Calorie restriction protects neural stem cells from age-related deficits in the subventricular zone

SVZ stem cell proliferation is transiently increased by calorie restriction

Calorie restriction was initiated at 14 weeks of age in a stepwise fashion with an initial 10% reduction of free feeding weight followed by a 25% reduction at 15 weeks and a 40% reduction at 16 weeks. Mice were maintained at a 40% reduction until euthanized at 6 months or 12 to 18 months. Age-matched controls were fed ad libitum, and maintained on a free feeding diet until sacrifice. We measured numbers of proliferating cells within the SVZ of young and aged mice fed either an ad libitum control diet or a calorie restricted diet by injecting mice with EdU and sacrificing them 2 hours later. SVZ proliferation was transiently enhanced in young calorie-restricted (CR) mice, but declined in aged mice, regardless of their feeding paradigm (Figure 1A-D), young ad libitum (AL) diet group: 98.8 ± 9.4, young CR: 159 ± 10.9, aged AL 41.7 ± 13.5, aged CR 32 ± 4.4, p < 0.05, F (1, 12) = 11.8). We detected no differences between male and female mice. This result suggests a temporally sensitive window during which calorie restriction can impact proliferation in the SVZ of aging mice.

Calorie restriction protects against the loss of neurogenesis in the aged SVZ

To begin to test whether CR preserves neuron production, we measured the number of neuroblasts (progenitor cells committed to a neuronal fate) in the SVZ. We used wholemount preparations to preserve the 3-dimensional structure of the SVZ niche for imaging studies (Figure 2). Wholemounts were immunostained for the neuroblast marker doublecortin (DCX) and imaged using confocal microscopy. As expected, we observed fewer neuroblasts in aged AL mice. However, there was no significant reduction in neuroblasts in the young and CR aged group (Figure 2A-E).

To test whether this survival of neuroblasts yielded more new neurons being born in the olfactory bulb, we measured olfactory bulb neurogenesis using the nucleotide analog BrdU, which was injected for 5 days. Mice were sacrificed two weeks later to allow BrdU-labeled cells time to reach the olfactory bulb. Based on mean numbers of BrdU-positive cells in the olfactory bulb, there were no differences in neuron production in the two groups of young mice. As expected, neurogenesis was significantly reduced in aged mice fed a normal diet. However, new neuron production was not reduced among aged mice in the CR group (Figure 2F), young AL: 153.6 ± 12.6, young CR: 179.3 ± 4.7, aged AL 64.3 ± 3.63, aged CR 141.7 ± 8.0, p < 0.05, F (1, 38) = 7.311). These data suggest that CR had no significant effect in young mice, but preserved neurogenesis in older mice.

Olfactory memory is enhanced by calorie restriction

In olfactory testing, young and aged mice fed a CR diet consistently outperformed young and aged mice fed an AL diet. All groups displayed equal sniffing times during the first odor exposure across trials, and retained olfactory memory following a 30-minute interval between odor exposures. However, during a 60-minute exposure interval, aged AL mice failed to retain olfactory memory, while all other groups recalled the odor with similar accuracy. Only mice fed a CR diet recalled an odor after a 120-minute interval between odor exposures, with the aged CR group outperforming the young AL group. No mice recalled odors after a 180-minute interval (Figure 3A).

During the first 5-minute exposure, all mice, regardless of age or diet, spent the same amount of time sniffing the odor-containing probe, indicating that the baseline was similar among mice. In addition, there was no effect of novelty between trials or testing days (Figure 3B). These data indicate that olfactory memory was not only protected, it was enhanced in mice fed a CR diet.

SVZ vasculature is remodeled with age despite calorie restriction

SVZ vascular density declined in all aged mice despite dietary intervention (Figure 4). This result suggests that the protective effects of calorie restriction are likely due to other modifiable factors in the microenvironment, such as inflammation, rather than by preserving vascular structure in the aging brain.

