Disruption of Lipid Profile, Glucose Metabolism, and Leptin Levels following Citalopram Administration and High-Carbohydrate and High-Cholesterol Diet in Mice

Patients with psychiatric diseases have increased morbidity and mortality, lower life expectancy, and lower quality of life compared to the general population. This is due to the mental disorder itself and associated comorbidities, often metabolic and cardiovascular in nature. The reason for the coexistence of these pathologies is likely related to subclinical inflammation and increased pro-inflammatory cytokine levels, but the exact mechanism is unknown [1, 2]. These comorbidities result from a combination of genetic predisposition, lifestyle factors, the psychiatric disease itself, and psychopharmacological therapy [3, 4].

Several psychopharmaceuticals have metabolic side effects such as weight gain, changes in lipid levels, glucose intolerance, and hypertension. These include selective serotonin reuptake inhibitors (SSRIs), which are often used as first-line drugs. Weight loss was initially observed in connection with SSRI antidepressants. However, prolonged use of these medications, exceeding 1 year, has been found to lead to weight gain and metabolic complications such as dyslipidemia, insulin resistance, and metabolic syndrome [5‒8].

There are several possible explanations for the combination of antidepressant therapy and metabolic syndrome. These include affinity for serotonin, histamine, and other receptors that influence food intake and energy expenditure. A common factor is also the similar pathophysiology of metabolic and mental disorders, both of which result in increased production of pro-inflammatory hormones. The action of antidepressants on adipokines may also contribute to metabolic imbalances. These hormones released from adipose tissue play a role in glucose utilization and control food intake in the hypothalamus. The expression of these hormones also changes with alterations in adipose tissue during the development of metabolic syndrome. Changes in adipokine levels have been observed in both depressive illness and antidepressant medication [9‒12].

Weight gain and other negative effects on energy metabolism are common side effects of many psychopharmaceuticals. Patients who take antipsychotics, especially second-generation ones, are most affected by these adverse reactions. Atypical antipsychotics have more pronounced and frequent metabolic effects compared to antidepressants [13]. However, since antidepressants are prescribed much more often than antipsychotics, the overall impact of the metabolic effects of antidepressants on the population is considerable. Therefore, our experiment focuses on citalopram, one of the most prescribed antidepressants.

The experiment aimed to observe changes in serum levels of selected metabolic parameters in relation to short- and long-term administration of citalopram and its long-term combination with a special diet in laboratory mice. In order to study the dynamics of these changes, the experiment was designed with a short-term group (1 week) and several long-term groups (5 weeks). The study monitored parameters of glucose and lipid metabolism, as well as levels of selected adipokines involved in the regulation of food intake and metabolism. The utilization of the special diet was specifically implemented to replicate the metabolic comorbidity and unhealthy way of life experienced by some patients.

The experiment was performed on 45 male C57BL/6N laboratory mice (obtained from Velaz s.r.o, Prague, Czech Republic). Male mice were selected for this study because the female estrous cycle can cause variations in metabolism, such as changes in energy intake and expenditure, adipocyte differentiation and white adipose tissue mass, lipid metabolism, and adipokine secretion [14‒18].

After a week of acclimatization, seven-week-old mice were randomly divided into 5 groups of 9 animals:

• 1.

  CTR – control group (n = 9)

• 2.

  CTR DIET – control group with special diet (n = 9)

• 3.

  CIT – citalopram-treated group (35 days) (n = 9)

• 4.

  CIT DIET – citalopram-treated group with special diet (35 days) (n = 9)

• 5.

  CIT 7 – citalopram-treated group (7 days) (n = 9)

Animals were housed in groups of 4–5 mice per cage under standard conditions (12 h natural light-dark cycle, 23°C ± 1°C, relative humidity 50–60%). Animals were given unrestricted access to water and feed, either standard mouse chow (Biostan Mypo, Biokron s r.o., Blucina, Czech Republic) or special diet depending on the group. Special diet consisted of standard mouse chow with added sucrose and cholesterol to increase carbohydrate and lipid content. Special diet was custom prepared by the manufacturer and its specific composition is detailed in the online supplementary Table (for all online suppl. material, see https://doi.org/10.1159/000541229). Food and water were available ad libitum. Animal weight and feed consumption were monitored twice a week.

