Anti-hypertensive treatment preserves appetite suppression while preventing cardiovascular adverse effects of tesofensine in rats†
Article first published online: 20 JUN 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 5, pages 985–992, May 2013
How to Cite
Bentzen, B. H., Grunnet, M., Hyveled-Nielsen, L., Sundgreen, C., Lassen, J. B. and Hansen, H. H. (2013), Anti-hypertensive treatment preserves appetite suppression while preventing cardiovascular adverse effects of tesofensine in rats. Obesity, 21: 985–992. doi: 10.1002/oby.20122
Disclosure: The authors declared no conflict of interest.
- Issue published online: 20 JUN 2013
- Article first published online: 20 JUN 2013
- Accepted manuscript online: 5 NOV 2012 04:23PM EST
- Manuscript Accepted: 10 SEP 2012
- Manuscript Received: 18 JAN 2012
Tesofensine is a novel triple monoamine reuptake inhibitor which is in development for the treatment of obesity. Preclinical and clinical data suggest that appetite suppression is an important mechanism by which tesofensine exerts its robust weight reducing effect. Notably, the strong hypophagic response to tesofensine treatment is demonstrated to be linked to central stimulation of noradrenergic and dopaminergic neurotransmission. The sympathomimetic mode of action of tesofensine may also associate with the elevated heart rate and blood pressure observed in clinical settings, and we therefore sought experimentally to address this issue.
Design and Methods:
The anorexigenic and cardiovascular effects of tesofensine were studied simultaneously in telemetrized conscious rats in a combined real-time food intake and cardiovascular telemetry monitoring system.
Acute administration of tesofensine caused a dose-dependent hypophagic effect as well as increased heart rate and blood pressure. Interestingly, combined treatment with metoprolol (b1 adrenoceptor blocker, 10-20 mg/kg, p.o.) fully prevented the cardiovascular sympathetic effects of tesofensine while leaving the robust inhibitory efficacy on food intake unaffected. Similarly, the angiotensin AT1 receptor antagonist telmisartan (1.0-3.0 mg/kg, p.o.) did not interfere with the anti-obesity effects of tesofensine, however, telmisartan only partially reversed the increase in systolic blood pressure and had no effect on the elevated heart rate induced by tesofensine.
These data suggests that tesofensine causes elevations in heart rate and blood pressure by increasing sympathetic activity, and that different adrenoceptor subtypes may be responsible for the anti-obesity and cardiovascular effects of tesofensine.
Tesofensine is a novel centrally acting triple monoamine reuptake inhibitor (MRI) with intrinsic inhibitory activity on noradrenaline, serotonin, and dopamine transporter function (1). When corrected for placebo and diet effects, long-term tesofensine treatment produces a weight loss of 10.6% in obese patients which is twice that achieved by currently marketed anti-obesity drugs (2, 3). The anti-obesity effect of tesofensine is likely explained by dose-dependent hypophagia due to stimulation of satiety, suggesting that tesofensine acts as an appetite suppressant to produce a negative energy balance (2, 4, 5). In addition, tesofensine is also demonstrated to increase nocturnal energy expenditure in human subjects (5). These findings have recently been corroborated and extended in preclinical settings, demonstrating that tesofensine induces a robust and sustained weight loss in a rat model of diet-induced obesity (DIO) of which the long-lasting drop in body weight is caused by appetite suppression with a gradually increase in energy expenditure (6). Notably, the hypophagic effect of tesofensine in DIO rats is critically dependent on stimulated α1 adrenoceptor activity, and to a less extend dopamine D1 receptor function, indicating that enhancement of central noradrenergic and dopaminergic neurotransmission constitute important mechanisms underlying the robust appetite-suppressing effect of tesofensine (7).
