Cellular Characteristics and Protein Signatures of Human Adipose Tissues from Donors With or Without Advanced Coronary Artery Disease

Figures

• 

  Figure A1. (A) A Gaussian wavelet (Φ) convolved with an adipocyte image taken from a PVAT sample at three representative wavelet size scales. (B) Maxima chains representing local maximal gradients, i.e., the wavelet-transform modulus maxima (WTMM), at three representative wavelet scales shown in green. (C) Magnified inserts from the white outlined area in panel B at three representative wavelet size scales to identify WTMM vectors (red) showing the magnitude and direction of local gradients of the maximum image density fluctuations. (D) Probability density functions (PDF), shown in red, of three representative wavelet scales generated from angles of local gradient vectors with the theoretical isotropic 1/2π line shown in black. The anisotropy factor (see equation in red, where P(X) is the PDF at one scale) is obtained by integrating the difference between each PDF and the theoretical 1/2π line.

• 

  Figure 1. A comparison of PVAT and SubQ between VR and CABG donors. (A) While there was no significant difference in age of donors from VR vs. CABG groups, the CABG group had higher body mass indices (BMIs) and circulating HbA1c levels. (B) Representative images of adipose tissue sections after hematoxylin/eosin staining for PVAT or subcutaneous adipose (SubQ) from donor groups. Scale bar = 50 μm. (C) Morphometric analysis was performed to quantify adipocyte number, size, and stromal area. Adipocytes were smaller in PVAT compared to SubQ regardless of disease incidence. VR PVAT n = 7, VR SubQ n = 7, CABG PVAT n = 9, CABG SubQ n = 11. (D) Sections were immunostained to detect MAC1 to quantify macrophages in adipose tissue in CABG PVAT (n = 8) and SubQ (n = 8). Representative images are shown, and quantification indicated higher macrophage presence in CABG PVAT compared to CABG SubQ. Scale bar = 750 µm. (E) Sections were stained with Masson’s trichrome stain to identify collagen. There was no detectable difference between CABG and VR PVAT or VR and CABG SubQ. Scale bar = 500 µm. VR PVAT n = 3, CABG PVAT n = 6, VR SubQ n = 4, CABG SubQ n = 4. (F) Sections from human adipose tissues from CABG…

• 

  Figure 2. A morphometric and anisotropy analysis of tissues with sliding window segmentation validate FIJI analysis. (A) A morphometric analysis of patient adipocytes from the sliding window segmentation analysis where patient median values from all their adipocytes are plotted. Quantified are area, solidity, circularity, and eccentricity. Significance, shown as asterisks, were calculated between patient tissue subtypes and across surgery subtypes using an unpaired Wilcoxon ranked sum test with an α of 0.05. (B) Anisotropy Factor plots of patient mean adipose tissue anisotropy factors and standard deviation across 50 wavelet size scales of VR PVAT vs. SubQ, CABG PVAT vs. SubQ, VR PVAT vs. CABG PVAT, and VR SubQ vs. CABG SubQ. All automated sliding window H&E 1024 × 1024 subregions that passed manual inspection of adipocyte content were used to calculate each of the patients’ mean anisotropy factors. Significance, shown as asterisks, were calculated at each wavelet scale using an unpaired Wilcoxon ranked sum test with an α of 0.05. The median adipocyte radii in microns of each category are shown as vertical bars with colors corresponding to the legend. (C) Each donor’s mean PVAT Anisotropy…

• 

  Figure 3. The diabetes status associated with few changes in CABG PVAT and SubQ. (A) Hematoxylin/eosin staining, and (B) morphometric comparisons of adipocyte size, number, and stromal area in CABG PVAT accounting for diabetes. Non-diabetic PVAT n = 7; diabetic PVAT n = 3; non-diabetic SubQ n = 8; diabetic SubQ n = 3. Scale bar 50 µM. (C) The MAC1 immunostaining in CABG and SubQ from donors with and without diabetes of stained area is shown in the graph on the right. Non-diabetic PVAT n = 4; diabetic PVAT n = 4; non-diabetic SubQ n = 4; diabetic SubQ n = 4. Scale bar 250 µm and 50 µm, respectively. Scale bar = 50 µm. (D) The quantification of tissue fibrosis using Masson’s trichrome Masson in non-diabetic and diabetic PVAT and SubQ. Non-diabetic PVAT n = 3; diabetic PVAT n = 4.