In the medicated groups, citalopram (Seropram^® inf, Lundbeck, Copenhagen, Denmark) was administered intraperitoneally at a dose of 10 mg/kg/day, while the control groups received the vehicle, i.e., saline, at the same time (daily at 9:00 a.m.). Citalopram or saline solution was applied for 5 weeks (except for the C7 group, which has been receiving citalopram only for the last 7 days before the end of the experiment).

All animal care and experimental procedures complied with the national laws respecting Directive 2010/63/UE and were approved by the Animal Care Committee of the University of Veterinary and Pharmaceutical Sciences Brno and by the Ministry of Education, Youth and Sports of the Czech Republic (ref.n.MSMT-13558/2016-3).

After the described time of administration, animals were fasted for 12 h, sacrificed and trunk blood was collected under general anesthesia induced by isoflurane (Forane^®; AbbVie, North Chicago, IL, USA). Collected blood was centrifuged (10 min at 1,917 g), the obtained serum was frozen and later used for spectrophotometric biochemical analysis of basic parameters (HDL, LDL, total cholesterol, triglycerides (TGLs), and glucose) using BioVendor kits and determination of serum levels of adipokines and cytokines by ELISA on Bio-Plex^® reader using Bio-Rad sets.

The resulting data are presented as means of the observed values and standard deviations (SD). Two-way ANOVA and subsequent Tukey post hoc test were used for statistical evaluation of groups of the 5-week experiment. Student’s t test was used to evaluate the difference between CTR (control) and CIT 7 (short-term 7-day citalopram without diet) groups. All analyses include 9 animals per group. Normality of the data was checked using histogram. Pearson correlation coefficient was used to report correlation. We considered p value <0.05 as the threshold for statistical significance. Data were analyzed using GraphPad Prism 6 (GraphPad Software, San Diego, CA, USA).

With an average start weight of 23.15 g ± 0.62 g and average end weight of 25.5 g ± 1.32 g, the body weight data are presented by relative weight gain at the end of the 5-week experiment compared to the start. We found significantly lower body weight gain due to citalopram (F(1, 31) = 10.24, p = 0.0032) and no significant change due to diet.

The cumulative food intake is presented due to housing conditions. It represents the total feed consumed by the group throughout the experiment. Although the diet groups consumed slightly more food, there was no significant difference. Table 1 and Figure 1 provide additional details on body weight changes and food intake.

We found significant changes in the lipid spectrum already after a week of citalopram administration. We observed an increase in total cholesterol (p = 0.002), LDL cholesterol (p < 0.0001), HDL cholesterol (p = 0.03), and TGLs (p = 0.03) compared to control. Cholesterol levels remained high also in a 5-week long experiment due to special diet – this applies to total (F(1, 32) = 144.7, p < 0.0001) and LDL cholesterol (F(1, 32) = 541.5, p < 0.0001). HDL cholesterol levels were lowered by special diet (F(1, 32) = 10.96, p = 0.0023) and increased by citalopram (F(1, 32) = 6.623, p = 0.0149). The atherogenic index (AI) was then calculated for each animal from the cholesterol data using the formula AI = (total cholesterol – HDL cholesterol)/HDL cholesterol. AI was significantly elevated by diet (F(1, 32) = 248.3, p < 0.0001) but unaffected by citalopram and positively correlated with LDL cholesterol levels (r = 0.9971, p = 0.0002).

TGL levels were lowered after a 5-week experiment due to the influence of diet (F(1, 32) = 92.53, p < 0.0001), but mainly we observed a significant interaction between the effects of diet and citalopram (F(1, 32) = 6.251, p = 0.0177). Subsequent simple main effect analysis showed that citalopram does not further affect TGL levels when diet is present (p = 0.0659) but lowers them without diet (p = 0.006), i.e., diet influences TGL levels more than citalopram. You can find further information in Table 1 and Figure 2.

