Aging entails distinct requirements for Rb at maintaining adult neurogenesis

Abstract

Cell cycle proteins play essential roles in regulating embryonic and adult neurogenesis in the mammalian brain. A key example is the Retinoblastoma protein (Rb) whose loss disrupts the whole neurogenic program during brain development, but only results in increased progenitor proliferation in the adult subventricular zone (SVZ) and compromised long-term neuronal survival in the adult olfactory bulb (OB). Whether this holds true of neurogenesis in the aged brain remains unknown. In this study, we find no evidence of irregular proliferation or early commitment defects in the mid-aged (12-month-old) and old-aged (20-month-old) SVZ following tamoxifen-inducible Rb knockout (Rb iKO) in mice. However, we highlight a striking defect in early maturation of Rb-deficient migrating neuroblasts along the rostral migratory stream (RMS), followed by massive decline in neuronal generation inside the aged OB. In the absence of Rb, we also show evidence of incomplete cell cycle re-entry (CCE) along with DNA damage in the young OB, while we find a similar trend towards CCE but no clear signs of DNA damage or neurodegenerative signatures (pTau or Synuclein accumulation) in the aged OB. However, such phenotype could be masked by the severe maturation defect reported above in addition to the natural decline in adult neurogenesis with age. Overall, we show that Rb is required to prevent CCE and DNA damage in adult-born OB neurons, hence maintain neuronal survival. Moreover, while loss of Rb alone is insufficient to trigger seeding of neurotoxic species, this study reveals age-dependent non-monotonic dynamics in regulating neurogenesis by Rb.

Keywords: Adult neurogenesis; Aging brain; NestinCreERT2 mice; Rb; Subventricular zone – olfactory bulb.

Fig. 1

Conditional loss of Rb has no major effect on SVZ-NSCs/progenitors’ proliferation in mid-aged (MA) and old-aged (OA) mice. Double immunostaining performed on brain sagittal sections in the aged SVZ with (A) YFP (green)-Ki67(red)-Hoechst (blue) in MA and OA Rb^+/− control and Rb^−/− mutant mice at 28dpt and 60dpt. Bar graphs showing normalized cells counts inside the SVZ of (B) YFP+ cells, (C) total Ki67+ cells including Ki67+ YFP+ cells and (D) % of Ki67+ cells in the total YFP population at various timepoints in both genotypes. Note the age-associated decrease in overall cell proliferation inside the SVZ between MA and OA mice as well as MA-28dpt and MA-60dpt in graphs C and D. The increase in % (YFP+ Ki67+) in total YFP population between OA-28dpt and OA-60dpt in D is likely due to the decrease in the size of YFP population associated with cell quiescence with age as seen in B. Asterisks indicate statistical significance of independent Student’s t-test: *p < 0.05, **p < 0.01, ***p < 0.001. Error bars correspond to standard error of the mean (SEM). Scale bar: 50 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Neuronal commitment in the SVZ does not seem to be affected by loss of Rb in MA and OA mice. Double immunostaining performed on brain sagittal sections of the aged SVZ with (A) YFP(green)-DCX(red), (D) YFP (green)-Sox2(red), and (E) YFP(green)-pH3(red) in MA and OA Rb^+/− control and Rb^−/− mutant mice at the timepoints indicated (Hoechst in blue). Bar graphs showing normalized cell counts inside the SVZ of (B) total DCX+ cells including YFP+ cells, (F) total Sox2+ cells including YFP+ cells, and (G) total PH3+ cells including YFP+ cells in both genotypes at indicated timepoints. (C) Graph showing % of DCX+ cells in total YFP population inside the SVZ. Independent Student’s t-tests showed no significant difference in the numbers of early stem/progenitors cells (YFP+ Sox2+) and neuroblasts (DCX+ YFP+) including dividing (Ki67+) and post-mitotic (pH3+) cells between genotypes in both age groups and timepoints studied. Error bars correspond to standard error of the mean (SEM). Scale bar: 50 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Defective neuronal maturation along the RMS in MA and OA mice following Rb iKO. Double immunostaining on brain sagittal sections with YFP (green)-DCX (red) and counterstained with Hoechst (blue), marking the three partitions of the RMS: VA, ventral arm; E, elbow; HA, horizontal arm in MA-28pt (A) Rb+/- control mice and (B) Rb-/- mutant mice. Insets in A and B show higher magnification images of the HA partitions, respectively. Bar graphs showing cell counts normalized to surface area in mm² in each RMS partition of (C) YFP+ cells and (D) (YFP+ DCX+) cells in both genotypes at timepoints indicated. (E) Graph illustrating % of (YFP+ DCX+) cells in total YFP population in each RMS partition in both genotypes at various timepoints in MA and OA mice. Compared with Rb+/- controls, Rb-iKO mice display a severe decline in the numbers of YFP+ cells and (YFP+ DCX+) cells detected in the VA and HA at MA and OA, primarily at 28dpt (A-D), with no difference in % of double positive cells in the total YFP population. This data highlights a defective neuronal maturation of migratory neuroblasts in the aged brain in the absence of Rb. Independent Student’s t-test were performed in C: *p < 0.05, **p < 0.01. Univariate ANOVA test was run in C and D: # p < 0.05, ## p < 0.01, ### p < 0.001. ‡ in D and E refers to data not assessed due to lack of matching levels in the RMS elbow at OA-28dpt, and, very low cell counts at OA-60dpt in both genotypes, respectively. Error bars correspond to S.E.M. Scale bar: 100 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

