Young blood reverses age-related impairments in cognitive function and synaptic plasticity in mice

Abstract

As human lifespan increases, a greater fraction of the population is suffering from age-related cognitive impairments, making it important to elucidate a means to combat the effects of aging. Here we report that exposure of an aged animal to young blood can counteract and reverse pre-existing effects of brain aging at the molecular, structural, functional and cognitive level. Genome-wide microarray analysis of heterochronic parabionts--in which circulatory systems of young and aged animals are connected--identified synaptic plasticity-related transcriptional changes in the hippocampus of aged mice. Dendritic spine density of mature neurons increased and synaptic plasticity improved in the hippocampus of aged heterochronic parabionts. At the cognitive level, systemic administration of young blood plasma into aged mice improved age-related cognitive impairments in both contextual fear conditioning and spatial learning and memory. Structural and cognitive enhancements elicited by exposure to young blood are mediated, in part, by activation of the cyclic AMP response element binding protein (Creb) in the aged hippocampus. Our data indicate that exposure of aged mice to young blood late in life is capable of rejuvenating synaptic plasticity and improving cognitive function.

Figure 1

Heterochronic parabiosis enhances dendritic spine number and synaptic plasticity in the aged hippocampus and elicits a plasticity-related expression profile. (a) Schematic depicting the parabiotic pairings. (b,c) Microarray analysis performed on the hippocampi of aged (18 months) isochronic and heterochronic parabionts 5 weeks after surgery. n = 4 mice per group. For all analyses, downregulated genes are shown in shades of blue, and upregulated genes are shown in shades of yellow. (b) Biological pathways involved in synaptic plasticity identified as part of the top signaling network (P < 0.05) using IPA software based on differentially expressed genes in isochronic and heterochronic parabionts. Inferred molecular interactions identified by IPA are shown in gray. (c) Heat map generated by unsupervised hierarchical clustering with a data set of genes differentially expressed between hippocampi of aged isochronic (Iso) and heterochronic (Hetero) parabionts using a cutoff at P < 0.01 and d score > 2 on the basis of Significance Analysis of Microarray (SAM). Color bars reflect the z scores. (d–j) Histological and electrophysiological analysis performed on aged (18 months) isochronic (n = 6) and heterochronic (n = 5) parabionts analyzed 5 weeks after surgery. (d) Immunohistochemical detection of Egr1, c-Fos and phosphorylated Creb (pCreb) protein in the DG of the hippocampi of aged isochronic and heterochronic parabionts. Arrowheads indicate individual cells. Scale bars, 100 μm (low magnification); 50 μm (high magnification). (e–g) Quantification of the immunostaining for Egr1 (e), c-Fos (f) and pCreb (g). Five sections per mouse were analyzed. (h,i) Representative Golgi stain image (h; scale bar, 5 μm) and quantification of dendritic spine density on tertiary branches (i). Five neurons per mouse were analyzed. (j) Population spike amplitude (PSA) recorded from the DG of aged parabionts. Representative LTP levels are shown for isochronic and heterochronic parabionts. All data are shown as the mean ± s.e.m. *P < 0.05, **P < 0.01, t test (e–g,i).

Figure 2

Administration of young blood plasma improves hippocampal-dependent learning and memory in aged mice. (a–c) Results from aged (18 months) mice that were cognitively tested after systemic treatment with young (3 months) or aged (18 months) plasma eight times over 24 d (100 μl per intravenous injection). n = 8 mice per group. (a) Schematic illustrating the chronological order used for plasma treatment and cognitive testing. (b,c) Hippocampal-dependent learning and memory assessed by contextual fear conditioning (b) and RAWM (c) after plasma treatment. (b) Percentage freezing time 24 h after training. (c) Number of entry-arm errors before finding the platform. (d) Cognitive testing by contextual fear conditioning in aged mice after systemic treatment with saline (n = 11), young plasma (n = 12) or heat-denatured (den.) young plasma (n = 10) (100 μl per intravenous injection). Results are shown as the percentage freezing time 24 h after training. All data are shown as the mean ± s.e.m. *P < 0.05, **P < 0.01, t test (b), repeated measures analysis of variance (ANOVA) with Bonferroni post-hoc test (c) or ANOVA with Tukey’s post-hoc test (d).

Figure 3

Creb mediates the enhancements in dendritic spine number and hippocampal-dependent learning and memory elicited by young blood in aged mice. (a–c) Western blot analysis (a) and quantification by optical intensity (b,c) of Creb overexpression and pCreb in isolated hippocampi of adult mice that were contralaterally locally infected with AAVs encoding either K-Creb or GFP. Hemi, hemisphere. (d) Schematic illustrating the chronological order used for AAV delivery of K-Creb or GFP and the parabiosis procedure. (e,f) Representative confocal image of an AAV-infected GFP-positive neuron and dendritic spines (scale bar, 25 μm; e) and quantification of spine density on tertiary branches (f). Five to seven neurons per mouse were analyzed. n = 8 isochronic and 6 heterochonic parabionts. (g) Quantification of immunostaining for pCreb in the DG of the hippocampi of aged animals systemically administered with young (3 months) or aged (18 months) plasma. n = 5 mice per group, and five sections per mouse were analyzed. (h–j) Cognitive testing of aged mice that were given bilateral stereotaxic injections of AAV encoding K-Creb (n = 8 per group) or GFP (n = 10 per group) into the DG followed by systemic treatment with young or aged plasma eight times over 24 d (100 μl per intraorbital injection). (h) Schematic illustrating the chronological order used for AAV delivery, plasma treatment and cognitive testing. Stx, stereotaxic. (i,j) Hippocampal-dependent learning and memory assessed by contextual fear conditioning (i) and RAWM (j) after plasma treatment. (i) Percentage freezing time 24 h after training. (j) Number of entry-arm errors before finding platform. All data are shown as the mean ± s.e.m. *P < 0.05, t test (b,c,g), ANOVA with Dunnett’s post-hoc test (i) or repeated measures ANOVA with Bonferroni post-hoc test (j).