An integrated approach for the in vitro dosimetry of engineered nanomaterials

Abstract

Background: There is a great need for screening tools capable of rapidly assessing nanomaterial toxicity. One impediment to the development of reliable in vitro screening methods is the need for accurate measures of cellular dose. We present here a methodology that enables accurate determination of delivered to cell dose metrics. This methodology includes (1) standardization of engineered nanomaterial (ENM) suspension preparation; (2) measurement of ENM characteristics controlling delivery to cells in culture; and (3) calculation of delivered dose as a function of exposure time using the ISDD model. The approach is validated against experimentally measured doses, and simplified analytical expressions for the delivered dose (Relevant In Vitro Dose (RID)f function) are derived for 20 ENMs. These functions can be used by nanotoxicologists to accurately calculate the total mass (RIDM), surface area (RIDSA), or particle number (RIDN) delivered to cells as a function of exposure time.

Results: The proposed methodology was used to derive the effective density, agglomerate diameter and RID functions for 17 industrially-relevant metal and metal oxide ENMs, two carbonaceous nanoparticles, and non-agglomerating gold nanospheres, for two well plate configurations (96 and 384 well plates). For agglomerating ENMs, the measured effective density was on average 60% below the material density. We report great variability in delivered dose metrics, with some materials depositing within 24 hours while others require over 100 hours for delivery to cells. A neutron-activated tracer particle system was employed to validate the proposed in vitro dosimetry methodology for a number of ENMs (measured delivered to cell dose within 9% of estimated).

Conclusions: Our findings confirm and extend experimental and computational evidence that agglomerate characteristics affect the dose delivered to cells. Therefore measurement of these characteristics is critical for effective use of in vitro systems for nanotoxicology. The mixed experimental/computational approach to cellular dosimetry proposed and validated here can be used by nanotoxicologists to accurately calculate the delivered to cell dose metrics for various ENMs and in vitro conditions as a function of exposure time. The RID functions and characterization data for widely used ENMs presented here can together be used by experimentalists to design and interpret toxicity studies.

Figure 1

Schematic Map for proposed integrated In Vitro Dosimetry Methodology. a, Proper dispersion preparation requires selection of appropriate sonication media (such as DI H₂O for metal oxide ENMs), and sonication above the critical sonication energy required to break ENMs down to the smallest possible agglomerates that are stable over time. Characterization of dispersion characteristics including agglomerate diameter and agglomerate effective density allow for accurate modeling of particokinetics in vitro, and determination of delivered dose metrics and the deposition fraction constant. b, Relevant In Vitro Dose functions (RID_f) provide a simplified tool for nanotoxicologists to quickly estimate delivered dose values for the ENMs investigated in this manuscript. Selection of the appropriate deposition fraction constant (α, listed in Table 1), allows nanotoxicologists to directly calculate relevant in vitro doses (RID) for any exposure duration, including delivered ENM mass (RID_M, μg), delivered particle number (RID_N, #), and delivered surface area (RID_SA, cm²), using the equations listed below. t is exposure duration (h), γ is ENM mass concentration (μg/ml), V is media volume applied to cells (ml), r_h is hydrodynamic radius (cm, listed in Table 1), and ρ_E is agglomerate effective density (g/cm³, listed in Table 1).

Figure 2

Validation of dosimetry methodology for two metal oxide ENMs. a. Validation of dosimetry approach for CeO₂ (d_XRD=28.4 nm) suspended in DMEM; b. Validation of dosimetry approach for SiO₂ coated CeO₂ (d_XRD=28.4 nm) suspended in DMEM. All experiments were done in triplicate, error bars represent standard deviation.

Figure 3

Comparison of time required to deliver 90% of the administered dose (t ₉₀ ) in hours (h), calculated following the described dosimetry methodology, for all materials investigated in two well plate configurations.