2015 Volume 38 Issue 8 Pages 1245-1249
Duloxetine is a widely used antidepressant worldwide. In the present report, we evaluated its capacity to induce micronucleated polychromatic erythrocytes (MNPEs) and micronucleated normochromatic erythrocytes (MNNEs) in mice. Two assays were performed, one with a single chemical administration and the other with daily chemical administration. In the first, we administered the antidepressant once to groups of 5 mice by the intragastric (i.g.) route (2, 20, and 200 mg/kg) and performed the analysis at 24, 48, and 72 h postadministration. A control group administered i.g. distilled water was included in the assay, as well as another treated with the micronuclei-inducing chemical daunorubicin (2.5 mg/kg, injected intraperitoneally (i.p.)). In this assay, we found significant damage induced by duloxetine starting from the first time evaluated, showing the highest MNPE increase at the end of the assay. We observed a saturation effect as well, suggested by a decreasing relative efficiency with respect to each tested dose. In a second assay, we administered the antidepressant i.g. every day for 5 weeks (2, 6, and 12 mg/kg), and micronuclei analysis was performed at the end of each week. In this study, we also found a significant increase in both MNPEs and MNNEs which was clear by the second week of administration. Our results suggest that short-term as well as cumulative damage is produced by duloxetine. Thus, confirmation of the observed genotoxic potential in other models seems advisable, as well as caution when prescribing this antidepressant.
Depression is a psychiatric disorder characterized by dysfunctional psychological processes, including alterations in self-referential schemas, cognitive biases, ruminations and processing mode (over general versus concrete); these cognitive processes are associated with altered functioning of brain systems, such as prefrontal areas and cingulate cortex, amygdala, lateral habenula and hyppocampus.1) Flexibility of the mood is lost in patients, thereby causing it to decrease and to no longer be influenced by favorable external events.2) The disease is a widely disseminated disorder and a leading cause of disability with approximately 350 million people affected worldwide.3) It is a complex disorder mainly managed with both psychosocial treatment and medication.
In light of the huge health and socio-economical problems addressed by depression, various generations of drugs have been designed and tested to control the disease. One of these is duloxetine, a potent serotonin and norepinephrine reuptake inhibitor which effectively desensitizes various autoreceptors and promotes neuroplasticity. The compound is used in major depressive disorder, stress urinary incontinence, and generalized anxiety disorder; however, due to its analgesic efficacy it is also prescribed for chronic pain conditions such as diabetic peripheral neuropathy, fibromyalgia, chronic low back pain, and osteoarthritis knee pain.4,5)
A number of reports have indicated that the medication is generally well tolerated and has significant efficacy, although a few reports have suggested inconclusive data.6–8) These evaluations have been mainly focused on the socio-economical impact as well as on the induction of co-lateral health damage; however, very limited reports have been published regarding the effect of duloxetine on the genetic material, which may be a significant biomarker for future disease development. In the genotoxicity area of research we only found a report that measured genotoxicity in mice with negative results in blood and brain cells.9) Therefore, in light of the scarce studies in this scientific field and of the fact that duloxetine can be used for long periods in persons of different ages and even in pregnancy, we feel genotoxic studies should be extended to better identify the effect of the medication. Our present contribution refers to evaluating the rodent micronuclei inducing capacity of duloxetine, a widely accepted biomarker to determine clastogenic/aneugenic effects. With this purpose in mind, we applied two models: one with a single administration of the compound and the microscopic analysis performed daily for three continuous days; another with a daily administration of the antidepressant and the analysis made every week along five weeks.
Duloxetine (Cymbalta® Eli Lilly & Co., Mex., Mexico) was obtained as the usually prescribed pharmaceutical medication. Daunorubicin, desipramine, methanol, the Giemsa stain, and phosphate buffered saline (PBS) were purchased from Sigma Chemicals (St. Louis, Mo., U.S.A.). Duloxetine, daunorubicin, and desipramine were dissolved in distilled water immediately before use.
Male mice ICR with a mean weight of 25 g were obtained from “Susana Cabrera S.A.” (Mex., Mexico) and adapted to laboratory conditions for a week before the experimental treatment. Mice were kept in polypropylene cages at 23°C with heat-treated hard wood bedding, in a 12 h dark–light cycle 50±10% humidity and with free access to food (Rodent Laboratory Chow 5001, Purina) and water. The proposed protocols were approved by the Committee of Ethics and Biosecurity of the National School of Biological Sciences.
Three-Day (Acute) Experimental StudyFive groups with five individuals each were used in the assay. Before the chemical administration we obtained blood from the tail of each mouse, extended it on clean slides, fixed the cells in methanol for 5 min, and stained them with a 4% Giemsa solution made in PBS at a pH 6.8.10) Then, mice were treated as follows: the control group was intragastrically (i.g.) administered water (0.3 mL per mouse), another group was intraperitoneally (i.p.) injected daunorubicin (2.5 mg/kg), and three more groups i.g. administered duloxetine; one with 2 mg/kg, another with 20 mg/kg, and the last one with 200 mg/kg. In the assay, the low dose corresponded to the therapeutic dose range prescribed in humans, and the highest dose was about 70% of the LD50 (282 mg/kg) obtained in our laboratory for mouse using the method of Lorke.11) At 24, 48, and 72 h after the chemical administration we obtained a few drops of blood from the tail of each mouse and followed the previously described procedure for fixing and staining the cells.
