Extracellular vesicles from human airway basal cells respond to cigarette smoke extract and affect vascular endothelial cells

Abstract

The human airway epithelium lining the bronchial tree contains basal cells that proliferate, differentiate, and communicate with other components of their microenvironment. One method that cells use for intercellular communication involves the secretion of exosomes and other extracellular vesicles (EVs). We isolated exosome-enriched EVs that were produced from an immortalized human airway basal cell line (BCi-NS1.1) and found that their secretion is increased by exposure to cigarette smoke extract, suggesting that this stress stimulates release of EVs which could affect signaling to other cells. We have previously shown that primary human airway basal cells secrete vascular endothelial growth factor A (VEGFA) which can activate MAPK signaling cascades in endothelial cells via VEGF receptor-2 (VEGFR2). Here, we show that exposure of endothelial cells to exosome-enriched airway basal cell EVs promotes the survival of these cells and that this effect also involves VEGFR2 activation and is, at least in part, mediated by VEGFA present in the EVs. These observations demonstrate that EVs are involved in the intercellular signaling between airway basal cells and the endothelium which we previously reported. The downstream signaling pathways involved may be distinct and specific to the EVs, however, as increased phosphorylation of Akt, STAT3, p44/42 MAPK, and p38 MAPK was not seen following exposure of endothelial cells to airway basal cell EVs.

Figure 1

Characterization of EVs produced by immortalized human airway basal cells in culture. (A), Representative size distribution histogram of EVs isolated from conditioned media of BCi-NS1.1 cells grown in culture. The NanoSight NS500 machine with Nanoparticle Tracking Analysis software was used to calculate the quantity of particles as well as their size distribution (mean, mode, etc.). A representative image of the particles is shown in the upper right. (B), Immunoblots (cropped for clarity and conciseness) of BCi-NS1.1 cell lysates and EVs produced from these cells with antibodies against proteins found in exosomes—CD63, HSP90, Hsc70, and TSG101—as well as Calnexin, which is not typically seen in exosomes. Equal amounts of total protein were loaded into the cell-lysate and EV lanes. As a negative control, BEGM without BPE was incubated for 48 h in flasks without cells and then processed by means of the same procedures used to isolate EVs. A volume equal to that of the EV preparation was loaded onto the lane marked “(—)”. (C), Representative image of BCi-NS1.1 EVs (marked by arrows) taken using electron microscopy.

Figure 2

CSE stimulates release of EVs from immortalized human airway basal cells in culture. (A), Floating bar graph depicting 3 replicate experiments quantifying EVs isolated from BCi-NS1.1 cells incubated with 6% CSE, compared with those isolated from cells with 0% CSE. The horizontal line inside the floating bar represents the mean of the 3 experiments. Roughly twice as many EVs per cell were seen after CSE exposure; the P value (two-tailed) was calculated using the ratio paired t test, with P ≤ 0.05 considered statistically significant. (B), Floating bar graph depicting the number of living BCi-NS1.1 cells just prior to EV isolation in the 3 experiments from A. The horizontal lines inside the floating bars represent the means of the 3 experiments. The P value (two-tailed) was calculated using the paired t test, with P ≤ 0.05 considered statistically significant. (C), Immunoblots (cropped for clarity and conciseness) of the exosomal protein Alix identified in BCi-NS1.1 cell EVs and in cell lysates after exposure to 0% or 6% CSE. Immunoblot of β-actin in cell lysates was used as a control. This experiment was repeated 3 times, with representative images shown. Equal amounts of total protein were loaded into each cell-lysate lane, and equal volumes of EV preparations were loaded into each EV lane. In equal volumes of EV preparations, more Alix protein was seen in those from CSE-exposed cells.

