Limitations of nta for extracellular vesicles measurements
Nanoparticle tracking analysis relies on measuring the position of a particle over time and using brownian motion theory to derive the diffusion coefficient of individual particles.
The limitations of nanoparticle tracking analysis are well known but purely discussed, especially when it comes to studying extracellular vesicles. The challenges can be summarized as follows:
- EV's are dim because their refractive index is similar to that of the medium in which they are suspended. Since they are dim, the signal-to-noise is limited, which leads to Localization uncertainty in NTA measurements.
- Most EV's are not just dim because of their optical properties, they are also relatively small (below $100\textrm{nm}$). Since they are small, their diffusion coefficient is large, which leads to even poorer localization accuracy.
- Dim particles that move fast lead to limited number of collected frames (see: Optimal track length for MSD measurements). I have seen papers using just 10 frames to characterize the size, which is very far from giving a reliable estimate.
- Since the scattering cross section scales as $d^6$ (or perhaps as $d^3$ if one considers that they are hollow), smaller particles are too dim for standard methods, which trims the detection at around $100\textrm{nm}$ diameter (even if Particle Metrix and NanoSight claim a lower limit of detection, they are normally misleading the reader.)
One of the biggest challenges is that NTA was built as a method to measure size, and when it comes to vesicles, the result is not reliable.
That's why people started using NTA to quantify the number of particles in a volume. NanoSight has an obscure algorithm to transform detections into concentration (as far as I know it's based on number of detections in a given frame). Particle Metrix does the same, but it's unclear if the algorithm is the same.
Since the performance of both instruments is so poor, most people started coupling size exclusion chromatography with NTA. In that way, each fraction is quantified and the size-profile is reconstructed from there.
Even if NTA is not the best method to measure concentration, it's a cheap one and it has the added benefit of working with fluorescence.
Myriade has released a product based on a different implementation of NTA, but with exactly the same limitations.
Backlinks
These are the other notes that link to this one.
Comment