Microscopic observation was carried out to quantify the number of micronucleated polychromatic erythrocytes (MNPEs) in 1000 polychromatic erythrocytes per animal. Micronuclei formed in these young erythrocytes showed recent damage induced by the evaluated compound.
Statistical evaluation of the obtained results was made with an ANOVA and the Dunnett’s tests. Besides, we determined the total genotoxic efficiency as well as the relative genotoxic efficiency by calculating the area beneath the curve.12)
Five-Week (Subchronic) Experimental StudyFor this assay, five groups with five individuals each were organized as follows: the control group was i.g. administered water daily (0.3 mL per mouse), another group was i.g. administered desipramine (10 mg/kg), and three more groups were i.g. administered duloxetine every day (one with 2 mg/kg, another with 6 mg/kg, and the last one with 12 mg/kg). All groups were allowed to freely consume food (Rodent Laboratory Chow 5001, Purina) and water. Desipramine was selected because of its high genotoxic capacity previously demonstrated in similar in vivo assays.13,14) Doses were selected after preliminary assays to avoid systemic toxicity. The low dose corresponded to the recommended therapeutic range for humans. In this assay we also registered the animal weight every third day.
With respect to the genotoxic evaluation, before the chemical administration we obtained blood from the tail of each mouse, extended it on clean slides, fixed the blood samples in methanol for 5 min, and stained them with a 4% Giemsa solution made in PBS at a pH 6.8. Then, at the end of each week we also repeated the above indicated procedure for obtaining, fixing and staining blood samples.
Microscopic observations to evaluate the genotoxic capacity of duloxetine were made in a similar way as those described for the acute assay. However, in this case, regarding micronuclei we registered the number of MNPEs in 1000 polychromatic erythrocytes per mouse, and we also identified the number of micronucleated normochromatic erytrhocytes (MNNEs) in 1000 normochromic erythrocytes per mouse. This last evaluation is indispensable in long term studies to detect the cumulative effect. Statistical evaluation of the obtained results was carried out with ANOVA and the Dunnett’s tests.
With respect to general toxicity, we found no deaths or other deleterious signs during the assay. Also, no significant changes in body temperature were detected in an independent assay made in a control group as well as in a group administered 200 mg/kg of duloxetine (5 mice each), where the body temperature was registered before the initiation of the experiment, and every 6 h along 72 h. Figure 1 shows the results obtained with respect to the induction of MNPEs. As expected, a low effect was observed in all groups of mice examined before the chemical administration, a pattern that was maintained in control mice throughout the assay. This result contrasted with the induction of MNPEs using the tested doses of duloxetine. The number of micronuclei induced with the three tested doses of the antidepressant was statistically significant with respect to their respective control value along the evaluated schedule; however, differences in the MNPEs induced among the three doses of duloxetine were not always significant. The high damage was found at 72 h of treatment. With regard to the positive selected agent (daunorubicin) we also observed a significant increase of MNPEs along the assay. Besides, through the calculation of the total genotoxic efficiency of duloxetine we determined a strong micronuclei induction from the first evaluated dose; moreover we also observed a decrease of the relative efficiency in regard to each of the applied doses, suggesting a dose-saturation effect (Table 1).
Dau=daunorubicin. Each bar represents the mean±S.E.M. obtained from 1000 polychromatic erythocytes per mouse. Five animals per group. *Statistically significant difference respect to the control value, ANOVA and Dunnett’s tests, p=0.05.
| Treatment | Area beneath curve (Total genotoxic efficacy) | Relative genotoxic efficacy | |
|---|---|---|---|
| A | B | ||
| Water | 7.2 | — | — |
| Dau (2.5 mg/kg) | 710.4 | 7.12 | 284.16 |
| DLX (2 mg/kg) | 340.8 | 3.7 | 170.4 |
| DLX (20 mg/kg) | 393.6 | 0.39 | 19.68 |
| DLX (200 mg/kg) | 446.4 | 0.045 | 2.232 |
Dau=daunorrubicin, DLX=duloxetine, A=maximum micronuclei induction/dose, B=area beneath curve/dose.
In regard to animal weight, a summary of the obtained results showed that before the experiment the mean weight found in all groups was 27.1 g, with values of 24.7 g and 28.2 as low and high weights, and at the end we found a mean of 29.1 g with 27.2 g and 30.6 g as the low and high weights. These data indicate a homogeneous weight pattern among groups, and suggest tolerability toward the tested doses regarding general systemic health. Besides, we observed no deaths or other signs of toxicity. Moreover, we made a 4-week study including two groups with 5 animals each: a control group daily administered with water by the oral route, as well as another with duloxetine (12 mg/kg). Rectal body temperature was measured before the assay and every third day, and the results showed a mean of 35.9°C with variations no greater than 0.5°C.
