Intestinal Gluconeogenesis Is a Key Factor
for Early Metabolic Changes after Gastric
Bypass but Not after Gastric Lap-Band in Mice
Stephanie Troy,1,11Maud Soty,2,3,4,11Lara Ribeiro,1,11Laure Laval,2,3,4Ste ´phanie Migrenne,5Xavier Fioramonti,5
Bruno Pillot,2,3,4Veronique Fauveau,6Roberte Aubert,1Benoit Viollet,7,8Marc Foretz,7,8Jocelyne Leclerc,7,8
Adeline Duchampt,2,3,4Carine Zitoun,2,3,4Bernard Thorens,9Christophe Magnan,5Gilles Mithieux,2,3,4,12,*
and Fabrizio Andreelli1,10,12,*
1Institut National de la Sante et de la Recherche Medicale, U695, Faculte ´ de Me ´decine Xavier Bichat, Universite Paris 7,
Paris, F-75870, France
2Institut National de la Sante et de la Recherche Medicale, U855, Lyon, F-69372, France
3Universite de Lyon, Lyon, F-69008, France
4Universite Lyon I, Villeurbanne, F-69622, France
5Centre National de la Recherche Scientifique, U7059, Universite ´ Paris 7, Paris, F-75251, France
6Institut Cochin, IFR Alfred JOST, Plate Forme de Microchirurgie, Faculte ´ de Me ´decine Cochin, F-75014, Paris, France
7Institut Cochin, Universite ´ Paris Descartes, CNRS (UMR 8104), Paris, France
8Inserm, U567, Paris, France
9Department of Physiology and Center for Integrative Genomics, University of Lausanne, Lausanne, Switzerland
10CHU Bichat Claude Bernard, Service de Diabe ´tologie-Endocrinologie-Nutrition, AP-HP, Paris, 75877, France
11These authors contributed equally to this work
12These authors contributed equally to this work
*Correspondence: email@example.com (G.M.), firstname.lastname@example.org (F.A.)
Unlike the adjustable gastric banding procedure
(AGB), Roux-en-Y gastric bypass surgery (RYGBP)
in humans has an intriguing effect: a rapid and sub-
stantial control of type 2 diabetes mellitus (T2DM).
We performed gastric lap-band (GLB) and entero-
gastro anastomosis (EGA) procedures in C57Bl6
mice that were fed a high-fat diet. The EGA proce-
dure specifically reduced food intake and increased
insulin sensitivity as measured by endogenous glu-
cose production. Intestinal gluconeogenesis in-
creased after the EGA procedure, but not after gas-
tric banding. All EGA effects were abolished in
GLUT-2 knockout mice and in mice with portal vein
denervation. We thus provide mechanistic evidence
that the beneficial effects of the EGA procedure on
food intake and glucose homeostasis involve intesti-
nal gluconeogenesis and its detection via a GLUT-2
and hepatoportal sensor pathway.
The exponential advance of the obesity epidemic has led to a re-
markable increase in surgical procedures for obesity (Stein-
brook, 2004). Two techniques are commonly used in treating
the morbidly obese. Laparoscopic adjustable gastric banding
(AGB) is an increasingly popular, purely restrictive bariatric pro-
cedure used extensively worldwide. This approach involves the
placement of a prosthetic band around the upper stomach to
partition it into a small, proximal pouch and a large, distal rem-
nant connected by a narrow constriction (Bo and Modalsli,
1983). Roux-en-Y gastric bypass (RYGBP) involves excising
approximately two-thirds of the stomach, and the small bowel
is divided 200 cm from the ileocaecal junction. The proximal
part of the small bowel is anastomosed to the stomach, and
the distal end is anastomosed to the ileum 100 cm from ileocae-
Reduction of the fat mass induced by bariatric surgery has
been generally accepted as the best explanation for the control
and, indeed, the reversal of the diabetes mellitus (Buchwald
et al., 2004; Sjostrom et al., 2004). However, RYGBP has an in-
triguing effect not observed following the AGB procedure: i.e.,
a rapid and substantial control of type 2 diabetes mellitus, often
within days (Pories, 2004). The dramatic speed at which type 2
diabetes mellitus resolves after RYGBP, but not after AGB, has
resistance may be independent of weight loss (Hickey et al.,
1998; Pories et al., 1995; Rubino and Marescaux, 2004; Scopi-
naro et al., 1998) and, indeed, are direct effects of the RYGBP
surgical procedure itself. How the RYGBP procedure can modify
insulin resistance and glucose tolerance so quickly remains
unclear. Recent reports suggest that changes in gut secretion
might, in part, explain the effects of RYGBP. However, caution
is required concerning this hypothesis, based on analysis of
the plasma levels of ghrelin and glucagon-like peptide-1 (GLP-
1), as indicated recently. Thus, in contrast to earlier reports, it
has been shown that active ghrelin levels were unaffected by
RYGBP procedure in obese patients and do not correlate with
daily caloric intake patterns after surgery (Korner et al., 2005).
Results in human studies about a possible beneficial effect of
weight loss on GLP-1 release are contradictory. Whereas GLP-
1 levels are decreased in overweight/obese subjects receiving
a very low-caloric diet (Adam et al., 2005), GLP-1 secretion has
Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc. 201
been reportedto beincreased(leRouxetal.,2007)ordecreased
after a RYGBP procedure (Reinehr et al., 2007).
It has become clear that the intestine is more than a digestive
tract. The small intestine can produce glucose and release it
into the portal blood in a process calledintestinal gluconeogene-
sis (Mithieux, 2005). The potential importance of this previously
unknown intestinal function has been pointed out by numerous
recent studies. Indeed, key enzymes of gluconeogenesis—glu-
cose-6-phosphatase (Glc6Pase) and phosphoenolpyruvate car-
boxykinase (PEPCK)—and their mRNAs are present in the small
intestine (SI) in both rat and human (Mithieux, 2005; Rajas et al.,
1999; Yanez et al., 2003). In addition, these genes are regulated
by nutrition in several species (Azzout-Marniche et al., 2007;
Cui et al., 2004; Kirchner et al., 2005). Intestinal gluconeogenesis
takes place notably when Glc6Pase and PEPCK genes are in-
duced, such as during fasting and high-protein feedings or
when diabetes occurs (Croset et al., 2001; Mithieux et al.,
utor to glucose production when the liver is deficient, as in mice
with invalidation of hepatic PEPCK (She et al., 2003) and in hu-
mansduringthe anhepatic phaseofliver transplantation(Battez-
zati et al., 2004). Lastly, portal sensing of intestinal gluconeogen-
esis induces hypophagia (Mithieux et al., 2005), and infusion of
glucose into the portal vein may modulate the whole-body glu-
cose disposal (Burcelin et al., 2000). This has led us to consider
the attractive hypothesis that the RYGBP procedure may cause
increased intestinal gluconeogenesis, which, in turn, may de-
crease food intake and restore glucose homeostasis.
To test this possibility, we performed gastric lap-band (GLB)
and entero-gastro anastomosis (EGA) procedures in C57Bl6
two most widely used bariatric surgical procedures.
