Fetal alcohol exposure leads to abnormal olfactory bulb development and impaired odor discrimination in adult mice

Abstract

Background: Children whose mothers consumed alcohol during pregnancy exhibit widespread brain abnormalities and a complex array of behavioral disturbances. Here, we used a mouse model of fetal alcohol exposure to investigate relationships between brain abnormalities and specific behavioral alterations during adulthood.

Results: Mice drank a 10% ethanol solution throughout pregnancy. When fetal alcohol-exposed offspring reached adulthood, we used high resolution MRI to conduct a brain-wide screen for structural changes and found that the largest reduction in volume occurred in the olfactory bulbs. Next, we tested adult mice in an associative olfactory task and found that fetal alcohol exposure impaired discrimination between similar odors but left odor memory intact. Finally, we investigated olfactory bulb neurogenesis as a potential mechanism by performing an in vitro neurosphere assay, in vivo labeling of new cells using BrdU, and in vivo labeling of new cells using a transgenic reporter system. We found that fetal alcohol exposure decreased the number of neural precursor cells in the subependymal zone and the number of new cells in the olfactory bulbs during the first few postnatal weeks.

Conclusions: Using a combination of techniques, including structural brain imaging, in vitro and in vivo cell detection methods, and behavioral testing, we found that fetal alcohol exposure results in smaller olfactory bulbs and impairments in odor discrimination that persist into adulthood. Furthermore, we found that these abnormalities in olfactory bulb structure and function may arise from deficits in the generation of new olfactory bulb neurons during early postnatal development.

Figure 1

Effect of fetal alcohol exposure on brain volume. High-resolution MRI was used to screen for volumetric differences between adult control (n = 10) and FAE (n = 10) mice in 62 brain regions. Negative and positive effect sizes indicate fetal alcohol exposure-induced decreases and increases in volume, respectively, with an effect size of ≥ 0.8 or ≤ -0.8 considered large (black bars).

Figure 2

Associative olfactory task. (a) Mice were trained to discriminate between two odors using an associative olfactory task in which the + odor but not the -odor was paired with sugar. (b) The + and -odors were composed of increasingly similar binary mixtures of R-carvone and S-carvone. (c) During training, mice spent more time digging at the + odor compared to the -odor regardless of odor similarity (100:0: n = 10; 90:10: n = 9; 80:20: n = 8; 70:30: n = 8; 60:40: n = 9; 50:50: n = 8). (d) During a probe test given 24 hours after training, mice spent more time digging at the + odor compared to the -odor when the odors were relatively distinct (100:0, 90:10, 80:20) but not when the odors were relatively similar (70:30, 60:40) or identical (50:50). (e) Mice had discrimination indices above chance level only when the odors were relatively distinct (100:0, 90:10, 80:20).

Figure 3

Effect of fetal alcohol exposure on associative olfactory task performance. (a) Adult control and FAE mice were trained to discriminate between two relatively distinct odors (100:0) or two relatively similar odors (80:20), with a 10 min (100:0: control n = 7, FAE n = 8; 80:20: control n = 9, FAE n = 9), 24 hour (100:0: control n = 8, FAE n = 8; 80:20: control n = 9, FAE n = 9), or 7 day (100:0: control n = 9, FAE n = 9; 80:20: control n = 10, FAE n = 9) delay between training and testing. (b) Both control and FAE mice spent more time digging at the + odor compared to the -odor during training. (c,d) Both control and FAE mice discriminated between + and -odors when the odors were relatively distinct (100:0 vs. 0:100), regardless of the delay between training and testing. (e,f) FAE mice failed to discriminate between + and -odors when the odors were relatively similar (80:20 vs. 20:80).

Figure 4

Associative olfactory task performance does not require the hippocampus. (a) Representative images of sham and hippocampal (HPC) lesions. (b) Both sham (n = 8) and HPC (n = 9) mice spent more time digging at the + odor compared to the -odor. (c) Both sham and HPC mice had discrimination indices above chance level. (d) HPC mice did not show habituation of activity levels in the open field. (e,f) HPC mice did not learn the location of a hidden platform in the water maze. (g) Density plots for grouped data showing where mice concentrated their searches for the hidden platform. The color scale represents number of visits.

Figure 5

Effect of fetal alcohol exposure on olfactory bulb neurogenesis. (a) Postnatal OB neurogenesis was assessed using 3 methods: (i) in vitro neurosphere assay, (ii) in vivo labeling with BrdU, and (iii) in vivo labeling with LacZ using a transgenic reporter system. (b,c) FAE mice (n = 9) had fewer in vitro neural precursor cells and fewer small neurospheres in the SEZ compared to control mice (n = 9) between P4 and P14. (d) FAE mice had fewer BrdU+ cells in the granule cell layer of the OB compared to control mice P7 (control n = 9, FAE n = 8) and P21 (control n = 8, FAE n = 9). (e) There was no difference in BrdU+ cells between control (n = 8) and FAE mice (n = 8) at P60. (f) FAE reporter mice had fewer LacZ+ cells in the granule cell layer of the OB at P30 (control n = 8, FAE (10%) n = 6, FAE (15%) n = 9).