We have recently reported that inhibition of transforming growth factor (TGF)-β in the brain reduced fat-related energy substrates concentrations in response to exercise. We investigated the relevance between the mobilization of fat-related energy substrates (nonesterified fatty acid and ketone bodies) during exercise and the effects of TGF-β in the brain. Low-intensity exercise was simulated by contraction of the hindlimbs, induced by electrical stimulation at 2 Hz in anesthetized rats (Sim-Ex). As with actual exercise, it was confirmed that mobilization of carbohydrate-related energy substrates (glucose and lactic acid) occurred immediately after the onset of Sim-Ex, and mobilization of fat-related energy substrates followed thereafter. The timing of mobilization of fat-related substrates corresponded to that of the increase in TGF-β in cerebrospinal fluid (CSF) in Sim-Ex. The level of TGF-β in CSF significantly increased after 10 min of Sim-Ex and remained elevated until 30 min of Sim-Ex. Intracisternal administration of TGF-β caused rapid mobilization of fat-related energy substrates. Meanwhile, there were no effects on the changes in carbohydrate-related substrates. The levels of catecholamines were slightly elevated after TGF-β administration, and, although not significantly in statistical terms, we consider that at least a part of TGF-β signal was transducted via the sympathetic nervous system because of these increases. These data indicate that TGF-β in the brain is closely related to the mobilization of fat-related energy substrates during low-intensity exercise. We hypothesized that the central nervous system plays a role in the regulation of energy metabolism during low-intensity exercise and this may be mediated by TGF-β.
- central nervous system
- energy metabolism
- ketone bodies
- nonesterifed fatty acid
it has been well-established that the predominant fuel for exercise shifts from carbohydrates to fat during the course of prolonged endurance exercise (30). This change appears to indicate that fat utilization in working muscles is intensified in place of carbohydrate utilization to spare glycogen in the skeletal muscles and liver (17, 28). However, the regulation of fat utilization during endurance exercise is not necessarily dependent solely upon the residual content of glycogen (8, 9). In addition to this, many other factors, such as the concentrations of nonesterified fatty acid (NEFA; see Ref. 27), insulin (7), and glucagon (13, 39) in plasma and the tissue levels of high-energy phosphate and the activity of AMP-activated kinase (14, 31, 40), have been reported to be involved in the regulation of the metabolism of carbohydrate and fat. To date, most studies on the regulatory mechanisms of the selection of suitable fuel in response to a particular exercise condition have been elucidated with respect to peripheral tissues. However, the brain continuously monitors the energy status of the whole body, makes a plan of exercise, and commands the execution of actual exercise. These facts led us to postulate that the central nervous system also plays an important role in the regulation of energy metabolism during exercise. The skeletal muscles, which drive actual exercise, appear to have a privilege in the consumption of energy fuel during exercise, and the liver plays a pivotal role in supply of energy substrates. Both organs are considered to autonomously regulate energy metabolism to a certain degree. Meanwhile, the brain comprehends the energy status of every part of the body, coordinates the energy demand of each part, and appropriately regulates the metabolism of the whole body. For example, the involvement of the hypothalamus and sympathetic nervous system in the regulation of energy metabolism at high-intensity exercise has already been reported (32–35). With a similar exercise, increases in the expression of corticotropin-releasing factor (CRF) mRNA have also been demonstrated (36). These findings imply that high-intensity exercise is a kind of stressor. On the other hand, the mechanisms of increased fat metabolism during prolonged low-intensity endurance exercise has not been elucidated in detail.