Overall, chronic tesofensine treatment is associated with minor adverse events with minimal cardiovascular effects, suggesting that tesofensine may be a well-tolerated long-term treatment for obesity. However, in this regard, dose-dependent elevations in heart rate and significant increases in blood pressure at the highest dose tested has been reported in obese individuals (2, 8). We, therefore, speculated whether the sympathomimetic effects of tesofensine might also associate with the reported effects on cardiovascular function. To address this hypothesis, we simultaneously monitored effects on food intake and body weight regulation in conjunction with cardiovascular parameters in telemetrized rats following administration of tesofensine alone, or in drug combinations of tesofensine with antihypertensive agents representing different mechanisms of action, i.e., metoprolol (β1 adrenoceptor antagonist) and telmisartan (angiotension AT1 receptor antagonist).
Materials and Methods
Animal care and housing
Five-month-old male normotensive Sprague-Dawley rats (508 ± 18 g, Harlan, Horst, The Netherlands) were housed in solid bottomed Plexiglas cages with dust-free wood chippings and a cardboard tube. Holding rooms were maintained under a 12-h light/dark cycle (lights off: 1500 h). Ambient temperature was 18.0-22.0°C and relative air humidity of 40-60%. The rats had ad libitum access to standard chow (Altromin GmbH, Lage, Germany) and water. All experiments were approved and conducted in accordance with the guidelines of the Animal Experimentation Inspectorate, Ministry of Justice, Denmark.
Simultaneous real-time feeding and cardiovascular monitoring in telemetrized rats
The rats were implanted at Harlan laboratories (Horst, The Netherlands) with Data Science International (DSI, St. Paul) Physiotel PA-C40 transmitters according to the manufacturer's description. In brief, the rats were anesthetized with isoflurane, ventilated, and laparotomy was performed under aseptic conditions. A pressure catheter was inserted and sealed in place with Vetbond (3M, St. Paul) into the isolated abdominal aorta. Finally, the transmitter was placed on top of the intestines, in parallel to the long axis of the body, and secured to the abdominal wall, where after the abdominal muscle layer and skin was closed with solvable sutures. The animals were allowed full postsurgical recovery before shipment. Blood pressure (systolic and diastolic arterial blood pressure) and heart rate (pulse rate) data were collected at a sampling rate of 500 Hz using Dataquest A.R.T (v.4.3) and Ponemah software (v.5.0) (DSI) using factory-provided calibration values for the individual transmitters and an Ambient Pressure Reference Monitor (DSI) to ensure accurate blood pressure measurements. Data were collected continuously for 48 h and binned in 5 s intervals.
Real-time feeding monitoring in telemetrized rats
Upon 2-3 weeks of postsurgery recovery, the rats were transferred to fully automated food intake monitoring cages (HM-2, MBRose, Faaborg, Denmark) modified to simultaneously determine individualized food intake (by microchip, see below) and cardiovascular condition (by telemetry). For combined telemetry analysis, two receivers (RPC-1, DSI) were placed in the bottom of each HM-2 food intake monitoring cages fully covering the cage surface area. The modified food intake monitoring cages were placed in a ventilated cabinet with lightproof doors and a light kit for control of the light-dark cycle (Scanbur BK, Karslunde, Denmark) being similar to that in the holding rooms. The animals were habituated to the food intake monitoring system for at least 7 days before initiation of drug treatment procedures. Before re-housing to the fully automated food intake monitoring cages, the rats were subcutaneuosly injected with a microchip (#402575, eVet, Haderslev, Denmark) to simultaneously identify and in real-time mode track feeding behavior of each individual animal throughout the entire experiment. Locomotor activity was detected by an integrated infrared sensor placed above the cage. Standard control unit settings are reported previously (7). All drugs were administered 30 min before dark onset and data were collected continuously for 48 h using a data reporting software (HMView, MBRose, Faaborg, Denmark). All rats received the same treatment in each individual experiment, i.e., a parallel study design was used, with a wash-out period of at least 5-7 days between treatments to assure normalization to baseline levels of food intake, locomotor activity, heart rate, and blood pressure. The home cage was removed from the food intake monitoring system during the drug administration procedure and returned immediately thereafter and automated monitoring of feeding behavior and cardiovascular parameters of each individual animal was resumed.