Analysis revealed no significant changes after 1 week of citalopram administration but some alterations in glucose metabolism in a 5-week long period. An ANOVA test showed changes in serum glucose levels due to two factors – citalopram (F(1, 32) = 6.038, p = 0.0196) and special diet (F(1, 32) = 4.609, p = 0.0395). In the comparison of individual groups by post hoc test, only the presence of both factors (CIT DIET group) in comparison with the CTR proved to be significant (p = 0.0135). We also found increased levels of glucose-dependent insulinotropic peptide (GIP) by citalopram (F(1, 31) = 5.177, p = 0.03) and increased glucagon by special diet (F(1, 29) = 8.429, p = 0.007). In the case of GIP and glucagon, significant ANOVA result was not followed by the significant result of the post hoc test, possibly due to the small group size. Insulin and GLP-1 levels did not show any significant changes. Additional details are provided in Table 1 and Figure 3.

We assessed serum levels of selected adipokines and hormones relevant to the metabolism of lipids and sugars and/or development of metabolic syndrome – namely leptin, ghrelin, resistin, and plasminogen activator inhibitor-1 (PAI-1). We have not found any significant difference in serum levels of the adipokines after 1 week of citalopram administration. Following an ANOVA test of groups in the 5-week long experiment, we found a significant increase in leptin levels due to a special diet (F(1, 31) = 8.260, p = 0.0073) and decreased leptin serum levels in groups with citalopram (F(1, 31) = 8.976, p = 0.0053). Additionally, we observed a highly significant increase in PAI-1 serum levels caused by a special diet (F(1, 30) = 49.67, p < 0.0001). No significant alterations were observed in ghrelin or resistin serum levels. Further information can be seen Table 1 and Figure 4.

Abdominal obesity and weight gain are defining features of metabolic syndrome and side effects of certain psychopharmaceuticals that can hinder therapy adherence. The impact of citalopram and other SSRI antidepressants on body weight is uncertain as they may initially reduce weight but are more prone to causing weight gain in long-term clinical studies [5, 6, 8, 19, 20].

In a preclinical setting, significantly reduced weight gain was observed after the administration of citalopram and fluoxetine (another drug of SSRI class) [21, 22]. In our experiment, we reached a similar conclusion as we observed a decrease in weight gain after 5 weeks of citalopram administration compared to the control group, without any significant changes in food intake. Therefore, studies carried out thus far (lasting up to 12 weeks), suggest that citalopram and other SSRIs lead to weight loss in rodents. However, these findings may not necessarily apply to longer periods of administration. Additionally, it is possible that this effect is also dependent on sex. A study conducted on Wistar rats found that fluoxetine administration did not result in weight gain in females, while males experienced only a partial decrease in weight gain [23].

Interestingly, a decrease in weight was observed due to the diet, although not significant. We hypothesize that the decrease may be attributed to the animals’ satiety-driven eating habits, resulting in a protein deficiency [24]. The special diet contains proportionally less protein. As a result, muscle growth occurs to a lesser extent, leading to less weight gain that is not compensated for by the potential growth of adipose tissue.

Dyslipidemia increases cardiovascular mortality and morbidity, and changes in the lipid spectrum have also been observed in patients treated with antidepressants. In our experiment, we studied the basic parameters of lipid metabolism – serum concentration of total, HDL and LDL cholesterol, and TGLs.

As expected, we observed significantly increased LDL and total cholesterol levels at the expense of HDL cholesterol and TGLs in groups with a diet. An increase in total cholesterol persists even after adjustment for HDL cholesterol (i.e., atherogenic index). The atherogenic index is one of the predictors of atherosclerosis and cardiovascular mortality. In our study, AI was elevated solely by a special diet and strongly correlated with LDL cholesterol levels. A special diet (with high sugar and lipid content), therefore, unsurprisingly results in higher cardiovascular risk. Citalopram administration for 7 days resulted in an increase in serum levels of both HDL and LDL cholesterol (thus AI was not affected) as well as TGLs. After 5 weeks of citalopram administration, cholesterol levels remained elevated at levels similar after 7 days of administration, while TGL levels decreased.

