Figure  1.  GDF11 changes the ratio of different subpopulations of EPCs-MA based on morphological features

(A) Light microscopic photographs showing mononuclear cells (MNCs) at day 10 in the culture with clusters of cells surrounded by spindle-like cells. Four morphologic subtypes were identified: cobblestone-like cells, spindle-like cells, drop like cells, and flatten round cells, which are marked with arrows, dark stars, arrowheads, and hollowed red stars, respectively. (B) Bar chart showing the percentage of spindle-like cells relative to other subtypes of EPCs. Statistical data on the ratio of spindlelike cell subpopulation are expressed as the mean values of percent spindle-like cells over total number of EPCs per EPCs clones (n = 2 batches in each group). (C) Representative fluorescence images with red staining depicting cells engaging in active Dil-Ac-LDL uptake from the culture medium and green staining for EPCs with lectin UEA-1 (lectin1) binding. The spindle-like cell is marked with black stars; Scale bar: (A) 100 µm and (C) 50 µm; ^**P < 0.01, ^★★ P < 0.01.

Figure  2.  Effects of GDF11 on EPCs-MA maturation

Immunofluorescence staining showing the expression and distribution of CDH5 (green) and CD31 (green) primarily in the plasma membrane. (A) CDH5 expressed in EPCs at varying levels and can be classified into low (red arrow), medium (yellow arrow), and high (white arrow) level based on the fluorescence intensity. The majority of EPCs expressed CD31 (yellow arrow), CD31 was negative in few cells (red arrows). (B) Bar chart showing percent cells expressing CDH5 at low, medium, and high levels. (C) Bar chart showing the percentage of CD31 positive and negative cells. Scale bar: 100 µm; ^**P < 0.01, ^★ P < 0.05, ^★★ P < 0.01.

Figure  3.  Effects of GDF11 on EPCs-MA migration in vitro and in vivo

(A) The experiment procedures for migration assays of EPCs: Prior to EPCs collection, the cells at day 8 in culture were treated with GDF11 for 48 h. The EPCs at day 10 were used for migration assays in vitro and in vivo. For migration assays in vivo, 8-10-weeks-old mice were subject to MI surgery and injection of EPCs via tail vein onehour post-MI. After 14 days, Doppler ultrasonography and histological assays were performed to evaluate cardiac function and measure the infarct size. Statistical data were obtained from 3 independent experiments (n = 3-4 per group). (B) The effect of GDF11 on the migration of EPCs-MA: Migrated cells were visualized after crystal violet staining in blue color. The bar chart present the percentages of migrated cell areas (n = 3 batches per group). Scale bar: 500 µm. (C) Echocardiographic assays depicting the changes of ejection fractions (EF) and fractional shortening (FS) of left ventricle (LV). Cardiac functional parameters of the MI mice and EPCs-Ytransplanted mice were used as negative and positive controls, respectively. n = 5-7/group. (D and E) The hearts were diseected into 3-4 parts from apex to base (D), and peri-infarct tissues (squared) were sectioned and stained with Masson trichrome for assessing infarct size. Fibrotic tissues were stained blue, and the viable tissues were stained red. Collagen (CoL) deposition was quantified with an automated image analyzer, and the data are expressed as percentage of tissue areas. The bar graphs show infarcts size in different groups. Scale bar sizes in Masson section: 2000 µm (E); ^*P < 0.05, ^**P < 0.01, ^★ P < 0.05.

Figure  4.  GDF11 restores the lost functional phenotypes of EPCs-MA

(A and C) Typical examples of immunofluorescence staining images depicting the expression and subcellular distribution of eNOS (red) and vWF (red). eNOS was primarily expressed in the nuclear. In most of EPCs-MA, vWF exhibited a diffuse cytoplasmic signal. vWF was expressed in EPCs at different levels, allowing for classifications into low, medium, and high levels based on the fluorescence intensities. The majority of EPCs expressed vWF at a medium level (yellow arrow), and only small numbers of cells expressed vWF at high (white arrow) and low levels (green arrow), respectively. (B and D) Bar chart showing the percentage of positive area of eNOS expression and cells expressing vWF at low, medium, and high levels, respectively. (E) Photographs captured with TEM showing EPCs with elongated WPBs (arrows) and trans-Golgi networks (TGN) containing enlarged parts where WPBs are formed (arrows heads). (F) Bar chart presenting the mean numbers of WPCs per EPCs, a pile of TNG was recorded as 1 WPB; Scale bar: (A and C) 50 µm; E, 0.5 µm; ^*P < 0.05, ^**P < 0.01, ^★ P < 0.05, ^★★ P < 0.01.

Figure  5.  GDF11 enhances autophagic clearance capability of EPCs-MA

(A) The effects of GDF11 on the conversion of endogenous LC3-Ⅰ to LC3-Ⅱ and p62 accumulation, as evidenced by by Western blot analysis. (B) Bar chart summarizing the ratio of LC3-Ⅰ/LC3-Ⅱ and relative protein levels of p62. GAPDH served as an internal control. (C) Autophagic vacuoles were counted under an electron microscope at 10, 000× magnification. All autophagic vacuoles present in each photograph of EPC were counted. Scale bar: 1 µm and 200 nm. (D) Bar chart showing the number of autophagic vacuoles in each EPCs (n = 3). (E) Representative images of LysoTracker Red staining depicting the effects of GDF11 on the lysosomal acidification in EPCs. (F) Bar chart summarizing the percentage of positive areas (accumulated acidic organelles) in EPCs. ^*P < 0.05, ^**P < 0.01.