Entero-Gastro Anastomosis and Inhibition
of Food Intake
In order to better understand the effects of bariatric surgery, we
performed GLB and EGA procedures in 6-month-old C57Bl6
male mice on a high-fat diet (Figures 1 and 2 and Supplemental
Figure 1. Effects of Bariatric Surgery in Mice
(A) Evolution of daily food intake by C57Bl6 high-fat diet mice before surgery (n = 15) and after bariatric surgery in pair-fed sham-operated (n = 15), gastric
lap-band (n = 15), and EGA mice (n = 15). Gastric lap-band mice died 11 days after surgery due to food accumulation above the lap-band and considerable
(B) Evolution of body weight in pair-fed sham-operated, gastric lap-band, and EGA mice (n = 15 per group).
(C) Body composition assessed by a biphotonic absorptiometry method in standard diet, high-fat diet, pair-fed sham-operated, gastric lap-band, and EGA mice
(n = 10 per group). Loss of fat mass was similar in pair-fed sham-operated, lap-band, and EGA groups 10 days postsurgery. *p < 0.05 for the difference between
high-fat diet and standard diet mice. NS: Nonsignificant when compared body weight or fat mass or lean mass in pair-fed sham-operated, lap-band, and EGA
groups. Data are expressed as means ± SEM.
Intestinal Gluconeogenesis and Gastric Bypass
202 Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc.
Data available online). After surgery, the high-fat diet was main-
mice, daily food intake recovered, 3 days after surgery, to the
level observed before surgery (3.5 g/day) (Figure 1A). The evolu-
tion of daily food intake after surgery differed between GLB and
EGA mice (Figure 1A). Food intake decreased sharply and simi-
larly in GLB and EGA mice (0.7 g/day) until 6 days after surgery.
Thereafter, food intake by GLB mice increased. Some GLB mice
died on day 11 due to food accumulation above the lap-band
and considerable esophagus dilatation. In contrast, the de-
creased food intake by EGA mice observed immediately after
surgery persisted at 0.7 g/day for almost 5 months (Figure 1A
and data not shown). To reduce the death rate in the GLB group,
amount equivalent to the daily food intake of EGA mice) through-
out the follow-up. During the follow-up after surgery and on this
diet, pair-fed sham-operated, EGA, and GLB groups similarly
lost body weight (Figure 1B) and fat mass (Figure 1C).
EGA Increases Insulin Sensitivity
Fasted and fed, high-fat diet (HFD) mice showed higher levels of
glucose, insulin, triglycerides, free fatty acids, TNF-a, resistin,
and leptin than standard diet mice (Table S1A). By 10 days after
surgery, pair-fed sham-operated, GLB, and EGA mice showed
similar levels of glycerol, free fatty acids, triglycerides, and b-hy-
droxy butyric acid, suggesting a similar induction of lipolysis and
ketogenesis by food restriction (Table S1B). Except for resistin
lin, glucagon, total adiponectin, TNF-a, and leptin levels were in
the same range in all groups after surgery (Table S1B).
The area under the curve (AUC) of glucose and insulin plasma
levels during an oral glucose tolerance test were significantly
higher in high-fat diet mice than standard diet mice (Figures
2A, 2B, and S3). By 10 days after surgery, AUC of glucose was
significantly lower in EGA mice than in pair-fed sham-operated
2A and S3). The improvement of glucose tolerance in EGA mice
Figure 2. Metabolic Effects of Bariatric Surgery in Mice
Evolution of glucose (A), insulin (B), and glucagon plasma levels (C) during an OGTT (3g/kg) in standard diet, high-fat diet, pair-fed sham-operated, gastric lap-
band, and EGA mice (n = 10 per group). EGA mice are characterized by higher levels of insulin during the OGTT when compared to the other groups, whereas
glucagon levels decreased similarly in all groups.
(D) Euglycaemic hyperinsulinaemic clamps were performed to measure whole-body insulin sensitivity as assessed by the glucose infusion rate (GIR), endoge-
nousglucoseproduction (EGP), and rateof disappearance of glucose (Rd) (n = 6per group). GIR and Rd were lower and EGP higherin high-fat diet than standard
diet mice. Rd and GIR were similar in pair-fed sham-operated, gastric lap-band, and EGA groups and were lower than in standard diet mice, indicating the per-
sistence of peripheral insulin resistance after bariatric surgery. In contrast, EGP was similar in pair-fed sham-operated and gastric lap-band mice and was sig-
nificantly lower only in EGA mice, suggesting an increase in hepatic insulin sensitivity in this group. *p < 0.05 for high-fat diet and standard diet mice. $p < 0.05 for
the difference between high-fat diet mice and pair-fed sham-operated, gastric lap-band, or EGA mice . xp < 0.05 for EGA and pair-fed sham-operated mice. #p <
0.05 for EGA and gastric lap-band mice. Data are expressed as means ± SEM.
Intestinal Gluconeogenesis and Gastric Bypass
Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc. 203
resulted from a substantial increase in insulin secretion (Figures
2A and 2B). Glucagon levels were similarly reduced during oral
glucose tolerance tests (OGTT) in all groups (Figure 2C).
Insulin sensitivity of peripheral glucose uptake and endoge-
nous glucose production (EGP) was assessed by the euglycae-
mic hyperinsulinaemic clamp method. As expected, in HFD
mice, the glucose infusion rate (GIR) and rate of disappearance
(Rd) were lower and endogenous glucose production (EGP)
was higher than in standard diet mice (Figure 2D). By 10 days af-
ter surgery, the rate of disappearance of glucose (Rd) was not
improved in pair-fed sham-operated, GLB, and EGA mice, sug-
gesting that food restriction by itself is unable to improve periph-
eral insulin sensitivity (Figure 2D). In contrast, EGA mice showed
a higher suppression of EGP by insulin than pair-fed sham-oper-
ated and GLB mice (Figure 2D). Indeed, the GIR of EGA mice in-
creased and reached the level observed in standard diet mice
(Figure 2D), and EGP was lower than that in GLB and pair-fed
sham mice. In summary, these findings indicate that EGA has
EGP in EGA mice correlated with the reduction in hepatic glu-
cose-6 phosphatase activity, suggesting a decrease in hepatic
gluconeogenesis (Figure S4). Lipotoxicity is a key mechanism
in insulin resistance. Surprisingly, the hepatic triglyceride con-
tent was similar in pair-fed sham-operated, GLB, and EGA
mice 10 days after surgery (Table S1B). In addition, triglycerides,
free fatty acid levels (Table S1B), and body composition
(Figure 1C) were also similar in these groups. Therefore, the
specific effect of EGA on hepatic insulin sensitivity is probably
not related to changes in the lipotoxic network.
Effect of GLP-1 Antagonist in EGA Mice
The increase of active GLP-1 levels has been proposed to be
a possible factor for the reduction of food intake and the meta-
bolic improvement observed after the RYGBP procedure. Rela-
tive to standard diet mice, the high-fat diet significantly reduced
GLP-1 levels both in the fasted state and at 40 min after initiation
of an OGTT (Figure S5). By 10 days after surgery, fasting active
GLP-1 levels were similar in pair-fed sham-operated, GLB, and
EGA mice and were increased at T40 min of the OGTT only in
EGA mice (Figure S5).