Transforming growth factor-β (TGF-β), a pleiotropic cytokine, regulates cell proliferation, differentiation, and apoptosis and has a key role in development and tissue homeostasis. However, the physiological role of TGF-β in the brain has been unclear. We previously reported the increase of active TGF-β in the cerebrospinal fluid (CSF) after endurance exercise in rats, and TGF-β in the brain might be involved in the manifestation of the feeling of fatigue (21, 24). In addition, intracisternal administration of TGF-β to sedentary rats resulted in a decrease in respiratory exchange ratio, an increase in the rate of whole body fat oxidation, and increases in the concentrations of fat-related energy substrates in the blood, i.e., ketone bodies and NEFA (41). These changes appear to be features that are similar to those observed in endurance exercise. Recently, we have demonstrated the inhibition of increase in fat oxidation during endurance exercise by blockade of TGF-β in rat brains (22). From these results, we hypothesized that TGF-β in the brain is related to the central regulation of fat metabolism during exercise. In this study, we utilized an exercise model in anesthetized rats, which elicited contraction of the skeletal muscle through electric stimulation, to simulate low-intensity endurance exercise and elucidated the time course of changes in the flux of energy substrates through the lower body and the liver and the changes in the concentration of active TGF-β in CSF. Furthermore, the changes in the flux of energy substrates caused by intracisternal administration of TGF-β were compared with those elicited by exercise simulation.
MATERIALS AND METHODS
This study was conducted in accordance with the ethical guidelines of the Kyoto University Animal Experimentation Committee and was in complete compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize the number of animals used and limit experimentation to that which was necessary to produce reliable scientific information. Male Wistar rats (8–9 wk old; CLEA Japan) were maintained on an inverse light-dark cycle (light on 1800 and light off 0600) for 1 wk to make them active during the experimental period (i.e., daytime). They were housed in standard plastic cages (26 × 42 × 21 cm; 2 or 3 rats/cage) at a room temperature of 22 ± 0.5°C and humidity of 55 ± 5%. Rats were fed a commercial diet (MF; Oriental Yeast, Tokyo, Japan) and water ad libitum except for the days of the operation or treatment.
TGF-β3 was purchased from R&D Systems (Minneapolis, MN). TGF-β3 (40 ng) was dissolved in 40 μl of artificial CSF (140 mM NaCl, 3 mM KCl, 1.5 mM Na2HPO4, 0.23 mM NaH2PO4, 1.5 mM CaCl2, 1.26 mM MgCl2, 3.4 mM d-glucose, and 0.1% BSA) containing 0.5 mM HCl and was administered in the cisterna magna of rats (TGF-β group). An equal volume of vehicle was used as a control (vehicle group). Both TGF-β and vehicle solution pH were adjusted to 7.4.
Implantation of Blood-Sampling Cannulas in the Target Vessels
Rats were anesthetized with isoflurane during the cannulation procedure. Hydrocoat catheters (Access Technologies) filled with 0.1% heparinized saline were inserted in the right atrium via the jugular vein, the hepatic vein, and the inferior vena cava, and the end of a catheter was located at the confluence of the right and the left femoral veins. With respect to the cannulation in the hepatic vein, the left hepatic lobe was jacked up, and the cannula was carefully inserted 5 mm in the left hepatic vein, which was connected with the left hepatic lobe and inferior vena cava up the liver. The composition of energy substrates in blood collected from the right atrium was considered to be similar to that in arterial blood. Blood collected from the inferior vena cava was designated as blood from the lower body, which mainly reflects that from the skeletal muscles and adipose tissues of the hindlimbs. After insertion of the cannulas, rats were housed in special small cages (Ballman immobilizing cage) so as to not disturb any of the implanted catheters. Any rats that ingested <60% of the normal food intake were excluded from the examination. The catheters were flushed with saline without heparin before the day of the experiment to avoid any confounding effects of heparin. The placement of each catheter was verified at autopsy.
Implantation of Cannulas in the Cisterna Magna
The surgical approach for implantation of a cannula for the collection of CSF and administration of TGF-β was described previously (41). Briefly, an anesthetized rat was fixed to a stereotaxic apparatus, and a stainless steel guide cannula (23 gauge) was set at 3.0 mm posterior to lambda. It was inclined anteriorly at angle of 60° and inserted 8.7 mm deep. The cannula was fixed with Loctite 454 (Loctite Japan, Yokohama, Japan). The rats were allowed to recover for 1 wk after surgery.
