Metabolism in adipose tissue in response to citalopram and trimipramine treatment – An in situ microdialysis study

Available online at www.sciencedirect.com

JOURNAL OF PSYCHIATRIC RESEARCH

Journal of Psychiatric Research 42 (2008) 578–586

www.elsevier.com/locate/jpsychires

Metabolism in adipose tissue in response to citalopram and trimipramine treatment – An in situ microdialysis study M. Flechtner-Mors a

a,*

, C.P. Jenkinson c, A. Alt b, G. Adler a, H.H. Ditschuneit

a

Department of Internal Medicine, University Ulm, Robert-Koch-Strasse 8, D-89081 Ulm, Germany b Department of Forensic Medicine, University Ulm, Germany c University of Texas Health Science Center at San Antonio, TX, USA Received 25 January 2007; received in revised form 29 May 2007; accepted 5 June 2007

Abstract The intake of antidepressants is often accompanied by weight gain. Antidepressants may influence lipid and carbohydrate metabolism that can result in metabolic changes and obesity. We investigated the effect of citalopram and trimipramine on interstitial glycerol, glucose and lactate concentration and blood flow in subcutaneous adipose tissue of obese subjects by means of the microdialysis technique. In addition, the effect of stimulation with norepinephrine on metabolic response was investigated. Each subject was compared to a control subject matched for BMI and age. Each group comprised 10 subjects. Circulating plasma triglyceride concentrations were higher in drug-treated groups. In subcutaneous adipose tissue, microdialysis experiments revealed a higher and prolonged glycerol release in the presence of norepinephrine, but not under basal conditions. In citalopram treated subjects, basal glucose and lactate concentrations were higher compared with controls or with the trimipramine treated group. Local administration of norepinephrine induced a decrease in glucose levels and an increase in lactate levels, but without significant differences between groups. Local adipose tissue blood flow decreased in control groups following norepinephrine application, but remained constant in the antidepressant groups. In conclusion, citalopram and trimipramine affected glucose and lipid metabolism in adipose tissue and resulted in enhanced release of glycerol and free fatty acids into the circulation.  2007 Elsevier Ltd. All rights reserved. Keywords: Antidepressant drugs; Adipose tissue; Lipolysis; Microdialysis

1. Introduction Body weight gain is frequently observed in subjects during both acute and long-term drug treatment of depression. The effect of antidepressant drugs on glucose and lipid metabolism could contribute to this effect, but the underlying mechanisms are poorly understood. Studies with tricyclic antidepressants (TCAs) and selective serotonin re-uptake inhibitors (SSRIs) on glucose metabolism have shown inconsistent results. In a clinical investigation the TCA, nortriptyline, was reported to *

Corresponding author. Tel.: +49 0 731 50044613; fax: +49 0 731 50024361. E-mail address: [email protected] (M. Flechtner-Mors). 0022-3956/$ - see front matter  2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jpsychires.2007.06.003

induce hyperglycemia that was not associated with weight gain (Lustman et al., 1997). Otherwise, TCAs have been shown to improve glucose homeostasis either without changes in weight (Okamura et al., 2000), or with weight gain (Himmerich et al., 2006). In animal models, TCAs increased blood glucose concentration (Chadwick et al., 2007; Erenmemisoglu et al., 1999). Further, hyperglycemia from acute TCA (imipramine) administration has been demonstrated, but the response was attenuated with chronic dosing (Gupta et al., 1992). The serotonergic antidepressant, fluoxetine, was reported to decrease glucose levels independently of its effect on body weight (Ghaeli et al., 2004; Maheux et al., 1997). However, short-term studies with citalopram revealed that this antidepressant was ineffective in changing

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

glucose homeostasis/carbohydrate metabolism, whether accompanied by either a small increase in body weight (Leinonen et al., 1999) or no change in body weight (Kauffman et al., 2005). Antidepressants may also affect serum lipid homeostasis. The reported effects of TCAs on serum triglyceride and cholesterol are inconsistent, with increased, unchanged, or reduced serum levels across disparate samples (Kopf et al., 2004; Pollock et al., 1994; Olusi and Fido, 1996; Diebold et al., 1998). Few studies have investigated the effect of SSRIs on lipid metabolism. Bilici et al. investigated the effect of SSRIs (i.e. citalopram, fluoxetine, sertraline, fluvoxamine) on lipid homeostasis as a secondary outcome safety measure in a sample of depressed individuals. It was reported that depressed patients exhibited non-significant lower mean plasma triglyceride and total cholesterol levels compared with baseline levels (Bilici et al., 2001). It has been suggested that the effects of antidepressants on lipid homeostasis are likely linked to alterations in body weight (McIntyre et al., 2006). Adipose tissue is a complex and highly active metabolic and endocrine organ (Stears and Byrne, 2001). Adipose tissue is the major store of lipid within the body and lipid energy is mainly released from adipose tissue as fatty acids which circulate as free fatty acids (FFAs). FFAs are substrates for hepatic synthesis of the triglyerides contained in very-low-density lipoproteins. Elevated FFA levels may induce hypertriglyeridemia (Byrne et al., 1991). Excessive influx of FFAs into muscle leads to insulin resistance (Grundy, 2004). Microdialysis is a technique to investigate local adipose tissue metabolism in vivo. This method allows continuous sampling and manipulation of the interstitial space of adipose tissue without influencing surrounding tissues and/or whole body function. Because of the microdialysis principle there is no drainage of fluid from the extracellular space. It is also possible to add metabolically active compounds to the ingoing dialysis solvent in order to study adipose tissue regulation without causing systemic effects of the agents because it is only delivered locally (Arner, 1995). We used the microdialysis technique to study glycerol, glucose, and lactate concentrations in abdominal subcutaneous adipose tissue of obese subjects who were under treatment with antidepressants, either trimipramine or citalopram, in comparison to matched non depressed controls. In addition the effect of catecholamines on metabolic parameters was measured by local application of norepinephrine to subcutaneous adipose tissue. 2. Research methods and procedures