Exendin (9–39) amide, a GLP-1 antagonist, is known to block
the whole-body GLP-1 action. To specify the physiological role
of GLP-1 in the metabolic adaptation after the gastric bypass
procedure, the intraperitoneal cavity of EGA mice was continu-
ously infused for 10 days with exendin (9–39) amide or NaCl
(0.9%) from an osmotic minipump. Exendin (9–39) amide or sa-
line infusion was started during the EGA procedure. Exendin
(9–39) amide had a small effect, relative to NaCl, on food intake
inhibition associated with the EGA procedure (Figure 3A). Body
Figure 3. Effects of Exendin (9–39) Amide on the Metabolic Effects of Bariatric Surgery
Evolution of daily food intake (A) and body weight (B) before and after the EGA procedure in saline- and exendin (9–39) amide-infused high-fat diet C57Bl6 mice.
(C) Evolution of glucose during an OGTT (3g/kg) in saline- and exendin (9–39) amide-infused high-fat diet C57Bl6 mice after an EGA procedure.
(D) Insulin plasma levels at T40 min of OGTT in saline- and exendin (9–39) amide-infused high-fat diet C57Bl6 mice.
(E) Evolution of glucose levels during an intraperitoneal insulin tolerance test (0.75 UI/kg) in high-fat diet C57Bl6 mice before surgery (HFD) and in saline- and
exendin (9–39) amide-infused high-fat diet C57Bl6 mice 10 days after an EGA procedure. For all figures, n = 5 mice per group. **p < 0.01 for saline- and exendin
(9–39) amide-infused mice. xp < 0.01 for high-fat diet C57Bl6 mice and saline-infused mice. #p < 0.01 for high-fat diet C57Bl6 mice and exendin (9–39) amide-
infused mice. Data are expressed as means ± SEM.
Intestinal Gluconeogenesis and Gastric Bypass
204 Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc.
or exendin (9–39) amide intraperitoneal infusion (Figure 3B). Dur-
ing the OGTT, exendin (9–39) amide-EGA mice failed to increase
icantly higher than that observed in NaCl-EGA mice (Figures 3C
and 3D). However, insulin sensitivity, assessed by an intraperito-
neal insulin injection, was equivalent in NaCl-EGA and exendin
(9–39) amide-EGA mice (Figure 3E). These data indicate that
GLP-1 may account for the enhancement of insulin secretion
during a glucose challenge but is not essential for the regulation
of food intake or the regulation of insulin sensitivity after EGA
EGA May Involve Enhanced Intestinal Gluconeogenesis
Glc6Pase and PEPCK, two major enzymes of liver gluconeogen-
esis, are present in the small intestine, exhibiting a decreasing
gradient of expression from the duodenum to the ileum (Mithieux
et al., 2004b). Glc6Pase and PEPCK are key regulatory enzymes
in the triggering of glucose release by the small intestine, which
takes place when both enzymes are induced (Mithieux et al.,
2004a, 2004b). So we first focused our attention on the expres-
sion of the Glc6Pase and PEPCK genes. Glc6Pase activity was
similar in segment 1 of all groups of mice, but it was markedly
higher in more distal segments (3 and 4) of the bowel of EGA
mice, compared to the same segments of pair-fed GLB, sham-
ies performed for both PEPCK and Glc6Pase enzymes strongly
suggested that increased expression of both proteins took place
in EGA mice and not in the other groups (Figure 4C).
We then tested whether increased expression of gluconeo-
genic enzymes translated into intestinal glucose release in EGA
mice. Because the intestine consumes glucose at a high rate,
we employed arterio-venous glucose balance determination
coupled with 3[3H]-glucose tracer dilution, which is required to
separate the actual uptake and release of glucose. In pair-fed
sham-operated mice, there was no difference in arterial and por-
tal venous 3[3H]-glucose-specific activity (SA) (Table 1). This in-
dicated that no newly synthesized glucose had been released by
the gut. In keeping with the absence of glucose release, the ve-
arterial plasma concentration (Table 1). This decrease reflected
the intestinal glucose removal. As a consequence, the fractional
extraction (FX) calculated from these data (i.e., the fraction of
3[3H]-glucose removed by the intestine) allowed us to calculate
(from arterial plasma glucose and portal blood flow) an intestinal
glucose uptake (IGU) comparable to the intestinal net glucose
release (IGR) calculated from these IGU and IGB was not differ-
ent than zero (Table 1). The results markedly differed in pair-fed
EGA mice. The portal 3[3H]-glucose SA was lower than the arte-
synthesized glucose had been released in the portal blood.
Figure 4. Changes in Intestinal Glc6Pase and PEPCK Enzymes following Bariatric Surgery
Intestinal Glc6Pase activity was assayed in segments n?1 (duodenum), n?3 (distal jejunum), and n?4 (distal ileum) in high-fat diet, pair-fed sham-operated, gastric
lap-band, and EGA groups (n = 6 mice per group) (A and B). In high-fat diet, pair-fed sham-operated and gastric lap-band groups, Glc6Pase activity was high in
segment 1and progressivelydecreased fromtheduodenumtothedistalileum.Incontrast, inEGAmice,Glc6Paseactivitywasstronglyexpressedinall intestinal
segments studied. Glc6Pase and PEPCK protein were analyzed by Western blotting in the ileal segment in the four groups of mice (C). *p < 0.05 for the difference
between EGA and high-fat diet or pair-fed sham-operated or gastric lap-band groups. Data are expressed as means ± SEM.
Intestinal Gluconeogenesis and Gastric Bypass
Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc. 205
Moreover, this glucose release counterbalanced the glucose up-
tration in the portal blood compared to the arterial blood (Table
1). In keeping with this rationale, the calculated IGR (32.4 ±
16.8 mol/kg/min) was comparable to the calculated IGU (Table
1). It must be noted that total EGP was not higher in EGA mice
previously reported suppression of hepatic glucose production
by portal glucose appearance (Mithieux et al., 2005; Sindelar
et al., 1997). We next questioned, using two approaches,
release might be a crucial link in the EGA effects on food intake
and insulin sensitivity.
cific glucose transporter (GLUT-2), and the effects of portal glu-
cose infusion are impaired in GLUT-2 knockout mice (Burcelin
et al., 2000). To analyze the contribution of GLUT-2 in EGA ef-
fects, we performed an EGA in GLUT-2 knockout mice after 4
mice, rescued withaglucose transporterexpressedin thebcells
under the control of the rat insulin promoter, which normalizes
glucose-stimulated insulin secretion (Thorens and Larsen,
2004). These mice are not diabetic but exhibit impaired
hepato-portal glucose sensing (Burcelin et al., 2000). Using
to strongly inhibit the food intake in GLUT-2 KO mice, as ob-
served in C57Bl6 high-fat diet mice (Figure 5A). As a conse-
quence, the body weight of GLUT-2 knockout mice was moder-
ately affected throughout the follow-up after surgery (Figure 5B).