Data were fed into a standard graphic and statistical analysis program (GraphPad Prism v.4.03). Body weight data were calculated as absolute values (grams) or daily body weight gain relative (control level = 100%) to the first day of drug administration (day 0). Body weight gain and food intake were expressed as means ± S.E.M of n individual animals. After acquisition of telemetry data, 12-h means were calculated using Microsoft Excel 2007. Finally, statistical analysis and data presentation (mean ± S.E.M.) were performed using GraphPad Prism v.4.03. All data were evaluated using a repeated-measure one-way ANOVA with Tukey's post hoc test was applied to perform statistical comparisons between treatment groups. A P-value less than 0.05 was considered statistically significant.
Tesofensine (8-Azabicyclo[3.2.1]octane,3-[3,4-dichlorophenyl)-2-(ethoxymethyl)-8-methyl-[1R-(2-endo,3-exo)]-2-hydroxy-1,2,3-propanetricarboxylate) is a derivative of an azabicyclooctane citrate, synthesized at the Department of Medicinal Chemistry, NeuroSearch A/S. Metoprolol and telmisartan were purchased from Sigma (St. Louis, MO). Tesofensine and metoprolol were dissolved in 0.9% saline solution, whereas telmisartan was dissolved in 1 N NaOH and subsequently titrated with 1 N HCl to pH 7.4. All drugs were administered p.o. (1.0 ml/kg). In drug combination experiments, tesofensine and the anti-hypertensive drug were administered simultaneously (<1 min apart) as separate drug solutions. Subthreshold level doses of telmisartan and metoprolol were used in the study in order to avoid hypotension during treatment. The doses were selected based on pilot studies and earlier findings (9-12).
Effects on food intake and body weight
Acute tesofensine administration robustly triggered a reduction of food intake in telemetrized rats (Figure 1A and 1B). The food intake in tesofensine-treated rats declined in a dose- and time-dependent fashion with the highest oral dose (5.0 mg/kg) at 12 h postdosing reducing food intake to ∼ 50% of the control level (P < 0.001). The hypophagic effect of tesofensine was sustained for up to 12 h (all doses), 24 h (3.0-5.0 mg/kg), and 48 h (5.0 mg/kg) after dosing, respectively, whereupon food intake returned to baseline levels (Figure 1B). The hypophagic effect of tesofensine was paralleled by a corresponding dose-dependent reduction in body weight (negative body weight gain) with the highest doses (3.0-5.0 mg/kg) producing a net body weight change of −1.0 to 1.5% (equivalent to 8-11 g, compared with the body weight of vehicle-treated rats) being evident for at least 48 h after drug administration (Figure 1C and 1D). The acute reduction in body weight is in accordance with previous studies in diet-induced obese rats (6, 7). Tesofensine also dose-dependently induced a significant, albeit short-lasting, increase in locomotor activity in the dose-range of 3.0-5.0 mg/kg (Figure 1E).
The intermediate dose (3.0 mg/kg) of tesofensine was selected for further characterization in acute drug combination studies with the antihypertensive agents metoprolol (Figure 2) and telmisartan (Figure 3), respectively. These drug interaction studies indicated that neither antihypertensive drug exhibited an effect on food intake or body weight regulation per se, and did also not affect tesofensine-induced reductions in food intake (Figures 2A, 2B, 3A, and 3B) and body weight (Figures 2C, 2D, 3C, and 3D). In contrast, metoprolol (Figure 2E, P < 0.05 compared with tesofensine alone), but not telmisartan (Figure 3E, P = 0.98 compared with tesofensine alone), completely prevented the locomotor activity inducing effect by tesofensine.
Effects on cardiovascular parameters
As expected, the telemetric monitoring of blood pressure and heart rate showed a clear diurnal variation (Figure 4), with higher blood pressure and heart rate observed during the active (nocturnal) period.