Figure 3

Human airway basal cell EVs promote the survival of HUVECs in culture. (A), Images of HUVECs in single wells of a 384-well microtiter plate following 48-h incubation with different concentrations of EVs isolated from the H1975 non-small cell lung cancer and N417 small cell lung cancer cell lines. (B), Images of HUVECs in single wells of a 384-well microtiter plate following 48-h incubation with different concentrations of BCi-NS1.1 cell EVs. The live cells were counted under fluorescence microscopy, and the results of 3 replicate experiments using 10 µg/mL EVs is summarized in the floating bar graph to the right. The horizontal lines inside the floating bars represent the means of the 3 experiments. Neg Ctrl, negative control (an equal volume of a preparation from BEGM without BPE incubated for 48 h in flasks without cells and then processed by means of the same procedures used to isolate EVs). Adjusted P values correspond to differences between the bars on either end of the horizontal line below them and were calculated using Tukey’s multiple comparisons test, with P ≤ 0.05 considered statistically significant. (C), The same experiment as in Fig. 4B was performed, except with EVs isolated from BCi-NS1.1 cells grown in BEBM, which is not supplemented with growth factors. (D), Floating bar graph of the numbers of living HUVECs after 48-h incubation with 10 µg/mL EVs from BCi-NS1.1 cells exposed to either 0% or 6% CSE. The horizontal lines inside the floating bars represent the means of 4 experiments, each performed with different batches of EV preparations. The P value (two-tailed) was calculated using the paired t test, with P ≤ 0.05 considered statistically significant.

Figure 4

VEGFR2, but not VEGFR1 or FGFR1, is involved in human airway basal cell EV–mediated survival of HUVECs. (A), Results of ELISAs, showing the presence of VEGFA in EVs isolated from BCi-NS1.1 cells. Negative control, an equal volume of a preparation from BEGM without BPE incubated for 48 h in flasks without cells and then processed by means of the same procedures used to isolate EVs. The floating bar graphs depict results from 3 batches of isolated EVs, with each batch run in duplicate. The horizontal lines inside the floating bars represent the mean concentrations from the 3 batches. The ELISAs were also performed after EVs were incubated with RIPA buffer to lyse them open. VEGFA was detected both with and without lysis. B–D, Floating bar graphs of the numbers of living HUVECs counted after 48-h incubation with either no EVs or 10 µg/mL EVs from BCi-NS1.1 cells in the presence of increasing concentrations of (B) VEGFR2-blocking antibody IMC-C11, (C) VEGFR1-blocking antibody IMC-18F, and (D) FGFR1 small-molecule inhibitor SSR128129E. The horizontal lines inside the floating bars represent the means of 3 experiments; adjusted P values correspond to differences between the bars on either end of the horizontal line below them and were calculated using Dunnett’s multiple comparisons test, with P ≤ 0.05 considered statistically significant.

Figure 5

Downstream effectors of human airway basal cell EV–mediated survival of HUVECs. (A) Immunoblot of phosphorylated and total VEGFR2, STAT3, Akt, p44/42 MAPK, and p38 MAPK proteins in HUVECs after incubation with BCi-NS1.1 cell EVs or controls for 1 h or 4 h. The experiment was performed in duplicate. BCi, incubation with EVs from BCi-NS1.1 cells; Ctr, incubation with a preparation from BEGM without BPE incubated for 48 h in flasks without cells and then processed by means of the same procedures used to isolate EVs; 1, 1µg of EVs or an equal volume of Ctr; 10, 10µg of EVs or an equal volume of Ctr; Mock, incubation with BEBM media alone. (B) Floating bar graph of the numbers of living HUVECs counted after 48-h incubation with either no EVs or 10 µg/mL EVs from BCi-NS1.1 cells in the presence of increasing concentrations of the Akt small-molecule inhibitor MK-2206. The horizontal lines inside the floating bars represent the means of 3 experiments; adjusted P values correspond to differences between the bars on either end of the horizontal line below them and were calculated using Dunnett’s multiple comparisons test, with P ≤ 0.05 considered statistically significant.