- Top of page
The common hepatic branch of the vagus nerve is a two-way highway of communication between the brain and the liver, duodenum, stomach and pancreas that regulates many aspects of food intake and metabolism. In this study, we utilized the afferent-specific neurotoxin capsaicin to examine if common hepatic vagal sensory afferents regulate lard intake. Rats implanted with a corticosterone pellet were made diabetic using streptozotocin (STZ) and a subset received steady-state exogenous insulin replacement into the superior mesenteric vein. These were compared with non-diabetic counterparts. Each group was then subdivided into those whose common hepatic branch of the vagus was treated with vehicle or capsaicin. Five days after surgery, the rats were offered the choice of chow and lard to consume for a further 5 days. The STZ-diabetic rats ate significantly less lard than the non-diabetic rats. Capsaicin treatment restored lard intake to that of the insulin-replaced, STZ-diabetic rats, but modified neither chow nor total caloric intake. This increased lard intake led to selective fat deposition into the mesenteric white adipose tissue depot, as opposed to an increase in all visceral fat pad depots evident after insulin replacement-induced lard intake. Capsaicin treatment also increased the levels of circulating glucose and triglycerides and negated the actions of insulin on these and free fatty acids and ketone bodies. Collectively, these data suggest that afferent signalling through the common hepatic branch of the vagus inhibits lard, but not chow, intake, directs fat deposition and regulates plasma metabolite levels.
A complex and interacting array of hormones and metabolites act both centrally and peripherally to regulate the amount and choice of food ingested. Our studies using manipulation of circulating glucocorticoid and insulin levels have revealed a stimulatory role on total caloric intake by the former and modulation of the choice of the caloric source ingested by the latter (La Fleur et al. 2004; Warne et al. 2006, 2007). In streptozotocin (STZ)-diabetic rats with fixed, elevated corticosterone concentrations, venous insulin replacement recovers voluntary lard, but not sucrose, ingestion to non-diabetic levels (Warne et al. 2006). The importance of the liver in mediating the effect of insulin on food choice has been suggested by our findings where, in STZ-diabetic high corticosterone rats, superior mesenteric insulin infusions do not necessarily result in elevated circulating insulin levels, yet consistently promote voluntary lard intake (Warne et al. 2006). The liver has been implicated as a key regulator of food intake in a number of other studies. For example, inhibition of liver fatty acid oxidation by mercaptoacetate increases food intake (Singer-Koegler et al. 1996).
The vagus nerve represents a two-way highway of communication between the brain and the periphery, as shown by the presence of afferent (sensory) and efferent (motor) fibres of parasympathetic origin (Berthoud, 2004). Afferent nerve fibres from the liver join those originating from the duodenum, pancreas, pylorus and distal gastric antrum (collectively known as the gastroduodenal branch) to form the common hepatic branch of the vagus. The majority of fibres within the common hepatic branch originate from the gastroduodenal branch; however, the liver is represented by fibres originating from the hepatic artery, portal vein and bile ducts, with little or no contribution of fibres from the hepatic parenchyma (Magni & Carobi, 1983; Carobi & Magni, 1985). Fibres at these locations are in prime positions to detect changes in hepatic output as well as sample the hormonal and metabolic milieu entering the liver.
Studies using hepatic branch vagotomy (HV) have shown the common hepatic branch of the vagus nerve to be important in the regulation of food intake. In STZ-diabetic high corticosterone clamped rats, HV promotes voluntary lard intake, suggesting the vagus nerve ordinarily has an inhibitory influence on lard intake, which can be overcome by insulin action (Warne et al. 2007). HV also prevents the lard-induced inhibition of food intake and changes in neuropeptide expression in STZ-diabetic rats (La Fleur et al. 2003, 2005b) and inhibits mercaptoacetate-induced food intake (Langhans, 2000). Hence, signals traversing the common branch of the hepatic vagus clearly regulate, and are responsive to, lard intake. However, it is unclear if the inhibitory signal for lard intake from the common hepatic branch of the vagus is afferent or efferent in nature.
