How to detect your EV cargo of interest using custom antibodies

EVs are lipid bilayer-enclosed particles released by all cell types and critical for cell-to-cell communication. Their unique ability to carry cargo biomolecules such as RNA/DNA, cytosolic or membrane-bound proteins, and other signaling factors, have put EVs in the spotlight as disease biomarker-packed particles and drug delivery vehicles^1-4. Current methods to study EV cargo and understand EV biogenesis and donor cell profile tend to rely on bulk measurements. Due to the high heterogeneity of EV samples, bulk assessments, such as western blot, dot blots and flow cytometry, can lead to poor characterization of EV subpopulations and cargo distribution.

Single Molecule Localization Microscopy (SMLM) is a super-resolution imaging technique that offers EV researchers the unique ability to resolve the molecular details of vesicles below 200 nm in size—the diffraction limit of traditional optical microscopes. This enables the precise profiling of EV populations down to the single-EV level, including quantification of EVs and their biomarkers, with 20 nm resolution. New end-to-end workflows, such as ONI’s EV Profiler 2 with AutoEV, simplify the steps needed to capture, label, detect, and quantify EV populations and biomarkers of interest, including internal EV cargo. While the workflow is compatible with both ONI-provided and user-provided reagents, some users struggle to validate and implement custom antibodies to confidently detect their targets of interest.

Using a Pan-EV stain to identify and size EVs

Probing for EV cargo is often challenging, and these challenges become more apparent when using an imaging-based approach such as SMLM. SMLM localizations must be clustered together before any downstream analysis occurs. Thus, a generic membrane stain is needed to ensure robust detection, morphology assessment, and sizing of EVs. Many commonly used membrane markers, such as wheat germ agglutinin (WGA) or concanavalin A (ConA), are lectins, making the resultant membrane visualization glycosylation-dependent. ONI’s EV Profiler 2 kit offers a membrane visualization method, termed Pan-EV stain, that functions independently of protein expression or membrane glycosylation, allowing for accurate sizing of many types of EVs and enabling quantification of the percentage of particles positive for a cargo marker without bias (Figure 1). The Pan-EV stain, in conjunction with the tetraspanin trio (CD9, CD63, and CD81) offered in the EV Profiler 2 kit allows users to confidently identify EVs in accordance with the latest MISEV guidelines⁵.

Figure 1. Schematic representation of the EVs captured and stained using the ONI’s EV Profiler 2 kit.

We suggest that a variety of control experiments are performed as part of SMLM experiments when validating a cargo marker for the first time, to ensure robust and reproducible data is obtained.

Controls to assess the effects of permeabilization on EVs

The small size of EVs, between 30 - 1000 nm in size, is also a challenge when detecting cargo, particularly on a single-EV basis. Due to the limited amount of membrane present, many traditional detergents create large pores relative to the size of the EV, which results in degradation of the vesicle structure. This occurrence, referred to as “over-permeabilization”, can lead to inaccurate internal cargo quantification, resulting in excessive background in SMLM due to membrane fragments scattered across the surface. We suggest the following controls to ensure that permeabilization does not result in the degradation of EVs.

1. No permeabilization
   A no-permeabilization control allows users to assess the appearance of non-permeabilized EVs and determine if the EVs are being degraded by permeabilization. If the EV size distribution or number of EVs captured changes substantially after permeabilization, over-permeabilization may be occurring, and some EVs are lost as a result. A no-permeabilization control also allows users to assess non-specific binding of antibodies targeting internal cargo molecules, as such antibodies should only exhibit staining in the permeabilized samples.
   
   Antibody signal in the non-permeabilized samples suggests that non-specific antibody binding is occurring. The abundance of the cargo molecule in the permeabilized sample should be compared to the level detected in the non-permeabilized samples. We recommend that the difference between cargo detected in permeabilized and non-permeabilized samples be statistically significant.
   
   
2. Lyzed EV
   In cases where the quantitative and qualitative results between the no permeabilization and permeabilization conditions are inconclusive, one may also perform a lyzed EV control. This is done using a high concentration of a common detergent (e.g. 1% saponin or Triton X-100) on the EV sample instead of the permeabilization step. This should result in the lyzing of the captured EVs, which will lead to a very sparse signal that should be difficult to cluster. If your lyzed EV results look similar to regular permeabilization results, you may be over-permeabilizing your EVs.

Controls to verify biomarker detection (even low-abundance ones)

EV cargo molecules are often low-abundance molecules and must be detected with high sensitivity. To ensure that your antibody is detecting your cargo molecule of interest, we recommend users perform the following controls when validating a cargo marker for the first time.

