Which exosome isolation kit yields intact exosome morphology suitable for electron microscopy and NTA?

With NTA, it is not so much the concentration that matters as the particle size distribution and degree of aggregation. Precipitation-based samples may still show a large-particle-size peak even after dilution; that is not exosomes but aggregates. It is recommended to take the same batch of MSC supernatant and divide it into several aliquots, then perform precipitation, ultracentrifugation, and SEC separately, using the same dilution factor, the same camera level, and the same detection parameters for NTA. Samples with good purity usually have a main peak at 100–200 nm, with D50 and D90 distributions that do not tail. Before loading, gently pipette up and down or briefly sonicate in a water bath for a few seconds, but do not repeatedly freeze-thaw. The purification strategy for both electron microscopy and NTA can be consistent: first ultrafiltration concentration, then passage through a size exclusion column, and finally elution with a small amount of PBS.

Precipitation methods for electron microscopy are prone to polymer micelles and co-precipitated proteins, negative staining backgrounds are dirty, and vesicles can aggregate due to high salt or polymers. MSC supernatant itself has a relatively large amount of exosomes, so it is recommended to directly use ultracentrifugation + density gradient/size exclusion chromatography. Ultrafiltration alone is unlikely to yield very pure exosomes, because contaminating proteins and free proteins will also be retained. In the magnetic bead method, magnetic bead particles in the background are difficult to distinguish from exosomes, so it is not recommended as a priority. If budget allows, consider IZON qEV or similar size exclusion columns: first concentrate 20–30 mL supernatant by ultrafiltration, then pass through SEC; the NTA particle size distribution usually becomes narrower. Electron microscopy fixation conditions also need to be optimized; excessively high glutaraldehyde concentration will disrupt the cup-shaped morphology.

Excessive impurities in electron microscopy images are sometimes not the kit's fault—it's due to too much cell debris or FBS exosome interference when collecting the supernatant. For MSC exosome isolation, it is recommended to use exosome-depleted serum or serum-free medium as much as possible; otherwise, bovine-derived small vesicles will contaminate the preparation. During sample collection, cell confluence and viability affect debris levels. After collecting the supernatant, gradient centrifugation at 300g, 2000g, and 10,000g is essential. Repeated freeze-thaw cycles before pelleting can cause vesicle rupture, releasing DNA and adhesion proteins, which aggravates aggregation. After switching to ultracentrifugation plus SEC, if the EM images are still dirty, go back and check fixation and staining: for negative staining, use 2% phosphotungstic acid or uranyl acetate, and keep both adsorption time and staining time short; otherwise, background will be heavy.

如果预算够，建议不要只买沉淀试剂盒，选排阻色谱柱（比如IZON qEV系列，或其他SEC预装柱）搭配超滤浓缩。qEV对去除可溶性蛋白和小分子效果好，样本回收的囊泡做NTA和电镜通常背景干净。具体货号按你起始体积选，比如上清浓缩后体积1 mL就选对应小柱。如果不想过柱，也可以试试膜亲和柱如exoEasy，但洗脱缓冲液可能含盐，电镜前要置换到PBS。无论用哪种商品化柱，刚换方法时先用一小批MSC上清验证，不要拿珍贵样本直接上。

If ultrafiltration is used for electron microscopy, I generally use it only as a pre-concentration step. Directly centrifuging with an ultrafiltration tube to a small volume will co-concentrate contaminating proteins and cytokines, and vesicles may also adsorb to the membrane or tube wall, affecting morphology. The magnetic bead method is even less suitable for electron microscopy, as there are many bound particles and the background is complex, and pH/salt changes during elution may also cause vesicle rupture. A relatively recommended workflow: low-speed to high-speed centrifugation of MSC supernatant to remove debris, 0.22 µm filtration (note the loss of large vesicles), 100,000g ultracentrifugation for 70 min, resuspension in PBS, followed by size exclusion chromatography. If an ultracentrifuge and angle rotor are available, this protocol offers better stability and purity than precipitation methods, and the cost is mainly time and consumables, so it is not expensive.