579

of antidepressant drug treatment. According to their medical records and medical history, the patients had major clinical depression with atypical or melancholic symptoms and were under treatment with either citalopram, 20– 40 mg/day, or trimipramine, 100–200 mg/day, during the previous 12 months. The past medical history and the background history provided by the patients suggested a possible association between drug intake and weight gain. The weight gain of the patients was >10% during the previous 12 months. During the 3 months prior to the study, body weight change was no greater than 3 kg. Subjects were in good health, as assessed by physical examination and laboratory tests. Comorbidities were well controlled and all drugs were discontinued 72 h before the experimental procedure, except citalopram and trimipramine, respectively. Subjects were able to understand the study protocol and informed consent was obtained. For each treated subject enrolled in the study, a non-depressed, untreated subject, matched for BMI and age, was also enrolled. Each study group (citalopram, trimipramine, and the two matched control groups) contained 10 obese female subjects. Clinical characteristics are given in Table 1, values are presented as mean ± SD. The study was approved by the Ethics Committee at the University of Ulm. 2.2. Microdialysis probe The microdialysis probes were constructed of cuprophane fibers (30 · 0.3 mm, molecular cut-off 3000 Da). One single fiber was glued to 50- and 100-mm-long sections of nylon tubing (Fa. Labakron, Sinsheim, Germany). The probes were sterilized by c-radiation. The microdialysis fibers were placed into abdominal subcutaneous adipose tissue, without anesthesia. The 100-mm long nylon tubing was connected to a microinjection pump (Perfusor VI; Braun Melsungen, Melsungen, Germany) and was perfused continously (2.5 lL/min) with isotonic saline. Changes in local adipose tissue blood flow were measured using the ethanol technique (Fella¨nder et al., 1996). For blood flow measurements, the perfusate contained ethanol at a concentration of 100 mM. The concentration of ethanol was measured in the outgoing dialysate, and the ethanol outflow-to-inflow ratio was calculated. The ethanol ratio reflects changes in the microcirculation, such that a reduced ratio indicates an increased loss of ethanol from the interstitial fluid, which is caused by an enhanced blood flow. Thus, a low ratio indicates an escape of ethanol from the dialysate that, in turn, mirrors a rise in local blood flow. No collection of the outgoing dialysate was made during the first 45 min of implantation of microdialysis probes. Thereafter, dialysate was collected every 15 min.

2.1. Subjects 2.3. Study protocol All subjects had been referred by a general practitioner to the University Hospital Obesity Center for the treatment of obesity. Study participants were selected with clinical depression who complained about weight gain after onset

After an overnight fast, patients were studied at 8:00 am while in a supine position. Body weight and height was determined while the patients wore underclothes

580

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

Table 1 Clinical characteristics and biomarkers of health of study subjects Citalopram (n = 10)

Control (n = 10)

p

Age (years) Body weight (kg) Body mass index (kg/m2) Percent body fat (%) Lean body mass (kg) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) HbA1 (%) Blood glucose (mg/dL) Insulin (lU/mL) Total cholesterol (mmol/L) HDL-cholesterol (mmol/L) Triglyceride (mmol/L)

58.4 ± 6.6 122.8 ± 31.9 44.8 ± 10.4 45.7 ± 4.4 69.7 ± 14.5 175.5 ± 35.0 97.4 ± 28 7.1 ± 0.9 118.1 ± 37.4 15.4 ± 12.4 5.5 ± 0.7 1.3 ± 0.3 2.0 ± 1.0

52.3 ± 8.2 123.5 ± 31.5 43.7 ± 10.5 42.6 ± 7.3 70.2 ± 15.2 150.0 ± 19.5 95.0 ± 11.7 6.7 ± 0.5 101.7 ± 17.5 19.7 ± 11.8 5.2 ± 0.8 1.3 ± 0.3 1.2 ± 0.3

ns ns ns ns ns ns ns ns ns ns ns ns < 0.05

Age (years) Body weight (kg) Body mass index (kg/m2) Percent body fat (%) Lean body mass (kg) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) HbA1 (%) Blood glucose (mg/dL) Insulin (lU/mL) Total cholesterol (mmol/L) HDL-C (mmol/L) Triglyceride (mmol/L)

