Hepatic branch vagotomy, like insulin replacement, promotes voluntary lard intake in streptozotocin-diabetic rats.

Abstract

Although high insulin concentrations reduce food intake, low insulin concentrations promote lard intake over chow, possibly via an insulin-derived, liver-mediated signal. To investigate the role of the hepatic vagus in voluntary lard intake, streptozotocin-diabetic rats with insulin or vehicle replaced into either the superior mesenteric or jugular veins received a hepatic branch vagotomy (HV) or a sham operation. All rats received a pellet of corticosterone that clamped the circulating steroid at moderately high concentrations to enhance lard intake. After 5 d of recovery, rats were offered the choice of lard and chow for 5 d. In streptozotocin-diabetic rats, HV, like insulin replacement, restored lard intake to nondiabetic levels. Consequently, this reduced chow intake without affecting total caloric intake, and insulin site-specifically increased white adipose tissue weight. HV also ablated the effects of insulin on reducing circulating glucose levels and attenuated the streptozotocin-induced weight loss in most groups. Collectively, these data suggest that the hepatic vagus normally inhibits lard intake and can influence glucose homeostasis and the pattern of white adipose tissue deposition. These actions may be modulated by insulin acting both centrally and peripherally.

Full text

Page 1

1
Title: Hepatic Branch Vagotomy, Like Insulin Replacement, Promotes Voluntary
Lard Intake in Streptozotocin-Diabetic Rats
Short Title: Hepatic Vagus, Insulin and Lard Intake
Authors: James P. Warne, Michelle T. Foster, Hart F. Horneman, Norman C.
Pecoraro, Abigail B. Ginsberg, Susan F. Akana and Mary F. Dallman
Affiliation: Department of Physiology, University of California San Francisco,
San Francisco, California.
Corresponding Author:
Dr. James Warne
Department of Physiology, Box 0444
University of California San Francisco
513 Parnassus Avenue
San Francisco
CA 94143, USA
Tel: 415-476-3861
Fax: 415-502-4848
E-mail: james.warne@ucsf.edu

Endocrinology. First published ahead of print April 4, 2007 as doi:10.1210/en.2007-0003
Copyright (C) 2007 by The Endocrine Society

Page 2

2
Disclosure Statement: The authors have nothing to disclose.
NIH Statement: This is an un-copyedited author manuscript copyrighted by The
Endocrine Society. This may not be duplicated or reproduced, other than for
personal use or within the rule of “Fair Use of Copyrighted Materials” (section
107, Title 17, U.S. Code) without permission of the copyright owner, The
Endocrine Society. From the time of acceptance following peer review, the full
text of this manuscript is made freely available by The Endocrine Society at
http://www.endojournals.org/. The final copy edited article can be found at
http://www.endojournals.org/. The Endocrine Society disclaims any responsibility
or liability for errors or omissions in this version of the manuscript or in any
version derived from it by the National Institutes of Health or other parties. The
citation of this article must include the following information: author(s), article title,
journal title, year of publication and DOI.
Abbreviations: B = corticosterone, FFA = free fatty acid, eWAT = epididymal
WAT, HV = hepatic branch vagotomy, IGF-I = insulin-like growth factor-I, Ins =
Insulin, Jug = Jugular, Mes = mesenteric, mWAT = mesenteric WAT, pWAT =
perirenal WAT, scWAT = subcutaneous WAT, STZ = streptozotocin, WAT =
white adipose tissue, Veh = vehicle.

Page 3

3
Abstract
Although high insulin concentrations reduce food intake, low insulin
concentrations promote lard intake over chow, possibly via an insulin-derived,
liver-mediated signal. To investigate the role of the hepatic vagus in voluntary
lard intake, streptozotocin-diabetic rats with insulin or vehicle replaced into either
the superior mesenteric or jugular veins received a hepatic branch vagotomy
(HV) or a sham operation. All rats received a pellet of corticosterone that
clamped the circulating steroid at moderately high concentrations to enhance lard
intake. After 5 days of recovery, rats were offered the choice of lard and chow
for 5 days. In streptozotocin-diabetic rats, HV, like insulin replacement, restored
lard intake to non-diabetic levels. Consequently, this reduced chow intake
without affecting total caloric intake, and insulin site-specifically increased white
adipose tissue weight. HV also ablated the effects of insulin on reducing
circulating glucose levels and attenuated the streptozotocin-induced weight loss
in most groups. Collectively, these data suggest that the hepatic vagus normally
inhibits lard intake and can influence glucose homeostasis and the pattern of
white adipose tissue deposition. These actions may be modulated by insulin
acting both centrally and peripherally.