Simulated Exercise by Electric Stimulation of Rat Hindlimb Skeletal Muscles
The skeletal muscles were stimulated electrically according to the method by Inoue et al. (19, 20) with a modification of the stimulation frequency. Needle-type electrodes, which were attached to an electrical stimulator (type SEN-2101; Nihon Kohden, Osaka, Japan), were inserted in both the gastrocnemius and the quadriceps muscles of both legs of rats under isoflurane anesthesia. Two electrodes were used for each muscle. The tibia and foot of each leg were fixed at right angles to each other with a knee clamp and a foot strap and stimulated electrically at 25 volts, 2 Hz with rectangular pulses of 2-ms duration for 30 min. Body temperature was maintained at 37°C with a heating pad.
Experiment 1: Determination of the flux of energy substrates through the liver and the lower body during simulated exercise.
Three days before the experiment, rats were implanted with cannulas in the cisterna magna. Two days before the experiment, cannulas were inserted in the jugular vein, the hepatic vein, and the inferior vena cava. On the experimental day, rats were fasted for 120 min before the onset of simulated exercise (Sim-Ex) and anesthetized with isoflurane and were then fixed to the apparatus for Sim-Ex.
Before and after 5, 10, 20, and 30 min of Sim-Ex, blood samples were successively collected from the right atrium, hepatic vein, and the inferior vena cava. Plasma and supernatants from deproteinized blood were isolated by centrifugation and stored at −70°C until analysis. After the completion of 30 min of Sim-Ex, the liver, the gastrocnemius, and quadriceps muscles were removed, freeze-clamped in liquid nitrogen, and stored at −70°C. Sham-stimulated rats were used as controls.
Experiment 2: Determination of the changes in the level of TGF-β in the CSF during Sim-Ex.
Six days before the experimental day, rats were implanted with an intracisternal cannula to collect CSF. On the experimental day, rats were fasted for 120 min before the onset of Sim-Ex, and 100-μl samples of the CSF were collected after 5, 10, 20, and 30 min of Sim-Ex. Three days before the experimental day, a CSF sample was collected from each rat. The level of TGF-β in this CSF sample was used as the basal (pre-Sim-Ex) value.
Experiment 3: Effects of intracisternal administration of TGF-β on the flux of energy substrates through the liver and the lower body.
Rats were treated identically as in experiment 1 before intracisternal administration of TGF-β. Before and 5, 10, 20, 30, and 60 min after intracisternal administration of TGF-β, blood samples were successively collected from the right atrium, hepatic vein, and the inferior vena cava. In a separate group of rats, whole blood, liver, gastrocnemius, and quadriceps muscles were collected before and 15, 30, and 60 min after intracisternal administration. Plasma was isolated by centrifugation and stored at −70°C until analysis.
In experiments 1 and 3, the flux of energy substrates through the liver and the lower body were calculated by subtracting the concentration in the right atrium from that of the hepatic vein and the inferior vena cava.
Experiment 4: Effects of intracisternal administration of TGF-β on the concentrations of catecholamines in plasma.
Three days before the experiment, rats were implanted with cannulas in the cisterna magna. On the experimental day, rats were treated identically as in experiment 3. Before and 5, 10, 30, and 60 min after intracisternal administration of TGF-β, blood samples were successively collected from the right atrium via cannulas in the jugular vein. Plasma was isolated by centrifugation and stored at −70°C until analysis.
Analysis of Energy Substrates and Catecholamines
The following substrates were measured in plasma by using appropriate assay kits: glucose (glucose CR-II; Wako Pure Chemical Industries, Osaka, Japan), NEFA (NEFA C; Wako Pure Chemical Industries), ketone bodies (ketone test; Sanwa Chemical Institute, Nagoya, Japan). Blood lactic acid was measured by using Determiner LA (Kyowa Medics, Tokyo, Japan). The muscle and liver glycogen contents were measured as glucose residues after hydrolysis of the samples in 1 M HCl at 100°C for 2 h according to the method of Hassid and Abraham (16). To measure the concentrations of catecholamines, plasma was treated with alumina at pH 8.6. After alumina were washed with water, catecholamines were eluted with 0.4 N perchloric acid and assayed using HPLC-ECD (HTEC-500; Eicom, Kyoto, Japan).