Trimipramine (n = 10) 55.1 ± 8.0 105.3 ± 16.7 39.6 ± 5.4 45.6 ± 3.6 58.8 ± 7.0 151.5 ± 20.1 90.5 ± 15.3 6.7 ± 0.6 102.5 ± 12.2 14.1 ± 10.1 5.4 ± 0.4 1.3 ± 0.2 2.1 ± 0.8

Control (n = 10) 50.7 ± 15.4 101.2 ± 13.00 37.8 ± 4.2 41.4 ± 3.8 58.8 ± 5.2 155.0 ± 21.2 94.0 ± 21.1 7.3 ± 0.9 102.9 ± 28.3 24.3 ± 20.7 5.2 ± 0.6 1.4 ± 0.3 1.4 ± 0.5

p ns ns ns ns ns ns ns ns ns ns ns ns < 0.05

without shoes and body composition was measured by bioelectrical impedance analysis (BIA). Blood was drawn from an antecubital vein without tourniquet. For the microdialysis experiments, two microdialysis probes were inserted in the abdominal subcutaneous adipose tissue. One microdialysis probe was perfused with saline for 180 min for basal measurements of glycerol, glucose and lactate over time. The second microdialysis probe was perfused with saline for basal measurement for 60 min. Thereafter, norepinephrine (Hoechst Marion Roussel, Bad Soden, Germany) at a concentration of 10 4 M was added to the perfusion solution for 120 min. Samples of dialysate were collected at intervals of 15 min. Interstital glycerol concentration reflects lipolysis, since glycerol is metabolized by the tissue to an insignificant extent (Arner and Bu¨low, 1993). Glycerol concentration were analyzed with a bioluminescence method (Bjo¨rkhelm et al., 1981). For glucose, lactate and ethanol measurements, two consecutive samples were combined. Glucose and lactate concentrations were determined electrochemically (Yellow Springs Instrument Co., Yellow Springs, OH). Ethanol concentrations were determined by gas-chromatography (Curry et al., 1966). 2.4. Statistical analysis Data are given as the mean ± SEM unless otherwise stated. Data on clinical characteristics were compared using an unpaired Student’s t test. Comparisons of time-related

changes in metabolite concentrations between group, treatment and norepinephrine stimulation were analysed using two-factor analysis of variance (ANOVA) for repeated measurements as appropriate. Post hoc analysis by Tukey’s honestly significant difference test was used to compare metabolic concentrations at different time points. A value of p < 0.05 was considered significant. All statistical evaluations were done using the SPSS statistical software package (SPSS for Windows 7.5, SPSS Inc, Chicago, IL). 3. Results Clinical characteristics and biomarkers of health of the study subjects are shown in Table 1. Data for all study participants were similar except for triglyceride concentrations which were significantly higher in subjects treated with the antidepressant medications citalopram or trimipramine (p < 0.05 for both). In subcutaneous adipose tissue, glycerol release remained constant in all study groups under basal conditions for 120 min (F(11,396) = 1.5, ns) (Fig. 1, Panels A and B). Glycerol release tended to be higher in subjects treated with citalopram. In the second microdialysis catheter, perfusion with norepinephrine (10 4 M) was initiated after 60 min of basal measurement (Fig. 1, Panels c and d). In the adipose tissue of all subjects, the concentration of glycerol in the dialysate increased significantly during the next 45 min (F(11,396) = 81.8, p < 0.001). Thereafter, differences in the glycerol kinetics were observed between

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

500

Glycerol (µmol/L)

Glycerol (µmol/L)

500 400 300 200 100

400 300 200 100

0

0 0

30

60

90

120

150

180

0

30

60

Time (min)

c

* * * *

*

*

90

120

150

180

Time (min)

d

* * * *

500

Glycerol (µmol/L)

500

Glycerol (µmol/L)

581

400 300 200

400 300 200 100

100

0

0 0

30

60

90

120

150

180

Time (min)

0

30

60

90

120

150

180

Time (min)

Fig. 1. Glycerol concentration in the dialysate of subcutaneous adipose tissue of obese subjects treated with either citalopram (d) or trimipramine (n), respectively, and control groups (citalopram control (s), trimipramine control (h)). Panels a and b: basal measurement. Panels c and d: basal measurement for 60 min, followed by addition of norepinephrine (10 4 M) as indicated by the arrows. * p < 0.05 vs control.