No significant change in glucose tolerance or in insulin sensitivity
was observed after EGA procedure (Figures 5C and 5D). These
findings strongly suggested that GLUT-2 and hepato-portal glu-
cose sensing are essential for the regulation of food intake and
the metabolic adaptation after EGA procedure in mice.
initiated by portal glucose appearance was dependent on the in-
tegrity of the autonomic nervous system around the portal vein
walls (Mithieux et al., 2005). C57Bl6 high-fat diet mice received
a local application of capsaicin (or saline as a control) around
the portal vein, as previously described (Mithieux et al., 2005),
at the time when the EGA procedure was performed. Capsaicin
is a neurotoxic agent inactivating selectively the afferent fibers of
the autonomic nervous system (Holzer, 1991). Within a few days,
EGA-capsaicin mice recovered a food intake comparable to
what was observed before the surgery (Figure 6A) and ceased
loosing weight (Figure 6B). Moreover, there was a marked atten-
uation of glucose tolerance and insulin sensitivity in capsaicin-
treated EGA mice compared to EGA-saline mice (Figures 6C
and 6D). This suggested that the integrity of portal nervous affer-
homeostasis. Taken together, our data demonstrated that the
specific EGA effect evidenced herein is dependent on a hep-
ato-portal glucose sensing of intestinal gluconeogenesis.
To elucidate the rapid metabolic improvement observed after
RYGBP in humans, we used a model of bariatric surgery in
mice. The RYGBP procedure performed in humans was impos-
sibleinmice, soweexcluded theduodenumand theproximal je-
junum from the alimentary tract by EGA, which also translated in
a direct access of food to the distal jejunum. Although the stom-
ach of the mice was not excluded as in RYGBP in humans, the
effects of EGA on food intake, glucose homeostasis, and GLP-
1 secretion were similar to those observed in humans. We also
wanted to compare the effects of EGA to the changes following
a GLB procedure. This comparison was important to provide
a better understanding of the specificities of the two procedures,
independently of body weight loss in pair-fed animals. We dem-
dure based on the exclusion of the duodenum and the proximal
tal jejunum) could induce a strong inhibition of food intake and
quickly improve glucose homeostasis. These effects were not
observed in pair-fed GLB mice, suggesting that neither the
food restriction nor the body weight loss are the main cause of
the metabolic effects observed in EGA.
In both GLB and EGA mice, insulin resistance in peripheral tis-
sues was unaffected by the surgery as demonstrated by clamp
studies. In contrast, EGA specifically improved the insulin sensi-
tivity of EGP. In consequence, we studied some of the mecha-
lin sensitivity. Thus, GLB and EGA groups had similar plasma
in the regulation of hepatic insulin sensitivity (Schaffler et al.,
2005). It was, therefore, possible that EGA might reduce inflam-
a plasma levels and macrophage infiltration in abdominal white
adipose (data not shown) were similar in EGA and GLB mice 10
days after the surgery. AMP-activated protein kinase (AMPK) is
an important metabolic sensor in various tissues (Viollet et al.,
Table 1. Plasma Glucidic Parameters and Intestinal Glucose Fluxes in Sham-Control Mice and EGA Mice
3[3H] Glucose SA (dpm/mml)Plasma Glucose (mmol/l)FXIGUIGBIGREGP
9.0 ± 1.1a
Pair-fed sham 244787 ± 35274 246667 ± 33761
360363 ± 51187 345915 ± 50986a8.7 ± 0.5
9.6 ± 1.2
0.04 ± 0.03 37.3 ± 22.1 49.2 ± 14.0 ?11.9 ± 18.4
0.05 ± 0.03 45.8 ± 25.3 13.4 ± 13.7
98.0 ± 11.0
8.6 ± 0.432.4 ± 16.8b94.5 ± 8.9
(FX) fractional extraction; (IGU) intestinal glucose uptake; (IGB) intestinal net glucose balance; (IGR) intestinal glucose release; (EGP) endogenous glu-
cose production. IGU, IGB, IGR, and EGP were expressed as mmole of glucose/min/kg. The results are expressed as mean ± SEM (n = 8 per group).
aValue in vein different than that in artery, p < 0.05 (Student’s t test for paired data).
bValue in EGA different than that in pair-fed sham, p < 0.05 (Mann Whitney’s test).
Intestinal Gluconeogenesis and Gastric Bypass
206 Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc.
ical activation of hepatic AMPK reduceshepatic glucose produc-
tion(Andreellietal., 2006;LongandZierath,2006).Wefound that
hepatic AMPK activity was similar in sham-operated, GLB, and
EGA mice (data not shown). As resistin levels were significantly
decreased only in EGA mice, this was an unexpected result be-
sistin knockout mice (Banerjee et al., 2004). Nevertheless, un-
changed hepatic AMPK activity after the EGA is in accordance
with our previous report showing that hepatic insulin sensitivity
is independent of hepatic AMPK (Andreelli et al., 2006).
It has been proposed that changes in gut hormone secretion
following RYGBP best explain the observed changes in appetite
and the rapid modification of whole-body insulin resistance
(Deacon, 2004; Gutzwiller et al., 1999). However, recent obser-
vations were in disagreement with this hypothesis (Korner
et al., 2005; le Roux et al., 2007; Reinehr et al., 2007). Here, we
confirm that active GLP-1 plasma levels increased after a glu-
cose challenge in EGA mice. More importantly, we show that
blockade of GLP-1 action by exendin (9–39) amide impaired
the effect of GLP-1 on insulin secretion but moderately affected
the inhibition of food intake and the changes in insulin sensitivity
following an EGA procedure. These observations argue against
the hypothesis that GLP-1 may be a critical regulator of food
intake and insulin sensitivity in our model.
The recent observation that the SI has the capacity to synthe-
size glucose and release it into the portal blood has constituted
an important breakthrough in the understanding of EGP (Croset
et al., 2001; Mithieux, 2005; Mithieux et al., 2004b). It is notewor-
of the Glc6Pase gene in rat and human SI and of the demonstra-
tion of gluconeogenesis in the rat intestine (Croset et al., 2001;
Rajas et al., 1999), the existence of this novel function of the
gut has received further support from several groups using dif-
ferent experimental approaches. This includes the confirmation
of the expression of Glc6Pase and/or PEPCK in the human SI
(Yanez et al., 2003) and their increased expression in the SI of
neonatal rats fed on a high-fructose diet (Cui et al., 2004), of pro-
tein-fed trout (Kirchner et al., 2005), and of protein-fed rats (Azz-
out-Marniche et al., 2007). Furthermore, SI glucose production
Figure 5. EGA Procedure Failed to Reduce Food Intake and Improve Glucose Tolerance and Insulin Sensitivity in High-Fat Diet GLUT2
Evolution of daily food intake (A) and body weight (B) in high-fat diet C57Bl6 mice and high-fat diet knockout GLUT2 mice before surgery and after the EGA
procedure (n = 6 for each group).
(C) Evolution of glucose levels during an OGTT (3g/kg) in high-fat diet C57Bl6 mice and high-fat diet knockout GLUT2 mice before surgery and after the EGA
procedure (n = 6 for each group).