In recent years, understanding of the importance of afferent and efferent signalling through the vagus has become apparent, which has led to the discovery of complex axes that regulate metabolism. For example, in addition to the classical direct actions of insulin on the liver, insulin can act within the brain to suppress liver gluconeogenesis via a vagal signal (Obici et al. 2002; Pocai et al. 2005a,b). Afferent signalling from the hepatic vagus and efferent sympathetic signalling to adipose tissue also regulates energy expenditure, glucose metabolism and fat distribution (Uno et al. 2006). Hence, determining the contribution of afferent and efferent vagal signalling in the regulation of food intake and metabolism is an important therapeutic consideration in pathological conditions where these processes are dysregulated.
Since the common branch of the hepatic vagus exhibits an inhibitory signal for lard intake that can be overcome by insulin action (Warne et al. 2007), we sought to examine if this inihibitory signal is mediated by afferent fibres. The following groups of rats were studied: non-diabetic citrate (the vehicle for STZ) injected and STZ-diabetic with either saline (‘STZ-saline’) or insulin (‘STZ-insulin’) infused into the superior mesenteric vein. These groups were further subdivided into those whose hepatic vagus was treated with the afferent-specific neurotoxin capsaicin and those that were treated with vehicle. All rats received a subcutaneous pellet of corticosterone. After 5 days of recovery, all rats were offered lard in addition to chow for a further 5 days. At the end of the experiment, plasma samples were taken to confirm concentrations of steady-state corticosterone and insulin. Biological actions of both corticosterone and insulin were confirmed by examining adrenal, thymus and spleen weights, liver glycogen content and the plasma levels of leptin, glucagon, glucose, triglycerides, total ketone bodies, free fatty acids (FFAs) and glycerol. White adipose tissue (WAT) depots were excised and weighed to gauge outcomes of lard intake.
- Top of page
In order to confirm the efficacy of capsaicin in the selective destruction of afferent fibres, the treated portion of the common hepatic vagus was processed for immunocytochemical detection of calretinin, a marker of vagal afferents (Dutsch et al. 1998; Berthoud, 2004), and TH, a marker of sympathetic nerve fibres (Foster & Bartness, 2006). Figure 1 shows calretinin (Fig. 1A and B) and TH (Fig. 1D and E) immunoreactivity in sections of the common hepatic branch of the vagus that were treated with either capsaicin or vehicle. Quantification showed that capsaicin treatment, irrespective of any other surgical manipulation, reduced the number of calretinin (P < 0.05, Fig. 1C) immunoreactive fibres, but had no effect on TH immunoreactive fibre number (Fig. 1F), compared with vehicle-treated vagi. This illustrates the selective nature of capsaicin treatment.
Figure 1. Calretinin (A–C) and tyrosine hydroxylase (TH; D–F) immunoreactivity of the common branch of the hepatic vagus after experimental manipulations Representative images from (A and D) vehicle and (B and E) capsaicin-treated vagi for each peptide are shown. Quantification revealed that capsaicin treatment selectively reduced calretinin (C) but not TH (F) immunoreactive fibre number. Scale bar, 10 μm. Arrows point to immunoreactive fibre bundles. Treatment groups with different letters above the error bars are significantly (P < 0.05) different from each other (i.e. a is significantly different to b), groups with the same letters above the error bars are not significantly different groups (i.e. a is not significantly different to a).