Acute treatment with tesofensine caused a dose-dependent increase in heart rate at all doses tested, lasting for up to 48 h after treatment (Figure 4A). Similarly, a dose-dependent modest increase in systolic blood pressure was observed up to 48 h after drug administration (3.0 mg/kg and 5.0 mg/kg) (Figure 4C). The effect of 3.0 mg/kg tesofensine on heart rate and systolic blood pressure thereby outlasted the hypophagic effects of 3.0 mg/kg tesofensine. A trend toward a dose-dependent rise in diastolic blood pressure was also observed, although the highest dose did not attain statistical significance (P = 0.204, Figure 4B).
Two combination drug studies were carried out in order to investigate if antihypertensive treatment could prevent or reduce the secondary hypertension and elevated heart rate caused by tesofensine. Co-treatment with tesofensine (3.0 mg/kg) and the β1 adrenoceptor antagonist metoprolol (10 mg/kg and 20 mg/kg) fully reversed tesofensine-induced tachycardia (Figure 5A). The heart rate lowering effect of metoprolol was, however, only observed during the first 24 h after administration, whereas the heart rate was normalized to control levels in the metoprolol + tesofensine combination groups (Figure 5A). The short-lasting effects of metoprolol reflect the pharmacokinetic properties in the rat (11). A normalization of the systolic blood pressure was also observed after co-treatment with metoprolol (20 mg/kg) for up to 24 h (Figure 5C). Similarly, the tesofensine-evoked increase in the diastolic pressure during the first light phase (12-24 h post-treatment) was reversed by co-treatment with metoprolol (20 mg/kg, Figure 5B). When administered alone, metoprolol (20 mg/kg) did not produce any significant effects on diastolic blood pressure in the first 24 h (Figure 5).
In a subsequent drug combinatorial study, tesofensine and the AT1 receptor antagonist telmisartan were investigated. As for the metoprolol study, a similar (3.0 mg/kg) dose of tesofensine was found to significantly increase heart rate. Co-treatment with telmisartan (1.0 and 3.0) did not revert the rise in heart rate after tesofensine administration, and with the highest dose of telmisartan combined with tesofensine, we observed a significant increase in heart rate as compared with tesofensine administration alone (Figure 6A). Although co-treatment with telmisartan was found to attenuate the increase in systolic blood pressure by tesofensine, the AT1 receptor antagonist did not lead to a significant prevention of tesofensine-induced hypertension (P > 0.05, compared with tesofensine alone, Figure 6B and 6C). Telmisartan alone (3.0 mg/kg) had no effect on heart rate and blood pressure (Figure 6).
Weight loss is often accompanied by an increase in perceived hunger and appetite sensations, which has been identified as an important predictor of weight relapse (13-15), and suppression of appetite function is, therefore, considered very important for the maintenance of weight loss. Recent clinical and preclinical reports have indicated that tesofensine acts as a strong appetite suppressant by triggering satiety and fullness sensations, which is believed to be a key mechanism underlying the robust anti-obesity effect of tesofensine (2, 4-7, 16). Hence, the present data on tesofensine-induced anorexia in telemetrized rats further supports this view. Tesofensine dose-dependently triggered a rapid hypophagic response lasting for up to 12-48 h, depending on the dose administered. The long-lasting anorexigenic effect of tesofensine suggests that the bioactive primary M1 metabolite (also being a triple MRI) of tesofensine contributed to the hypophagic and weight-lowering effect in rats, as the M1 metabolite exhibits significantly higher steady-state concentrations and longer T1/2 in rodents (1). In contrast, the human steady-state plasma concentrations of M1 are ∼ 60% lower as compared with those of tesofensine (17), implying that the contribution of M1 to the overall activity might be lower in humans. In addition, it is suggested that increased energy metabolism may potentially contribute to the robust weight loss induced by tesofensine. Accordingly, a recent respiratory calorimetry study indicated a moderate rise in fat oxidation and nocturnal thermogenesis after short-term tesofensine treatment in overweight or moderately obese men (5). Also, while DIO rats show long-term sustained reductions in body weight during chronic tesofensine treatment regimens, hypophagia is most pronounced during the first week of treatment followed by a gradual development of tolerance to the anorexigenic effect of tesofensine (6), thus being in indirect agreement with the clinical findings.