These findings align with previous studies. Lipid spectrum imbalances with altered cholesterol levels and higher TGL levels have also been observed in clinical practice [20, 25, 26]. However, mouse lipid metabolism is different from human and animal models show decreased levels of TGLs due to SSRIs, specifically fluoxetine [23, 27]. The rapidity of drug-induced changes suggests that metabolic syndrome is not the sole cause of these side effects. This claim is supported by the findings of the intervention of antidepressants in the processes of lipolysis and lipogenesis via interference with sterol regulatory element-binding proteins [28, 29].

Contrary to expectations, we observed elevated TGL levels in the special diet groups. A reduction in triacylglycerol levels is less probable with a high-carbohydrate, high-cholesterol diet as such a diet tends to elevate TGL levels. However, there are specific circumstances or conditions under which a reduction in TGL levels may occur, particularly if other factors are involved. First of all, diets high in fructose, cholesterol, and lipids have been demonstrated to cause non-alcoholic hepatic steatosis [30]. This can subsequently result in the inhibition of hepatocyte nuclear factor 4α (HNF4α), which in turn leads to a decrease in TGL levels [31, 32]. Additionally, the elevated glucagon levels observed in the special diet groups (though not statistically significant) tend to result in reduced TGL levels and possibly high polysaccharide content (due to the grain-based diet, which is rich in fiber and a small amount of starch is added to the special diet) may help reduce TGL by slowing carbohydrate absorption and improving insulin sensitivity [33].

Besides glucose itself, we determined the serum concentrations of hormones that affect glucose utilization and homeostasis, namely insulin, glucagon, glucagon-like peptide-1 (GLP-1), and GIP. Insulin is a pancreatic hormone that improves glucose utilization and facilitates its transport into cells. Glucagon is a peptide hormone produced by alpha cells in the pancreas. It helps maintain glucose levels in the body through gluconeogenesis and glycogenolysis and thus prevents hypoglycemia. GIP and GLP-1 are short-acting incretin hormones that are secreted by endothelial cells in response to food passage. Among other things, they stimulate insulin secretion in the pancreas, affect glucagon levels (GIP increases, GLP-1 decreases), and regulate glucose homeostasis [34].

The effect on glucose metabolism is ambiguous in citalopram (or overall SSRI), according to clinical studies. In our experiment, we observed several trends (an increase in GIP levels due to citalopram, an increase in serum glucagon due to diet, increased insulin levels and a decrease in glucagon levels from citalopram and an increase in GLP-1 levels due to both factors), although none of them were significant. The large interindividual variability and short biological half-life of the studied parameters (mainly GLP-1 and glucagon, but also e.g., ghrelin), may have contributed to the high standard deviations observed and thus to this outcome. Additionally, the small size of the experimental groups could also be a contributing factor. Interestingly, the reduction in serum glucose levels in the presence of both diet and citalopram (CIT DIET group) was shown to be statistically significant compared to the control. However, since the immediate glucose state does not indicate a longer-term state of energy homeostasis, this finding alone cannot be conclusive as evidence of impaired glucose metabolism due to citalopram and/or a diet.

A study performed in the same animal model (C57BL/6 mouse) found significant changes in metabolic parameters (mainly glucose and insulin) due to a lipid-rich diet and an increase in serum insulin concentration after 4 weeks of escitalopram administration [35]. Our results are similar but not significant. This difference probably stems from the different composition of the diet in the two experiments. In the referenced study, a diet with a higher fat content was used. The diet’s higher fat content (administered for a prolonged period) leads to faster development of impaired glucose tolerance and insulin resistance [36, 37].

In addition to the above-mentioned parameters of glucose and lipid metabolism, we also determined the serum concentrations of some adipokines involved in energy homeostasis – leptin, resistin, PAI-1, and ghrelin. Leptin is an adipokine that is involved in maintaining energy homeostasis. At the hypothalamic satiety center, it reduces food intake and promotes energy expenditure. In obese individuals, its level is elevated because of leptin resistance and thus impaired weight regulation [6]. Our study found that after 5 weeks a lipid-rich diet increased leptin levels, while citalopram decreased them. Leptin levels increase in the special diet control group was accompanied by slightly higher feed consumption. In the group with citalopram and a special diet (CIT DIET), the effect of the citalopram weight reduction was outweighed by the effect of a diet. A statistically significant reduction in leptin levels following citalopram administration correlates with clinical experience, although the cases described relate to indications of citalopram other than depression [9, 38]. Citalopram is thus likely to cause leptin levels to decrease in animal models and humans.

