Cellular and behavioral effects of cranial irradiation of the subventricular zone in adult mice

Abstract

Background: In mammals, new neurons are added to the olfactory bulb (OB) throughout life. Most of these new neurons, granule and periglomerular cells originate from the subventricular zone (SVZ) lining the lateral ventricles and migrate via the rostral migratory stream toward the OB. Thousands of new neurons appear each day, but the function of this ongoing neurogenesis remains unclear.

Methodology/principal findings: In this study, we irradiated adult mice to impair constitutive OB neurogenesis, and explored the functional impacts of this irradiation on the sense of smell. We found that focal irradiation of the SVZ greatly decreased the rate of production of new OB neurons, leaving other brain areas intact. This effect persisted for up to seven months after exposure to 15 Gray. Despite this robust impairment, the thresholds for detecting pure odorant molecules and short-term olfactory memory were not affected by irradiation. Similarly, the ability to distinguish between odorant molecules and the odorant-guided social behavior of irradiated mice were not affected by the decrease in the number of new neurons. Only long-term olfactory memory was found to be sensitive to SVZ irradiation.

Conclusion/significance: These findings suggest that the continuous production of adult-generated neurons is involved in consolidating or restituting long-lasting olfactory traces.

Figure 1. Focal irradiation decreased DCX immunoreactivity in the SVZ.

(A, B) Focal gamma-ray irradiation of the SVZ. Adult mice were anesthetized and placed in a stereotaxic frame for cranial irradiation. A lead shield protected their body during exposure of the SVZ to gamma rays. A total dose of 15 Gray was delivered in three equal fractions administered at two-day intervals. H, hippocampus. (C) DCX staining of neuroblasts in a coronal section of the SVZ from a sham-treated mouse (left) and from an irradiated mouse 7 months after SVZ irradiation (right). Note the weaker DCX staining in the SVZ of the irradiated (IRR) mouse. LV, Lateral ventricle. (Scale bar: 100 µm.). (D) Densitometric analysis of DCX immunoreactivity in the SVZ of sham-treated and irradiated mice 7 months after irradiation. OD, optical density. Student's t test; *** p<0.0001 (n = 6 ).

Figure 2. Irradiation decreased the number of DCX⁺ cells in the OB.

(A) Representative images showing DCX⁺ cells in coronal sections of the OB, 7 months after SVZ irradiation. The GL, EPL, GCL and RMSob are indicated. (Scale bar: 100 µm). (B and C) Densitometric analysis of DCX immunoreactivity in total OB (B), including the RMSob (C, left), GCL (C, middle) and GL, (C, right) of sham-treated and irradiated mice 7 months after irradiation. OD, optical density. Student's t test; *** p<0.0001. ** p<0.01 (n = 12). (D) Densitometric analysis of DCX staining along the rostrocaudal axis of the OB, in sham-treated mice and irradiated mice 7 months after irradiation (n = 12). All cell layers along the entire rostrocaudal axis of the OB were equally affected by SVZ irradiation. (E) Immature neurons visualized by DCX staining in the GCL and GL of sham-treated and irradiated mice, 7 months after irradiation. (Scale bar: 20 µm.)

Figure 3. Irradiation reduced the recruitment of new neurons.

(A, B) New cells were labeled with BrdU 3 days after the last focal irradiation and survival was determined 11 days later. Photomicrographs show OB coronal sections labeled with BrdU, for sham-treated and irradiated mice. (Scale bar: 30 µm). (C) New cells were labeled with BrdU 8 or 120 days after the first session of irradiation. The mean number of BrdU⁺ cells in the entire OB was determined for sham-treated and irradiated mice 11 days after the final BrdU injection. Student's t test; ** p<0.01. * p<0.05 (n = 3–6 mice). (D) BrdU⁺ cell number in the OB, including the GCL, EPL and GL, for sham-treated and irradiated mice, with BrdU injected 8 days after the first session of irradiation. Student's t test; ** p<0.01. * p<0.05 (n = 4–6).

Figure 4. Spontaneous discrimination was not affected by irradiation.

