|

CAD-LB: Biomarker Detection Validation – ICAM-1 HUVEC Stimulation Model

BySam Walker

Introduction

This post is another update in my series of experiments aiming to both characterize and validate the CAD-LB platform. We have now shifted our focus to validating CAD-LB biomarker detection which will be done using multiple experimental models. The most recent model, and the topic of this post, uses HUVEC cells to validate the detection of ICAM-1 on HUVEC-derived EVs. We know that following stimulation HUVECs exhibit a marked upregulation of ICAM-1; in these experiments we investigated if CAD-LB could detect increased percentages of ICAM-1+ EVs in the conditioned cell culture media of stimulated HUVECs. Following these experiments, we found that within the EV compartment CAD-LB indeed recapitulated the ICAM-1 expression relationship seen in parent cell counterparts.

Experimental Approach

HUVEC cells were cultured in both T25 flasks and 12-well plates and split into two groups. One group was stimulated with 1 ng/mL TNF-α (stimulated) while the other group was left untreated (unstimulated). After 24 hours, HUVECs in 12-well plates were stained for ICAM-1 expression using ICC methods while the conditioned cell culture media was collected from T25 flasks and used for EV analysis. CAD-LB was used to identify ICAM-1+ EV subsets in the conditioned media of stimulated and unstimulated HUVECs using our typical fluorescent dyes (CFSE & conjugated antibody targeting biomarker). Secondary antibody-only samples were used in ICC experiments for background correction. Label-only (no EV) and isotype controls were employed in CAD-LB experiments to quantify background labeling/capture and nonspecific labeling levels, respectively.

Results & Discussion

Representative images of stimulated and unstimulated HUVECs are shown in Figure 1A-B. As expected, stimulated HUVECs exhibited a large upregulation of ICAM-1; ICC reported an approximate 37-fold increase in mean fluorescence intensity (MFI) (Figure 1C).

Figure 1. HUVEC ICAM-1 Expression: Images & Quantification. (A-B) Representative epifluorescence images (333 µm x 333 µm) taken following HUVEC treatment. (A) and (B) correspond with HUVECs stimulated with 1 ng/mL TNF-α for 24 hours and unstimulated HUVEC controls, respectively. Following treatment, HUVECs were stained for ICAM-1 expression using ICC methods. Scale bars = 20 µm. (C) ICAM-1 mean fluorescence intensity (MFI) summary for stimulated and unstimulated HUVECs following background correction using secondary antibody-only control samples. All bars represent mean ± SEM (n = 3). ** p < 0.01, by Student’s t-test.

The results for EV experiments are presented in Figure 2. CAD-LB successfully detected ICAM-1+ EVs in the conditioned media of both stimulated (Figure 2A) and unstimulated (Figure 2B) HUVECs. Indeed, CAD-LB reported the same ICAM-1 expression relationship as ICC methods identifying approximately 28% and 4% ICAM-1+ EVs in stimulated and unstimulated HUVEC conditioned media samples, respectively (Figure 2C). Compared to cellular changes, the EV compartment was altered to a lesser degree with stimulated HUVEC-derived EVs presenting 7.5-fold higher ICAM-1 expression.

Figure 2. HUVEC-derived EV ICAM-1 Expression: Images & Quantification. (A-B) Representative regions of interest (ROI = 71 µm x 71 µm) in confocal microscope images taken following EV capture. HUVEC conditioned media was labeled with CFSE then antibodies (ICAM-1/mouse [Ms] IgG isotype control), prefiltered (0.5 µm), and diluted to prepare an input of 10^4 EVs. (A) and (B) correspond with the merged signals (CFSE & ICAM-1 antibody) for a representative stimulated and unstimulated HUVEC conditioned media sample, respectively. The green, red, and yellow arrows in (A) and (B) indicate a lone CFSE signal, a lone ICAM-1 signal, and an instance of colocalization, respectively. Scale bars = 5 µm. (C) ICAM-1 detection summary for stimulated and unstimulated HUVEC-derived EVs following isotype control correction. All bars represent mean ± SEM (n = 3). ** p < 0.01, by Student’s t-test.
It should also be noted that CAD-LB detected significantly more EVs in stimulated HUVEC conditioned media samples (Figure 3A). Additionally, label-only (no EV) control samples generated CFSE counts orders of magnitude lower than either experimental group (Figure 3A-B). Importantly, isotype controls returned colocalization levels orders of magnitude lower for both stimulated and unstimulated samples (Figure 3C: representative stimulated isotype control) indicating that ICAM-1 labeling was specific.  

Figure 3. HUVEC-derived EV Counts & Control Samples. (A) CFSE count summary for stimulated and unstimulated HUVEC conditioned media samples (n = 6) as well as label-only (no EV) controls. (B-C) Representative regions of interest (ROI = 71 µm x 71 µm) in confocal microscope images taken for control samples. (B) Merged signals (CFSE & ICAM-1 antibody) for a representative label-only (no EV) control sample. (C) Merged signals (CFSE & Ms IgG isotype control) for a representative stimulated HUVEC conditioned media sample. Scale bars = 5 µm. All bars represent mean ± SEM (n = 3), unless otherwise indicated. * p < 0.05, by Student’s t-test.

These experiments validate CAD-LB biomarker detection and should instill confidence in the platform. Furthermore, stimulation model results provide evidence that CAD-LB can detect EV biomarkers that capture protein expression shifts and possibly functional changes in parent cells. The detection of ICAM-1+ EVs following HUVEC stimulation is a phenomenon that has been previously reported in literature and carefully verified [1]. The finding that stimulation increases EV concentration in conditioned media is also in accordance with published results; both CAD-LB and NTA [1] report similar fold increases in EV concentration following TNF-α treatment. These experiments also highlight an advantage of CAD-LB, namely, the ability to process nonpurified EV samples. With minimal sample preparation and low EV input requirements (104) CAD-LB can provide a rapid and accurate assessment of EV biomarkers.

References

[1] Hosseinkhani, B., et al., Direct detection of nano-scale extracellular vesicles derived from inflammation-triggered endothelial cells using surface plasmon resonance. Nanomedicine, 2017. 13(5): p. 1663-1671.

Similar Posts

Leave a Reply