Disparate Central and Peripheral Effects of Circulating IGF-1 Deficiency on Tissue Mitochondrial Function

Abstract

Age-related decline in circulating levels of insulin-like growth factor (IGF)-1 is associated with reduced cognitive function, neuronal aging, and neurodegeneration. Decreased mitochondrial function along with increased reactive oxygen species (ROS) and accumulation of damaged macromolecules are hallmarks of cellular aging. Based on numerous studies indicating pleiotropic effects of IGF-1 during aging, we compared the central and peripheral effects of circulating IGF-1 deficiency on tissue mitochondrial function using an inducible liver IGF-1 knockout (LID). Circulating levels of IGF-1 (~ 75%) were depleted in adult male Igf1^f/f mice via AAV-mediated knockdown of hepatic IGF-1 at 5 months of age. Cognitive function was evaluated at 18 months using the radial arm water maze and glucose and insulin tolerance assessed. Mitochondrial function was analyzed in hippocampus, muscle, and visceral fat tissues using high-resolution respirometry O2K as well as redox status and oxidative stress in the cortex. Peripherally, IGF-1 deficiency did not significantly impact muscle mass or mitochondrial function. Aged LID mice were insulin resistant and exhibited ~ 60% less adipose tissue but increased fat mitochondrial respiration (20%). The effects on fat metabolism were attributed to increases in growth hormone. Centrally, IGF-1 deficiency impaired hippocampal-dependent spatial acquisition as well as reversal learning in male mice. Hippocampal mitochondrial OXPHOS coupling efficiency and cortex ATP levels (~ 50%) were decreased and hippocampal oxidative stress (protein carbonylation and F₂-isoprostanes) was increased. These data suggest that IGF-1 is critical for regulating mitochondrial function, redox status, and spatial learning in the central nervous system but has limited impact on peripheral (liver and muscle) metabolism with age. Therefore, IGF-1 deficiency with age may increase sensitivity to damage in the brain and propensity for cognitive deficits. Targeting mitochondrial function in the brain may be an avenue for therapy of age-related impairment of cognitive function. Regulation of mitochondrial function and redox status by IGF-1 is essential to maintain brain function and coordinate hippocampal-dependent spatial learning. While a decline in IGF-1 in the periphery may be beneficial to avert cancer progression, diminished central IGF-1 signaling may mediate, in part, age-related cognitive dysfunction and cognitive pathologies potentially by decreasing mitochondrial function.

Keywords: Cognitive function; IGF-1; Learning and memory; Mitochondria; Oxidative stress; ROS.

Fig. 1

Liver IGF-1 deficiency (LID) increases insulin resistance. a Experimental timeline of circulating IGF-1 deficiency induced by liver-specific AAV8-TBG-Cre (LID) or AAV8-TBG GFP (GFP) illustrating the timepoints for induction and functional analyses. b Circulating IGF-1 in the serum is significantly reduced in the LID (n = 12) compared with GFP (n = 6) mice at 18 months of age. c Body mass was not different between GFP (green) and LID (blue) mice at 18 months and 24 months and was comparable with young WT (black) mice (6-month reference group; n = 6). d Glucose tolerance test (GTT) and e insulin tolerance test (ITT) in 18-month GFP and LID mice and f area under the curve (AUC) for insulin tolerance test (n = 6–8). GTT/ITT were analyzed by two-way ANOVA with Sidak’s post hoc test (*p < 0.05). Serum IGF-1 and ITT AUC were compared with two-tailed student’s t test (*p < 0.05). Data are represented as the mean ± SEM. Radial arm water maze (RAWM); oxidative phosphorylation (OXPHOS)

Fig. 2

IGF-1 deficiency does not alter age-related changes in muscle mass or mitochondrial function. Decrease in a raw muscle mass and b muscle mass normalized to body mass in gastrocnemius and quadriceps femoris muscles from 24-month GFP and LID mice compared to 6-month WT (n = 6). Simultaneous c oxygen consumption rate (OCR) and d hydroperoxide production rate (ROS) in permeabilized red gastrocnemius fibers from 6-month WT, 24-month GFP, and 24-month LID mice measured in the O2K with Amplex UltraRed and normalized to wet fiber mass (n = 6). e Linear regression between gastrocnemius fiber OXPHOS capacity and normalized gastrocnemius mass. f Amplex UltraRed Reaction Rate (ROS) as a percentage of oxygen consumption rate (OCR) in gastrocnemius fibers. Statistical significance determined by ordinary one-way ANOVA with Tukey’s Multiple Comparison Test (*p < 0.05). Box plots depicted as mean ± SEM. Six-month WT (black); 24-month GFP (green); 24-month LID (blue). Glutamate (Glu); malate (Mal); adenosine diphosphate (ADP); rotenone (Rot); ascorbate (Asc); N,N,N′,N′-tetramethyl-p-phenylenediamine dihydrochloride (TMPD); antimycin A (AmA); ETC complex I (CI); ETC complex II (CII); ETC complex IV (CIV)

