Hmga2 promotes neural stem cell self-renewal in young but not old mice by reducing p16Ink4a and p19Arf Expression

Abstract

Stem cells persist throughout life in diverse tissues by undergoing self-renewing divisions. Self-renewal capacity declines with age, partly because of increasing expression of the tumor suppressor p16(Ink4a). We discovered that the Hmga2 transcriptional regulator is highly expressed in fetal neural stem cells but that expression declines with age. This decrease is partly caused by the increasing expression of let-7b microRNA, which is known to target HMGA2. Hmga2-deficient mice show reduced stem cell numbers and self-renewal throughout the central and peripheral nervous systems of fetal and young-adult mice but not old-adult mice. Furthermore, p16(Ink4a) and p19(Arf) expression were increased in Hmga2-deficient fetal and young-adult stem cells, and deletion of p16(Ink4a) and/or p19(Arf) partially restored self-renewal capacity. let-7b overexpression reduced Hmga2 and increased p16(Ink4a)/p19(Arf) expression. Hmga2 thus promotes fetal and young-adult stem cell self-renewal by decreasing p16(Ink4a)/p19(Arf) expression. Changes in let-7 and Hmga2 expression during aging contribute to the decline in neural stem cell function.

Figure 1. Hmga2 expression is elevated in stem cells in the hematopoietic and nervous systems but declines with age, in contrast to p16^Ink4a and p19^Arf

Hmga2 expression by microarray (A) and qPCR (B) in E14.5 fetal liver HSCs (Thy-1^lowSca-1⁺Lineage⁻Mac-1⁺ cells), P60 and P720 bone marrow HSCs (Thy-1^lowSca-1⁺Lineage⁻c-kit⁺ cells), and P60 bone marrow CD45⁺ hematopoietic cells. Microarray signal intensities represent mean±SD from 3 independent samples of each cell type (*P<0.01; D=significantly above background; U=not detectable above background). Differences were confirmed in independent samples by qPCR. C) qPCR in freshly dissected E14.5 telencephalon, P0 VZ, P30 SVZ, P360 SVZ, and P720 SVZ (expressed as fold change relative to P360 SVZ; mean±SD for 4–5 mice per stage). Hmga2 expression significantly declined with age (*P<0.01,**P<0.05). p16^Ink4a and p19^Arf expression could not be detected (U) in fetal, newborn, or young adult samples but significantly (**P<0.05) increased with age in P360 and P720 SVZ samples. D) Hmga2 and Hmga1 were elevated in SVZ versus differentiated layers of the cortex by qPCR in P30 wild-type mice (mean±SD of 3 independent experiments; *P<0.01,**P<0.05). E) Western blot of E14.5 telencephalon, P0 VZ, and P600 SVZ. Consistent with results by qPCR (D) and in situ hybridization (F), Hmga2 protein levels (E) were elevated in undifferentiated cells within the VZ as compared to differentiated cells in the cortex but declined with age. F) In situ hybridization for Hmga2 on sections of E11.5, E14.5, P0 and P600 forebrains (* indicates lateral ventricle). Hmga2 transcript levels were greatly elevated in the VZ/SVZ, but declined with age. T-tests were paired except in C.

Figure 2. Hmga2 promotes the self-renewal of fetal and young adult CNS stem cells

A) Typical neurospheres that formed after 9 days in non-adherent cultures from E14.5 telencephalon cells cultured at very low cell density (~1cell/μl). B) The percentage of cells from E11.5 telencephalon, E14.5 telencephalon, P0 VZ, and P49-56 SVZ of Hmga2^−/− mice and littermate controls that gave rise to multipotent neurospheres in culture, the diameter of these neurospheres, and their self-renewal potential (number and percentage of cells from individual primary neurospheres that gave rise to multipotent secondary neurospheres upon subcloning). Hmga2 deficiency significantly reduced self-renewal at all of these stages and stem cell frequency at P0 and P49-56 (mean±SD for 5–7 independent experiments per stage, *P<0.01, **P<0.05). C) P0 VZ cells were dissociated and plated in adherent cultures at clonal density (0.33cells/μl). Wild-type (+/+) stem cell colonies contained significantly more cells than Hmga2^−/−(−/−) colonies (3 independent experiments). The frequency of dying cells within multilineage colonies after 10 days in culture was examined based on activated caspase-3 staining (D), or the frequency of condensed, fragmented nuclei identified by 4′,6-diamino-2-phenylindole dihydrochloride (DAPI) staining (E). Only rare cells underwent cell death in Hmga2^+/+ and Hmga2^−/− colonies (3 independent experiments). F) A significantly lower percentage of cells within Hmga2^−/− multilineage colonies (22±11%, mean±SD) incorporated a 20 minute pulse of BrdU as compared to Hmga2^+/+ colonies (42±10%; 3 experiments; *P<0.01). G) P0 Hmga2^−/− VZ cells were infected with either GFP-only control lentivirus or with Hmga2+GFP lentivirus. Hmga2 over-expression significantly increased the size and self-renewal of Hmga2^−/− neurospheres (3 experiments: P<0.01, **P<0.05). All T-tests were paired.