Page 4

4
Introduction
The brain regulates food intake by integrating hormonal and metabolic signals,
both through direct stimulation of the brain and indirectly through afferent
signaling from other organs. Two important hormones in this regard are insulin
and corticosterone (B) that appear to have reciprocal effects on both brain and
body (1). Insulin directly serves to inhibit orexigenic and to excite anorexigenic
neurons in the arcuate nucleus (2, 3), consequently inhibiting food intake. B, on
the other hand, appears to increase the drive for sucrose (4), saccharin (5) and
lard (6). Both hormones clearly exert actions directly on the brain and on key
metabolic tissues throughout the body, often resulting in antagonistic effects.
When there is a choice of caloric sources (chow and lard), B induces an increase
in total caloric intake, whereas the choice of what calories are derived from which
food source is strongly influenced by the prevailing insulin levels (6). In
adrenalectomized rats, B replacement produces a dose-dependent increase in
lard, but not chow, intake whereas in adrenalectomized rats with streptozotocin
(STZ)-diabetes, B replacement produces a dose-dependent increase in chow,
but not lard, intake. Lard intake is restored in the STZ treated rats in a dose-
dependent fashion by increasing circulating insulin levels (6). Furthermore, in
STZ-diabetic rats with moderately elevated B concentrations, venous insulin
replacement results in the exclusive recovery of lard, but not sucrose, intake with
consequent inhibition of chow intake, when all three sources of calories are
offered ad libitum (7). Low concentrations of insulin secreted in anticipation of

Page 5

5
meals have also been shown to increase ingestion of palatable, but non-caloric
foods in humans (8). Venous insulin replacement into the superior mesenteric
and right external jugular veins both restore voluntary lard intake in STZ-diabetic
rodents. However, since superior mesenteric insulin infusions more naturally
replace insulin of pancreatic origin, and also do not necessarily result in
significantly elevated insulin levels in the systemic circulation, despite having
clear actions on hepatic glucose metabolism, these data pointed to the liver as a
key regulatory site for lard intake (7).
There is clear and reciprocal communication between the brain, notably the
hypothalamus, and the liver that regulates many key aspects of metabolism (9).
For example, insulin acting on the CNS can regulate hepatic glucose output,
through vagally-mediated suppression of gluconeogenesis (10). However,
afferent signals from the liver can affect the brain regulation of energy
homeostasis as well. For example, inhibition of fatty acid oxidation by
mercaptoacetate increases food intake (11), an effect that is most likely triggered
by a signal from the liver since the effects of mercaptoacetate are abolished by
hepatic branch vagotomy (HV) (12). HV also prevents the lard-induced inhibition
of food intake and changes in neuropeptide expression in STZ-diabetic rats (13,
14). Since the liver is a major target organ for insulin, which inhibits
glycogenolysis and gluconeogenesis, and promotes fat synthesis (15), there is
therefore major potential for an insulin-stimulated generation of metabolic signals
to affect hepatic vagal afferents and, consequently, affect food intake.

Page 6

6
In these studies, we assessed the contribution of an insulin-stimulated, hepatic
vagus-derived signal in mediating voluntary lard ingestion in STZ-diabetic, B-
clamped rats. Two subdivisions of rats, based on the experimental
manipulations, making a total of 10 groups, were examined. The first subdivision
consisted of 5 groups: non-diabetic control, STZ-diabetic with vehicle infused into
the jugular (Veh-Jug) or superior mesenteric vein (Veh-Mes), STZ-diabetic with
insulin infused into the jugular (Ins-Jug) or mesenteric vein (Ins-Mes). Each
group was then further subdivided into two groups, one was sham HV (sham),
and the other underwent HV. At the end of the experiment, plasma and liver
samples were taken to confirm levels of B and insulin, as well as insulin-sensitive
hormones, metabolites and liver glycogen levels to confirm biological action.
White adipose tissue (WAT) depots were excised and weighed to gauge
outcomes of lard intake and to delineate any differences between the two venous
infusion sites, which have been previously shown to have different effects on
different WAT depots (7).