Download figure to PowerPoint
Food intake is presented in Fig. 2. As shown in Fig. 2A, after surgery, chow intake rose in all groups until day 3; thereafter the STZ-treated groups continued to increase chow intake with time, whereas the citrate-treated (hence non-diabetic) control groups stabilized chow intake. Presentation of lard on day 5 caused a reduction in chow intake in all groups (day 6 measures). Thereafter, the citrate-treated groups exhibited reduced chow intake compared with previous, no-choice intake. The STZ-treated groups, however, diverged from day 7 onwards; this depended on the infusion into the superior mesenteric vein. STZ-treated rats infused with saline progressively increased chow intake back to pre-choice levels, whereas those rats infused with insulin ate less chow than these groups. Total chow intake for the 5 days of choice illustrated a significant (P < 0.001) difference between the insulin-manipulated groups, but no effect of vagal manipulation. Specifically, the STZ-treated groups ate significantly (P < 0.001) more than the citrate-treated groups. Insulin infusion into the STZ-treated rats reduced chow intake, compared with saline infusion, but hyperphagia remained evident.
Figure 2. Intake of chow (A and B), lard (C and D) and total calories (E and F) throughout the experiment, shown as the daily intake (A, C and E) and the summed intake over the last 5 days of the experiment (choice period; B, D and F) Data are calculated by the sum of the daily caloric intake (in kcal) per 100 g of body weight. STZ reduced lard intake, an effect that was partially prevented by either capsaicin treatment of the common hepatic vagus or by insulin replacement. Chow and consequently total caloric intake was increased by STZ, an effect that was attenuated by insulin replacement and was not affected by capsaicin treatment. Treatment groups with different letters above the error bars are significantly (P < 0.05) different from each other (i.e. a is significantly different to b), groups with the same letters above the error bars are not significantly different groups (i.e. a is not significantly different to a or ab).
Download figure to PowerPoint
Lard intake is shown in Fig. 2C and D. As illustrated in Fig. 2C, presentation of lard on day 5 caused all rats to consume the lard equally during the first day of availability (day 6 data). Thereafter, the insulin- and vagus-associated patterns were evident. Total 5 day intake of lard is presented in Fig. 2D. In the citrate-injected groups, capsaicin treatment of the vagus did not modify lard intake. STZ injection significantly (P < 0.05) reduced lard intake, which was prevented by capsaicin treatment, but not to the level of the non-diabetic controls. Examination of the pattern of lard intake of the STZ-injected, capsaicin-treated group showed that lard intake was comparable on the second, forth and fifth days of lard presentation (days 7, 9 and 10; Fig. 2C), at levels slightly lower than the insulin-replaced counterparts. Lard intake on the third day of lard availability (day 8) was low, and comparable to the STZ-injected, vehicle-treated group. Insulin infusion into the superior mesenteric vein elevated lard intake to levels of non-diabetic controls. Capsaicin treatment did not modify lard intake in the insulin-infused groups.
Total caloric intake is presented in Fig. 2E and F. Similar to chow intake, STZ induced a significant (P < 0.01) increase in total caloric intake. As revealed by 2-way ANOVA, insulin replacement significantly (P < 0.05) reduced total intake (Fig. 2F). Post hoc analysis showed significant (P < 0.05) differences between the saline- and insulin-infused groups that received capsaicin treatment, but not those treated with vehicle on the common branch of the hepatic vagus.
All rats lost body weight throughout the study (Fig. 3), in keeping with previous observations (Warne et al. 2006). STZ injection further exacerbated the weight loss, compared with citrate-injected (non-diabetic) controls. The STZ-induced weight loss was partially prevented by insulin infusion, an action that was curtailed by capsaicin treatment of the common branch of the hepatic vagus. Body weight loss was halted and reversed by provision of lard on day 5.
Figure 3. Body weight changes over the 10 day period Data are presented as the total data (A; error bars are omitted for clarity), then direct vehicle (open symbols) versus capsaicin (filled symbols) comparisons for the citrate (B), STZ-saline (C) and STZ-insulin (D) groups. All rats, at least initially, lost body weight. Body weight loss of the citrate-treated groups was initially lower than the STZ-treated groups. After introduction of lard, body weights of all rats stabilized and eventually increased. Capsaicin treatment had no effect on body weight. *P < 0.05 citrate-injected groups versus STZ-saline and STZ-insulin, capsaicin- (but not vehicle-) treated groups.
