Metabolic

Control Analysis (Cornish-Bouden, A., and Cárdenas, M. L., eds) pp. 207-214,

Kluwer Academic Publisher, The Netherlands

5. Fernández-Novell, J. M., Bellido, D., Vilaró, S., and Guinovart, J. J. (1997)

Biochem. J. 321, 227-231

6. García-Rocha, M., Roca A, de La Iglesia, N., Baba, O., Fernández-Novell, J.

M., Ferrer, J. C., and Guinovart, J. J. (2001) Biochem J. 357, 17-24

7. Fernández-Novell, J. M., López-Iglesias, C., Ferrer, J. C., and Guinovart, J. J.

(2002) FEBS Lett. 531, 222-228

8. Agius, L., Peak, M., Newgard, C. B., Gómez-Foix, A. M., and Guinovart, J. J.

(1996) J. Biol. Chem. 271, 30479-30486

9. Brown, K. S., Kalinowski, S. S., Megill, J. R., Durham, S. K., and Mookhtiar,

K. A. (1997) Diabetes 46, 179-186

10. de la Iglesia, N., Veiga-da-Cunha, M., Van Schaftinguen, E., Guinovart, J. J.,

and Ferrer, J. C. (1999) FEBS Lett., 456, 332-338

11. Toyoda, Y., Miwa, I., Kamiya, M., Ogiso, S., Nonogaki, T., Aoki, S., and

Okuda, J. (1995) Histochem. Cell. Biol. 103, 31-38

12. Fernández-Novell, J. M., Castel, S., Bellido, D., Ferrer, J. C., Vilaró, S., and

Guinovart, J. J. (1999) FEBS Lett., 459, 211-214

13. Seoane, J., Gómez-Foix, A. M., O’Doherty, R. M., Gómez-Ara, C., Newgard,

C. B. and Guinovart, J. J. (1996) J. Biol. Chem. 271, 23756-23760

14. Gomis, R. R., Cid, E., García-Rocha, M., Ferrer, J. C., and Guinovart, J. J.

(2002) J. Biol. Chem. 277, 23246-23252

15. Gustafson, L. A., Neeft, M., Reijngoud, D. J., Kuipers, F., Sauerwein, H. P.,

Romijn, J. A., Herling, A. W., Burger, H. J., and Meijer, A. J. (2001) Biochem.