subjects treated with antidepressant medication and matched controls, and between the two drug treatment groups. Glycerol concentration in adipose tissue of subjects treated with citalopram increased to a higher peak concentration than in controls, followed by a similar gradual decline in both groups over the course of the study. In the citalopram treated subjects, the glycerol level was higher at every time point. During norepinephrine administration, the glycerol level was higher in citalopram treated subjects (F(1,18) = 5.4, p < 0.05). In trimipramine treated subjects, the glycerol level increased in a manner similar to the matched control group but remained at the elevated level until the end of the study period. A gradual decrease in glycerol concentration was observed only in the control group. The difference between trimipramine treated subjects and controls was significant (F(1,18) = 2.9, p < 0.01). The apparent difference in the kinetics of glycerol outflow between citalopram and trimipramine treated subjects did not reach statistical significance. The basal glucose concentration decreased slightly in the dialysate of‘ adipose tissue in all study groups during the study period (F(5,180) = 11.1, p < 0.001) (Fig. 2, Panels A and B). The basal glucose level in adipose tissue was significantly higher in subjects treated with citalopram, compared with controls (F(1,18) = 5.7, p < 0.05) (Fig. 2, Panel a) and compared with subjects under trimipramine treat-

ment (F(1,18) = 7.0, p < 0.05). The basal glucose concentration also declined significantly in all groups prior to norepinephrine perfusion (F(5,180) = 12.0, p < 0.001) (Fig. 2, Panels c and d). The addition of norepinephrine to the microdialysis perfusate tended to affect the interstitial glucose concentration in the citalopram group compared with controls (F(1,18) = 3.2, p = 0.086). Interestingly, in the citalopram treated subjects the glucose concentration declined strongly during the first 60 min due to norepinephrine stimulation, and increased thereafter back to basal levels over the next 60 min. This pattern of change in glucose levels was less marked in the control subjects. No significant difference was observed between the citalopram and trimipramine groups over the entire study period (F(1,18) = 1.7, ns). The basal lactate concentration increased during the saline perfusion for 180 min in the adipose tissue of all subjects (F(5,180) = 19.8, p < 0.001) (Fig. 3, Panel a and b). Basal lactate was significantly higher in subjects treated with citalopram compared with controls (F(1,18) = 4.6, p < 0.05) (Fig. 3, Panel a). This effect was not observed in trimipramine treated subjects (Fig. 3, Panel b). Similarly, after the addition of norepinephrine, lactate concentration increased in all study groups (F(5,180) = 18.9, p < 0.001) with no significant difference between study groups (Fig. 3, Panel c and d).

582

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

*

*

*

*

*

0.8

* Glucose (mmol/L)

Glucose (mmol/L)

0.8

0.6

0.4

0.2

0.6

0.4

0.2

0

30

60

90

120

150

180

0

30

60

c

120

150

180

150

180

d

*

*

0.8

Glucose (mmol/L)

0.8

Glucose (mmol/L)

90

Time (min)

Time (min)

0.6

0.4

0.6

0.4

0.2

0.2

0

30

60

90

120

150

180

Time (min)

0

30

60

90

120

Time (min)

Fig. 2. Glucose concentration in the dialysate of subcutaneous adipose tissue of obese subjects treated with either citalopram (d) or trimipramine (n), respectively, and control groups (citalopram control (s), trimipramine control (h)). Panels a and b: basal measurement. Panels c and d: basal measurement for 60 min, followed by addition of norepinephrine (10 4 M) as indicated by the arrows. * p < 0.05 vs control.

Blood flow remained constant under basal conditions in all study groups (F(5,180) = 1.09, ns) (Fig. 4, Panels a and b). In response to norepinephrine stimulation, the ethanol ratio increased in the control groups, indicating a decline in blood flow velocity, but this was not observed in the drug treatment groups (Fig. 4, Panels c and d). This differential effect was significant in the trimipramine treated group, compared with controls (F(1,18) = 4.9, p < 0.05), but in the citalopram vs. control group comparison it did not reach significance (F(1,18) = 0.901, ns). 4. Discussion Citalopram and trimipramine caused a high and sustained release of glycerol in subcutaneous adipose tissue of obese subjects in response to adrenergic stimulation. The kinetic pattern was somewhat different in both drugtreated groups, with a greater release in citalopram treated subjects. Usually in the presence of intensive adrenergic stimulation, glycerol outflow increases to a peak level and declines rapidly thereafter. This phenomenon has been shown in vivo by microdialysis technique in subcutaneous adipose tissue of lean (Arner et al., 1991; Stallknecht et al., 1997) and obese (Flechtner-Mors et al., 2002) sub-

jects, and suggests that sustained adrenergic stimuli or chronic administration of adrenergic agents may induce desensitization of adrenergic receptors (Hausdorff et al., 1990). Catecholamines regulate fat cell function using various adrenergic receptors. Beta-adrenoceptors (b1, b2, b3) exert a lipolysis stimulating effect, whereas a2-receptors mediate lipolysis inhibiting effects (Large et al., 2004). In addition, it recently has been shown that a1-adrenoceptors exert lipolytic activity in subcutaneous adipose tissue of obese subjects (Boschmann et al., 2002; Flechtner-Mors et al., 2002). In the present study the sustained release of glycerol could not be accounted for by the available catecholamine concentration, but is likely due to changes in receptor or post-receptor mechanisms due to antidepressant treatment. These findings are consistent with recent animal experimental studies, where it has been shown that acute and long-term administration of citalopram desensitizes a2receptors (Grandoso et al., 2006). Further, the a1-adrenergic system and responsiveness is affected by antidepressants (Maj et al., 1998). Repeated administration of trimipramine increased the response to a1-adrenoceptor stimulation and thus induced a1-adrenergic up-regulation (Maj et al., 1998). Similarly, an increase in brain a1-adrenoceptor