Download figure to PowerPoint
WAT weight showed effects of STZ, insulin replacement and capsaicin treatment, as shown in Fig. 4. Capsaicin treatment of the vagi of the citrate-injected controls had no effect on the weight of any WAT depot examined. STZ treatment (saline infused) served to significantly (P < 0.05) reduce the weight of all fat pads examined. Insulin infusion into STZ-treated rats served to significantly (P < 0.05) elevate mWAT weight above non-diabetic levels (Fig. 4A), restore eWAT weight (Fig. 4C) and partially restore pWAT weight (Fig. 4D) to that of the citrate-injected controls, without affecting scWAT weight (Fig. 4B), in keeping with previous observations (Warne et al. 2006, 2007). Capsaicin treatment prevented the effects of STZ (saline-infused group) on mWAT weight (Fig. 4A).
Figure 4. White adipose tissue (WAT) weights of the mesenteric (mWAT) (A), subcutaneous (scWAT) (B) epididymal (eWAT) (C) and perirenal (pWAT) (D) WAT depots, normalized to grams body weight (bw), at the end of the study Insulin replacement of diabetic rats generally increased weight of all fat pads except scWAT. Capsaicin treatment of the STZ-saline-treated rats only served to increase mWAT weight and did not modify any effects of insulin. Treatment groups with different letters above the error bars are significantly (P < 0.05) different from each other (i.e. a is significantly different to b or c), groups with the same letters above the error bars are not significantly different groups (i.e. a is not significantly different to a).
Download figure to PowerPoint
Figure 5 shows plasma hormone levels and hormone-sensitive tissue weights. Plasma insulin (Fig. 5A) and leptin (Fig. 5B) levels were significantly (P < 0.001) reduced by STZ treatment with saline infusion, compared with citrate-injected controls. Insulin infusion into the STZ-injected rats elevated circulating insulin and leptin levels. Capsaicin treatment of the hepatic vagus had no effect on these hormones. Corticosterone levels (Fig. 5C) were unaffected by any insulin or hepatic vagal experimental manipulation, showing the efficacy of the corticosterone pellet.
Figure 5. Plasma insulin (A), leptin (B) and corticosterone (C) levels and adrenal (D), spleen (E) and thymus (F) weights STZ treatment increased spleen weight and reduced insulin and leptin levels, this reduction being partially attenuated by insulin replacement. Corticosterone levels, and consequently adrenal and thymus weights, were generally unaffected by any manipulation. Treatment groups with different letters above the error bars are significantly (P < 0.05) different from each other (i.e. a is significantly different to b and c), groups with the same letters above the error bars are not significantly different groups (i.e. a is not significantly different to a or ab).
Download figure to PowerPoint
The adrenal weights (Fig. 5D) were typically low, showing the efficacy of the corticosterone treatment; however, significant differences (P < 0.05) were observed between STZ-insulin capsaicin-treated group and the citrate and the STZ-saline vehicle-treated groups. No other differences were observed. Spleen weight (Fig. 5E) was significantly (P < 0.05) elevated by STZ injection and unaffected by insulin infusion or capsaicin treatment. In contrast, thymic weight (Fig. 5F) generally showed no differences among groups. However, the citrate-injected, capsaicin-treated group was significantly (P < 0.05) heavier than the citrate-injected vehicle-treated, STZ-saline vehicle-treated and STZ-insulin capsaicin-treated groups. Collectively, the organ weights presented in Fig. 5, together with the total caloric intake presented in Fig. 2 and body weight changes presented in Fig. 3, show the long-term efficacy of corticosterone treatment, STZ-induced diabetes and insulin replacement throughout the experiment and are validated by the terminal plasma measures of corticosterone and insulin (Fig. 5).