Determination of the Level of Active TGF-β in the CSF
TGF-β is produced as an inactive form that is referred to as latent TGF-β (23). The amino-terminal part of a protein called a latency-associated protein (LAP) produces the latent TGF-β by masking the active site of TGF-β. The dissociation of LAP by an as yet undetermined mechanism leads to the activation of the latent form of TGF-β. We determined the concentration of active TGF-β in CSF samples.
To determine the concentration of TGF-β in the CSF, we conducted a bioassay using TGF-β-responsive mink lung epithelial cells (TMLCs; see Ref. 1). This cell line was comprised of mink lung epithelial cells that were stably transfected with the TGF-β-responsive human plasminogen activator inhibitor 1 promoter fused to a luciferase reporter gene (kindly provided by Dr. M. Abe, Department of Nanomedicine, Tokyo Medical and Dental University and Dr. D. Rifkin, Department of Cell Biology, New York University Medical Center). All samples were diluted with DMEM containing 0.1% BSA. To establish a standard curve, purified TGF-β (R&D Systems) was used. TMLCs suspended in DMEM with 10% FBS were seeded on 96-well plates (1 × 104 cells/well) and allowed to attach for 6 h; the medium was then replaced with a 100-μl aliquot of the sample solution. After the medium-sample exchange (16 h), luciferase activity was measured using a Veritasmicroplate luminometer (Promega) according to the manufacturer's instructions. All experiments were performed in triplicate. The levels of active TGF-β in the CSF are expressed as relative luminescence units.
Measurement of Blood Flow Rate
Blood flow rates of the aorta and the inferior vena cava were determined by a Transit Time Ultrasound blood flowmeter T402 (Transonic Systems) with 1.0 and 0.5 mm diameter flow probes at 37°C. The blood flow rate of the aorta was designated as the arterial blood flow. The blood flow rate was determined using a different subset of rats from which blood was not collected because simultaneous blood flow rate determination and blood collection from a single rat was not appropriate for the precise measurement. The blood flow rates were examined for 30 min after intracisternal administration of TGF-β.
Data are expressed as means ± SE. Statistical analyses of the differences between before and after treatment (Sim-Ex and TGF-β administration) in the same group were performed with one-way repeated-measures ANOVA followed by post hoc Dunnett's test (experiment 1) and Dunnett's multiple-comparison test (experiment 3). Comparisons of different groups were conducted using two-way repeated-measures ANOVA followed by a post hoc Student's t-test at each time point (experiment 1 and experiments 3 and 4). In the measurement of the levels of active TGF-β, paired t-test was used for statistical analysis of differences between before and during Sim-Ex in the same rats (experiment 2). Analyses were performed using the Prism software package (Graph pad software), and P values < 0.05 were considered significant.
Experiment 1: Effects of Sim-Ex on the Flux of Energy Substrates Through the Liver and the Lower Body
Figure 1 presents the changes in the concentrations of atrium lactic acid in the blood during Sim-Ex by electric stimulation of hindlimb muscles at 2 Hz. In the present experimental condition, the basal concentration of atrium lactic acid was 1.12 ± 0.15 mM. The concentration of atrium lactic acid increased rapidly, ∼2.5-fold, with the onset of Sim-Ex, and this increase was sustained throughout the duration of the exercise period (P < 0.01 by one-way repeated-measures ANOVA; Fig. 1). The atrium lactic acid reached 2.69 ± 0.30, 2.93 ± 0.38, 2.92 ± 0.45, and 2.61 ± 0.39 mM after 5, 10, 20, and 30 min of Sim-Ex, respectively.