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

583

0.6 0.6

Lactate (mmol/L)

*

*

*

*

*

0.4 0.3 0.2

Lactate (mmol/L)

0.5

0.5

0.4 0.3 0.2 0.1

0.1

0.0

0.0 0

30

60

90

120

150

0

180

30

60

c

d

0.6

120

150

180

0.6 0.5

0.5

*

0.4

*

Lactate (mmol/L)

Lactate (mmol/L)

90

Time (min)

Time (min)

0.3 0.2

0.4 0.3 0.2 0.1

0.1

0.0

0.0 0

30

60

90

120

150

180

Time (min)

0

30

60

90

120

150

180

Time (min)

Fig. 3. Lactate concentration in the dialysate of subcutaneous adipose tissue of obese subjects treated with either citalopram (d) or trimipramine (n), respectively, and control groups (citalopram control (s), trimipramine control (h)). Panels a and b: basal measurement. Panels c and d: basal measurement for 60 min, followed by addition of norepinephrine (10 4 M) as indicated by the arrows.

affinity for the a1-agonist phenylephrine was induced by citalopram (Mogilnicka et al., 1987). Antidepressants also affect b-receptors. In subcutaneous adipose tissue b-receptors have been shown to exhibit tachyphylaxis (Arner et al., 1991). Down-regulation of b-receptors has been described as a common biochemical effect of chronic treatment with many antidepressant drugs (Holoubek et al., 2004). However, this has not been reported for citalopram (Holoubek et al., 2004) and trimipramine (Kopanski et al., 1983; Hauser et al., 1985). Therefore the inhibiting effects on a2-receptors, the stimulation of a1-receptors and unchanged b-receptor activity may cause enhanced lipolysis. Thus, it is suggested that changes in adrenoceptor function or distribution on adipocytes may have induced the enhanced and prolonged lipolysis rate in the presence of norepinephrine. It is possible that, in citalopram treated subjects, the observed higher glucose level in adipose tissue may contribute to prolonged lipolysis. Although the role of glucose on the regulation of lipolysis is still unclear, glucose seems to be necessary for release of hormone sensitive lipase (HSL) and HSL activity (Raclot et al., 1998). Prolonged treatment of adipocytes with high glucose and insulin concentrations increases basal and stimulated lipolysis associated with a 40% increase in the level of HSL (Botion and Green, 1999).

In the present study glucose concentration in subcutaneous adipose tissue was higher in citalopram treated subjects compared to the similar levels observed in both the trimipramine and control groups. Interestingly, this finding was mirrored by the slightly elevated plasma glucose level in the citalopram-treated subjects. It has been shown, using the microdialysis technique, that basal interstitial glucose concentration closely reflects that in venous blood (Jannson et al., 1988; Simonsen et al., 1994). Our finding of increased glucose levels with citalopram was unexpected, since in general SSRIs are reported to reduce blood glucose levels (Erenmemisoglu et al., 1999; Ghaeli et al., 2004). However, a decrease in blood glucose concentration is mostly caused by weight loss. Our study subjects had gained weight during long-term treatment with SSRIs that may have caused higher blood glucose concentrations. It is a common clinical finding that weight gain occur after one year of treatment with SSRIs (Harvey and Bouwer, 2000). Subjects treated with trimipramine also reported weight gain, but their subcutaneous basal glucose levels were similar to controls. Thus, the two antidepressants appear to have different effects on systems that influence glucose metabolism. It is conjectured that both SSRIs and TCAs affect glucose homeostasis via their effects on catecholamines, which modulate insulin activity and glucose homeostasis (McIntyre et al., 2006a; Goodnick, 2001).

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

1.0

1.0

0.9

0.9

Ethanolratio

Ethanolratio

584

0.8 0.7

0.8 0.7

0.6

0.6

0.5

0.5 0

30

60

90

120

150

180

0

30

60

Time (min)

c

d

1.0

120

150

180

1.0

*

*

*

90

120

150

0.9

Ethanolratio

0.9

Ethanolratio

90

Time (min)

0.8 0.7

0.8 0.7

0.6

0.6

0.5

0.5 0

30

60

90

120

150

180

Time (min)

0

30

60

180

Time (min)

Fig. 4. Ethanol ratio in subcutaneous adipose tissue of obese subjects treated with either citalopram (d) or trimipramine (n), respectively, and control groups (citalopram control (s), trimipramine control (h)). Panels a and b: basal measurement. Panels c and d: basal measurement for 60 min, followed by addition of norepinephrine (10 4 M) as indicated by the arrows. *p < 0.05 vs control.