Plasma metabolites were affected by STZ treatment, insulin replacement and capsaicin treatment (Fig. 6). Capsaicin treatment did not affect any of the plasma metabolites or liver glycogen in the citrate-injected rats. STZ injection with saline infusion resulted in significant (P < 0.05) elevations in triglycerides (Fig. 6A), FFA (Fig. 6C) and total ketone bodies (Fig. 6D). Capsaicin treatment of the STZ-saline group further elevated triglycerides compared with STZ-saline, vagal vehicle-treated rats. In the vehicle-treated STZ-diabetic rats, insulin infusion reduced plasma triglycerides, FFA and total ketone bodies. Capsaicin treatment significantly (P < 0.05) attenuated the actions of insulin on FFA and total ketone bodies and partially attenuated the effects on plasma triglycerides.
Figure 6. Plasma triglycerides (A), glycerol (B), free fatty acid (FFA) (C) and total ketone bodies (D) STZ elevated circulating triglycerides, FFA and total ketone bodies. Insulin replacement reduced plasma triglycerides, FFA and total ketone bodies, effects of which were prevented by capsaicin treatment of the common hepatic vagus. Treatment groups with different letters above the error bars are significantly (P < 0.05) different from each other (i.e. a is significantly different to b and c), groups with the same letters above the error bars are not significantly different groups (i.e. a is not significantly different to a).
Download figure to PowerPoint
Plasma glucagon (Fig. 7A), liver glycogen (Fig. 7B) and plasma glucose (Fig. 7C) showed interrelated patterns of change with STZ, insulin and capsaicin treatments. Both plasma glucagon and glucose were significantly (P < 0.01) elevated and liver glycogen content was significantly (P < 0.01) reduced by STZ treatment. Capsaicin treatment of the common hepatic vagus further significantly (P < 0.05) increased plasma glucose and glucagon levels, whilst further reducing liver glycogen content. Insulin replacement partially reduced glucose levels and fully reduced glucagon levels to non-diabetic, citrate-injected, levels whilst significantly (P < 0.05) elevating liver glycogen content. Capsaicin treatment in this instance partially attenuated (P < 0.05) the actions of insulin on liver glycogen and glucagon and completely attenuated the actions of insulin on plasma glucose levels.
Figure 7. Plasma glucagon (A), liver glycogen content (B) and plasma glucose levels (C) STZ treatment elevated plasma glucagon and glucose levels whilst decreasing liver glycogen content. Capsaicin treatment further exacerbated these effects. Insulin replacement reduced both plasma glucose and glucagon whilst elevating liver glycogen content, an effect that was attenuated by capsaicin treatment of the common hepatic vagus. Treatment groups with different letters above the error bars are significantly (P < 0.05) different from each other (i.e. a is significantly different to b and c), groups with the same letters above the error bars are not significantly different groups (i.e. b and c are not significantly different to bc).
Download figure to PowerPoint
Table 1 compares the results obtained in this study with that of a previous study that utilized a similar model, in which the common hepatic branch of the vagus was cut (HV) as opposed to treated with capsaicin. This enables dissection of the effects of afferent and efferent fibres in the regulation of food intake, fat deposition and plasma variables, as outlined in the discussion.
Table 1. Comparison between the effects of capsaicin treatment of the common hepatic branch of the vagus and hepatic branch vagotomy (HV) on food intake, white adipose tissue (WAT) weights, hormones and metabolites in streptozotocin (STZ)-treated, high corticosterone rats infused with either saline or insulin
- Top of page
Our results show that capsaicin treatment of the common hepatic branch of the vagus promotes lard intake without affecting chow intake in STZ-diabetic, corticosterone-clamped rats. This suggests that afferent signalling through the hepatic vagus inhibits lard intake, an effect that can be overcome by insulin acting outside of the hepatic vagus since capsaicin did not modify the lard intake of non-diabetic rats with unmanipulated insulin or STZ-diabetic, insulin-replaced rats. The variable daily pattern of lard intake of the STZ-diabetic, capsaicin-treated group is probably associated with the few remaining afferent nerve fibres after capsaicin treatment that might produce a significantly attenuated inhibitory hepatic vagal signal. Our study also shows that this capsaicin-induced increase in lard intake leads selectively to fat deposition in the mWAT depot, as opposed to the increase in all omental fat depots evident after insulin replacement-induced lard intake. Capsaicin treatment also increased the levels of circulating glucagon, glucose and triglycerides and negated the actions of insulin on FFAs and total ketone bodies. These findings can be compared with those obtained using HV (Warne et al. 2007; see Table 1), to reveal the contribution of afferent versus efferent signalling through the common hepatic branch of the vagus.
