3.2 Experimental Animal Data
3.2.2 Postnatal Developmental Studies
Bastos et al. (155) examined the immediate effects of chronic fluoxetine treatment on the development and lesion-induced plasticity of retinotectal axon projections in Lister Hooded rats. Two exposure periods were used. In one group, rat pups were injected i.p. with 0 (0.9% saline) or 7.5 mg/kg bw/day fluoxetine on PND 1 – 10. In the second group, the rats were i.p. injected with 0 or 10 mg/kg bw/day fluoxetine [purity not specified] on PND 14 – 28. On PND 21, a lesion was induced in the left retina of some of the rats in the second group. To allow for tracing of the retinotectal pathway, the right eye was injected with horseradish peroxidase on PND 9 or 27 in the 2 groups, respectively. On the day following the tracer injection, the animals in each group were killed for removal and sectioning of the brain. [A total of 57 animals were examined but the numbers treated and examined within most treatment groups were not specified.] In vehicle-treated rats receiving either the early or later treatments, uncrossed retinotectal pathways were arranged in discrete clusters of terminal labeling in the rostral portion of the tectum. Rearrangements in these pathways were observed in 33% (4 of 12) of rats treated with fluoxetine from days 14 to 28 and an unspecified percentage of rats treated from days 1 to 10. The changes were characterized by decreased density of terminal rostral tectum labeling and abnormal spreading of retinal terminal fields along the rostra-caudal axis. These results suggest that fluoxetine treatment induced an active reorganization of the retinotectal axons. Fluoxetine treat-ment was also found to increase plasticity of retinotectal axon projections following the induction of retinal lesions. Following lesion induction in vehicle-treated rats, there was a small reorganization of intact uncrossed projections with only a few terminals invading the denervated tectal surface. In 53%
(8 of 15) of rats in the PND 14 – 28 fluoxetine-treated group, amplified reorganization characterized by the obvious spreading of uncrossed retinal axons into denervated areas was noted following lesion induction. The study authors interpreted the data as suggesting that fluoxetine treatment induces axonal rearrangements and amplifies neural plasticity in the CNS of young rats.
Appendix II
Strengths/Weaknesses: Because Bastos et al. (155) do not clarify the number of animals in each group and assay, it is difficult to determine the reliability of the findings. The study authors’ use of the terms
“neonatal” and “juvenile” to describe the 2 treatment periods, PND 1 – 10 and 14-28, respectively, may not be fully accurate for the specific windows of treatment.
Utility (Adequacy) for CERHR Evaluation Process: The study by Bastos et al. (155) is of low utility for use in a CERHR review for two reasons. First, the fluoxetine-associated changes in retinotectal axon development in juvenile rats is not clearly a model for human risk. Had fluoxetine been shown to alter neuronal architecture in a manner leading to cognitive impairment, the applicability to human risk would have been clearer. Second, the i.p. injection of what may have been a high dose of fluoxetine decreases the relevance to oral exposure in humans. It is also noted that the early postnatal period in this study corresponds to the fetal period in humans, making the dose route and size potentially less relevant to humans.
Wegerer et al. (35) studied the effects of fluoxetine treatment on serotonergic and noradrenergic system development in rats administered fluoxetine during prepubertal and pubertal stages. Male Wistar rats were administered 0 or 5 mg/kg bw/day fluoxetine [purity not specified] in drinking water for 2 weeks starting at 25 or 50 days of age. The dose was said to be 10 times higher than the human dose; however, it has been shown to be the minimum dose that produces serotonin reuptake inhibition in adult rats.
Weight gain was monitored daily. Six rats per treatment group were sacrificed at various time periods.
The day 25 – 39 treated group of rats was killed at either PND day 50 (n = 6 animals) or PND 90 (n = 6), either 10 days or 8.5 weeks, respectively, following discontinuation of dosing. The day 50 – 64 treatment group was killed at 90 days of age. Brains were removed and homogenized for an examination of
3H-paroxetine and 3H-nisoxetine binding to serotonin and noradrenaline transporters, respectively.
