Nanomaterials
are the subject of a range of biomedical, commercial,
and environmental investigations involving measurements in living
cells and tissues. Accurate quantification of nanomaterials, at the
tissue, cell, and organelle levels, is often difficult, however, in
part due to their inhomogeneity. Here, we propose a method that uses
the distinct optical properties of a heterogeneous nanomaterial preparation
in order to improve quantification at the single-cell and organelle
level. We developed “hyperspectral counting”, which
employs diffraction-limited imaging via hyperspectral
microscopy of a diverse set of fluorescent nanomaterials to estimate
particle number counts in live cells and subcellular structures. A
mathematical model was developed, and Monte Carlo simulations were
employed, to improve the accuracy of these estimates, enabling quantification
with single-cell and single-endosome resolution. We applied this nanometrology
technique with single-walled carbon nanotubes and identified an upper
limit of the rate of uptake into cellsapproximately 3,000
nanotubes endocytosed within 30 min. In contrast, conventional region-of-interest
counting results in a 230% undercount. The method identified significant
heterogeneity and a broad non-Gaussian distribution of carbon nanotube
uptake within cells. For example, while a particular cell contained
an average of 1 nanotube per endosome, the heterogeneous distribution
resulted in over 7 nanotubes localizing within some endosomes, substantially
changing the accounting of subcellular nanoparticle concentration
distributions. This work presents a method to quantify the cellular
and subcellular concentrations of a heterogeneous carbon nanotube
reference material, with implications for the nanotoxicology, drug/gene
delivery, and nanosensor fields.