Figure 2 presents the effects of Sim-Ex on the flux of energy substrates in plasma. The output of lactic acid from the lower body increased significantly after 5 min of Sim-Ex (P < 0.05 by one-way repeated-measures ANOVA; Fig. 2A). The output of lactic acid from the lower body gradually decreased but remained slightly higher than that before Sim-Ex. The uptake of lactic acid in the liver also increased significantly after 5 min of Sim-Ex (P < 0.05 by one-way repeated-measures ANOVA; Fig. 2B). The output of glucose from the liver increased significantly after 10 and 20 min of Sim-Ex (P < 0.05 by one-way repeated-measures ANOVA; Fig. 2C). Although the output of ketone bodies from the liver decreased transiently after 5 min of Sim-Ex, it gradually increased from 5 to 30 min (Fig. 2D). However, this increase was not significant. The output of NEFA from the lower body did not change up to 20 min of Sim-Ex, but significantly increased at 30 min (P < 0.05 by one-way repeated-measures ANOVA; Fig. 2E). The influence of 30 min of Sim-Ex on the glycogen content in the liver and the hindlimb muscles is shown in Table 1. Sim-Ex by electric stimulation at 2 Hz for 30 min significantly attenuated the glycogen content in the quadriceps and tended to reduce that in the liver and the gastrocnemius (P < 0.05 by Student's t-test; Table 1). These results suggest that Sim-Ex by electric stimulation of the hindlimb muscles at 2 Hz was almost comparable to low-intensity endurance exercise.
Experiment 2: Effect of Sim-Ex on the Level of Active TGF-β in the CSF
Under the same conditions as experiment 1, CSF samples were collected after 5, 10, 20, and 30 min of Sim-Ex. Figure 3 presents the effect of Sim-Ex on the levels of active TGF-β in the CSF. The mean basal concentration of TGF-β in the CSF was 39.73 ± 7.81 pg/ml. Sim-Ex by electric stimulation gradually increased TGF-β in the CSF. The levels of TGF-β in the CSF after 10, 20, and 30 min of Sim-Ex were significantly higher than basal (before Sim-Ex) in each rat (P < 0.05 by paired t-test; Fig. 3). Mean increments of the level of TGF-β compared with basal were as follows: 5 min, 2 ± 11%; 10 min, 52 ± 18%; 20 min, 116 ± 23%; and 30 min, 108 ± 39%.
Experiment 3: Effects of Intracisternal Administration of TGF-β on the Flux of Energy Substrates Through the Liver and the Lower Body
With respect to the flux of carbohydrate-related energy substrates, intracisternal administration of TGF-β elicited no effect, and the changes in the output of glucose from the liver and the uptake of glucose in the lower body were similar in both TGF-β-administered and vehicle groups (Fig. 4, A and B). The uptake of lactic acid in the liver and the output of lactic acid from the lower body were also similar in both groups (Fig. 4, C and D). On the other hand, with respect to the flux of fat-related energy substrates, the output of ketone bodies from the liver in the TGF-β group increased significantly after 5 and 10 min compared with the vehicle group (P = 0.053 by two-way repeated-measures ANOVA: TGF-β vs. vehicle; P < 0.05 at 5 and 10 min by Student's t-test: vs. vehicle; Fig. 5A). There was no change in the uptake of ketone bodies in the lower body and the uptake of NEFA in the liver between the two groups (Fig. 5, B and C). The output of NEFA from the lower body tended to increase in the TGF-β group, but this increase was not statistically significant (Fig. 5D). Table 2 presents the changes in the concentrations of energy substrates in plasma after intracisternal administration of TGF-β under isoflurane anesthesia. There was no change in the glucose concentration between the two groups (Table 2). The plasma concentration of ketone bodies in the TGF-β group at 15 min increased significantly compared with that before TGF-β administration and that in the vehicle group at 15 min (P < 0.05 at 15 min by Dunnett's multiple-comparison test: vs. before administration; P < 0.05 at 15 min by Student's t-test: vs. vehicle; Table 2). The plasma concentration of NEFA increased significantly in the TGF-β group at 30 min compared with that before TGF-β administration and increased significantly in the TGF-β group after 15 and 30 min compared with the vehicle group at 15 and 30 min (P < 0.05 at 30 min by Dunnett's multiple-comparison test: vs. before administration; P < 0.05 at 15 and 30 min by Student's t-test: vs. vehicle; Table 2). Intracisternal administration of TGF-β elicited no changes in the glycogen content of the liver, gastrocnemius, and the quadriceps (Table 3) and blood flow rate in the aorta and the inferior vena cava (Table 4).
