Distinct, common and synergistic effects of insulin and IGF-1 receptors on healthy murine ageing

Abstract

Objective: Reduced IGF-1 signalling is an evolutionarily conserved mediator of longevity, yet the magnitude of this effect is substantially larger in organisms retaining a common insulin and IGF-1 receptor. Whether this reflects the failure to simultaneously reduce IGF-1 and insulin signalling in mammalian model systems remains unexplored, as is the associated impact on markers of healthy ageing. We set out to address these uncertainties.

Methods: We compared the duration of healthy life (healthspan) in male mice with haploinsufficiency of the insulin receptor (IRKO), IGF-1 receptor (IGF-1RKO), or both (DKO), versus wildtype (WT) littermates. Cognitive performance was defined using nesting studies at 3- and 24-months of age. Brain transcriptome was characterised at 3- and 18-months of age using RNA-seq.

Results: Healthspan was longer in DKO versus WT, with IRKO and IGF-1RKO being intermediate. At 2 years of age, DKO also exhibited preserved nesting behaviour in contrast with all other genotypes. Differential insulin sensitivity or weight gain during ageing did not explain the preserved healthspan of DKO, since these were comparable to IRKO littermates. Brain transcriptomics at 18 months of age revealed lower expression of canonical ageing-associated genes in DKO versus WT, although many of these findings were replicated in IRKO versus WT or IGF-1RKO vs WT.

Conclusions: Reduced insulin and IGF-1 receptor expression have both common and synergistic effects upon elements of healthy mammalian ageing, suggesting future ageing studies should consider targeting both insulin and IGF-1 signalling.

Keywords: Ageing; Healthspan; IGF-1; Insulin; Metabolism.

Fig. 1

Metabolic characterisation during ageing A) Body mass during ageing (n = 15/genotype); B) Body mass at 3 months (ANOVA p < 0.001; n = 15,14,15,14); C) Body mass at 18 months (ANOVA p < 0.001; n = 15,11,14,13); D,E) Glucose tolerance testing at 20 months, quantified by area under curve (ANOVA p = 0.03; n = 10,10,13,13); F) Correlation between area under glucose tolerance test curve and body mass (p < 0.001; n = 46); G,H) Insulin tolerance testing at 20 months, quantified by area under curve (ANOVA p = 0.01; n = 11,10,13,13); I) Correlation between area under insulin tolerance test curve and body mass (p < 0.001; n = 47). In panels A–E and G-H data are presented as mean (SEM). * denotes p < 0.05. AU denotes arbitrary units.

Fig. 2

Healthspan is extended in DKO mice A) Mean nesting score at 24 months (Kruskal-Wallis p = 0.01; n = 4,5,7,11); B) Mean nesting score at 3 months (Kruskal-Wallis p = 0.42; n = 5,3,3,9); C) Kaplan-Meier curve illustrating healthspan (Log rank p = 0.04; n = 15/genotype). D) Summary data from Kaplan-Meier curves and pairwise log-rank tests between specified genotypes. In panels A–B data are presented as mean (SEM). * denotes p < 0.05. AU denotes arbitrary units.

Fig. 3

Immunoblotting of whole brain from 3-month old WT, IRKO, IGF-1RKO and DKO A) Immunoblots of insulin receptor (IR), IGF-1 receptor (IGF-1R), Serine-473 phosphorylated Akt (pAkt), total Akt and beta-actin from wild type (WT), insulin receptor knockout (IRKO) IGF-1R knockout (IGF-1RKO) and double knockout (DKO) mouse brain at 3-months of age (n = 3,3,3,2). B) Densitometry quantification of IR expression normalised to beta-actin; C) Densitometry quantification of IGF-1R expression normalised to beta-actin; D) Densitometry quantification of total Akt expression normalised to beta-actin; E) Densitometry quantification of pAkt expression normalised to total Akt. In panels B–E data are presented as mean (SEM); * denotes p < 0.05 (Student's t-test), whilst p values approaching statistical significance are specifically quantified above the relevant comparator bars; AU denotes arbitrary units.

Fig. 4

Serum insulin and IGF-1 in 18-month old WT, IRKO, IGF-1RKO and DKO Serum non-fasting insulin (A) and IGF-1 (B) concentrations defined by ELISA in wild type (WT), insulin receptor knockout (IRKO) IGF-1R knockout (IGF-1RKO) and double knockout (DKO) mice at 18-months of age. Data are presented as mean (SEM). There are no statistically significant between-group differences (Mann-Whitney test).

Fig. 5

Brain transcriptomic characterisation A) Volcano plot illustrating differentially expressed genes (DEGs) in 18-month old DKO versus 18-month old WT mice. Log₂ fold change data below zero denote lower expression in DKO than WT. Red data points denote 33 DEGs with FDR-adjusted p value < 0.05, with the top 10 labelled by gene name. B) Heatmap of 33 DEGs between 18-month old DKO versus WT, with IRKO and IGF-1RKO data presented to illustrate shared and distinct patterns using hierarchical clustering of rows. Data are normalised by gene with expression measured in arbitrary units. Unadjusted data are presented in Table 1. C) Venn diagram illustrating common and unique DEGs from comparisons of DKO vs WT, IRKO vs WT and IGF-1RKO vs WT.