(D) Evolution of glucose levels during an intraperitoneal insulin tolerance test (0.75 UI/kg) in high-fat diet C57Bl6 mice and high-fat diet knockout GLUT2 mice
before surgery and after the EGA procedure (n = 6 for each group). *p < 0.05 for high-fat diet C57Bl6 mice and high-fat diet knockout GLUT2 mice before surgery.
xp < 0.01 for high-fat diet C57Bl6 mice and high-fat diet knockout GLUT2 mice after surgery. Data are expressed as means ± SEM.
Intestinal Gluconeogenesis and Gastric Bypass
Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc. 207
from glutamine has been suggested to be an important contrib-
utor to EGP when hepatic gluconeogenesis is strongly blunted,
such as in mice with specific invalidation of liver PEPCK, or ab-
sent, as occurred in the anhepatic phase of liver transplantation
in humans (Battezzati et al., 2004; She et al., 2003).
A key observation herein was the marked induction of both
Glc6Pase and PEPCK enzymes in segments 3 and 4 of the SI
(distal jejunum and ileum, respectively), specifically in EGA
parts is related to their position along the anterior-posterior axis.
Important redifferentiation may thus occur in response to
a change of this position (Beck, 2002; Traber and Silberg,
1996). Accordingly, placing the distal parts of the intestine (seg-
ments 3 and4) ina position closeto the site of high nutrientavail-
ability (the stomach) might induce their redifferentiation into
a ‘‘duodenal’’-like intestine involved in the absorption of nutri-
ents. In agreement with this hypothesis, the duodenum is the
thieux et al., 2004a). The absence of Glc6Pase change in the by-
passed segments, devoid of nutrients in EGA mice, is intriguing.
The Glc6Pase gene expression is, indeed, increased in the SI in
response to fasting (Rajas et al., 1999). Actually, Glc6Pase activ-
pared to mice fed on a standard diet (G.M., unpublished data).
The reason why it was also higher in segment 1 of the other
groups of mice is unclear. The SI Glc6Pase gene is suppressed
by insulin in the normal situation (Croset et al., 2001; Rajas et al.,
1999). Thus, Glc6Pase activity could be increased in duodenal
segment 1 in the other groups of mice because of insulin resis-
tance due to HF feeding.
Using a glucose tracer dilution approach, we observed that
this marked increase in expression of Glc6Pase and PECPK en-
zymes translated into a significant glucose release by the SI in
the postabsorptive situation. It must be noted that we calculated
that the IGR in EGA mice might represent approximately one-
third of total EGP (see Table 1). However, we would like to point
out that one must be very cautious regarding the quantitative as-
pect of these calculations. The combination of a glucose tracer
Figure 6. EGA Procedure Failed to Reduce Food Intake and Improve Glucose Tolerance and Insulin Sensitivity in C57Bl6 High-Fat Diet Mice
after Disruption of Hepato-Portal Signaling by Capsaicin
Evolution of daily food intake (A) and body weight (B) before and after the EGA procedure in C57Bl6 high-fat diet mice and capsaicin-treated C57Bl6 high-fat diet
(C) Evolution of glucose concentration during an OGTT (3g/kg) before and after the EGA procedure in C57Bl6 high-fat diet mice and capsaicin-treated portal vein
C57Bl6 high-fat diet mice.
(D) Evolution of glucose levels during an intraperitoneal insulin tolerance test (0.75 UI/kg) before and after the EGA procedure in C57Bl6 high-fat diet mice
and in capsaicin-treated C57Bl6 high-fat diet mice. For all figures, n = 5mice per group. Data are expressed as means ± SEM. Symbols for statistical significance
as in Figures 2 and 4.
Intestinal Gluconeogenesis and Gastric Bypass
208 Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc.
dilution and an arterio-venous glucose balance determination is
a unique method to determine glucose release from a glucose-
utilizing organ. However, a weakness of the approach is its low
accuracy. A crucial data set, indeed, relates to the determination
of glucose SA in the portal vein and the artery. Because of inter-
animal differences and also because of the fact that there is high
blood flow through the SI, differences are difficult to determine
with accuracy. However, this is critical to demonstrate the exis-
tence of glucose release when it takes place. In the absence of
an increase in glucose concentration in the portal blood (this is
the case when the release does not exceed the uptake), the de-
crease in portal glucose SA is the only evidence indicating that
glucose release has occurred. Portal and arterial glucose SA
are obtained from the same animal. Statistical analyses may,
thus, involve a paired test. This allowed us to determine differ-
ences even if the means are very close and obtained from vari-
abledata(see Table1).Inaddition, theSAvalueshavetobemul-
tiplied or divided by other parameters, which are themselves
variable, to estimate the final fluxes. One must, therefore, be
conscious that, even when differences in glucose SA indicate
estimates and not accurate values. Irrespective of the exact
value of IGR, the latter was sufficient to compensate for the SI
glucose utilization so that portal and arterial glucose concentra-
tions were comparable (Table 1). In our previous study, a SI glu-
portal signal-dependent mechanism (Mithieux et al., 2005). Por-
tal glucose appearance could also modulate hepatic glucose
production (Sindelar et al., 1997) or whole-body glucose homeo-
stasis (Thorens and Larsen, 2004). It is, therefore, likely that the
portal detection of intestinal glucose production in the post-
absorptive state (from a gluconeogenic pathway) might be
a key mechanistic link in the decreased food intake and in-
creased insulin sensitivity in EGA mice. Strongly supporting
this proposal, no effect on food intake or insulin sensitivity was
detectable in EGA mice in the absence of functional glucose
sess intestinal glucose production in GLB mice. Unfortunately,
we failed to obtain a steady state of plasma glucose concentra-
tion and glucose-specific activity in these mice, a condition that
isrequired for the validity of the tracer approach. We have no de-
finitive explanation for this pitfall. This might be due to the possi-
bility that small amounts of food were persistent within the stom-
ach (as also revealed by the presence of gastric juice), probably
because of different characteristics of gastric emptying. How-
ever, because there was no induction of Glc6Pase and PEPCK
in GLB mice, as it was observed in pair-fed sham-operated
mice (see Figure 4), it seems unlikely that they could release in-
testinal glucose from a gluconeogenic origin.
In conclusion, our study provides new findings about the
mechanisms by which the gastric bypass rapidly improves glu-
cose homeostasis. Using a gastric bypass model in mice, our
coneogenesis and stimulate the hepatoportal glucose sensor via
procedure quickly modifies the insulin sensitivity of hepatic glu-
cose production and food intake independently of body weight
loss and GLP-1 action. This leads us to propose that intestine
glucose metabolism, especially through its gluconeogenic func-
tion, may be a crucial actor not only in the control of food intake
but also for the regulation of glucose homeostasis.
Animals and Diet
Two-month-old male C57Bl6 mice (Janvier, Le Genest Saint Isle, France) and
RIPGLUT1 3 GLUT2?/?mice backcrossed for seven generations with C57BL/
6 mice (provided by B. Thorens) were acclimated to our animal house under
controlled temperature (22?C) and light conditions (light/dark, 12 hr/12 hr)
and were fed ad libitum a high-fat diet (45 kcal% fat, 35 kcal% carbohydrate,
20 kcal% protein) (Research Diets, New Brunswick, NJ) for 16 weeks. At 6
months of age, mice underwent the surgical procedures as described below.