Isotype control OR detection of antigen not found in EVs

Isotype controls serve to check non-specific antibody binding. Ideally, a sample stained with a species- and isotype-matched irrelevant antibody control should result in little staining; and the amount of staining should be similar between permeabilized and non-permeabilized samples. If the amount of EVs positive for the isotype control is similar to that of your cargo molecule of interest, it may indicate that your cargo is not present at detectable levels or that your antibody lacks specificity. If desired, an antibody targeting a protein not found in EVs (such as Nup98, a nuclear pore complex component) may be used instead of a specifically-raised isotype control.

Cell staining

Users may perform staining in a relevant cell culture model before proceeding with EV staining and imaging, to ensure that an antibody targeting a cargo molecule of interest is specific. A knockout or knockdown line for the protein of interest may be used to confirm antibody specificity. Note that antibodies might work well for other protein-detection techniques, such as western blot, might not work equally in fluorescent labeling. This is uncommon but needs to be considered.

Verifying EVs are detected by SMLM after permeabilization

To confirm the effects of the EV Profiler 2 workflow including permeabilization on EVs, PANC1-derived EVs were captured using the kit’s phosphatidylserine (PS) Capture, and labeled using the Pan-EV + Tetraspanin Trio + custom marker detection modality. ALG-2 interacting protein X (ALIX), a well-characterized cargo protein contained in many EVs types and implicated in recruiting tetraspanins⁶, was detected using a recombinant rabbit monoclonal antibody from EV Profiler 2. Staining Solution from ONI’s dSTORM Training Kit was used as the rabbit anti-Nup98 control, and a rabbit isotype control was purchased (ThermoFisher). All cargo antibodies were directly conjugated to AlexaFluor™ 647, and imaging was performed on a Nanoimager Mark II S microscope using NimOS software. Analysis was performed using ONI’s CODI software with preset “Pan-EV + custom marker” analysis settings. The DBSCAN algorithm was used to cluster localization data, using localizations from the Pan-EV and Tetraspanin Trio stains.

The data demonstrate no significant difference in the clustering based on the Pan-EV and Tetraspanin Trio stain, indicating no loss of EVs upon permeabilization, regardless of which protein is stained for (Figure 2A). Additionally, a significant signal was only detected in the 640 channel in samples permeabilized and stained for ALIX (Figure 2B).

Figure 2. Permeabilization does not significantly change the number of EVs detected. A) Quantification of detected clusters for each condition, each point represents a FOV, the mean ± SD is plotted. B) Quantification of clusters detected for each experimental condition, plotted with biomarker distribution. All FOVs are represented within each bar; each biomarker bar represents the mean, with SD shown for all FOVs from that condition.

The EV Profiler 2 kit offers researchers a new validated permeabilization modality to ensure that EVs remain intact and that low-abundance cargo, such as ALIX, can be detected and quantified successfully, particularly when including relevant controls. Using the Nanoimager microscope and CODI software, we show a comprehensive framework for implementing cargo marker detection involving EV permeabilization using the EV Profiler 2 kit. The controls detailed here will allow users to gather critical data on EV size, EV biomarker distribution, and quantify EV cargo of interest using custom antibodies while ensuring experiment reproducibility and reliability.

Interested in learning more about how super-resolution can boost your EV research?

References

1. Dixson, A.C., Dawson, T.R., Di Vizio, D. et al. Context-specific regulation of extracellular vesicle biogenesis and cargo selection. Nat Rev Mol Cell Biol 24, 454–476 (2023). doi: 10.1038/s41580-023-00576-0
2. Kim, J., Lee, S.K., Jeong, SY. et al. Cargo proteins in extracellular vesicles: potential for novel therapeutics in non-alcoholic steatohepatitis. J Nanobiotechnol 19, 372 (2021). doi: 10.1186/s12951-021-01120-y
3. Katsuda, T., Kosaka, N., Ochiya, T., The roles of extracellular vesicles in cancer biology: Toward the development of novel cancer biomarkers. Proteomics 14, 412-425 (2014). doi: 10.1002/pmic.201300389
4. Hurwitz, S.N., Rider, M.A., Bundy, J.L. et al. Proteomic profiling of NCI-60 extracellular vesicles uncovers common protein cargo and cancer type-specific biomarkers. Oncotarget. 7(52):86999-87015 (2016). doi: 10.18632/oncotarget.13569
5. Welsh JA, Goberdhan DCI, O'Driscoll L., et al. Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J Extracell Vesicles. 13(2):e12404 (2024). doi: 10.1002/jev2.12404.
6. Larios, J., Mercier, V., Roux A. et al. ALIX- and ESCRT-III–dependent sorting of tetraspanins to exosomes. J Cell Biol. 219 (3) (2020). doi: 10.1083/jcb.201904113