PAI-1, an adipokine and endothelial hormone, functions as an antifibrinolytic and prothrombotic agent. Increased PAI-1 levels are linked to atherosclerosis risk and are associated with obesity, insulin resistance, metabolic syndrome, and higher pro-inflammatory cytokine levels. At the end of the experiment, we observed a significant increase in PAI-1 levels in the special diet groups, like other proatherogenic factors such as LDL cholesterol. Studies conducted in clinical practice and in animal models have shown a decreased level of PAI-1 after the administration of SSRIs in depressed patients and in animals with a depressive-like phenotype [39, 40]. Adjustment of elevated PAI-1 levels due to antidepressant therapy suggests a possible role for this hormone in the pathophysiology of mental disorders. In our experiment without the depression model, there were no significant changes in PAI-1 levels after citalopram administration.

Resistin, an adipokine studied for its potential role in insulin resistance, has uncertain effects on glucose metabolism. It may serve as a biomarker for inflammation in metabolic syndrome due to a positive correlation with the occurrence of insulin resistance and the level of pro-inflammatory cytokines. Isolated clinical studies in depressed patients show the potential of some antidepressants to reduce resistin levels, but only in patients who respond to pharmacological treatment. However, resistin levels remain unchanged in patients who do not achieve remission [41]. The results of our animal study show that there was no significant change in resistin levels. Thus, it can be concluded that neither a special diet nor citalopram administration has a significant effect on resistin levels in C57BL/6 mice.

Ghrelin is a gastrointestinal hormone that stimulates appetite in the hypothalamus. Ghrelin increases during fasting and decreases after a meal. This leads to increased food intake and body weight. During our experiment we found a slight increase in serum ghrelin concentration in the citalopram groups and a slightly greater increase in ghrelin levels in the special diet groups, accompanied by a slight increase in feed consumption. However, these changes were not statistically significant. Clinical studies investigating the effect of antidepressant therapy on human plasma ghrelin concentrations did not show any significant change [42]. These results suggest that citalopram is unlikely to interfere with ghrelin metabolism and secretion.

Citalopram is generally considered metabolically inert or beneficial in the short term. The observed reduction in weight gain compared to the control in our study supports this claim. However, we also observed dyslipidemia induced by both diet and citalopram after just 1 week of drug administration. When citalopram and diet are combined, the effect of the diet predominates, and the drug does not increase or modify (either improve or worsen) this negative effect. Based on our observations, it can be concluded that citalopram affects some metabolic parameters in the C57BL/6 mouse. To gain further insight into the observed alterations, it would be intriguing to track metabolic parameters over an extended period, during which citalopram is more closely linked to the development of metabolic syndrome. Additionally, monitoring stress hormone levels and pro-inflammatory cytokines would be valuable.

All animal care and experimental procedures complied with the national laws respecting Directive 2010/63/UE and were approved by the Animal Care Committee of the University of Veterinary and Pharmaceutical Sciences Brno and by the Ministry of Education, Youth and Sports of the Czech Republic (ref.n.MSMT-13558/2016-3).

The authors declare no competing interests.

This research was supported by project IGA 316/2017/FAF of the University of Veterinary and Pharmaceutical Sciences Brno.

T.H. and H.K.: conceptualization; methodology; investigation; formal analysis; resources; funding; acquisition; and writing – original draft, review, and editing. J.P.: conceptualization, methodology, validation, investigation, formal analysis, and writing – original draft. M.K.: methodology, resources, formal analysis, and writing – review and editing.

Tomáš Hammer and Hana Kotolová contributed equally to this work.

Data are not publicly available due to ethical reasons. Further inquiries can be directed to the corresponding author.