J. 358, 665-671

16. Scott, D.K., O'Doherty, R. M., Stafford, J. M., Newgard, C.B., and Granner, D. K.

(1998) J. Biol. Chem. 273, 24145-24151

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

21

17. Trinh, K., Minassian, C., Lange, A. J., O'Doherty, R. M., and Newgard, C. B.

(1997) J. Biol. Chem. 272, 24837-24842

18. Massagué, J., and Guinovart, J. J. (1977) FEBS Lett. 82, 317-320

19. Chan, T. M., and Exton, J. H. (1976) Anal. Biochem. 71, 96-105

20. Lang, G., and Michal, G. (1974) Methods of Enzymatic Analysis (Bergmeyer, H.U.,

ed) Vol. 3, pp. 1238-1242, Academic Press, New York

21. Seoane, J., Trinh, K., O'Doherty, R. M., Gomez-Foix, A. M., Lange, A. J.,

Newgard, C. B., and Guinovart, J. J. (1997) J. Biol. Chem. 272, 26972-26977

22. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254

23. Thomas, J. A., Schlender, K. K., and Larner, J. A (1968) Anal. Biochem. 25,

486-499

24. Kuwajima, M., Newgard, C. B., Foster, D. W., and McGarry, J. D. (1986) J.

Biol. Chem. 261, 8849-8853

25. Alegre, M., Ciudad, C. J., Fillat, C., and Guinovart,J. J. (1988) Anal. Biochem.

173, 185-189

26. Baba, O. (1993) Kokubyo Gakkai Zasshi 60, 264-287

27. Tobwin, H., Staehelin, T., and Gordon, J. (1992) Biotechnology 24, 145-149

28. O'Doherty, R. M., Lehman, D. L., Seoane, J., Gómez-Foix, A. M., Guinovart. J.

J., Newgard, C.B. (1996) J. Biol. Chem. 271, 20524-20530

29. Valera, A., and Bosch, F. (1994) Eur. J. Biochem. 222, 533-539

30. Ciudad, C. J., Carabaza, A., and Guinovart, J. J. (1986) Biochem. Biophys. Res.

Commun. 141, 1195-1200

31. Fernández-Novell, J. M., Roca, A., Bellido, D., Vilaró, S., and Guinovart, J. J.

(1996) Eur. J. Biochem. 238, 570-575

32. Carabaza, A., Ciudad, C. J., Baqué, S., and Guinovart, J. J. (1992) FEBS Lett.

296, 211-214

33. Bergans, N., Stalmans, W., Goldman, S., Vanstapel, F. (2000) Diabetes 49,

1419-1426

34. Aiston, S., Hampson, L., Gomez-Foix, A. M., Guinovart, J. J., and Agius, L.

(2001) J. Biol. Chem. 276, 23858-66

35. Ciudad, C. J., Massagué, J., and Guinovart J. J. (1979) FEBS Lett. 99, 321-324

36. Bosch, F., Gómez-Foix, A. M., Ariño, J., and Guinovart, J. J. (1986) J. Biol.

Chem. 261, 16927-16931

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

22

37. Shulman, G. I., Rothman, D. L., Chung, Y., Rossetti, L., Petit, W. A., Barrett, E.

J., and Shulman, R. G. (1988) J. Biol. Chem. 263, 5027-5029

38. David, M., Petit, W. A., Laughlin, M. R., Shulman, R. G., King, J. E., and

Barrett, E. J. (1990) J. Clin. Invest. 86, 612-617

39. Gustafson, L. A., Jumelle-Laclau, M. N., Van Woerkom, G. M., Van

Kuilenburg, A. B. P. and Meijer, A. J. (1997) Biochim. Biophys. Acta 1318,

184-190

40. van Dijk, T. H., van der Sluijs, F. H., Wiegman, C. H., Baller, J. F. W.,

Gustafson, L. A., Burger, H. J., Herling, A. W. , Kuipers, F., Meijer, A. J., and

Reijngoud, D. J. (2001) J. Biol. Chem. 276, 25727-25735

41. Massillon, D., Bollen, M., De Wulf. H., Overloop, K., Vanstapel, F., Van

Hecke, P., and Stalmans, W. (1995) J. Biol. Chem. 270, 19351-19356

42. Klein, P. S., and Melton, D. A. (1996) Proc. Natl. Acad. Sci. USA 93, 8455-

8459

43. Orena, S.J., Torchia, A. J., and Garofalo, R. S. (2000) J. Biol. Chem. 275,

15765-15772

44. Fernández-Novell, J. M., Ariño, J., Vilaró, S., Guinovart, J. J. (1992) Biochem.

J. 281, 443-448

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

23

FOOTNOTES

This study was supported by Grant PB98-0992 from the Dirección General de

Enseñanza Superior (Ministerio de Educación y Cultura, Spain), Grant 992310 from the

Fundació La Marató de TV3, and the Juvenile Diabetes Foundation International.

¶ R.G.: Recipient of a Doctoral Fellowship (Formación de Profesorado Universitario)

from the Spanish Government (Ministerio de Educación y Cultura). C.F.: Recipient of a

Postdoctoral Fellowship from the Argentinean Government (Consejo Nacional de

Investigaciones Científicas y Técnicas).

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

24

FIGURE LEGENDS

FIG. 1. Glycogen content, Glc-6-P concentration and active GS levels in

hepatocytes incubated with DHA. After 16 h in DMEM, hepatocytes were incubated

for 2 h with 1, 5, 10 or 25 mM DHA. Cells were then collected and glycogen content

(A), intracellular Glc-6-P concentration (B) and active GS (C) levels were determined

as described under “Experimental Procedures”. Data represent the mean ± S.E.M. of

four independent experiments.

FIG. 2. Subcellular distribution of GS and pattern of glycogen deposition in

hepatocytes incubated with glucose or DHA. Confocal microscopy images of cultured

hepatocytes treated for 2 h with the indicated concentrations of glucose or DHA.

Control cells were incubated in DMEM without glucose and DHA. At the end of the

incubations, hepatocytes were processed for immunofluorescence analysis with the L1

anti-GS antibody (A) or with an anti-glycogen antibody (B) as indicated under

“Experimental Procedures”. In (B), actin was also stained, with FITC-conjugated

phalloidin, to delineate the contour of the cells. The scale bar indicates 10 µm.

FIG. 3. Effects of GS, GK and G6Pase overexpression on glycogen content, Glc-6-P

concentration and active GS levels in hepatocytes incubated with glucose or DHA.

Hepatocytes were treated with AdCMV-RLGS (hatched bars), AdCMV-GK (grey bars),

AdCMV-G6Pase (black bars), or left untreated (open bars). After 16 h in DMEM,

hepatocytes were incubated for 2 h with 10 mM DHA or with 5 mM or 25 mM glucose

and glycogen content (A), Glc-6-P concentration (B) and active GS (C) were measured

as described under “Experimental Procedures”. Data represent the mean ± S.E.M. of

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

25

four independent experiments. *p<0.05 and **p<0.01 vs. the respective uninfected

control.

FIG. 4. Correlation of glycogen content and endogenous active GS vs. Glc-6-P

concentration in hepatocytes incubated with glucose or DHA. The mean values of

glycogen content (A) or endogenous active GS (B) shown in FIG. 1 and FIG. 3 were

plotted against the corresponding mean values of Glc-6-P concentration. Open symbols

correspond to hepatocytes incubated glucose and closed symbols to hepatocytes

incubated with DHA. Cells treated with AdCMV-GK (triangles), AdCMV-G6Pase

(squares), or left untreated (circles). Regression coefficients are indicted in the plots.