Page 7

7
Methods
Male rats (Sprague-Dawley, Simonsen, Gilroy, CA) weighing 302 ± 2g were
housed individually in hanging wire cages in a temperature (22°C) and light
(lights on 0600h-1800h) controlled room. Rats were allowed to adapt to their
new environment for 4 days before experimentation. All experimental
procedures were approved by the University of California San Francisco
Institutional Animal Care and Use Committee. The rats had ad libitum access to
pelleted rat chow (Purina Chow #5008, Purina, St. Louis, MO; 13.84 kJ/g) and
water throughout the experiment. When noted, all rats were also provided with
ad libitum supply of lard (Armour, Omaha, NB; 37.62 kJ/g) in a metal cup.
Surgical Procedures and Treatments:
The experimental design expanded on that previously reported (7), with the same
treatment schedule but in this instance, additional surgical manipulations of the
hepatic vagus. There were 5 rats in each of the 10 groups. All procedures were
performed in one surgery (day 0). All rats were anesthetized using ketamine
(75mg/kg, i.m.) and xylazine (10mg/kg, i.m.). Ketoprofen (10mg/kg, s.c.) was
provided as an analgesic after surgery but prior to the rat regaining
consciousness. All rats received incisions into the right of the neck, for access to
the right external jugular vein (catheter inserted if appropriate for that group) as
well as to the left side of the body, for access to the hepatic vagus and, if
appropriate, the superior mesenteric vein. During all procedures, the opened

Page 8

8
abdominal cavity was bathed in sterile saline to prevent drying of the viscera. All
rats received a subcutaneous pellet of B. After all appropriate surgical
manipulations were performed, all incisions were closed using silk suture.
The rats were allowed 5 days to recover after surgery, during which incisions,
body weight and food and water intake were monitored daily at 1000h. All rats
were also presented with lard ad libitum on day 5 at 1000h. Body weight and
solid and liquid intakes were monitored daily for a further 5 days. On day 10 at
1000h, final day food and water intake measures were taken. All rats were then
killed by decapitation and samples were collected.
Hepatic Branch Vagotomy. The common hepatic vagal branch was visualized
through an incision into the left side of the body by gently moving aside
surrounding tissues, which were held out of the field of view with saline-soaked
sterile gauze. For the HV groups, the common hepatic vagal branch was
located as it separates from the left vagal trunk and cut. The sham-operated
groups underwent all procedures except transection of the neural tissue (13).
The hepatic branch transection was visualized at the end of the experiment. In
all HV operated rats, the common hepatic vagal branch was still transected.
STZ-Induced Diabetes. Diabetes was induced by a subcutaneous injection of
STZ (Sigma Chemicals, St. Louis, MO; 65mg/kg in citrate buffer pH 4.2) whilst
the rats were still unconscious. Control rats were injected with citrate buffer

Page 9

9
(2ml/kg). Diabetes was confirmed by the presence of marked glucosuria
(Multistix® 9 SG, Bayer Corp, Elkhart, IN) on day 3.
Insulin Replacement. The same, low, dose of insulin (3U/day; Humulin R®
U500, Eli Lilly and Company, Indianapolis, IL) or saline was infused in the STZ-
treated animals at one of two locations (jugular or superior mesenteric veins) via
the insertion of catheters (PE5 tubing, 1.5cm, fused to PE60 tubing, 1.5cm)
attached to osmotic minipumps (Alzet, model 2002, Alza, Palo Alto, CA) as
previously described (7). For the jugular infusions, the right external jugular vein
was accessed from an incision into the side of the neck. The vein was exposed,
gently elevated, a small incision was made and the catheter was inserted. The
vein was sealed using sterile glue (Vetabond, 3M Animal Care Products, St.
Paul, MN). A subcutaneous pocket from the neck to the back was then created
to hold the catheter and attached osmotic pump. For the superior mesenteric
infusions, the cecum was externalized from the incision made into the left side of
the body and placed onto gauze soaked in sterile saline and the superior
mesenteric vein was gently exposed. The catheter was inserted into the vein
and immediately sealed into place using sterile glue. The cecum and osmotic
minipump were then quickly internalized, such that the minipump nestled close to
the cecum and small intestine. The presence of an osmotic mini-pump at either
site did not cause any obvious signs of discomfort for the rats.