Figure 2 shows the results in relation to MNPEs. We found a constant low value in control mice along the assay, as well as a constant increase induced by duloxetine during the five weeks of the study. The two high doses revealed the stronger damage with respect to the control value (water administered mice), showing an increase of more than nine times at the end of the study. At this time, the potency of duloxetine was even higher than desipramine. Regarding the effect of duloxetine on the number of MNNEs. Figure 3 shows similar damage to that described before. However, in this case the effect induced by duloxetine was more clearly shown by the second week of the assay, with a somewhat stronger effect at the end of the study, when there were more than 20 micronuclei induced with the two high doses (similar to those exerted by desipramine); furthermore, a clear effect with the low dose was manifested in the last three weeks of the study. These data seem congruent with the cumulative genotoxic effect registered in normochromatic cells.
Des=desipramine. Each bar represents the mean±S.E.M. obtained from 1000 polychromatic erythocytes per mouse. Five animals per group. *Statistically significant difference respect to the control value, ANOVA and Dunnett’s tests, p=0.05.
Des=desipramine. Each bar represents the mean±S.E.M. obtained from 1000 normochromic erythocytes per mouse. Five animals per group. *Statistically significant difference respect to the control value, ANOVA and Dunnett’s tests, p=0.05.
The lack of a complete understanding as to why depression occurs introduces a complexity in its treatment although it opens the door for the development of drugs with novel mechanisms of action. In this context, given that duloxetine is a highly prescribed antidepressant, studies to confirm its clinical safety are mandatory. In the toxicological field, a number of mild or moderate effects have been described including dry mouth, nausea, insomnia, somnolence, dizziness, headache, constipation, and sexual dysfunction.15) More severe affections such as microscopic or lymphocytic colitis, pseudolymphoma, hyponatremia, various types of hepatic injury, and major malformations16) have also been sporadically registered.
Regarding preclinical and clinical security, genotoxic studies constitute a highly relevant approach because these studies can demonstrate the capacity of the compound to induce genic and chromosome alterations, which can be the basis of future disease development, such as various chronic degenerative illnesses and cancer. Moreover, with respect to antidepressants, previous reports by our laboratory as well as by other authors have shown positive results, and duloxetine in particular has been reported to induce embryotoxicity in aquatic organisms.13,14,17,18) Regarding duloxetine, researchers report having examined mouse blood and brain cells with the comet assay and observed negative results.9) This finding is contrary to our present results and could be related with the different endpoint measured and the applied test, or because DNA observations were made 24 h after duloxetine withdrawal, a time when DNA repair or the detoxification process could have been started. Nevertheless, as evaluation of the compound’s genotoxic potency requires various tests that identify different lesions, in our research we applied the micronucleus test as a contribution to this purpose. Micronuclei represent chromosome fragments, or whole chromosomes abnormally segregated during cell division. Its quantification is usually recommended in genotoxic test batteries due to the fact that it is a relatively easy and economical assay and because of its known high efficiency to show chromosome damage.19,20) Our study revealed a capacity of the tested antidepressant to increase the number of micronuclei in recently formed erythrocytes as shown in the acute assay, where bone marrow damage should have occurred 1–2 d prior to sampling.21) In this assay, a significant elevation of MNPE was observed with the three tested doses. Moreover, the evaluation of the total genotoxic efficiency indicated high micronuclei induction since the first tested dose.
Such genotoxic effect was confirmed in our subchronic assay where we observed an increase in young MNPE, as well as in mature cells (MNNE) which correspond to erythrocytes with a life time around 30 d, and whose identification suggests cumulative damage in this cell population throughout the exposure time.21,22) This is a potentially relevant observation because of the possible implications in long term treatments of depression, where such repetitive and cumulative damage could have some significance for the future health status. Our micronuclei results showed no lineal dose response among the induction exerted by the three doses of duloxetine, a result that seems to be in line with the saturation effect observed in the acute assay, and whose subjacent explanation can be related with an overwhelming of the enzymatic rate involved in processes such as DNA repair, biotransformation, or detoxification of duloxetine.
Although no information about health damaging effect has been reported so far, our data suggested the need to be careful in the treatment of the disease, including the potential synergism between antidepressants. Our genotoxic results also suggested the pertinence of performing studies with other experimental models to reach a solid conclusion on the issue. Moreover, as no explanation has been proposed for DNA/chromosomal damage induced by duloxetine and other antidepressants, the detection of specific molecular lesions or the involved biochemical routes is a matter of research. Also, secondary, external factors, such as the hypothermia induced by various antipsychotic drugs are known to alter the number of micronuclei in mice.23) In the present study we found no significant body temperature modification in a separate assay that registered such physiological aspect; however, a simultaneous measurement of micronuclei and temperature is a more adequate approach. Therefore, studies of this type can be carried out to explore the origin of micronuclei, as well as others directed to molecular oxidation, which can be involved in depression,24) or to modification on the activity of the duloxetine metabolizing enzyme (CYP2D6),25) among other aspects.
The authors declare no conflict of interest.