Statistical significance of data from the control vs. treated group was determined by ANOVA followed by two-tailed post-hoc t-test. Fluoxetine and norfluoxetine levels in blood were analyzed by HPLC and UV detection. Plasma levels of fluoxetine and norfluoxetine were similar in both age groups of rats and ranged from 27 to 29.9 ng/mL and 242.4 to 271.8 ng/mL, respectively. There was no effect on body weight gain and no obvious behavioral changes. The earlier fluoxetine treatment that started at 25 days of age resulted in a significantly (~20%) increased density of 3H-paroxetine binding sites in the frontal cortex when measured at 50 and 90 days of age. This persistent effect was not seen in 90-day-old rats that received fluoxetine treatment starting at 50 days of age. In neither of the two ages evaluated were effects seen on the density of 3H-paroxetine binding sites in other brain regions examined, including the parietal cortex, occipital cortex, hypothalamus, and midbrain. Further, fluoxetine treatment had no effect on dissociation constants for 3H-paroxetine or 3H-nisoxetine or density of 3H-nisoxetine binding sites. The study authors postulated that the 2-week fluoxetine treatment beginning at day 25 may have caused serotonin-induced production and release of astrocytic growth factor. The study authors indicated that the biologic significance of these effects in rodents is not known, and this lack does not permit extrapolation to humans. However, they cautioned that this study suggests that fluoxetine exposure during the development of the serotonergic system is capable of inducing persistent changes in the brain’s structural architecture that are not produced following treatment of the mature brain.
Strengths/Weaknesses: In the study by Wegerer et al. (35), it appears that six rats/condition were used.
While this number is low, it appears acceptable for publication standards within this area of work. The route of administration (drinking water) may not be relevant to human medication exposure, and the
Appendix II
use of a single dose level precludes a dose-response evaluation.
Utility (Adequacy) for CERHR Evaluation Process: The study by Wegerer et al. (35) is adequate for use in the CERHR evaluation. Although the increase in serotonergic but not adrenergic projections in the frontal cortex in 25-day-old rats is intriguing, the significance of this alteration in rats for human risk is difficult to predict.
Norrholm and Ouimet (156) examined hippocampal dendritic spine density in juvenile Sprague-Dawley rats receiving acute or chronic fluoxetine treatment. In the acute study, rats were i.p. injected with a single dose of 5 mg/kg fluoxetine HCl [purity not specified] on PND 21. One control group was i.p. injected with 0.9% saline and a second control group was not handled. The rats were sacrificed 24 hours later, at PND 22. In the chronic study, rats were treated in the same manner as rats in the acute study, but dosing was continued (i.p. injection of 5 mg/kg bw/day) for 3 weeks.
Half of the animals were killed 24 hours after the last injection (PND 42), while the remaining animals were killed 21 days following the last injection (PND 62). Brain samples were prepared for a determination of dendritic spine density in the CA1 region of the hippocampus and the dentate gyrus.
Each treatment group contained three or four rats. Data were analyzed by two-tailed Student’s t-test.
The only significant effects observed in the acutely fluoxetine-treated rats were a 25.9% increase in total number of secondary dendrites and an 18.9% increase in summed dendritic length, which were significant when compared to a pooled group of saline and non-handled controls. Chronic fluoxetine treatment inhibited the age-related increase in CA1 dendritic spine density that was observed in saline and nonhandled controls between PND 22 and 62. CA1 dendritic spine density in fluoxetine-treated animals was significantly lower than in saline and non-handled control groups 24 hours after the chronic treatment ended (17.1 and 25.5% lower than saline and non-handled controls, respectively) and after the 3-week recovery period following chronic treatment (20.0 and 23.6% lower than saline and non-handled controls, respectively). Dendritic spine length in the CA1 was not affected by chronic fluoxetine treatment. No effects occurred in the dentate gyrus following acute or chronic treatment with fluoxetine. [These region-specific effects on spine density immediately after acute treatment, and 3 weeks following chronic treatment suggest that the development of dendritic spines was arrested during a period in which rapid growth would normally occur.] The study authors suggest that these results may reflect interference with either the formation or retention of new spines, which typically occurs during the second postnatal month in the rat hippocampus. Additional drugs were also examined and results are reported in the study but will not be reviewed here.
Strengths/Weaknesses: The study by Norrholm and Ouimet (156) uses a small sample size, as is common with these types of studies. The i.p. route of administration and the use of only a single dose level of fluoxetine are important weaknesses in the study.
Utility (Adequacy) for CERHR Evaluation Process: The study by Norrholm and Ouimet (156) is of low utility in a CERHR review. It is difficult to predict whether the fluoxetine-associated effects on formation or retention of dendritic spines can be extrapolated to humans.