All procedures were performed in accordance with the principles and guide-
lines established by the European Convention for the Protection of Laboratory
C57Bl6 and RIPGLUT1 3 GLUT2?/?high-fat diet mice undergoing surgery
were fasted overnight and anaesthetized with 2% isoflurane (Abbott, Rungis,
France) and air/oxygen. The gastric lap-band was made from a piece of poly-
ethylene catheter, positioned around the upper stomach, closed, and then su-
tured to the abdominal wall. For the EGA procedure, the pyloric sphincter was
ligatured, followed by an entero-gastric anastomosis allowing the exclusion of
the duodenum and the proximal jejunum of the alimentary tract. Sham-oper-
ated mice underwent the same duration of anesthesia as GLB or EGA mice.
After surgery, sham-operated, GLB, and EGA mice were pair-fed with the
same high-fat diet used before surgery and studied 10 days after the surgery.
For exendin (9–39) amide infusion experiments, the intraperitoneal cavity of
high-fat diet C57Bl6 mice was continuously infused for 10 days with exendin
(9–39) amide (at the rate of 2 pmol 3 kg?13 min?1) or NaCl (0.9%) by an
osmotic minipump (Alzet Model 2004; Alza, Palo Alto, CA) (Cani et al., 2006).
Exendin (9–39) amide and saline infusions were started during the EGA proce-
gauze compress moistened with NaCl (0.9%) or 80 ml of a solution of capsaicin
(10 mg/ml) in water: ethanol: tween 20 (8: 1: 1, vol/vol) was applied for 10 min
around the portal vein during the EGA surgery procedure, as previously
described (Mithieux et al., 2005).
Food Intake Measurement and Body Composition Analysis
For food intake measurements, mice were individually housed with food and
water ad libitum. Food consumption was monitored every day for 15 days.
Body composition analysis was performed on anaesthetized living mice by
(PIXImus Lunar; GE Medical Systems, Madison, Wisconsin, USA).
Blood waswithdrawn fromthe tailveinforbothfed and fasted experiments us-
ing EDTA-aprotinin as the anticoagulant. In the fed state, blood was collected
at 23:00 hr. For fasting experiments, food was removed at 18:00 hr, and the
mice were kept in a different clean cage for 5 hr before collecting blood. For
the oral glucose tolerance test, blood glucose levels were evaluated using
a glucometer (Glucotrend II; Roche Laboratories, Indianapolis, IN). Serum in-
sulin, glucagon, active GLP-1, leptin, total adiponectin, TNF-a, and resistin
concentrations were assessed by Lincoplex assays (Linco Research, St.
Charles, MO). Serum concentrations of triglycerides, free fatty acids (FFAs),
ketone bodies, and glycerol were determined using an automated Monarch
device (CEFI, IFR02, Paris, France) as described previously (Viollet et al.,
Glucose and Insulin Tolerance Tests
A glucose tolerance test (3 g/kg body weight) was performed on mice fasted
for 12 hr. Blood glucose levels were determined at 0, 20, 40, and 60 min. For
the insulin tolerance test, animals fasted for 5 hr were injected intraperitoneally
with 0.75 units of insulin/kg body weight, and glucose levels were measured 0,
15, 30, and 60 min postinjection.
Intestinal Gluconeogenesis and Gastric Bypass
Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc. 209
Glucose Turnover Analysis during Hyperinsulinemic-Euglycaemic
To determine the rate of glucose use, a catheter was indwelled into the jugular
vein under anesthesia, sealed under the back skin, and glued onto the top of
the skull (Viollet et al., 2003). The mice were then housed individually. The
micewere allowed torecoverfor 4–6 days,and, after 2 days, they showed nor-
mal feeding behavior and motor activity. On the day of the experiment, the
mice were fasted for 6 hr. The whole-body glucose use rate was determined
in hyperinsulinaemic euglycaemic conditions. Under the physiological hyper-
insulinaemic condition, insulin was infused at a rate of 4 mU/kg 3 min for
3 hr, and 3-(3H) glucose was infused at a rate of 30 mCi/kg 3 min, higher
than for the basal condition, to ensure a detectable plasma 3-(3H) glucose en-
richment. Throughout the infusion, the blood glucose concentration in blood
samples (3.5 ml) collected as appropriate from the tip of the tail vein was mon-
itored with a glucose meter. Euglycaemia was maintained by periodically ad-
justing a variable infusion of 16.5% (weight/volume) glucose. Plasma glucose
concentrations and 3-(3H) glucose-specific activity were determined in 5 ml of
blood sampled from the tip of the tail vein every 10 min during the last hour of
theinfusion.Miceshowing variations inspecificactivitygreaterthan15%were
excluded from the study. Tritiated H2O and 3-(3H) glucose enrichments were
determined in total blood after deproteinization as follows and as described
previously (Andreelli et al., 2006). Five microliters of tail venous blood were
mixed with 250 ml of 0.3 M ZnSO4. Then, 250 ml of 0.3 M Ba(OH)2were added
to precipitatethe proteins and blood cells, and the precipitatewas spun down.
The supernatant was evaporated to dryness at 50?C to remove tritiated water.
The dry residue was dissolved in 0.5 ml water to which 10 ml Aqualuma plus
scintillation solution wasadded (Lumac LSC, Groningen, Netherlands), and ra-
dioactivity was determined in a Packard Tri-Carb 460C liquid scintillation sys-
tem (Rabalot, France). In a second aliquot of the same supernatant, the glu-
cose concentration was assayed by the glucose oxidase method (Trinder;
Sigma, St. Louis, MO). Under the conditions of hyperinsulinemic-euglycaemic
clamp, the rate of endogenous glucose production was equal to the glucose
disposal rate (Rd, reflecting glucose utilization) and glucose infusion rate. Rd
was calculated according to the formula Rd = EGP + GIR = [3-(3H)] glucose in-
fusion rate (disintegrations per minute [dpm/min]) divided by blood glucose
specific activity (dpm/mg) during the last 20 min of the glucose clamp
(50–70 min after the onset of insulin infusion).
Intestinal Glucose Flux Determinations
Anesthetized mice in the postabsorptive state (6 hr after food removal) were
fitted with two catheters in the left carotid artery and the right regular vein
and infused with 3-(3H) glucose in the regular vein at a rate of 8 kBq/min. A lap-
arotomy was performed to allow access to the portal vein at the time of blood
removal. After 90 min, a time when a steady state of glucose SA was obtained,
blood was gently sampled simultaneously in the carotid artery and the portal
vein as previously described (Mithieux et al., 2005). Total EGP was calculated
from the 3-(3H) glucose infusion rate and the arterial glucose SA. The fractional
extraction of glucose across the intestine (Fx) was calculated as: Fx = ((3-[3H]
glucose concentrationportal vein)) / (3-[3H] glucose SAartery3 glucose concen-
trationartery).The total intestinal bloodflow (IBF, considered tobe equivalent to
the portal blood flow) was determined from the same mice (6 hr fasted) some
daysbefore the 3-(3H) glucose infusion experiment fromaPulse WaveDoppler
echography approach using a Visualsonics (Vevo 770 High-Resolution Imag-
ing System) apparatus. Mice were anaesthetized with 2% isoflurane and air/
oxygen, and the portal vein blood flow was determined at an angle of 30?
