Circulating Growth Differentiation Factor 11/8 Levels Decline With Age

Abstract

Rationale: Growth differentiation factor 11 (GDF11) and GDF8 are members of the transforming growth factor-β superfamily sharing 89% protein sequence homology. We have previously shown that circulating GDF11 levels decrease with age in mice. However, a recent study by Egerman et al reported that GDF11/8 levels increase with age in mouse serum.

Objective: Here, we clarify the direction of change of circulating GDF11/8 levels with age and investigate the effects of GDF11 administration on the murine heart.

Methods and results: We validated our previous finding that circulating levels of GDF11/8 decline with age in mice, rats, horses, and sheep. Furthermore, we showed by Western analysis that the apparent age-dependent increase in GDF11 levels, as reported by Egerman et al, is attributable to cross-reactivity of the anti-GDF11 antibody with immunoglobulin, which is known to increase with age. GDF11 administration in mice rapidly activated SMAD2 and SMAD3 signaling in myocardium in vivo and decreased cardiac mass in both young (2-month-old) and old (22-month-old) mice in a dose-dependent manner after only 9 days.

Conclusions: Our study confirms an age-dependent decline in serum GDF11/8 levels in multiple mammalian species and that exogenous GDF11 rapidly activates SMAD signaling and reduces cardiomyocyte size. Unraveling the molecular basis for the age-dependent decline in GDF11/8 could yield insight into age-dependent cardiac pathologies.

Keywords: Gdf11 protein, mouse; Mstn protein, mouse; aging; intercellular signaling peptides and proteins; transforming growth factor-β.

Figure 1. Growth differentiation factor 11 (GDF11)/8 processing schematic diagram

Schematic representation of GDF11/8 protein structure and the corresponding processing leading to the mature and active ligands. The active ligand is reduced to 2 ≈12.5-kDa monomers in the presence of reducing agent. SP indicates signal peptide.

Figure 2. Serum levels of growth differentiation factor 11 (GDF11)/8 decrease with age in mouse, rat, horse, and sheep

Western analysis for GDF11/8 protein (5 ng, positive control) on 5 µL of serum from mouse (A), rat (B), horse (C), and sheep (D) from different ages. Results using the anti-GDF11/8 antibody (Abcam) show age-dependent decline of the ≈12.5-kDa band and increase of the ≈25-kDa band. Coomassie stained gels (bottom) for the corresponding Western blots are shown below each Western analysis. E, Taqman quantitative real-time polymerase chain reaction analysis of Gdf11 mRNA–detected reduced levels in the spleen and kidney but not skeletal muscle, in aged (24-month old) when compared with young (2-month old) mice. Gdf11 gene expression levels were normalized to hypoxanthine-guanine phosphoribosyltransferase. Each dot represents results from an individual mouse (n=6–9 mice), with mean±SD overlaid. *Empty well. M indicates molecular marker.

Figure 3. The 25-kDa band in mouse serum identified by the monoclonal antibody (Abcam) is consistent with IgG light chain and is absent in IgG-deficient mice

A, Western analysis on purified mouse IgG protein (10 µg) and on sera from young (2-month old; lanes 1, 2, and 3) and old (22-month old; lanes 5, 6, and 7) mice (left) and rats (right). B, Western analysis (top) and Coomassie stained gel (bottom) on sera from a young mouse and an old mouse and a titration of purified mouse IgG protein (IgG heavy chain [H.C.] and IgG light chain [L.C.]), showing identification on Western analysis of the IgG light chain at levels seen by Coomassie staining. C, Western analysis of sera obtained from 7-month-old IgG-deficient mice (Rag1 knockout [KO]; lanes 1, 2, and 3) and wild-type (WT) littermates (lanes 4 and 5) showing the absence of the ≈25-kDa band in the Rag1 KO group. All Western analyses were performed with anti–Growth differentiation factor 11 (GDF11)/8 antibody (Abcam). These data are consistent with our interpretation that the ≈25-kDa band detected by the anti-GDF11/8 antibody is IgG light chain. M indicates molecular marker.

Figure 4. IgG depletion from mouse serum reduces the 25-kDa band, but the 12.5-kDa band remains unchanged in old and young mice

A, The ≈25-kDa band in serum from an old mouse is significantly removed after IgGs are depleted first with protein-G sepharose resin (IgG depl.) and further with a monoclonal antibody against mouse IgG (IgG depl.+anti-mIgG). Elution of IgG bound to protein G sepharose is shown in the far right lane. B, Western analysis showing that IgG depletion in sera from 3 pooled young mice (2-month old) significantly decreases the ≈25-kDa band, whereas the signal at ≈12.5 kDa remains unchanged. The corresponding Coomassie gels are shown for both A and B (bottom). C, Serum pooled from 3 old mice (2 µL) was run on a reducing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gel and Coomassie stained. The protein band at ≈25-kDa band was excised and submitted for mass spectrometry analysis. D, The top 20 protein hits identified by mass spectrometry reveal predominantly IgG light chain in the ≈25-kDa band in old mouse. All Western analyses were performed using the anti–growth differentiation factor 11/8 antibody (Abcam) under reducing conditions. *Empty well. LC-MS/MS indicates liquid chromatography-mass spectrometry/mass spectrometry; and M, molecular marker.

Figure 5. Growth differentiation factor 11 (GDF11) activates cardiac SMAD signaling in vivo

A, Western analysis was performed on protein samples harvested 1 hour post administration of saline or GDF11 protein (1 mg/kg) via tail vein injection in young (left; 5, 6, 7, and 8) and old (right; 13, 14, 15, and 16) mice. Phospho- and total-SMAD2 levels together with GAPDH (loading control) were analyzed in whole-heart protein lysates. B, Comparison of SMAD2 phosphorylation on young (17, 18, 19, and 20) and old mice (left; 21, 22, 23, and 24; n=2 per group) was performed on cardiac samples harvested 1 hour after injection of saline (young, 17 and 18; old, 21 and 22) or 1 mg/kg of GDF11 protein (young, 19 and 20; old, 23 and 24). C, SMAD3 phosphorylation on GDF11 or saline stimulation was determined by immunofluorescence (white arrows) staining on transverse cardiac section from mice treated with saline or GDF11 protein (1 mg/kg) for 1 hour via tail vein injection. Scale bar, 50 µm. Inset scale bar, 10 µm.

Figure 6. Growth differentiation factor 11 (GDF11) reduces heart weight and cell size in young and old mice in 9 days

A, Bar graphs showing dose-dependent effects of GDF11 in young (2-month old; left) and old (22-month old; right) mice using different doses of GDF11 protein during a period of 9 days. B, Representative fluorescent images of cardiomyocyte cross-sectional area from young and aged mice administered daily with GDF11 protein (1 mg/kg per day) or saline for 9 days. GDF11 treatment significantly reduced cardiomyocyte cross-sectional area (right bar graph) in both young and old mice (n=7 per group). All data presented as mean±SEM. Scale bar, 20 µm. DAPI indicates 4',6-diamidino-2-phenylindole; n.s., not significant; and WGA, wheat germ agglutinin.