Calorie restriction prevents increased inflammation and senescence within the aged SVZ

The neural stem cell microenvironment can influence stem cell activity across the lifespan. We have previously shown that the SVZ microenvironment becomes inflammatory with age, correlating with declining neural stem cell function. We also found that proliferation and differentiation potential of neural stem cells is influenced by the activation state of microglia, the immune cells of the brain [19].

To begin to investigate inflammatory markers in young and aged mice fed either AL or CR diets, wholemount SVZ preparations were immunostained for ionized calcium binding adaptor molecule 1 (IBA-1), a marker of microglia, and cluster of differentiation 68 (Cd68), an inflammation-associated glycoprotein expressed on microglia. Wholemounts were imaged using confocal microscopy and the total number of IBA-1+ microglia were counted. IBA-1+ microglia were significantly more numerous in aged AL mice relative to young AL and CR mice; however, this was not the case in aged mice on a CR diet compared to young mice (Figure 5A). IBA-1+ microglia were also larger in aged AL mice relative to young AL, young CR, and aged CR, indicating an increased inflammatory phenotype with age that was not observed in aged mice fed a CR diet (Figure 5B). Similarly, microglia expression of the activation marker CD68 was also lower in the CR mice compared to those fed an AL diet (Figure 5C). Further, levels of the inflammation-associated cytokines interleukin-6 (IL-6) and interleukin-1-beta (IL-1β) were increased in aged AL SVZ, but not in aged CR mice compared to young control mice (Figure 6A). Together, these data suggest that microglia are directly impacted by calorie restriction, and this intervention decreases the inflammatory profile in the SVZ of aged mice.

We examined the effect of calorie restriction on senescence in the SVZ of young and aged mice by measuring β-galactosidase activity. Aged mice fed ad libitum displayed two-fold more senescent cells in the SVZ compared with young AD controls. Calorie restriction prevented this increase in senescent cells in the aged SVZ (Figure 6C). This protective effect of calorie restriction on neural stem cells is likely due in part to reduced inflammation and senescence in the SVZ niche.

Animals

Male and female C57Bl/6 mice (aged 6-18 months) were obtained from the National Institute of Aging calorie restricted colony. Mice were divided into young (6 to 7 months old) and old (12-18 months old) groups. Calorie restricted mice received less chow in a stepwise fashion to 40% of free feeding weight by 16 weeks (10% restriction at 14 weeks, 25% restriction at 15 weeks, and 40% restriction at 16 weeks) of age. Age-matched controls were fed ad libitum, and maintained on this diet until sacrifice. All animals were provided continuous free access to water. Animal handling and sacrifice were performed in compliance with IACUC guidelines.

Data from male and female mice were combined for this report, since we observed no differences in key measurements, including microglia activation and neurogenesis. This strategy also allowed us to use fewer mice, since this colony of calorie-restricted mice has been discontinued and mice were in limited availability.

SVZ wholemount dissection

SVZ tissue was dissected from the striatal wall as described previously [22]. Wholemounts were processed with immunohistochemistry as described below and positioned onto slides for further confocal analysis.

EdU and BrdU labeling and detection

Mice were injected with EdU (10 ng/ml solution) intraperitoneally (50 mg/kg) 2 hours before they were sacrificed and perfused to assess cell proliferation. For identification of newborn cells in the olfactory bulb, mice were injected daily for 4 days with BrdU. Two weeks after the last injection, mice were perfused with cold 1x PBS followed by 4% paraformaldehyde. After cryosectioning, brain sections were permeabilized with 0.3% PBST for 45 minutes at room temperature. The sections were then incubated in 2N HCl at 37° C for one hour followed by two 5-minute washes in borate buffer. Primary and secondary incubation proceeded according to immunohistochemistry protocol. The total number of EdU or BrdU positive cells were counted on all sections containing the SVZ and olfactory bulbs and averaged across sections.