Interaction of antidepressants with the hypothalamic– pituitary–adrenal (HPA) axis may also contribute to the divergent glucose findings in this study. It has been reported that cortisol production was higher in depressed women compared to non-depressed women, but did not diminish with citalopram treatment (Kauffman et al., 2005). In contrast, in trimipramine treated subjects HPA activity was decreased (Frieboes et al., 2003). Since glucocorticoid hormones inhibit glucose utilization (Plested et al., 1987) this could explain the increased glucose concentration in citalopram treated subjects. In all study groups glucose concentration in the dialysate of adipose tissue decreased in response to local catecholamine perfusion. This finding is consistent with similar results from previous studies in lean (Boschmann et al., 2002) and obese subjects (Flechtner-Mors et al., 2004). In adipose tissue, glucose is metabolized to glycogen, CO2, glyceride-glycerol, glyceride fatty acids, pyruvate and lactate (DiGirolamo, 2000). The increase in lactate concentration in adipose tissue observed in our study is in agreement with previous reports (Hagstro¨m et al., 1990; Jannson et al., 1994). Lactate production is also stimulated by norepinephrine (Crandall et al., 1983; FlechtnerMors et al., 2005). The higher basal lactate concentration in citalopram treated subjects is possibly due to the higher basal glucose concentration. Since there was no difference

in lactate synthesis in adipose tissue in the various groups of patients, it appears that lactate metabolism was not altered by the antidepressant drugs. Microdialysis measurements reflect adipose tissue metabolism that could, in principle, be influenced by changes in interstitial local blood flow. Since basal blood flow remained constant in all study groups, both antidepressants appeared to have no effect on changes in subcutaneous adipose tissue blood flow. However, differences were observed in the presence of catecholamines. Blood flow in antidepressant-treated groups remained constant, but in the presence of norepinephrine blood flow decreased in the control groups. Thus, the adrenergic effect of norepinephrine on blood flow is affected by citalopram and trimipramine, respectively. The effect of antidepressants on peripheral adipose tissue blood flow has not been extensively studied. The selective norepinephrine reuptake inhibitor reboxetine has been shown to decrease systemic vascular resistance (Mayer et al., 2006). In subcutaneous adipose tissue of lean subjects reboxetine increased blood flow, as measured by the microdialysis technique (Boschmann et al., 2002a). Adrenergic receptors are expressed on vascular cells and catecholamines are thought to stimulate vasodilatation (Crandall et al., 1997), but further studies are warranted to assess the effect of antidepressant treatment on the regulation of peripheral blood flow.

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

We investigated metabolism in subcutaneous adipose tissue of obese subjects undergoing either citalopram or trimipramine treatment for depression, compared with treated obese subjects compared to obese, but otherwise healthy controls. Depression per se may change metabolism and affect lipid and glucose homeostasis, autonomic nervous system and hypothalamic pituitary adrenal axis functioning. These metabolic activities could confound the effects of chronic antidepressant treatment on metabolism in adipose tissue and therefore may limit the conclusions to be drawn from this study. In summary, we observed a significant impact of citalopram and trimipramine on norepinephrine stimulated lipolysis in subcutaneous adipose tissue in obese subjects. The prolonged release of glycerol and associated free fatty acids could contribute to an increase in hepatic synthesis of triglycerides and very-low-density lipoproteins and thus induce lipid and glucose metabolic dysfunction which could contribute to the development of insulin resistance and obesity. Acknowledgement We thank Katja Huber for excellent technical and laboratory assistance. Sigrid Geiger is gratefully acknowledged for coordinating the study. The study was funded from the University of Ulm. References Arner P, Kriegholm E, Engfeldt P. In vivo interactions between beta-1 and beta-2 adrenoceptors regulate catecholamine tachyphylaxia in human adipose tissue. Journal of Pharmacology and Experimental Therapeutics 1991;3001:229–33. Arner P, Bu¨low J. Assessment of adipose tissue metabolism in man: comparison of Fick and microdialysis techniques. Clinical Science 1993;85:247–56. Arner P. Techniques for the measurement of white adipose tissue metabolism: a practical guide. International Journal of Obesity 1995;19:435–42. Bilici M, Efe H, Koroglu MA, Uydu HA, Bekaroglu M, Deger O. Antioxidative enzyme activities and lipid peroxidation in major depression: alterations by antidepressant treatments. Journal of Affective Disorders 2001;64(1):43–51. ¨ stman J. Sensitive kinetic bioluminesBjo¨rkhelm J, Arner P, Thore A, O cent assay of glycerol release from human fat cells. Journal of Lipid Research 1981;22:1142–7. Boschmann M, Krupp G, Luft FC, Klaus S, Jordan J. In vivo response to a1-adrenoceptor stimulation in human white adipose tissue. Obesity Research 2002;10(6):555–8. Boschmann M, Schroeder C, Christensen NJ, Tank J, Krupp G, Biaggioni I, et al. Norepinephrine transporter function and autonomic control of metabolism. Journal of Clinical Endocrinolgy and Metabolism 2002a;87(11):5130–7. Botion LM, Green A. Long-term regulation of lipolysis and hormonesensitive lipase by insulin and glucose. Diabetes 1999;48:1691–7. Byrne CD, Brindle NP, Wang TW, Hales CM. Interaction of nonesterified fatty acid and insulin in control of triacylglycerol secret ion by Hep G2 cells. Biochemical Journal 1991;280:99–104. Chadwick W, Wilson G, van de Venter M, Oelofsen W, Roux S. Shifts in metabolic parameters surrounding glucose homoeostasis