(Huck, 2005). The mean intestinal blood flow was not different in EGA mice
(2.1 ± 0.3 ml/min) and pair-fed sham control mice (2.0 ± 0.2 ml/min). The
intestinal glucose uptake was calculated as: IGU = IBF 3 glucose concentra-
tionartery3 Fx. The intestinal glucose balance was calculated as: IGB = IBF 3
(glucose concentrationartery? glucose concentrationportal vein). The intestinal
glucose release was deduced from IGU and IGB according to the equation:
IGB = IGU ? IGR.
Hepatic Triglyceride Content
Lipids were extracted by the Folch method in a mixture of 2:1 chloroform/
methanol (vol/vol) as previously described (Villena et al., 2004). The extract
was washed with 0.2 volumes of saline (NaCl 0.9%) and centrifuged at
2,000 rpm for 10 min. The organic phase was then recovered, and triglyceride
content was determined using the Infinity triglyceride reagent (Sigma, St.
Glc6Pase and PEPCK Analyses
Hepatic and intestinal Glc6Pase enzyme activities were measured in 6 hr
fasted sham-operated, GLB, and EGA mice as previously described (Andreelli
et al., 2006; Rajas et al., 1999). In summary, small intestine and liver samples
frozen in liquid nitrogen were powdered and homogenized by sonication in
20 mM HEPES and 0.25 M sucrose (pH 7.3) (100 mg of wet tissue per milliliter).
Homogenates were diluted 1 in 10 and Glc6Pase at maximal velocity (20
mmol/liter glucose-6 phosphate) determined at 30?C on complex formation
of Pi produced from Glc6Pase. Specific Glc6Pase activity was cleared of the
contribution of nonspecific phosphohydrolase activities by subtracting the ac-
tivity toward b-glycerophosphate (20 mmol/liter) (Andreelli et al., 2006; Rajas
et al., 1999). Glc6Pase and PEPCK protein amounts were studied by western
blotting according to previously described procedures (Croset et al., 2001;
Mithieux et al., 2005; Rajas et al., 1999).
Data are expressed as means ± SE. The statistical significance of differences
between groups was assessed using two-tailed Student’s t test for unpaired
values. Student’s t test for paired samples and the Mann Whitney test were
used in the studies relating to intestinal glucose fluxes, as appropriate.
Supplemental Data include five figures and one table and can be found online
This work was supported by INSERM (Institut National de la Sante ´ et de la Re-
cherche Me ´dicale), ALFEDIAM (Association pour l’Etude du Diabe `te et des
Maladies Me ´taboliques), and Institut Benjamin Delessert. The authors are
grateful to Professor Marre for stimulating discussions and Dr. Granner for
kindly providing the PEPCK antibody used in western blotting studies. We
thank J. Bauchet, Nicolas Sorhaindo, and Houda Beltaief (Centre d’Explora-
tions Fonctionnelles Inte ´gre ´, CEFI, IFR Xavier Bichat, Paris) for the determina-
tion of blood parameters in mice, and we thank Nicole Kubat for her insightful
review of the manuscript. The authors declare that there is no conflict of
interest in relation to this work.
Received: May 29, 2007
Revised: April 11, 2008
Accepted: August 11, 2008
Published: September 2, 2008
Adam, T.C., Jocken, J., and Westerterp-Plantenga, M.S. (2005). Decreased
glucagon-like peptide 1 release after weight loss in overweight/obese
subjects. Obes. Res. 13, 710–716.
Andreelli, F., Foretz, M., Knauf, C., Cani, P.D., Perrin, C., Iglesias, M.A., Pillot,
B., Bado, A., Tronche, F., Mithieux, G., et al. (2006). Liver adenosine mono-
phosphate-activated kinase-alpha2 catalytic subunit is a key target for the
control ofhepatic glucoseproductionbyadiponectin and leptin but notinsulin.
Endocrinology 147, 2432–2441.
Azzout-Marniche, D., Gaudichon, C., Blouet, C., Bos, C., Mathe, V., Huneau,
prandial amino acid excess in high-protein fed rats? Am. J Physiol. Regul.
Integr. Comp. Physiol. 292, R1400–R1407.
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., et al. (2004). Regulation of
fasted blood glucose by resistin. Science 303, 1195–1198.
Battezzati, A., Caumo, A., Martino, F., Sereni, L.P., Coppa, J., Romito, R., Am-
matuna, M., Regalia, E., Matthews, D.E., Mazzaferro, V., and Luzi, L. (2004).
Intestinal Gluconeogenesis and Gastric Bypass
210 Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc.
Nonhepatic glucose production in humans. Am. J. Physiol. Endocrinol. Metab. Download full-text
Beck, F. (2002). Homeobox genes in gut development. Gut 51, 450–454.
Bo, O., and Modalsli, O. (1983). Gastric banding, a surgical method of treating
morbid obesity: preliminary report. Int. J. Obes. 7, 493–499.
Buchwald, H., Avidor, Y., Braunwald, E., Jensen, M.D., Pories, W., Fahrbach,
K., and Schoelles, K. (2004). Bariatric surgery: a systematic review and meta-
analysis. JAMA 292, 1724–1737.
Burcelin, R., Dolci, W., and Thorens, B. (2000). Glucose sensing by the hepa-
toportal sensor is GLUT2-dependent: in vivo analysis in GLUT2-null mice.
Diabetes 49, 1643–1648.
lin, R. (2006). Improvement of glucose tolerance and hepatic insulin sensitivity
byoligofructose requires afunctional glucagon-like peptide 1 receptor. Diabe-
tes 55, 1484–1490.
Croset, M., Rajas, F., Zitoun, C., Hurot, J.M., Montano, S., and Mithieux, G.
(2001). Rat small intestine is an insulin-sensitive gluconeogenic organ. Diabe-
tes 50, 740–746.
Cui, X.L., Soteropoulos, P., Tolias, P., and Ferraris, R.P. (2004). Fructose-
responsive genes in the small intestine of neonatal rats. Physiol. Genomics
Deacon, C.F. (2004). Therapeutic strategies based on glucagon-like peptide 1.
Diabetes 53, 2181–2189.
Gutzwiller, J.P., Goke, B., Drewe, J., Hildebrand, P., Ketterer, S., Handschin,
D., Winterhalder, R., Conen, D., and Beglinger, C. (1999). Glucagon-like
peptide-1: a potent regulator of food intake in humans. Gut 44, 81–86.