For the majority of variables examined, non-diabetic rats showed no effect of capsaicin treatment. However, these rats had high, steady-state corticosterone levels which, consequently, elevated circulating insulin levels in comparison with those expected for intact, naïve rats with freely secreted corticosterone of a similar age and at a similar time of day (typically around 2 ng ml−1, e.g. La Fleur et al. 2005a). Tipping the important corticosterone: insulin balance (Strack et al. 1995) by manipulating insulin levels in groups of rats that otherwise share similar, steady-state corticosterone concentrations is probably the key factor in revealing the role of the hepatic vagus in the maintenance of metabolic homeostasis.
In addition to our findings, other studies have shown that afferent signalling is important in regulating fat intake. For example, rats with general autonomic afferent denervation by capsaicin over-eat high fat on at least the first test (Chavez et al. 1997). Furthermore, jejunal and portal infusions of lipids increase the activity of hepatic vagal afferents (Randich et al. 2001). This has been suggested to complement, or be an alternative to, celiac vagal afferent-mediated, lipid-induced inhibition of food intake (Cox et al. 2000, 2004; Randich et al. 2000). In this and our previous study (Warne et al. 2007; Table 1), chow intake balance changes with lard intake such that total intake is unaffected by any manipulations of the hepatic vagus. This supports other studies that have shown that the hepatic vagus does not influence total caloric intake (e.g. Louis-Sylvestre et al. 1980; Bellinger et al. 1984; Bellinger & Williams, 1988). Hence, the hepatic vagal afferents are responsive to dietary lipids and can signal to modify composition of the diet, but not total caloric intake.
Since capsaicin treatment of the common hepatic vagus does not modify lard intake in either non-diabetic or STZ-diabetic, insulin-replaced rats, this suggests insulin is acting via a mechanism independent of the hepatic vagus. A direct, central site of insulin action is a possible candidate, since there is widespread presence of the insulin receptors throughout the brain (Havrankova et al. 1978). Notably, two important sites involved in reward, namely the ventral tegmental area and the substantia nigra pars compacta, express insulin receptors (Figlewicz et al. 2003). Hence, there is capacity for central insulin to affect rewarding behaviour, such as choosing palatable food intake especially under the motivational influence of high corticosterone levels (Pecoraro et al. 2006). Indeed, choice is important considering that intracerebroventricular administration of high insulin concentrations can reduce palatable food intake (Figlewicz et al. 2004), which is not evident after a choice diet of saturated macronutrients (Van Dijk et al. 1997). In contrast, the arcuate nucleus is clearly the epicentre for the inhibitory actions of insulin, as well as leptin, on food intake (Niswender & Schwartz, 2003). Insulin and leptin actions at this site probably account for the reduction in total caloric intake of STZ-diabetic rats by insulin replacement, which also accounts for why total intake was unaffected by capsaicin treatment of the hepatic vagus.
Chronic stress and glucocorticoids result in elevated pleasurable food intake in both rodents (la Fleur et al. 2004; Pecoraro et al. 2004) and humans (Epel et al. 2001), although total caloric intake might be unaffected or even reduced (Gibson, 2006). In our model, high steady-state corticosterone levels were utilized in this study that mimic those observed with chronic stress. STZ-diabetes itself is a stressor, as evidenced by similar behavioural, autonomic, endocrine and neuroendocrine characteristics to chronic stress in rodents (Scribner et al. 1993); therefore it was essential to equalize corticosterone concentrations in all groups. In the presence of high corticosterone levels, but defective insulin production due to STZ treatment, pleasurable food intake is not stimulated. Hence, elevated insulin levels in concert with elevated corticosterone are probably responsible for the increased pleasurable food intake associated with chronic stress.