585

resulting from tricyclic antidepressant therapy: implications of insulin resistance? Journal of Pharmacy and Pharmacology 2007;59(1): 95–103. Crandall DL, Fried SK, Francendes AA, Nickel M, DiGirolamo M. Lactate release from isolated rat adipocytes. Influence of cell size, glucose concentration, insulin and epinephrine. Hormone and Metabolic Research 1983;15:326–9. Crandall DL, Hausman GJ, Kral JG. A review of the microcirculation of adipose tissue: anatomic, metabolic, and angiogenic perspectives (Review). Microcirculation 1997;4(2):211–32. Curry AS, Walker GW, Simpson GS. Determination of ethanol in blood by gas chromatography. The Analyst 1966;91:742–3. Diebold K, Michel G, Schweizer J, Diebold-Dorsam M, Fiehn W, Kohl B. Are psychoactive-drug-induced changes in plasma lipid and lipoprotein levels of significance for clinical remission in psychiaric disorders? Pharmacopsychiatry 1998;31(2):60–7. DiGirolamo M. Measurements of glucose conversion to its metabolites. In: Ailhaud G, editor. Adipose Tissue Protocols. Totowa, New Jersey: Humana Press; 2000. p. 181–92. Erenmemisoglu A, Ozdogan UK, Saraymen R, Tutus A. Effect of some antidepressants on glycaemia and insulin levels of normoglycaemic and alloxan-induced hyperglycaemic mice. Journal of Pharmacy and Pharmacology 1999;51(6):741–3. Flechtner-Mors M, Jenkinson CP, Alt A, Adler G, Ditschuneit HH. In vivo alpha (1)-adrenergic lipolytic activity in subcutaneous adipose tissue of obese subjects. Journal of Pharmacology and Experimental Therapeutics 2002;301:229–33. Flechtner-Mors M, Jenkinson CP, Alt A, Biesalski HK, Adler G, Ditschuneit HH. Sympathetic regulation of glucose uptake by the alpha1adrenoceptor in human obesity. Obesity Research 2004;12(4):612–20. Flechtner-Mors M, Jenkinson CP, Alt A, Biesalski HK, Adler G, Ditschuneit HH. Studies of phosphodiesterase effects on adipose tissue metabolism in obese subjects by the microdialysis technique. Journal of Physiology and Pharmacology 2005;56:355–68. Fella¨nder G, Linde B, Bolinder J. Evaluation of the microdialysis ethanol technique for monitoring of subcutaneous adipose tissue blood flow in humans. International Journal of Obesity and Related Metabolic Disorders 1996;20(3):220–6. Frieboes RM, Sonntag A, Yassouridis A, Eap CB, Baumann P, Steiger A. Clinical outcome after trimipramine in patients with delusional depression – A pilot study. Pharmacopsychiatry 2003;36(1):12–7. Ghaeli P, Shahsavand E, Mesbahi M, Kamkar MZ, Sadeghi M, DashtiKhavidaki S. Comparing the effects of 8-week treatment with fluoxetine and imipramine on fasting blood glucose of patients with major depressive disorder. Journal of Clinical Psychopharmacology 2004;24(4):386–8. Goodnick PJ. Use of antidepressants in treatment of comorbid diabetes mellitus and depression as well as in diabetic neuropathy. Annals of Clinical Psychiatry 2001;13:31–41. Grandoso L, Garcia L, Pineda J, Ugedo L. Acute and long-term administration of citalopram desensitizes alpha2-adrenoceptors in the rat vas deferens. Journal of Pharmacy and Pharmacology 2006;58(3):367–73. Grundy SM. Obesity, metabolic syndrome, and cardiovascular disease. Journal of Clinical Endocrinology and Metabolism 2004;89(6):2595–600. Gupta B, Awasthi A, Jaju BP. Effect of acute and chronic imipramine treatment on glucose homeostasis. Indian Journal of Medical Research 1992;96:65–71. Hagstro¨m E, Arner P, Ungerstedt U, Bolinder J. Subcutaneous adipose tissue: a source of lactate production after glucose ingestion in humans. American Journal of Physiology 1990;258:E888–93. Harvey BH, Bouwer CD. Neuropharmacology of paradoxic weight gain with selective serotonin reuptake inhibitors. Clinical Neuropharmacology 2000;23(2):90–7. Hausdorff WP, Caron MG, Lefkowitz RJ. Turning off the signal: desensitization of b-adrenergic receptor function. The FASEB Journal 1990;4:2881–9.