Experiment 4: Effects of Intracisternal Administration of TGF-β on the Concentrations of Catecholamines in Plasma
Figure 6 presents the time course changes in the concentrations of atrium norepinephrine and epinephrine in plasma before and after intracisternal administration of TGF-β. Norepinephrine and epinephrine concentrations in plasma did not change significantly after TGF-β administration compared with vehicle and before administration (Fig. 6).
We have previously reported the increase in active TGF-β in rat CSF during exercise (21, 24). Intracisternal administration of TGF-β enhances fat metabolism (41). These results led us postulate that TGF-β in the brain is related to the regulation of fat metabolism by the central nervous system. We have already demonstrated suppression of the increase in fatty acid oxidation during treadmill running by immunoneutralization of TGF-β in rat brains (22).
In this study, we elucidated whether TGF-β elicits changes in energy metabolism similar to those during endurance exercise by measuring the changes in the concentration of energy substrates in blood. In particular, we paid attention to metabolic changes in the skeletal muscles and the liver, which are important in execution of exercise. Specifically, we examined the changes in uptake and release of energy substrates and metabolites through these organs as a result of intracisternal administration of TGF-β compared with those caused by simulation of actual exercise (Sim-Ex).
In experiment 1, the level of atrium lactic acid was sustained at a higher concentration (∼2.8 mM) during Sim-Ex (Fig. 1). Because this level of lactic acid is slightly lower than the lactate threshold (5, 38), the exercise state in Sim-Ex appeared to correspond to low-intensity exercise. Therefore, we considered the changes in the concentrations of the energy substrates to be similar to those observed in low-intensity exercise. The output of lactic acid from the lower body would be derived primarily from the electrically contracting muscles, and its concentration reached the highest point immediately after the onset of Sim-Ex. By contrast, the uptake of lactic acid in and the output of glucose from the liver reached a maximum immediately after the onset of Sim-Ex. These increases are considered to reflect the enhancement of gluconeogenesis and glycogenolysis. Meanwhile, with respect to fat-related substrates, the output of NEFA from the lower body gradually increased ∼10 min after the onset of Sim-Ex (Fig. 2E). These NEFA would be derived from the adipose tissues distributed within the lower body. The output of ketone bodies from the liver decreased transiently immediately after the onset of Sim-Ex and reverted to the initial level thereafter (Fig. 2D). It is also known that the output of ketone bodies increases after prolonged medium-intensity exercise (18, 25). Although the time course of the changes in the concentration of ketone bodies from the lower body was slightly different, the characteristic pattern of increases in energy substrates, i.e., the delayed recruitment of fat-related substrates compared with that of carbohydrate-related substances, was essentially reproduced in Sim-Ex.
With respect to the flux of energy substrates, we should consider not only the changes in concentration of each substrate in blood but also the changes in blood flow. In medium-intensity exercise, no changes were reported in the renal and portal venous blood flow in rats (18) and that in the hepatic artery and the portal vein in dogs (3). In our Sim-Ex investigation, electric stimulation of rat hindlimb muscles at 5 Hz, which is higher than the 2 Hz used in this study, increased the heart rate by 30 beats/min, and the percentage of increase was only 5% of the resting heart rate (Mizunoya W, Okabe Y, Shibakusa T, Inoue K, and Fushiki T, unpublished observation). Therefore, the changes in blood flow during low-intensity exercise appear to be smaller than the changes in the concentration of the energy substrates. No changes were observed in the blood flow of the aorta and the inferior vena cava after intracisternal administration of TGF-β (Table 4). According to these data, we could consider the energy flux through organs with respect to the differences in the concentration of each substrate without considering blood flow.