Sham-treated and irradiated mice were tested daily, in successive sessions of habituation/dishabituation and memory tests. Histograms indicate the mean time spent investigating an odorant during 90 seconds of exposure, with a two-minute period of rest between consecutive exposures. (A) Habituation with 6 successive exposures to linalool. Both groups showed progressively less interest in investigating the same odorant in repeated exposures, a process called habituation. (B–G) Sessions of habituation (4 successive exposures to the odorants indicated) followed by dishabituation (a single period of exposure to an odorant similar to that used for habituation), recall of habituation (two successive exposures to the odorant used for habituation) and a final dishabituation with a single exposure to a dissimilar odorant. The time spent investigating the test odorant is shown. The extent of dishabituation was similar for the 2 groups: all mice detected even small differences between similar odorants (3B). Neither irradiated nor sham-treated mice could distinguish spontaneously between the limonene and terpinene enantiomers (3C and D). No significant effects of irradiation were observed (Two-way ANOVA; p>0.05, n = 9–10 mice). (H) 30-minute olfactory memory was not affected by irradiation. A mint odorant was introduced into the cage for five minutes. Two minutes later, the same odorant was introduced again for five minutes. The odorant was introduced into the cage for a final two-minute period after a 30-minute rest period (memory test). Histograms indicate the mean time of investigation. No effect of irradiation was observed (two-way ANOVA; p>0.05 n = 9–10 mice).

Figure 5. Irradiation has no effect on performance in reinforced discrimination tasks.

(A) Odor detection thresholds were not altered by irradiation. Accuracy (% of correct responses) is shown for the detection of successively lower concentrations of (+)-carvone (10 blocks of 20 trials). (+)-carvone was the rewarded (S+) stimulus and the solvent, mineral oil (MO), was the non-rewarded (S-) stimulus. Water-deprived mice were first trained to distinguish between a high concentration of (+)-carvone and MO. They were then subjected to daily blocks of trials in which they were exposed to progressively lower concentrations. Acquisition rate was similar for the 2 groups. A score of 50% corresponds to the success rate expected on the basis of chance alone (dashed line, A–C). No significant differences were observed for irradiated mice (p>0.05, n = 7). (B) The acquisition of discrimination ability in separate 2-odorant discrimination tasks and performance in the corresponding 8-odorant task were not affected by irradiation. The accuracy of performance in the discrimination tasks is shown as a % of correct responses for 8 blocks of 20 trials for odorant pair A (1% anisole, S+ vs 1% cineole, S-), odorant pair B (0.1% n-amyl acetate, S+ vs 1% linalool, S-), odorant pair C (1% butanoic acid, S+ vs 1% beta-ionone, S-), odorant pair D (1% (+)-limonene, S+ vs 1% (+)-carvone, S-) and 4 blocks of 40 trials for 8-odorant tasks. In the two-odorant tests, the stimuli, A, B, C and D (giving eight possible permutations) were introduced in a random order. No effect of SVZ irradiation was observed (two-way ANOVA; p>0.05, n = 9–10). (C) The ability to distinguish between pairs of mixtures of two odors was not affected by irradiation. Mixtures contained 1% (+)-carvone (indicated by (+)-C or S+) and 1% (−)-carvone (indicated by (−)-C or S-). Five alternating mixtures with different ratios were used and animals were rewarded only when a go-response was observed in the presence of mixtures in which (+)-carvone was the dominant compound. Concentrations (%) of odors are given for the following pairs of mixtures: 8/2 vs 2/8: 0.8% (+)-C+0.2% (−)-C (S+) vs 0.2% (+)-C+0.8% (−)-C (S-); 7/3 vs 3/7: 0.7% (+)-C+0.3% (−)-C (S+) vs 0.3% (+)-C+0.7% (−)-C (S-); 6/4 vs 4/6: 0.6% (+)-C+0.4% (−)-C (S+) vs 0.4% (+)-C+0.6% (−)-C (S-); 5.2/4.8 vs 4.8/5.2: 0.52% (+)-C+0.48% (−)-C (S+) vs 0.48% (+)-C+0.52% (−)-C (S-). Performance accuracy is shown as a % of correct responses for 10 blocks of 20 trials. No effect of SVZ irradiation was observed (two-way ANOVA; p>0.05, n = 7).

Figure 6. Impaired long-term memory in irradiated mice.

Mice underwent 8 blocks of 20 trials every day for 4 days, to train them to distinguish between 1% anisole (rewarded odorant) and 1% cineole (non-rewarded odorant). Mice were tested on the same task after a rest period of one month, in one block of 20 trials, but with no reward given for a correct response. Representative results for experiments performed in triplicate are shown. (A–C) Mean values (%) for correct responses in the last block of training (acquisition) and in the first block of testing (memory test) are shown for each sham-treated (A) and irradiated mouse (B) and for all mice (C). (D) Means of errors in trials 1 to 20 of the memory test session. Two-way ANOVA followed by unpaired or paired Student's t tests, as appropriate; * p<0.05 (n = 9–10).