In the present study, tesofensine dose-dependently increased locomotor activity during the first 12 h dark phase, and it may thus be postulated that the augmented locomotor activity may have caused changes in food-seeking behavior. However, the hypophagic effect of tesofensine was more potent and longer lasting (up to 48 h) as compared with the capacity of tesofensine to induce locomotor activity (up to 12 h). In this regard, it is likely that the different temporal pharmacodynamics on food intake and locomotor activity is associated with the pharmacokinetics of tesofensine. In comparison to tesofensine, the M1 metabolite has a longer T1/2 (see above) with a fourfold to fivefold lower in vivo potency on dopamine reuptake transporter inhibition (1), which argues for the metabolite did not contribute significantly to tesofensine-induced locomotor activity. Also, the evidence that metoprolol completely prevented the locomotor stimulatory effect of tesofensine without affecting tesofensine's efficacy on hypophagia, indicates that the moderate increase in locomotor activity had no influence of food-seeking behavior. Hence, we infer that locomotor effects had no influence on the appetite suppressing effect of tesofensine. In addition, it may be speculated that metoprolol antagonized tesofensine-induced locomotor activity by indirect action on striatal dopaminergic neurotransmission, as various β1 blockers are reported to inhibit rat striatal dopamine release (18).
The preclinical finding of cardiovascular effects of tesofensine in awake and freely moving rats is in accordance with clinical findings, also showing significant dose-dependent elevations in heart rate at lower dose levels than required to raise diastolic and systolic blood pressure (2, 5). Notably, the cardiovascular effects outlasted the hypophagic effects following acute administration of tesofensine. Because tesofensine and the M1 metabolite show equipotent inhibition of noradrenaline reuptake in vitro (1), it is likely that the primary metabolite contributed to the sympathetic cardiovascular effects of tesofensine. In this regard, β1 adrenoceptor blockade by metoprolol fully prevented the cardiovascular effects of tesofensine which strongly indicates that tesofensine's noradrenergic reuptake inhibitory profile is far the most important denominator for its cardiovascular effects. In contrast, telmisartan did only marginally reduce tesofensine-evoked rise in blood pressure without concomitant modification of the increased heart rate. Telmisartan did not affect tesofensine-stimulated heart rate, hence reflecting the clinical observation that AT1 receptor blockade only effectively reduces hypertensive, not tachycardia, conditions (19). Surprisingly, telmisartan (1-3 mg/kg) co-treatment only moderately attenuated the tesofensine-induced rise in blood pressure, suggesting that telmisartan in this dose range insufficiently counteracted tesofensine's sympatho-stimulatory effects. Whether higher doses of telmisartan would significantly prevent the increase in blood pressure after tesofensine administration needs further investigation. The lack of effect on heart rate per se is in accordance with other studies on the effects of telmisartan in freely moving rats (9).
In addition, it is unlikely that telmisartan or metoprolol, when co-administered with tesofensine, would cause potential drug-drug interactions and thus influence individual drug pharmacokinetics and pharmacodynamics. Accordingly, the most important route of metabolism of tesofensine is N-demethylation via the cytochrome P450 CYP3A4 enzyme and tesofensine is also highly protein bound in the plasma (20). By contrast, telmisartan has been found to be exclusively metabolized by conjugation to glucuronic acid with the lack of cytochrome P450-dependent metabolism being distinct for telmisartan, as other nonpeptide angiotensin II receptor antagonists are oxidized to varying degrees (21). In comparison, metoprolol is almost exclusively metabolized by CYP2D6 (22).
Whether tesofensine would affect blood pressure and heart rate differently in obese rats is not addressed in the present report and must await further studies. It should also be noted that the present observations are restricted to the acute effects of tesofensine, and do not exclude that the change in cardiovascular parameters in telemetrized rats after chronic tesofensine treatment may closer mimic clinical findings. Also being in good agreement with clinical (2, 4, 5) and preclinical (6, 7) reports, tesofensine produced an acute strong hypophagic response with a corresponding negative net change in body weight in telemetrized rats. Overall, the experimental in vivo settings used in the present study facilitates advanced synchronous monitoring of cardiovascular and food intake parameters in real-time mode, representing a valid and rational method for simultaneously studying clinically relevant anti-obesity and vital sign effects of novel drugs.