With respect to the technical consideration of anesthesia, we used isoflurane in this study. Nishiyama et al. (26) indicated that isoflurane has effects on plasma glucose and catecholamine levels during prolonged operation and found significant increases at 5 h and thereafter in human. However, in our experimental condition, basal concentrations of plasma glucose were slightly higher (experiment 3: atrium glucose of vehicle-administrated group, 0 min, 195.3 ± 10.8 mg/dl), but no significant increase was found at the end of experiment (30 min, 205.2 ± 18.3 mg/dl), and the concentrations of catecholamine were not changed (Fig. 6).
The time course of the change in concentration of active TGF-β in CSF under this condition was very intriguing. The concentration of active TGF-β in CSF increased significantly 10 min after the onset of Sim-Ex (Fig. 3), and this timing corresponded to the increase in the output of NEFA from the lower body and recovery of the output of ketone bodies from the liver. Moreover, this timing was also consistent with the manifestation of inhibitory effects of anti-TGF-β antibodies in the brain on fatty acid oxidation during treadmill running of rats (22). However, at present, it is unclear what induced the elevation in TGF-β during exercise. We can consider several possibilities as the main cause of the elevation by Sim-Ex, as follows: 1) the changes in the plasma concentrations of energy substrates (e.g., the elevated lactic acid or reduced ketone bodies concentrations), 2) the changes in the tissue levels of high-energy phosphate, the activity of AMP-activated kinase, or the concentration of glycogen in working muscles, or 3) the alterations in the intensity or duration of muscle contraction. Considering the effect of intracisternal administration of TGF-β on the enhancement of fatty acid metabolism (41), we investigated whether the increase in the concentration of TGF-β in CSF during Sim-Ex was related to the recruitment of fat-related energy substrates in Sim-Ex.
Under the same condition as Sim-Ex, TGF-β was administered intracisternally to the rats, instead of inducing skeletal muscle contraction by electric stimulation. No changes in carbohydrate-related substrates, i.e., flux of glucose and lactic acid through each organ (Fig. 4, A and D) and glycogen content in the liver (Table 3), were observed. On the other hand, with respect to fat-related substrates, the output of ketone bodies from the liver (Fig. 5A) and that of NEFA from the lower body (Fig. 5D) increased immediately after the administration of TGF-β. As for ketone bodies, because there was no change in NEFA incorporation in the liver (Fig. 5C), the rate of production of ketone bodies from NEFA would be caused by the enhancement of β-oxidation of fatty acids. The increase in the output of NEFA from the lower body was affected by individual variation, and we were not able to detect statistical significance. We then confirmed the changes in the concentration of energy substrates in the blood samples obtained by decapitation under the same condition of administration (Table 2). Similar to experiment 3, intracisternal administration of TGF-β elicited no changes in the concentration of blood glucose but caused increases in the concentrations of ketone bodies and NEFA. Concerning NEFA, the samples collected by decapitation reflect the responses of whole adipose tissues in the body, and these changes may be cumulative compared with the partial response of the lower body. These data also suggested that the intracisternal administration of TGF-β affected the whole body. Whether NEFA is reesterified to triacylglycerol or degraded by β-oxidation in the liver is determined by the energy level of the whole body. In such a mechanism, TGF-β in the brain may function as a signal for the enhancement of fatty acid oxidation. However, the mechanism by which this signal is transmitted to the liver has not been elucidated yet. To compare the difference in the effects between Sim-Ex and TGF-β on the recruitment of fat-related substrates, the time course change in the output of ketone bodies from the liver and that of NEFA from the lower body was examined (Fig. 7), with the onset of Sim-Ex and the administration of TGF-β set at time 0. The timing of the output of fat-related substrates following the intracisternal administration of TGF-β was clearly indicated to be much earlier than that with Sim-Ex. These data indicate that intracisternal administration of TGF-β stimulated utilization of fat-related energy substrates and suggested that the increase in the concentration of TGF-β in the brain would be related to the recruitment of these substrates in Sim-Ex.