Page 10

10
B-Treatment. Circulating B levels were maintained at steady-state
concentrations by placing a 100mg pellet of B (100%; Steraloids Inc., Newport,
RI) subcutaneously through a small incision in the back. Previous studies have
shown this treatment produces sustained moderate elevations (~150 ng/ml) in
the concentrations of B (16).
Sample Collection. After rats were killed by decapitation, trunk blood was
collected into chilled tubes containing 100µl EDTA (65mg/ml). Tubes were
centrifuged, plasma collected and stored at -80°C. Liver biopsies (100mg) were
quickly collected from the same lobe (lobus sinister lateralis), snap frozen and
stored at -80°C. The rest of the body was put onto ice for subsequent dissection
and weighing of the white adipose tissue (WAT) fat pads (left and right
subcutaneous [scWAT], epididymal [eWAT], perirenal [pWAT] and mesenteric
[mWAT]), the thymus and spleen. At this time, the position of the catheters and
osmotic mini-pumps was verified. In all cases, one end of the catheter was
securely inserted into the desired vein, the other attached to the mini-pump. In
addition, the hepatic vagus was visualized and in all cases appeared severed.
Plasma and Liver Assays. Plasma B, insulin, leptin and insulin-like growth
factor-I (IGF-I) concentrations were assessed by radioimmunoassays at half
volumes (MP Biomedicals, Orangeburg, NY; Linco Research Inc., St. Charles,
MO; Diagnostic Systems Laboratories Inc., Webster, TX), whereas plasma
glucose, triacylglycerols, glycerol, total ketone bodies and free fatty acids (FFA)

Page 11

11
were measured colorimetrically on a plate reader using kits (Mega Diagnostics,
Los Angeles, CA; Sigma-Aldrich, St. Louis, MO; Wako Chemicals, Neuss,
Germany), all as previously described (6, 7). Liver glycogen content was
assessed by a colorimetric plate assay outlined previously (17) and was
standardized to mg wet weight.
Statistical Analyses:
All data are presented as the mean ± standard error of the mean. Body weight
data were analyzed by repeated measures ANOVA followed by a repeated
measures t-test to compare HV versus sham operated for each group.
Due to the nature of the experimental design, all other data were analyzed in two
ways. First, the STZ-treated groups (Veh-Mes ± HV, Ins-Mes ± HV, Veh-Jug ±
HV, Ins-Jug ± HV) were analyzed by 3-way ANOVA to test for significant
(p<0.05) effects of and interactions between the condition of the hepatic vagus
(HV vs. sham), insulin replacement (vehicle vs. insulin) and site of replacement
(superior mesenteric vs. right external jugular). The results of the 3-way ANOVA
are summarized in table 1. Data from all groups were subsequently analyzed by
2-way ANOVA to compare data to non-diabetic, but still moderate-B, controls.
The factors for this statistical test were condition of the hepatic vagus (HV vs.
sham) and insulin manipulation (Control, Veh-Mes, Ins-Mes, Veh-Jug, Ins-Jug).
Significant (p<0.05) effects after both 3-way and 2-way ANOVAs were followed
by post-hoc tests of individual group differences (Tukey’s Test), the results of

Page 12

12
which are presented on the figures, where different letters indicate statistically
significant differences between groups (e.g. a is different than b, but neither is
different than ab).

