Characterization of multiple myeloma vesicles by label-free relative quantitation

Abstract

Multiple myeloma (MM) is a hematological malignancy caused by a microenviromentally aided persistence of plasma cells in the bone marrow. The role that extracellular vesicles (EVs), microvesicles and exosomes, released by MM cells have in cell-to-cell communication and signaling in the bone marrow is currently unknown. This paper describes the proteomic content of EVs derived from MM.1S and U266 MM cell lines. First, we compared the protein identifications between the vesicles and cellular lysates of each cell line finding a large overlap in protein identifications. Next, we applied label-free spectral count quantitation to determine proteins with differential abundance between the groups. Finally, we used bioinformatics to categorize proteins with significantly different abundances into functional groups. The results illustrate the first use of label-free spectral counting applied to determine relative protein abundances in EVs.

Keywords: Cell biology; Exosomes; LC-MS/MS; Label free; Microvesicles.

Figure 1

Cryo-transmission electron microscopy (cryo-TEM) images of the A) MM.1S cell-derived extracellular vesicles and B) U266 cell-derived extracellular vesicles, indicated by the black. The cryo-TEM carbon support grids (white arrows) are also seen these images. C) Number distributions of MM.1S and U266 extracellular vesicle diameters derived from Dynamic Light Scattering (DLS) measurements.

Figure 2

Venn diagram renderings of overlapping and unique protein identifications. A) MM.1S vesicles (EV) v. global cell lysate (CL). B) U266 vesicles (EV) v. global cell lysate (CL). C) MM.1S vesicles (EV) v. U266 vesicles (EV). Data shows many overlapping protein identifications while also harboring unique IDs in each comparison.

Figure 3

Clustering of LC-MS/MS spectral counts and pie chart illustrations of the PANTHER gene ontology annotations for molecular function and biological process for those proteins with significantly different abundances between the cell line vesicles when compared to the parent cell lysate. A) Clustering of spectral count data for the MM.1S vesicles and parent cell lysate. B) Enlarged clustering of spectral count data for those proteins selected for validation from MM.1S cell line. C) PANTHER gene ontological annotations for the molecular function and biological process for proteins with statistically different abundances from the MM.1S cell line. D) Clustering of spectral count data for the U266 vesicles and parent cell lysate. E) Enlarged clustering of spectral count data for those proteins selected for validation from the U266 cell line. F) PANTHER gene ontological annotations for the molecular function and biological process for proteins with statistically different abundances from the U266 cell line. Results suggest variable abundance of specific proteins, which can confer different potential biological processes and molecular functions.

Figure 4

EdgeR label-free analysis for differentially expressed proteins from the MM.1S derived vesicles and the U266 derived vesicles. A) Hierarchal clustering of LC-MS/MS spectral count data. B) Enlarged clustering of spectral count data for those proteins selected for validation. C) Smear plot of the log fold change (log FC) by log counts-per-million (log CPM) for the vesicle data from each cell line. D) Pie chart illustrations of the PANTHER gene ontology annotations for molecular function and biological process for those proteins differentially expressed between the MM derived vesicles. Data suggests vesicular protein abundance distinguishes between the cells of origin allowing for differential functional potential.

Figure 5

Smear plot of log fold change by log fold change for the vesicles and global cell lysates. A) Proteins with no significant change in abundance between vesicles and global lysates changes in the same direction. B) Proteins with significant abundance differences in opposite directions between the vesicle and global lysate samples. C) Overlay of A & B. Those proteins with the greatest independent differences in abundance are labeled with Uniprot Accession numbers.

Figure 6

Validation of protein identifications and spectral count relative quantitation from the proteomic analysis of the MM.1S and U266 derived vesicles and global cell lysates by immunoblot. A) Immunoblot of vesicle identified proteins. Blot was probed for CD9, IgGκ LC, Nucleolin and GAPDH. B) Immunoblot for comparison of protein relative abundance between vesicles and cell lysates. Blots were probed for CD9, CD44, MHC Class I, BST-2, GAPDH and Actin. Data confirms the LC-MS/MS protein identifications and relative abundances observed in the data sets. HeLa (CD9) and ARH77 (IgGκ LC) global lysates were used as positive controls.