Hickey, M.S., Pories, W.J., MacDonald, K.G., Jr., Cory, K.A., Dohm, G.L.,
Swanson, M.S., Israel, R.G., Barakat, H.A., Considine, R.V., Caro, J.F., and
Houmard, J.A. (1998). A new paradigm for type 2 diabetes mellitus: could it
be a disease of the foregut? Ann. Surg. 227, 637–643.
Holzer,P.(1991).Capsaicin:cellular targets, mechanisms of action,and selec-
tivity for thin sensory neurons. Pharmacol. Rev. 43, 143–201.
Huck, K. (2005). Gepulste Dopplersonographie und Duplexsonographie. In
Kursbuch Doppler und Duplex Sonographie, K. Huck, ed. (Stuttgart: Georg
Thieme Verlag), pp. 87–91.
Kirchner, S., Seixas, P., Kaushik, S., and Panserat, S. (2005). Effects of low
protein intake onextra-hepatic gluconeogenic enzymeexpressionand periph-
eral glucose phosphorylation in rainbow trout (Oncorhynchus mykiss). Comp.
Biochem. Physiol. B. Biochem. Mol. Biol. 140, 333–340.
Korner, J., Bessler, M., Cirilo, L.J., Conwell, I.M., Daud, A., Restuccia, N.L.,
J. Clin. Endocrinol. Metab. 90, 359–365.
le Roux, C.W., Welbourn, R., Werling, M., Osborne, A., Kokkinos, A., Laure-
nius, A., Lonroth, H., Fandriks, L., Ghatei, M.A., Bloom, S.R., and Olbers, T.
(2007). Gut hormones as mediators of appetite and weight loss after Roux-
en-Y gastric bypass. Ann. Surg. 246, 780–785.
Long, Y.C., and Zierath, J.R. (2006). AMP-activated protein kinase signaling
in metabolic regulation. J. Clin. Invest. 116, 1776–1783.
Mithieux, G. (2005). The new functions of the gut in the control of glucose
homeostasis. Curr. Opin. Clin. Nutr. Metab. Care 8, 445–449.
Mithieux, G., Bady, I., Gautier, A., Croset, M., Rajas, F., and Zitoun, C. (2004a).
Induction of control genes in intestinal gluconeogenesis is sequential during
fasting and maximal in diabetes. Am. J. Physiol. Endocrinol. Metab. 286,
Mithieux, G., Rajas, F., and Gautier-Stein, A. (2004b). A novel role for glucose
6-phosphatase in the small intestine in the control of glucose homeostasis. J.
Biol. Chem. 279, 44231–44234.
Mithieux, G., Misery, P., Magnan, C., Pillot, B., Gautier-Stein, A., Bernard, C.,
Rajas, F., and Zitoun, C. (2005). Portal sensing of intestinal gluconeogenesis
is a mechanistic link in the diminution of food intake induced by diet protein.
Cell Metab. 2, 321–329.
Pories, W.J., Swanson, M.S., MacDonald, K.G., Long, S.B., Morris, P.G.,
Brown, B.M., Barakat, H.A., deRamon, R.A., Israel, G., Dolezal, J.M., et al.
(1995). Who would have thought it? An operation proves to be the most effec-
tive therapy for adult-onset diabetes mellitus. Ann. Surg. 222, 339–350.
Rajas, F., Bruni, N., Montano, S., Zitoun, C., and Mithieux, G. (1999). The glu-
cose-6 phosphatase gene is expressed in human and rat small intestine:
regulation of expression in fasted and diabetic rats. Gastroenterology 117,
Reinehr, T., Roth, C.L., Schernthaner, G.H., Kopp, H.P., Kriwanek, S., and
Schernthaner, G. (2007). Peptide YY and glucagon-like peptide-1 in morbidly
Rubino, F., and Marescaux, J. (2004). Effect of duodenal-jejunal exclusion
in a non-obese animal model of type 2 diabetes: a new perspective for an
old disease. Ann. Surg. 239, 1–11.
Schaffler, A., Scholmerich, J., and Buchler, C. (2005). Mechanisms of disease:
adipocytokines and visceral adipose tissue–emerging role in nonalcoholic
fatty liver disease. Nat. Clin. Pract. Gastroenterol. Hepatol. 2, 273–280.
Scopinaro, N., Adami, G.F., Marinari, G.M., Gianetta, E., Traverso, E., Fried-
man, D., Camerini, G., Baschieri, G., and Simonelli, A. (1998). Biliopancreatic
diversion. World J. Surg. 22, 936–946.
She, P., Burgess, S.C., Shiota, M., Flakoll, P., Donahue, E.P., Malloy, C.R.,
Sherry, A.D., and Magnuson, M.A. (2003). Mechanisms by which liver-specific
PEPCK knockout mice preserve euglycemia during starvation. Diabetes 52,
Sindelar, D.K., Chu, C.A., Neal, D.W., and Cherrington, A.D. (1997). Interaction
of equal increments in arterial and portal vein insulin on hepatic glucose pro-
duction in the dog. Am. J. Physiol. 273, E972–E980.
Sjostrom, L., Lindroos, A.K., Peltonen, M., Torgerson, J., Bouchard, C., Carls-
style, diabetes, and cardiovascular risk factors 10 years after bariatric surgery.
N. Engl. J. Med. 351, 2683–2693.
Steinbrook, R. (2004). Surgery for severe obesity. N. Engl. J. Med. 350, 1075–
Thorens, B., and Larsen, P.J. (2004). Gut-derived signaling molecules and
vagal afferents in the control of glucose and energy homeostasis. Curr.
Opin. Clin. Nutr. Metab. Care 7, 471–478.
Traber, P.G., and Silberg, D.G. (1996). Intestine-specific gene transcription.
Annu. Rev. Physiol. 58, 275–297.
Villena, J.A., Viollet, B.,Andreelli, F.,Kahn, A.,Vaulont, S.,and Sul,H.S. (2004).
Induced adiposity and adipocyte hypertrophy in mice lacking the AMP-
activated protein kinase-alpha2 subunit. Diabetes 53, 2242–2249.
Viollet,B.,Andreelli, F.,Jorgensen,S.B.,Perrin,C.,Geloen, A.,Flamez,D.,Mu,
J., Lenzner, C., Baud, O., Bennoun, M., et al. (2003). The AMP-activated pro-
tein kinase alpha2 catalytic subunit controls whole-body insulin sensitivity. J.
Clin. Invest. 111, 91–98.
Viollet, B., Foretz, M., Guigas, B., Horman, S., Dentin, R., Bertrand, L., Hue, L.,
and Andreelli, F. (2006). Activation of AMP-activated protein kinase in the liver:
a new strategy for the management of metabolic hepatic disorders. J. Physiol.
Yanez, A.J., Nualart, F., Droppelmann, C., Bertinat, R., Brito, M., Concha, I.I.,
and Slebe, J.C. (2003). Broad expression of fructose-1,6-bisphosphatase and
phosphoenolpyruvatecarboxykinaseprovideevidence for gluconeogenesis in
human tissues other than liver and kidney. J. Cell. Physiol. 197, 189–197.
Intestinal Gluconeogenesis and Gastric Bypass
Cell Metabolism 8, 201–211, September 3, 2008 ª2008 Elsevier Inc. 211