This and our previous studies have shown that STZ treatment reduces the weight of all WAT depots and insulin replacement into the superior mesenteric vein prevented, partially or totally, the decrease in mWAT, pWAT and eWAT weight, compared with saline-infused counterparts (Warne et al. 2006). The finding that insulin replacement increases mWAT weight to above that of non-diabetic levels suggests that there is an increased propensity towards deposition of fat into this depot after lard intake. A combination of local insulin action and corticosterone are probably responsible for this. Corticosterone redistributes fat from subcutaneous to omental (preferentially mWAT) depots in rats and humans (Thakore et al. 1998; Bell et al. 2000; Dallman et al. 2004, 2007). Our findings that scWAT was unaffected by insulin replacement support this notion. Since superior mesenteric-infused insulin traverses through mWAT, local action could prime this depot for receipt of lipids above the other omental depots. In contrast, the exclusive elevation in mWAT after capsaicin-induced lard intake suggests that this is the first site for fat deposition, probably due to its proximity to the sites of absorption. The lack of insulin ‘priming’ of other WAT depots might account for their lack of change in weight with lard intake under those conditions.
Capsaicin treatment of STZ-treated rats, irrespective of saline or insulin infusion, elevated plasma glucose, triglyceride and glucagon levels and reduced liver glycogen content. The overall levels were influenced, however, by prevailing insulin levels that regulate glucose and triglyceride uptake. Glucagon reduces liver glycogen content, consequently elevating glucose output (Greenberg et al. 2006), and stimulating adipocyte lipolysis (Schade et al. 1979), hence triglyceride release. Since neural innervation of the pancreas can regulate α-cell glucagon output (Havel & Taborsky, 1989), and the common branch of the hepatic vagus receives fibres from the pancreas (Berthoud, 2004), afferent signalling through the common hepatic branch of the vagus could signal to the brain for consequent manipulation of glucagon levels. This might complete a homeostatic feedback loop that limits extreme glucose and triglyceride output. An intact efferent signalling pathway from the brain through the hepatic vagus is clearly important since HV, which transects both afferent and efferent fibres, does not result in significant changes in plasma glucose or liver glycogen content (Warne et al. 2007; Table 1).
As shown in this study and others (e.g. Bartness & Rowland, 1983), insulin acts to prevent the elevation of circulating levels of FFA and total ketone bodies evident in STZ-diabetes. Capsaicin treatment negated the actions of insulin on these metabolites, which suggests that a component of afferent hepatic vagal signalling regulates these processes, corroborated by similar findings after HV for FFA (Warne et al. 2007; Table 1). Both release of FFA and inhibition of ketogenesis can be achieved by stimulation of sympathetic nerves (Beuers et al. 1986; Bartness et al. 2005). Insulin, acting via an afferent signal through the hepatic vagus, could signal the brain to modify these efferent sympathetic signals. Hence, the loss of afferent, parasympathetic signalling might eliminate a feedback loop by which insulin reduces FFA release and ketogenesis when these glucose alternative energy supplies are not needed.
Collectively, these data suggest that afferent signalling through the common hepatic branch of the vagus inhibits lard, but not chow, intake and regulates plasma metabolite levels depending on the insulin: corticosterone balance of the rat. Considering that chronic stressors are an omnipresent part of our current life and work styles (Wardle & Gibson, 2002; Kunz-Ebrecht et al. 2003) and might contribute to the obesity epidemic (Dallman et al. 2006), vagal manipulation could therefore represent a novel therapeutic avenue to regulate dysregulated or unhealthy (but palatable) food intake.