Interestingly, the present results suggests a different pharmacodynamic profile of tesofensine and β1 blocker combinational therapy as compared with sibutramine, a dual serotonin and noradrenaline reuptake inhibitor (23, 24). Sibutramine produces a rather modest weight loss and significantly elevates heart rate and blood pressure in obese patients which constitutes a major concern in the clinical utility of sibutramine (25). A clinical study in obese hypertensive patients indicated that sibutramine treatment with combined Ca2+ channel antagonist + ACE inhibitors or metoprolol + hydrochlorothiazide treatment, respectively, significantly attenuated sibutramine's anti-obesity effects (26). The latter combination most negatively affected sibutramine's weight-reducing efficacy which may be explained by the common observation that β-blockers can induce weight gain per se (27). In contrast, metoprolol therapy did not significantly interfere with sibutramine's anti-obesity and metabolic effects in a study on normotensive obese patients, leaving it so far unresolved whether combined β1 blocker treatment is feasible to reduce cardiovascular adverse effects of sibutramine in obese subjects (28). In this context, it should be noted that anorexigenic effects of sibutramine are believed to be closely associated with stimulated α1- and β1-adrenoceptor function, as sibutramine-induced hypophagia is antagonized by prazosin and metoprolol, respectively (29). The implications from these studies may be that anti-obesity drugs with noradrenergic activity will potentially have less anti-obesity efficacy when combined with β-blockers to ameliorate any sympathetic cardiovascular effects. However, the present study suggests that this may not to be the case for tesofensine, because combined treatment with metoprolol did not affect the anti-obesity effects of tesofensine. Hence, this observation indicate a clear pharmacodynamic separation between two distinct and important mechanisms of action of tesofensine, namely, the anti-obesity effects associated with α1 adrenoceptor stimulation (7) and cardiovascular effects linked to augmented β1 adrenoceptor function. The α1 adrenoceptor effect of tesofensine is suggested to be secondary to a blockade of hypothalamic synaptic noradrenaline reuptake leading to inhibition of intrahypothalamic appetite signaling circuits to evoke satiety responses (7, 30). In contrast, it is most conceivable that the cardiovascular effects of tesofensine are being mediated via increased peripheral noradrenergic tonus. Also being in contrast to sibutramine, the anorexigenic effect of tesofensine requires stimulation of both α1 adrenoceptor and dopamine D1 receptor function to obtain full appetite-suppressing activity in DIO rats, hence indirectly pointing to the possibility that tesofensine treatment leads to recruitment of dopaminergic neurotransmission. This is relevant for the effect in obese individuals having reduced striatal dopamine function. Data from magnetic resonance imaging studies indicate that obese individuals overeat to compensate for striatal dopamine hypofunctioning (31). In human volunteers, tesofensine blocks the striatal dopamine transporter at doses causing weight loss (unpublished results), implying that tesofensine may correct imbalanced dopamine function in obese individuals.
In conclusion, we demonstrate that combined tesofensine and metoprolol treatment preserves tesofensine's appetite suppressing efficacy while also preventing elevations in heart rate and blood pressure in rats. These findings invite the possibility that combined antihypertensive treatment with tesofensine would also be effective in obese patients without affecting tesofensine's robust anti-obesity efficacy.
The authors thank MBRose (Henrik Johansen, Søren Ellegaard) and DSI (Thomas Penning) for skillful technical assistance. They also thank DSI for kindly providing telemetry hardware and software.
- 8Tesofensine—a novel potent weight loss medicine. Evaluation of: Astrup A, Breum L, Jensen TJ, Kroustrup JP, Larsen TM. Effect of tesofensine on bodyweight loss, body composition, and quality of life in obese patients: a randomised, double-blind, placebo-controlled trial. Lancet 2008; 372:1906–1913. Expert Opin Investig Drugs 2009; 18: 1043–1046..