Page 13

13
Results
The effects of the experimental manipulations on body weight are shown in figure
1. All rats initially lost weight during the study, because of the catabolic effects of
the moderately elevated B (18). In the sham operated rats (Fig 1A), STZ
treatment initially further exacerbated this weight loss in all groups except those
which received insulin into the jugular vein (p<0.05). Introduction of lard on day 5
curtailed further weight loss in all groups such that only the veh-jug group
remained significantly lower (p<0.05) than the non-diabetic controls. HV curtailed
the initial differences between the treatment groups such that no significant
differences existed (Fig 1B).
When comparing sham and HV operated rats of each group by repeated
measures ANOVA, the non-diabetic controls (Fig 1C) and those rats receiving
vehicle into the superior mesenteric vein (Veh-Mes; Fig 1D) showed no
significant differences between the two groups. In contrast, HV attenuated the
weight loss on days 2, 3 and 4 of the study in those rats receiving insulin
replacement into the superior mesenteric vein, compared to sham operated
counterparts (Ins-Mes; Fig 1E). The STZ-treated rats that received vehicle into
the jugular vein showed significant (p<0.05) time, treatment and time-treatment
interaction effects (Veh-Jug; Fig 1F). Specifically, weight loss was greater in the
sham operated compared to HV group on days 8, 9 and 10 of the study. Rats
receiving insulin into the jugular vein also displayed significant (p<0.05) time,

Page 14

14
treatment and time-treatment interaction effects; however, in this instance the
sham-HV operated rats lost less weight than the HV-operated counterparts, with
significance evident on days 4 and 5 of the study (Ins-Jug; Fig 1G).
The effects of HV, STZ-diabetes and site-specific insulin replacement on caloric
intake are shown in figure 2. Chow intake during the 5 days of recovery after
surgery rose steadily in the STZ-treated groups (data not shown). Provision of
lard on day 5 resulted in all groups consuming the new food equally over the
chow during the first day of availability. The data presented in figure 2 and table
1 for 5-day food intake reflect consumption from the second day of ad libitum
access until the end of the study. Three-way ANOVA of the STZ-treated groups
(Table 1) revealed a significant (p<0.05) effect of insulin and a significant
(p<0.05) insulin-HV surgery interaction, but no effect of the site of venous
infusion, on lard intake (Fig 2A). Subsequent post-hoc analysis showed, in those
rats receiving superior mesenteric venous infusions, that HV increased lard
intake in the saline infused rats to the level of their insulin infused counterparts.
This amount of lard intake was similar to that observed in the non-diabetic control
rats. HV did not modify the effects of venous insulin infusion. Those rats that
had jugular venous infusions showed a pattern like that of the superior
mesenteric infused rats, with insulin infusion increasing lard intake in the sham
HV operated rats. However, the HV operated rats with either saline or insulin
infusions were not significantly different from the sham operated, saline infused
rats.

Page 15

15
Three-way ANOVA revealed that both chow (Fig 2B) and total caloric (Fig 2C)
intake were significantly attenuated by insulin replacement, irrespective of site, in
STZ-diabetic rats. Post-hoc tests also revealed that chow intake was reduced by
HV in the group that received vehicle into the superior mesenteric vein, an effect
not evident in the corresponding jugular group that received vehicle infusion.
Both chow and total caloric intake were significantly (p<0.05) higher than that of
the non-diabetic controls.
As shown in figure 3 and table 2, STZ-diabetes resulted in a significant reduction
(p<0.05) in the weight of all of the fat pads examined. This effect, however, did
not reach statistical significance in the mWAT of the sham-HV operated rats that
received vehicle into the superior mesenteric vein, when mWAT weight was
adjusted per g body weight (Fig 3C). Analysis within the STZ-treated groups of
both absolute (Table 2) and body weight adjusted (Figure 3) WAT weights by 3-
way ANOVA revealed several significant (p<0.05) effects. Left scWAT weight
showed a significant (p<0.05) HV surgery-replacement site interaction. The
jugular insulin-induced maintenance of left scWAT weight was attenuated by HV
such that it was not significantly different from the STZ-depleted weights of all
other groups. In contrast, right scWAT weight was unaffected by any
experimental manipulation within the STZ-treated groups.