Figure 3. Hmga2 deficiency does not affect the ability of CNS stem cells to undergo multilineage differentiation in culture but does reduce proliferation in vivo

A) P0 VZ cells were dissociated and cultured adherently at clonal density for 7 days in a medium that promotes self-renewal (see Methods). Hmga2^+/+ and Hmga2^−/− colonies contained few cells that exhibited signs of neuronal (TuJ1⁺) or astrocytic (GFAP⁺) differentiation and no oligodendrocytes (O4⁺), suggesting Hmga2 deficiency does not cause premature differentiation. B) Some colonies were cultured for another 7 days in a medium that favors differentiation (see Methods). Hmga2^+/+ and Hmga2^−/− stem cell colonies contained similar proportions of neurons (Tuj1⁺), astrocytes (GFAP⁺), and oligodendrocytes (O4⁺). C) The percentage of P49-56 SVZ cells that incorporated a two-hour pulse of BrdU in vivo was significantly reduced in Hmga2^−/− mice (−/−) as compared to wild type (+/+) controls (5 mice per genotype, 6–10 sections per mouse,*P<0.01). D) Cross sections through the P0 lateral ventricle. The lateral wall (white arrows and bars) was significantly thinner in the Hmga2^−/− brain as compared to littermate controls (**P<0.05; 3 brains per genotype, 6–8 sections per brain). E) Transverse sections through the P49 distal ileum. HuC/D⁺ neurons in myenteric plexi are indicated by white brackets. F) Hmga2^−/− mice had significantly fewer HuC/D⁺ neurons per transverse gut section than wild-type mice (N=5 guts for wild-type, N=4 for Hmga2^−/−, 8–10 sections counted per gut; *P<0.01). All T-tests were paired.

Figure 4. Restricted neural progenitors from the CNS and PNS proliferate normally in the absence of Hmga2

(A–D) E14.5 telencephalon cells, P0 VZ cells, or P49-56 SVZ cells were dissociated and cultured at clonal density for 12 days. (E–H) E14.5 gut cells, or P0 or P49-56 gut plexus/outer muscle layer cells were dissociated and cultured at clonal density for 12 days. A) Typical neuron-only or astrocyte-only colonies formed by E14.5 telencephalon cells. (E) Typical neuron-only or glia-only colonies formed by E14.5 gut cells. Neither the number of cells per neuron-only colony, nor the number of cells per glia-only colony, nor the frequency of cells that formed these colonies differed between wild-type (+/+; blue) and Hmga2^−/− (−/−; burgundy) mice in the CNS (B–D) or PHS (F–H) (3–4 independent experiments per stage; error bars always represent SD). All T-tests were paired.

Figure 5. Hmga2 negatively regulates p16^Ink4a and p19^Arf expression in CNS and PNS stem/progenitor cells from fetal and young adult mice but not from old adult mice

(A,C) CNS neurospheres were cultured from wild-type or Hmga2^−/− mice of different ages (E14.5 telencephalon, P0 VZ, P49-56 SVZ, or P570-600 SVZ). (B,D) PNS neurospheres were cultured from wild-type or Hmga2^−/− mice of different ages (E14.5 gut, or P0, P49-56, or P570-600 plexus/outer muscle layer). p16^Ink4a and p19^Arf transcript levels were determined by qPCR in primary CNS (A) or PNS (B) neurospheres. Each bar shows the fold increase in Hmga2^−/− as compared to wild-type neurospheres (error bars represent SD, 5–6 independent experiments per stage; *P<0.01, **P<0.05). Western blot for p16^Ink4a, p19^Arf, or β-actin (loading control) in CNS (C) or PNS (D) neurospheres. E) Western blot for p16^Ink4a, p19^Arf, or β-actin in freshly dissected E14.5 telencephalon, P49-56 SVZ, and P570-600 SVZ cells. In vitro and in vivo, p16^Ink4a and p19^Arf expression levels were elevated in Hmga2^−/− cells at the fetal and young adult, but not old adult, stages. F) P0 SVZ cells were dissociated from Hmga2^+/+ (+/+) or Hmga2^−/− animals (−/−), infected with GFP-only control virus or Hmga2+GFP virus, and allowed to form neurospheres. Hmga2 over-expression in Hmga2^−/− cells was sufficient to reduce p16^Ink4a to approximately wild-type levels by Western blot. All T-tests were unpaired.