Fig. 3

IGF-1 protects against age-related increase in adipose mass and decrease in adipose mitochondrial function. Change in a raw fat mass and b fat mass normalized to body mass in epididymal white adipose tissue (eWAT) and subcutaneous white adipose tissue (sWAT) fat pads, c and ratio of visceral (eWAT) to subcutaneous (sWAT) fat from 24-month GFP and LID mice compared with 6-month WT (n = 6). Simultaneous d oxygen consumption rate (OCR) and e hydroperoxide production rate (ROS) production in eWAT sections from 6-month WT and 24-month GFP and LID mice measured in the O2K with Amplex UltraRed and normalized to wet tissue mass (n = 6). Linear regression between f eWAT OXPHOS Capacity and normalized eWAT mass and g CIV OCR and normalized eWAT mass (n = 6). h eWAT Amplex UltraRed Reaction Rate (ROS) as a percentage of oxygen consumption rate (OCR) (n = 6). i Lipid hydroperoxides measured using F₂-isoprostanes in eWAT from 6-month WT and 24-month GFP and LID mice (n = 6). Statistical significance determined by ordinary one-way ANOVA with Tukey’s Multiple Comparison Test (*p < 0.05). Box plots depicted as mean ± SEM. 6-month WT (black); 24-month GFP (green); 24-month LID (blue). Glutamate (Glu); malate (Mal); adenosine diphosphate (ADP); rotenone (Rot); ascorbate (Asc); N,N,N′,N′-tetramethyl-p-phenylenediamine dihydrochloride (TMPD); antimycin A (AmA); ETC complex I (CI); ETC complex II (CII); ETC complex IV (CIV)

Fig. 4

Circulating IGF-1 deficiency impairs hippocampal-dependent spatial acquisition and extinction. Mice (control and LID) were tested in the RAWM for spatial memory with 8 trials per day for 3 days (days 1–3 acquisition). One week post acquisition (day 10), mice were tested for memory of the learned target location (probe) followed by a reversal learning (extinction) of a new target location on day 11. During the acquisition phase, LID mice show increased path length to reach target (a), increased entries (b), and duration (c) into incorrect arms (errors) compared with controls. LID mice also show increased latency (d), albeit there were no differences in velocity between groups (e). f Memory of the learned target tested on day 10 was not different between groups. g LID mice made significantly more errors during reversal learning compared with controls. h Representative heat maps at the end of the third day of acquisition show that LID mice make more errors in spatial learning as measured by the aforementioned parameters

Fig. 5

Reduction in circulating IGF-1 levels decreases brain mitochondrial efficiency and increases stress response and oxidative damage. a Oxygen consumption rate (OCR) and b OXPHOS coupling efficiency (1 – LEAK/OXPHOS Capacity) of hippocampus mitochondria isolated from 18-month GFP and LID mice measured in the O2K with Amplex UltraRed and normalized to μg mitochondrial protein (n = 8–10). c NADP+/NADPH ratio and d ATP concentration of cortex from 6-month WT and 18-month GFP and LID (n = 5–8). e Relative protein abundance and f principal component analysis (PCA) of metabolic and stress response proteins of isolated cortex protein from 18-month GFP and LID mouse brains by targeted mass spectrometry and selected reaction monitoring (n = 6). g Ratio of reduced (GSH) to oxidized (GSSG) glutathione and h level of F₂-isoprostanes in the cortex of 6-month WT and 18-month GFP and LID mice (n = 5–8). i Protein carbonylation (FITC conjugated) normalized to Coomassie gel staining in the cortex of 18-month GFP and LID mice (n = 7–9). Statistical significance for experiments with two groups determined by two-tailed student’s t test (*p < 0.05), and statistical significance for experiments with three groups determined by ordinary one-way ANOVA with Tukey’s multiple comparison test (*p < 0.05). Box plots depicted as mean ± SEM. Six-month WT (black); 18-month GFP (green); 18-month LID (blue). Glutamate (Glu); malate (Mal); adenosine diphosphate (ADP); rotenone (Rot); ascorbate (Asc); N,N,N′,N′-tetramethyl-p-phenylenediamine dihydrochloride (TMPD); antimycin A (AmA); ETC complex I (CI); ETC complex II (CII); ETC complex IV (CIV)