Immunohistochemistry and measurement of senescent cells

Brain sections were fixed in 4% paraformaldehyde for 1 hour at room temperature and permeabilized/washed in 0.3% PBST. SVZ whole mounts were fixed in 4% paraformaldehyde overnight at 4°C and permeabilized/washed with 2% PBST. Cell cultures were fixed in 4% PFA for 15 minutes at room temperature and permeabilized/washed with 0.1% PBST. Prior to primary antibody incubation, tissues were incubated in blocking solution containing 10% normal donkey serum in the appropriate PBST concentration overnight at 4° C. Incubation in primary antibodies (anti-BrdU, 1:250, Novus Biologicals, #NB-500-169, Littleton, Colorado; anti-IBA1, 1:200, Wako, #019-19187, Richmond, Virginia; anti-Cd68, 1:250, ABD Serotech, #MCA1957, Oxford, UK; anti-GFAP, 1:500, Dako/Agilent, #Z0334, Santa Clara, California; anti-laminin, 1:250, Sigma Aldrich, #L9393, St. Louis, Missouri; anti-DCX, 1:200, Santa Cruz, #SC8066 Dallas, Texas; proceeded under blocking solution overnight at 4° C. The tissue was washed 3 times in PBST solution after incubation in primary antibodies. Under the same blocking solution, secondary antibodies were added and incubation proceeded overnight at 4° C. The tissue was washed 3 times in PBST solution after incubation in secondary antibodies. Nuclear staining consisted of 10 minutes of incubation in 1:1000 DAPI in PBS. Species-appropriate secondary antibodies were purchased from Jackson ImmunoResearch and used at a concentration of 1:250. The number of positive cells was measured in 10 confocal z-stacks for each mouse. Senescent cells were detected using The Calbiochem Senescence Detection Kit (EMDMillipore # QIA117) using the protocol in the kit. Beta-galactosidase positive cells were counted in the entire SVZ.

Preparation of mRNA and real-time PCR

Subventricular zone tissues were microdissected and processed using the RNeasy mini kit for RNA purification (Qiagen, Hilden, Germany, cat. # 74104). RNA amplification was performed using a 7500 Real-Time PCR System Thermal Cycler (Applied Biosystems, Foster City, CA) with Sybr Green PCR master mix (Applied Biosystems, cat. # 4309155) for amplification detection. Primer sequences were as follows: IL-6 Fwd: CAACGATGATGCACTTGCAGA Rev:GGTACTCCAGAAGACCAGAGG, IL-1β Fwd: GAGAACCAAGCAACGACAAAATA Rev: TGGGGAACTCTGCAGACTCAAAC. The ΔΔ Ct protocol was followed for quantification as previously described [39].

Olfactory figure

Olfactory memory is a neurogenesis-dependent process distinctly different from hippocampal-dependent memory [30]. In tests of olfactory memory, a novel, non-food odor is presented to a mouse, and sniff time is recorded across a 5-minute exposure period. The odor is removed for a time interval (30, 60, 120, or 180 minutes), and then presented to the mouse again for a second 5-minute exposure period. A relative decrease in sniff time during the second exposure indicates an olfactory memory of that odor. We assessed olfactory memory as previously described [30]. Briefly, mice were exposed to a 0.5 μl sample of a novel odor (Aura Cacia essential oils: cinnamon, clove, rosewood, lemon, thyme, or citronella, www.auracacia.com) blotted onto a 10 mm x 10 mm piece of Whatman filter paper. Each mouse was exposed to a control probe with 0.5 μl distilled water blotted onto the filter paper to familiarize the mouse with the procedure prior to the first testing session. The paper was secured in a disposable pipette and suspended in the home cage for 5 minutes, and time spent sniffing and rearing near the scented probe was recorded. Mice were exposed to the same scented probe following a randomly selected time interval (30, 60, 120, 180, or 240 minutes); sniffing and rearing times were again measured. Each mouse was tested in each time interval, with at least 24 hours between testing sessions. Different odors were used on each test interval so that no mouse was exposed to a single scent outside of the two testing sessions. Mice that did not move during the testing session were not included in the analysis.

Statistical analysis

Differences among experimental groups were analyzed using a two-way ANOVA with a Newman-Keuls correction for multiple comparisons. Between-group analyses were performed using a one-way ANOVA with a Bonferroni correction for multiple comparisons. All comparisons utilized an alpha value of 0.05.