Figure 6. Deletion of p16^Ink4a and/or p19^Arf partially rescues the defects in neural stem cell frequency and self-renewal potential as well as SVZ proliferation in Hmga2^−/− mice

Images show typical CNS neurospheres after 9 days culture of P49-56 SVZ cells at clonal density. p16^Ink4a/p19^Arf deficiency (A; 4–6 mice/genotype in 4 independent experiments), p16^Ink4a deficiency (B; 4–5 mice/genotype in 3 independent experiments), or p19^Arf deficiency (C; 3–5 mice/genotype in 3 independent experiments) did not affect the percentage of wild-type SVZ cells that formed multipotent neurospheres, the diameter of these neurospheres, or their self-renewal potential (absolute number or percentage of primary neurosphere cells that gave rise to multipotent secondary neurospheres upon subcloning of individual neurospheres) but did significantly increase the frequency, diameter, and self-renewal of Hmga2^−/− neurospheres. All data represent mean±SD (*, significantly different (P<0.05) from wild-type; §, significantly different from Hmga2^+/+p16^Ink4a/p19^Arf−/− mice (A) or Hmga2^+/+p16^Ink4a−/− mice (B) or Hmga2^+/+p19^Arf−/− mice (C); #, significantly different from Hmga2^−/−p16^Ink4a/p19^Arf+/+ mice (A) or Hmga2^−/−p16^Ink4a+/+ mice (B) or Hmga2^−/−p19^Arf+/+ mice (C)). D) BrdU⁺ cells in the SVZ of Hmga2/p16^Ink4a/p19^Arf compound mutant mice. E) p16^Ink4a/p19^Arf deficiency partially rescued the reduction in SVZ proliferation within young adult Hmga2^−/− mice without affecting proliferation in wild-type littermates (3 mice per genotype, 6–10 sections per mice). All T-tests were unpaired.

Figure 7. Hmga2 is not required for the self-renewal of CNS stem cells from old mice, and increasing let-7 microRNA expression with age contributes to the decline in Hmga2 expression

A) Cells were isolated from the SVZ of P570-600 wild-type (+/+) or Hmga2^−/− (−/−) mice, and cultured 9 days to generate neurospheres. The frequency of SVZ cells that formed multipotent neurospheres, their diameter, and their self-renewal potential were unaffected by Hmga2 deficiency (3 independent experiments). B) Typical photos of P570 SVZ showing infrequent BrdU⁺ cells in the SVZ, irrespective of genotype (* indicates lateral ventricle). 6.3±1.7% of wild-type SVZ cells and 5.9±1.3% of Hmga2^−/− SVZ cells from old mice incorporated a pulse of BrdU (mean±SD for 3 brains, 7–10 sections per brain). C) let-7 family microRNA expression was compared by qPCR in freshly dissected E14.5 telencephalon, P0 VZ, P360 SVZ, and P720 SVZ (expressed as fold change relative to E14.5 telencephalon; each bar represents mean±SD for 3–4 mice per stage). let-7b expression significantly increased with age (*P<0.01, **P<0.05). D) P0 wild-type SVZ cells were infected with GFP-only control lentivirus or with let-7b+GFP lentivirus. let-7b over-expression significantly decreased the diameter of the neurospheres and their self-renewal (3 experiments: *P<0.01). E) P600 wild-type SVZ cells were infected with GFP-only virus, Hmga2+GFP virus, or 3′-UTR truncated Hmga2 (lacking let-7 binding sites in the 3′UTR)+GFP virus. Only 3′-UTR truncated Hmga2 over-expression significantly increased the diameter and self-renewal of neurospheres (3 experiments: **P<0.05). F) Protein from neurospheres examined in (D) and (E) was subjected to Western blot. Over-expression of let-7b reduced Hmga2 expression and increased p16^Ink4a expression. Over-expression of 3′-UTR truncated Hmga2 increased Hmga2 expression in P600 CNS neurospheres and reduced p16^Ink4a expression. Over-expression of wild-type Hmga2 had little effect. All T-tests were paired except in C.