Mendes-da Silva et al. (157) examined the effects of neonatal fluoxetine exposure on forced-swim behavior in Wistar rats. The forced-swim procedure has been widely used as a rodent model of learned helplessness or depression. Beginning 1 day following birth (PND 1) and continuing to PND 21, 26
Appendix II
rats/group [gender not specified] received saline or 10 mg/kg bw/day fluoxetine [purity not specified]
by s.c. injection. Body weight gain was monitored and data were analyzed by Students t-test. Body weight gain was significantly reduced from PND 9 to 21 in the fluoxetine group; however, by PND 60, body weights were equivalent in the 2 treatment groups. At 60 days of age, the rats were subjected to a forced-swim test. For the test, rats were placed in a tank of water from which they could not escape and were forced to swim for 15 minutes. One day later, the rats were returned to the tank for 5 minutes and latency to the first escape attempt and duration of behavioral immobility were measured. Swim-test data were evaluated by the Mann-Whitney two-tailed test. The study authors stated that fluoxetine-treated rats displayed reduced depressive behavior, as evidenced by an increased latency to escape and decreased behavioral immobility; however, these statements are not consistent with the tabular data. [Based on tables in the study it appears that the opposite is true. Latency to escape attempt was smaller in fluoxetine-treated (97.5 seconds) vs. control rats (154.5 seconds) and behavioral immobility was increased in the fluoxetine (24.5 seconds) vs. control group (9 seconds).]
Strengths/Weaknesses: The study by Mendes-da Silva et al. (157) contains ambiguous results.
Additional weaknesses are the s.c. route of administration and the use of only a single dose level of fluoxetine.
Utility (Adequacy) for CERHR Evaluation Process: The study by Mendes-da Silva et al. (157) has no utility for a CERHR review due to the ambiguity in the presentation of the results.
Dow-Edwards (158) treated pre-weaning Sprague-Dawley rats with 25 mg/kg fluoxetine [purity unspecified] s.c. on days 11, 13, 15, 17, and 19 (the morning that pups were discovered with dams was day 1). Pups (5 males and 5 females/litter) were reared by their dams and weaned on day 21. All pups within a litter received the same treatment and it is suggested that 15 litters were exposed to each of 3 treatment regimens (cocaine, fluoxetine, or vehicle). On day 75 and 76, animals were tested for auditory-startle reactivity and habituation. Behavioral data were examined with data collapsed across members of the same litter and same sex. The study was performed primarily to determine whether cocaine’s effects on development of the nervous system are consistent with effects upon 5HT reuptake inhibition. Thus, a fluoxetine group was used to directly explore the effects of serotonin reuptake inhibition independent of cocaine’s other possible mechanisms. At the time of behavioral testing, body weights did not differ between the fluoxetine-treated animals and the controls. Upon initial auditory-startle measurement, the fluoxetine-treated males but not females showed increased startle amplitudes in the latter trial blocks of the session (P = 0.062). This finding was interpreted as increased sensitization, a phenomenon that has been seen following lesion of raphe nuclei as well as in response to increased background noise or stimulus intensity. On the second day of testing, the fluoxetine-treated males were also more reactive to the startle stimulus. The authors interpreted these findings as consistent with a subtle reduction in function of neurons in the raphe complex, pathways known to have an inhibitory influence on startle responding. Similar findings have been reported in a tactile-startle paradigm following acute fluoxetine administration to adult rats (Geyer and Tapson, 1988; cited in Dow-Edwards (158)).
Strengths/Weaknesses: The study by Dow-Edwards (158) was well done. Although all members of a litter received the same treatment, data were reduced across the male vs. female members of each litter.
Thus, it is suggested that an effective number of 15 animals per group may have been used. The group
Appendix II
number is large compared to other studies presented in this section of the report. Weaknesses of this study are the use of the s.c. route of administration and the use of a single dose level of fluoxetine.
Utility (Adequacy) for CERHR Evaluation Process: The study by Dow-Edwards (158) is adequate for use in a CERHR review, but the conclusions are of low utility given the P value of 0.062.
[In conclusion, most studies in this section appear to use solid measurement techniques. How-ever, measurements were generally made on small numbers of subjects, which introduces sta-tistical power concerns. In addition, the i.p. or s.c. routes of exposure and the single-dose level designs are important limitations in an evaluation of human risk.]