OB neurogenesis is severely reduced in the mid-aged brain following Rb iKO. (A) YFP immunostaining (green) counterstained with Hoechst (blue) on brain sagittal sections in the GCL (granule cell layer) and GL (periglomerular layer) of the OB in young aged YA, MA and OA Rb^+/− and Rb^−/− mice at 28dpt and 60dpt. Bar graphs designate YFP+ cell counts normalized to surface area in mm² in the (B) GCL and (C) GL at indicated timepoints in Rb^+/− and Rb^−/− mice. Note the sharp boost in OB neurogenesis (YFP+ cells) detected in Rb^+/− mice at MA-28dpt compared with YA-28dpt and OA-28dpt (A-C). Loss of Rb negatively affects this increase at MA-28dpt in both OB layers. (D) Scheme showing downregulation of DCX and upregulation of CR during normal neuronal maturation inside the OB. Panel in D shows double immunohistochemistry of DCX (green) and CR (red) in the GCL of Rb+/- mice. Bar graphs showing CR+ cell counts normalized to GCL surface area in mm², and % of DCX+ cells in total CR population in the GCL between age groups in Rb+/- control mice. Note the delayed neuronal maturation in MA-control brains as indicated by the high percentage of DCX+ co-labeled with CR (immature neurons) and the low number of CR+ cells only (mature neurons). Independent Student’s t-test; *p < 0.05, **p < 0.01, ***p < 0.001. Error bars correspond to S.E.M. Scale bar: 50 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Rb-iKO triggers cell cycle re-entry (CCE) and DNA damage response (DDR) in adult-born neurons inside the young adult OB. (A) Bar graph showing normalized YFP cell counts in the GCL at 28dpt in two Nestin-CreERT2-YFP transgenic lines used here with significantly higher recombination efficiency in Line 2 compared with Line-1. (B) Triple immunohistochemistry performed on OB coronal sections showing YFP (green), PCNA (red) and Ki67 (blue) in the GCL in YA Rb^+/− and Rb^−/− mice, and bar graph comparing percentages of PCNA+, Ki67+ and PCNA+ Ki67+ cells in the total YFP population in the GCL between genotypes. Note the increase in all three populations upon loss of Rb as a sign of CCE. (C) YA mice were subjected to birthdating studies with BrdU (refer to methods for detail). Triple staining for YFP (green), PCNA (red) and BrdU (blue) at 28dpt in both genotypes, bar graphs showing percentage of BrdU+ cells in total GCL YFP+ cells, percentages of YFP+, PCNA+ and YFP+ PCNA+ in total GCL BrdU+ cells, and percentage of PCNA+ in (YFP+ BrdU+) cell population in the GCL, respectively. Note the sharp increase in the percentage of newborn neurons (YFP+ BrdU+) co-expressing PCNA, confirming the presence of CCE inside the OB. (D) Triple immunostaining for YFP (green), γH2A.X (red) and pH3 (blue) at 28dpt in the GCL at YA in both genotypes with bar graph showing percentages of γH2A.X+, pH3+ and γH2A.X+ pH3+ in the total GCL-YFP cell population. This data shows DDR in Rb-null newborn YFP+ neurons as indicated by the high % of (γH2A.X+ YFP+) cells upon loss of Rb. White arrows indicate co-labeled YFP+ cells. Independent Student’s t-test: *p < 0.05, **p < 0.01, ***p < 0.001. Error bars correspond to S.E.M. Scale bar: 50 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6

Rb-iKO mice show no signs of CCE, DDR or neurodegeneration along the SVZ-OB axis at MA-210dpt. (A) Bar graphs showing normalized cell counts inside the SVZ of total YFP+ cells, total Ki67+ cells including (YFP+ Ki67+), and total PCNA+ cells including (YFP+ PCNA+) cells in Rb^+/− and Rb^−/− mice at MA-210dpt. (B) Double immunostaining showing expressions of YFP (green) and H3K9me3 (red) with no significant difference in signal between genotypes inside the SVZ at MA-210dpt. (C) Bar graphs showing, left, YFP+ cell counts normalized to surface area in GCL and GL in Rb^+/− and Rb^−/− brains; middle, percentages of PCNA+, Ki67+ and (PCNA+ Ki67+) cells out of total GCL YFP+ cells; and, right, percentages of γH2A.X+, pH3+ and (γH2A.X+ pH3+) out of total GCL YFP+ cells with no significant difference between genotypes at MA-210dpt. (D) Double immunohistochemistry showing YFP (green) and pTau (red) expressions with normalized counts of pTau+ and (YFP+ pTau+) cells in the GCL of both genotypes at MA-210dpt. (E) Double immunohistochemistry showing YFP (green) and Synuclein (red) expression with normalized counts of (YFP+ Syn+) cells in the GCL of both genotypes at MA-210dpt. Data indicates no major change in the expression patterns of pTau and Syn (or accumulation of aggregates) between genotypes. Error bars correspond to standard error of the mean (SEM). Scale bar: 50 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7

Summary of the distinct roles played by Rb during adult neurogenesis with age. Aging entails distinct requirement(s) for Rb in the regulation of adult neurogenesis inside the OB. Loss of Rb in the young adult brain leads to enhanced progenitor proliferation with no major defects in neuronal differentiation or maturation, but a severe compromise in long-term survival of newborn OB neurons , which is due to high predisposition for CCE and DDR. In contrast, in the mid-aged brain, conditional deletion of Rb does not influence progenitor proliferation or neuronal commitment but negatively affects maturation of neuroblasts, and potentially their survival (sharp decrease in OB neurogenesis). In the old aged brain, loss of Rb seems to be associated with a similar phenotype as in the mid-aged brain, with signs of CCE in Rb-null newborn OB neurons. Of note, phenotypic assessment in OA mice could be masked by the overall decline in the level of neurogenesis with age.

Supplementary figure 1

Supplementary figure 2