586

M. Flechtner-Mors et al. / Journal of Psychiatric Research 42 (2008) 578–586

Hauser K, Olpe H-R, Jones RSG. Trimipramine, a tricyclic antidepressant exerting atypical actions on the central noradrenergic system. European Journal of Pharmacology 1985;111:23–30. Himmerich H, Fulda S, Schaaf L, Beitinger PA, Schuld A, Pollma¨cher T. Changes in weight and glucose tolerance during treatment with mirtazapine. Diabetes Care 2006;29:170–1. Holoubek G, Nolder M, Treiber K, Mu¨ller WE. Effect of chronic antidepressant treatment on beta-receptor coupled signal transduction cascade. Which effect matters most? Pharmacopsychiatry 2004;37(Supp 2):S113–9. Jannson PA, Fowelin J, Smith U, Lo¨nnrothe P. Characterization by microdialysis of intercellular glucose level in subcutaneous tissue in humans. American Journal of Physiology 1988;255:E218–20. Jannson P-A, Larsson A, Smith U, Lo¨nnroth P. Lactate release from the subcutaneous adipose tissue in lean and obese men. Journal of Clinical Investigation 1994;93:240–6. Kauffman RP, Castracane VD, White DL, Baldock SD, Owens R. Impact of the selective serotonin reuptake inhibitor citalopram on insulin sensitivity, leptin and basal cortisol secret.ion in depressed and nondepressed euglycemic women or reproductive age. American Journal of Obstetrics and Gynecology 2005;21(3):129–37. Kopanski C, Tuerck M, Schultz JE. Effects of long-term treatment of rats with antidepressants on adrenergic-receptor sensitivity in cerebral cortex; structure activity study. Neurochemistry International 1983;5:649–59. Kopf D, Westphal S, Luley CW, Ritter S, Gilles M, Weber-Hamann B, et al. Lipid metabolism and insulin resistance in depressed patients. Journal of Clinical Psychopharmacology 2004;24(5):527–31. Large V, Peroni O, Let exier D, Ray H, Beylot M. Metabolism of lipids in human white adipocytes (Review). Diabetes Metabolism 2004;30:294–309. Leinonen E, Skarstein J, Behnke K, Agren H, Helsdingen JT. Efficacy and tolerability of mirtazapine versus citalopram: a double-blind, randomized study in patients with major depressive disorder. Nordic Antidepressant Study Group. International Clinical Psychopharmacology 1999;14:329–37. Lustman PJ, Griffith LS, Clouse RE, Freedland KE, Eisen SA, Rubin EH, et al. Effects of nortriptyline on depression and glycemic control in diabetes: results of a double-blind, placebo-controlled trial. Psychosomatic Medicine 1997;59:241–50. Maj J, Rogo´z_ Z, Skuza G, Margas W. Repeated trimipramine induces dopamine D2/ D3 and a1-adrenergic up-regulation. Journal of Neural Transmission 1998;105:329–42.

Maheux P, Ducros F, Bourque J, Garon J, Chiasson JL. Fluoxetine improves insulin sensitivity in obese patients with non-insulin-dependent diabetes melltus independently of weight loss. International Journal of Obesity 1997;21(2):97–102. Mayer AF, Schroeder C, Heusser K, Tank J, Diedrich A, Schmieder RE, et al. Influences of norepinephrine transporter function on the distribution of sympathetic activity in humans. Hypertension 2006;48(1):120–6. McIntyre RS, Soczynska JK, Konarski JZ, Kennedy SH. The effect of antidepressants on lipid homeostasis: a cardiac safety concern (Review)? Expert Opinion on Drug Safety 2006;5(4):523–37. McIntyre RS, Soczynska JK, Konarski JZ, Kennedy SH. The effect of antidepressant on glucose homeostasis and insulin sensitivity: synthesis and mechanisms (Review). Expert Opinion on Drug Safety 2006a;1(5):157–68. Mogilnicka E, Zazula M, We˛dzony K. Functional supersensitivity to the a1-adrenoceptor agonist after repeated treatment with antidepressant drugs is not conditioned by b-down-regulation. Neuropharmacology 1987;26:1457–61. Okamura F, Tashiro A, Utumi A, Imai T, Suchi T, Tamura D, et al. Insulin resistance in patients with depression and its changes during the clinical course of depression: minimal model analysis. Metabolism 2000;49(10):1255–60. Olusi SO, Fido AA. Serum lipid concentrations in patients with major depressive disorder. Biological Psychiatry 1996;40:1128–31. Plested CP, Taylor E, Brindley DN, Vernon RG. Interactions of insulin and dexamethasone in the control of pyruvate kinase activity and glucose metabolism in sheep adipose tissue. Biochemical Journal 1987;247(2):459–65. Pollock BG, Perel JM, Paradis CF, Fasiczka AL, Reynolds 3rd CF. Metabolic and physiologic consequences of nortriptyline treatment in the elderly. Psychpharmacology Bulletin 1994;2(30):145–50. Raclot T, Dauzats M, Langin D. Regulation of hormone-sensitive lipase in 3T3-F442A adipocytes. Biochemical Biophysical Research Communications 1998;245:510–3. Simonsen L, Bu¨low J, Madsen J. Adipose tissue metabolism in humans determined by vein catheterization and microdialysis technique. American Journal of Physiology 1994;29:E357–65. Stallknecht B, Bu¨low J, Frandsen E, Galbo H. Desenstitization of human adipose tissue to adrenaline stimulation studied by microdialysis. The Journal of Physiology 1997;500(Pt1):271–82. Stears AJ, Byrne CD. Adipocyte metabolism and the metabolic syndrome (Review). Diabetes Obesity and Metabolism 2001;3:129–42.