The slopes of the two lines in (A) are 44 and 19 µg of glycogen/nmol Glc-6-P for

hepatocytes incubated with glucose and DHA, respectively.

FIG. 5. Western blot analysis of active and total GP in hepatocytes incubated with

glucose or DHA. (A) After 16 h in DMEM, hepatocytes were incubated for 2 h with 30

mM glucose or 100 µM forskolin, as indicated. The blot was performed in the presence

or in the absence of the peptide used for immunization (Peptide 1:

9EKRRQI[pSer]IRGI19) or the analogous non-phosphorylated peptide (Peptide 2:

9EKRRQISIRGI19), as indicated. M designates the lane of molecular mass markers. (B)

After 16 h in DMEM, hepatocytes were incubated for 2 h with 10 mM DHA or with 5

mM or 25 mM glucose, as indicated. Cell homogenates were subjected to Western blot

using an anti-rat liver GP antibody (total GP) or a chicken antibody that specifically

recognizes phosphorylated GP (p-Ser GP).

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

26

TABLE I. Glucose-phosphorylating activity, G6Pase and total GS activity in GS-, GK-,

and G6Pase-overexpressing hepatocytes.

Cultured rat hepatocytes were treated with AdCMV-GK or AdCMV-G6Pase at a

moi of 4, with AdCMV-RLGS at a moi of 10, or were left untreated. Cells were

incubated for 16 h with DMEM without glucose, treated for 2 h with 25mM glucose and

finally collected for measurement of GK and HK, G6Pase and total GS activities as

described under “Experimental Procedures”. Activities are expressed in mU/106 cells

and as a percentage of control values. Data represent the mean ± S.E.M. of four

independent experiments.

GK HK G6Pase GS

Untreated 9.3 ± 2.6 3.3 ± 0.8 9.8 ± 0.2 4.4 ± 0.4 AdCMV-GK 27.7 ± 4.9 - - - (296 ± 43 %) AdCMV-G6Pase - - 25.2 ± 1 - (257 ± 10 %) AdCMV-RLGS - - - 18.5 ± 2 (420 ± 45 %)

- indicates no significant differences vs. control values

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0

2

4

6

8

10

12

14

16

18

20

0 5 10 15 20 25 30

DHA (mM)

µµg G

lyco

gen

/ 1

06 C

ells

Figure 1A

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

0 5 10 15 20 25 30

DHA (mM)

nm

ols

Glc

6-P

/10

6 c

ells

Figure 1B

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0,50,60,70,80,91,01,11,21,31,41,51,61,71,81,9

0 5 10 15 20 25 30

DHA (mM)

Acti

ve G

lyco

gen

Syn

thase

mU

/10

6 c

ells

Figure 1C

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0

10

20

30

40

50

60

70

80

90

10mM DHA 5mM Glucose 25mM Glucose

µµg o

f G

lyc

og

en

/10

6 c

ell

s

*

**

**

**

**

**

**

Figure 3A

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

10mM DHA 5mM Glucose 25mM Glucose

nm

ol

Glc

6-P

/10

6 c

ell

s

*

*

*

*

Figure 3B

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

10mM DHA 5mM Glucose 25mM Glucose

Acti

ve G

lyco

gen

Syn

thase

mU

/10

6 C

ells

*

*

***

**

**

Figure 3C

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0

10

20

30

40

50

60

70

80

0,0 0,5 1,0 1,5 2,0

Glc 6-P

nmol/106 cells

µµg o

f g

lyco

gen

/10

6 c

ells

r=0,97

r=0,97

Figure 4A

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

0

1

2

3

4

5

0,0 0,5 1,0 1,5 2,0

Glc 6-P

nmol/106 cells

Acti

ve G

lyco

gen

Syn

thase

mU

/10

6 c

ells

r=0,88

Figure 4B

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

M M

Glucose - - + - - + - - +

Forskolin - + - - + - - + -

Peptide 1 + + + - - - - - -

Peptide 2 - - - - - - + + +

Figure 5A

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

P-Ser GP

Glc (mM) - - 5 25

DHA (mM) - 10 - -

GP

Figure 5B

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from

;Ferrer and Joan J. GuinovartRoger R. Gomis, Cristián Favre, Mar García-Rocha, Josep Mª Fernàndez-Novell, Juan C.

effective in activating hepatic glycogen synthaseGlucose 6-phosphate produced by gluconeogenesis and by glucokinase is equally

published online January 7, 2003J. Biol. Chem. 

  10.1074/jbc.M212151200Access the most updated version of this article at doi:

 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

by guest on March 11, 2020

http://ww

w.jbc.org/

Dow

nloaded from