Nanomaterials including extracellular vesicles and synthetic nanoparticles are of broad diagnostic and therapeutic application such as disease biomarkers, drugs, and drug delivery vehicles (Cheng and Hill, 2022, Mitchell et al., 2021). The size, shape, and surface characteristics of the nanomaterials are critical for their pharmacokinetics, biodistribution, and interaction with the biological system either in cellular level or in organ level (Davis et al., 2008, Jiang et al., 2008, Poon et al., 2019). The dose of the nanoparticles recently has been shown to be critical for their tumor delivery efficiency (Ouyang et al., 2020). Attention is also brought into the effect of extracellular vesicle dosing on their biological efficacy (Gupta et al., 2021). Therefore, the accurate size and concentration determination is critical for the diagnostic and therapeutic application of nanomaterials (Caputo et al., 2019, Faria et al., 2018, Théry et al., 2018).

Single particle analysis methods including nanoparticle tracking analysis (NTA) (Gardiner et al., 2013, Malloy and Carr, 2006), resistive pulse sensing (RPS) (Vogel et al., 2016), and nanoflow cytometry (Lian et al., 2019) were developed to determine the size and concentration of the particles. Nanoparticle tracking analysis (NTA) based on the laser scattering and Brownian motion of the particles enables simultaneous, multiparametric analysis of nanoparticles in liquid suspension including size distribution, concentration, and direct and real-time visualization (Malloy and Carr, 2006) (Figure S1a). It is now widely used to determine the size and concentration of various samples in different fields (Maguire et al., 2017) such as extracellular vesicles (Auger et al., 2022, Gardiner et al., 2014), virus particles (Kramberger et al., 2012), gold nanoparticle (James and Driskell, 2013, Zhang et al., 2020), and therapeutic proteins formulations (Bai et al., 2017, Tian et al., 2016, Vasudev et al., 2015). The resistive pulse sensing (RPS) strategy (DeBlois and Bean, 1970, DeBlois et al., 1977, Liu, 2024, Liu et al., 2019), the so-called coulter principle, has been extensively exploited in cell sorting and counting for decades and is emerging as new methods for particle sizing and counting (Figure S1b).

To perform NTA analysis, the sample in liquid solution needs to be injected into a sample chamber illuminated by a laser, and the motion of the particles are individually recorded by tracking the light scattering centers of the particles with a camera in either charge-coupled device (CCD) or scientific complementary metal oxide semiconductor (sCMOS) mode (Malloy and Carr, 2006). The Brownian motion of the particles is thus tracked in real-time by the camera, and each particle is simultaneously visualized and tracked by specific image tracking and the data can be processed and analyzed by a customized software (Malloy and Carr, 2006). The diffusion coefficient of each tracked particle could be determined and the particle size in hydrodynamic diameter of each particle was calculated according to the 2-D Stokes-Einstein equation assuming the particles were spherical (Malloy and Carr, 2006, Saveyn et al., 2010). Because the scattering center of each particle is individually tracked, it is possible to generate particle size distribution (PSD) that reflects the actual number of particles, which is superior to the intensity-weighted z-average or polydispersity index produced by ensemble measurement methods such as dynamic light scattering (DLS) (Malloy and Carr, 2006). The NTA reports not only the average particle size which called mean size, but also the frequent particle size which called modal size. Moreover, particle concentrations could be produced as each and every particles are separately analyzed in the sample cell within a given illuminated volume (Malloy and Carr, 2006).

Compared with DLS, NTA has better size resolution in determining the size of unimodal and bimodal nanoparticles (Anderson et al., 2013, Bell et al., 2012, Caputo et al., 2019, Filipe et al., 2010), but NTA is not able to resolve three or more particle populations in the multimodal mixtures with high resolution(Anderson et al., 2013, Filipe et al., 2010). The interlaboratory comparison (ILC) strategy has identified critical parameters in modal size measurement by NTA including: sample shipping, sample handle and preparation, sample storage, video capture (camera level and detection threshold), software, and data acquisition and analysis (Hole et al., 2013). Accurate and reproducible particle mode size measurement (coefficient of variance, CV reduced to 4.4 % from 38.5 %) was significantly improved following SOPs with these parameters optimized (Hole et al., 2013). The algorithms used for data processing greatly impact the particle size measurement output of NTA and DLS analysis as well (Kim et al., 2019a, Kim et al., 2019b). However, the accurate particle concentration measurement by NTA is much more challenging and concentration overestimation by NTA has been reported in several studies, the overestimation is possibly associated with subjective choice of measurement parameters and heterogeneity on particle size and optical property of the sample (Bachurski et al., 2019, Grabarek et al., 2019, Vogel et al., 2021). The NTA is based on the laser scattered by the particles, which varies significantly with the physicochemical property of the particles such as size and refractive index, is critical for the accuracy of particle size and concentration measurement (van der Pol et al., 2014). The size detection limit of NTA is related to accurate concentration determination (van der Pol et al., 2014). Change of camera level (CL) and detection threshold (DT) significantly changed the concentration measurement results (Defante et al., 2018, Gross et al., 2016, Maas et al., 2015, Vestad et al., 2017). The capture time should be long enough to catch enough particles but should avoid decrease of measured concentration with time due to adherence or sedimentation of the particles (Krueger et al., 2016). Increase of the number of video capture replicates also reduced the variance of measurement results (Parsons et al., 2017). The instrument settings including the type of camera, laser wavelength, depth of laser beam, and cleanness/wear of the metalized glass optical plate surface were also reported to significantly affect the concentration measurement results (Bachurski et al., 2019, Gardiner et al., 2013, Vestad et al., 2017). For the NTA instrument integrated with syringe pump, the flow rate is critical for the accurate size and concentration measurement (Krueger et al., 2016, Tong et al., 2016, Zhou et al., 2015). Being aware of the factors affecting the accurate measurement of size and concentration and potential subjective bias on setting the measuring parameters, all the aforementioned efforts are devoted to standardize the measuring protocol to increase the comparability of the data and support the quality control in basic research and regulatory approval (Caputo et al., 2021, Varga et al., 2014, Witwer et al., 2013). However, limited quantitative research has been published on how to optimize the parameters to set up standardized protocols for accurate concentration determination. To minimize the potential operator subjective bias and increase the comparability of measurement results, we aim to further develop a streamlined and standardized procedures with which all the aforementioned critical parameters are optimized. In this research, we first investigated the effect of particle concentration, particles per frame, particle size, detection threshold, camera level, and syringe pump speed on the accurate size and concentration determination in a quantitative way using the widely adopted NTA instrument NS300. As the refractive index of polystyrene (PS) and SiO2 standard particles is different, we included both as testing reference materials to search critical parameters for standard operating procedures (SOPs). We then evaluate the SOPs by determining size and concentration of the monomodal and mixture (bimodal to quadrimodal) of polystyrene or SiO2 particles. After that, we further tested the SOPs by measuring the size and concentration of biological extracellular vesicles from mammalian cells and edible plants. As orthogonal methods are recommended for particle size and concentration determination (Caputo et al., 2019), we also compared the results by NTA to those determined by transmission electron microscopy (TEM), dynamic light scattering (DLS), and RPS.