CRF is closely related to the stress response (2), and this factor has also been reported to be involved with exercise (36). Intracerebroventricular administration or microinjection in the hypothalamus of CRF and its related peptide urocortin caused an increase in the blood concentrations of energy substrates and catecholamines and the oxygen consumption (10–12, 15). Scheurink et al. (32–35) reported the hypothalamic and sympathetic regulation of energy metabolism during high-intensity exercise. In their exercise model, the rapid increases in glucose and NEFA in blood were observed, and it is anticipated that CRF is involved. On the other hand, we demonstrated that intracisternal administration of TGF-β was different from the effects of CRF and urocortin, in that it did not affect the carbohydrate metabolism and only enhanced fat metabolism. In addition, we also reported no effect of intracisternal administration of TGF-β on oxygen consumption (41). Thus the effects of TGF-β in the brain would be different from that of CRF. The amount of fuel for exercise and from which substrate energy will be acquired are determined by the relative intensity of the exercise (29). CRF will play a pivotal role even in high-intensity exercise because it is known to function in severely stressful conditions. On the other hand, TGF-β in the CSF has been observed to increase during relatively low-intensity exercise, such as Sim-Ex, and it is considered to function for the mobilization of only fat-related energy substrates; therefore, TGF-β in the brain may contribute to the energy regulation during low-intensity endurance exercise.
It is now known that TGF-β holds key roles in the regulation of neuron survival and orchestration of repair processes in the central nervous system (4). In animals and humans, TGF-β is widely expressed in almost all brain regions. TGF-β provokes their cell-type-specific responses through the ligand-induced formation of a heteromeric receptor complex between TGF-β receptor I (TβR-I) and TGF-β receptor II (TβR-II). Similarly, throughout the central nervous system, these TGF-β receptors can be found in neurons and glial cells (37). It is very interesting to examine how TGF-β in the brain causes enhancement of fatty acid metabolism in the whole body. However, the sites of action of TGF-β in the brain and the signal transduction pathway to peripheral tissues have not yet been elucidated. Although TGF-β caused slight increases in plasma norepinephrine and epinephrine levels at 10 min after administration compared with those at 0 min, no significant differences in plasma catecholamine concentrations were observed between the TGF-β and vehicle groups (Fig. 6). We considered that at least a part of TGF-β signal was transducted via the sympathetic nervous system because the levels of plasma catecholamines were elevated although they were not statistically significant. When sympathetic postganglionic nerves are stimulated at frequencies low enough to be comparable to physiological conditions, very little norepinephrine overflows in the circulation (6). The signal caused by TGF-β through the sympathetic nervous system might not be too strong to be reflected as the significant increase in the concentrations of catecholamines in the blood. In addition, the recruitment of fat-related energy substrates by intracisternal administration of TGF-β was relatively acute. The result also implies the possibility that the effects of TGF-β on peripheral tissues may be transmitted directly via the sympathetic nervous system, not via any hormonal factors. However, it is interesting that intracisternal administration of TGF-β did cause solely the mobilization of fat-related substrates, not that of carbohydrate-related substrates. Scheurink et al. (32) reported that infusion of phentolamine, an α-adrenoceptor antagonist, in the ventromedial hypothalamus (VMH) markedly enhanced the exercise-induced increase in plasma NEFA but reduced the blood glucose concentration. TGF-β might elicit its effect via specific activation of β-adrenoceptor or inactivation of α-adrenoceptor in the VMH. In our unpublished study, TβR-I and TβR-II can be found in the hypothalamus, including VMH, arcuate hypothalamus, and especially paraventricular hypothalamus; these regions are known to be related to control of energy metabolism.
A more detailed study should be conducted to elucidate the mechanism of mobilization in fat-related substrates by TGF-β in the brain.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society