• 

  Figure 4. SWATH proteomics reveals differential protein expression in proteins involved in adipogenesis, secretion, and insulin resistance. VR 201 donor had shunt placed ~16 years prior to the VR surgery. (A) PCA plot showing differences in 4 SWATH groups: CABG PVAT (n = 4, green), VR PVAT (n = 6, purple), CABG SubQ (n = 3, black), and VR SubQ (n = 5, yellow). (B) Volcano plots highlight the proteins of interest that are expressed differently between CABG and VR PVAT. (C) Volcano plots highlight the proteins of interest that were significantly differently expressed (pink dots) between CABG and VR SubQ. (D) STRING analysis highlighted gene ontology cellular component terms in all differentially expressed proteins between the PVAT. Terms are classified by false discovery rate (FDR), strength (internal STRING measurement), and observed gene count. (E) STRING analysis highlighted GO cellular component terms in all differentially expressed proteins between the SubQ groups. Terms were classified by false discovery rate (FDR), strength, and observed gene count. (F) Proteins that are classified in gene ontology terms “extracellular space”, “extracellular region”, “extracellular exosome”, and…

• 

  Figure 5. SWATH proteomics of non-diabetic and diabetic CABG PVAT. (A) From the first SWATH data set, the volcano plot shows significant protein expression differences (pink dots) between CABG PVAT and SubQ. In this data set, there was a mix of diabetic and non-diabetic samples. (B) STRING analysis of proteins differentially expressed between CABG PVAT and SubQ displaying enriched cellular component gene ontology terms. (C) Proteins that are classified in gene ontology terms “extracellular space”, “extracellular region”, “extracellular exosome”, and “vesicle” in PVAT/SubQ comparison were classified by p-value and log fold change for CABG PVAT/CABG SubQ. (D) PCA showed a clear distinction between non-diabetic and diabetic CABG PVAT. (E) Volcano plot identified ten proteins higher in non-diabetic than diabetic CABG PVAT. (F) Proteins significant in MRM proteomics, displayed by log fold change. Each bar represents an individual ion.

• 

  Figure 6. The detection of RAB27A and NOTCH3 in human adipose tissue. Antigens were selected for validation for the proteomics. (A) Shown are representative images of immunofluorescence staining in two CABG donor tissues, with matching PVAT or subcutaneous (SubQ) adipose tissue from each donor. The negative control (NEG) was a secondary antibody treatment only. Scale bar = 20 μm. (B) The same samples were analyzed for NOTCH3 protein localization. Non-immune IgG staining is shown in the bottom row. Scale bar = 20 μm.

• 

  Figure 7. The differentiation of human preadipocytes and impact on mitochondria. Various preadipocyte populations from CABG donors were either left under standard growth conditions or differentiated as described for nine days. (A) LipidTOX was used to detect lipids based on fluorescence intensity. (B) Immunoblot of PLIN1 (62 kD), PPARG (57, 53 kD), and RAB27A (27 kD) following differentiation (Diff) compared to undifferentiated (UN) controls. Quantification of the immunoblots are shown in the graphs on the right. (C) Mitotracker was used to stain mitochondria as indicated (red) in undifferentiated or differentiated populations. Scale bar = 25 μm. (D) Quantification of the mitotracker stain from groups shown in (C). p-value * ≤ 0.05, ** p-value ≤ 0.01. (E) Human PVAT-derived preadipocytes were grown in monolayer culture and differentiated into adipocytes for eleven days. Nuclei were stained with DAPI (blue), and Oil Red O (red) was utilized to detect lipids. (F) Human preadipocytes were obtained from PVAT from donors undergoing CABG and grown in 3D spheroid culture as described. Cells were either left undifferentiated in a standard growth medium or differentiated for ten or eighteen days in…