- Split View
-
Views
-
Cite
Cite
Paul Borm, Frederick C. Klaessig, Timothy D. Landry, Brij Moudgil, Jürgen Pauluhn, Karluss Thomas, Remi Trottier, Stewart Wood, Research Strategies for Safety Evaluation of Nanomaterials, Part V: Role of Dissolution in Biological Fate and Effects of Nanoscale Particles, Toxicological Sciences, Volume 90, Issue 1, March 2006, Pages 23–32, https://doi.org/10.1093/toxsci/kfj084
- Share Icon Share
Abstract
Dissolution, translocation, and disposition have been shown to play a key role in the fate and effects of inhaled particles and fibers. Concepts that have been applied in the micron size range may be usefully applied to the nanoscale range, but new challenges are presented based on the small size and possible change in the dissolution:translocation relationship. The size of the component molecule itself may be on the nanoscale. Solute concentration, surface area, surface morphology, surface energy, dissolution layer properties, adsorbing species, and aggregation are relevant parameters in considering dissolution at the nanoscale. With regard to the etiopathology caused by these types of particulates, the metrics of dose (particle number, surface area, mass or shape) is not yet well defined. Analytical procedures for assessing dissolution and translocation include chemical assay and particle characterization. Leaching of substituents from particle surfaces may also be important. Compartmentalization within the respiratory tract may add another dimension of complexity. Dissolution may be a critical step for some nanoscale materials in determining fate in the environment and within the body. This review, combining aspects of particle toxicology, material science, and analytical chemistry, is intended to provide a useful basis for developing relevant dissolution assay(s) for nanoscale particles.
INTRODUCTION
During the past decade, a challenge for modern particle research has emerged by the increasing use of materials with a dimension in the range of 1–100 nm in nanoscience and nanotechnologies (Borm, 2002; Borm and Kreyling, 2004). Nanoscale particles (NSP) have three or perhaps two dimensions in the range of 1–100 nm, and may be considered to be a toxicological entity in their interactions with the environment, cells, and molecules within the cell. At these dimensions, there is increased concern that nanoscale particles may translocate from the respiratory tract or other portal of exposure, with potential for effects in other organ systems. In addition, little is known about their environmental distribution, accumulation, and degradation (Colvin, 2003). Dissolution, translocation, and deposition have been shown to play a key role in the fate and effects of conventional fibers and particles. The concepts that have been applied to particles in the micron size range may be usefully applied to the nanoscale range, but with new challenges based on the very small size and possible changes in the dissolution:translocation relationship.
Determining the colloidal and surface properties of particles in this size range can be complex, but this state is likely to play a key role in biological fate and effects. This review seeks to combine materials science principles and analytical approaches to nanoscale particle dissolution, with an objective that these could be applied to assessment of dissolution as part of a screen for environmental and or toxicological evaluation of nanoscale materials.
PARTICLE TOXICOLOGY: BASIC CONCEPTS
Apart from the specific chemical surface reactivity of nanoparticles, a particle load or burden in the lung can induce toxicological response(s) that differ principally from soluble or non-particulate toxicants. For the interpretation of inhaled particle effects, dose, deposition, dimension, durability and defense are important determinants of response. First of all, the dose at a specific site (in the lungs) determines the potential toxicity of particles. Obviously, this deposited dose is dependent on the inhaled concentration as well as the dimensions of the particle. As a result of diffusion processes, the deposition probability of nanoparticles in the respiratory tract markedly increases for ultrafine particles above that for larger particles where convective flow dominates deposition. The dose metric for nanoscale particles (NSP) is an additional complexity, as particle number, surface area, and shape, among other factors, may play a role in addition to the traditional mass-based metric (Donaldson et al., 2002; Oberdorster et al., 2005).
The lung has extensive, location-specific defense systems such as mucociliary clearance in the upper airways and macrophage clearance in the lower, unciliated part of the respiratory tract. Particle transport by macrophages from the alveolar region toward the larynx is rather slow in humans, even under normal conditions, and it eliminates only a fraction of the deposited particles in the peripheral lung. The remainder may accumulate unless the particles are biodegradable or cleared by simple chemical dissolution in lung surfactant or interstitial fluid. Therefore, the same deposition of NSP with a different durability can lead to a different cumulative dose. The kinetics of dissolution of inhaled particulates determine whether a low-toxicity particle, such as amorphous silica, will dissolve in the epithelial lining fluid or whether such particles as carbon blacks or iron oxides are engulfed by alveolar macrophages. Thus, solubility is a key driver for the clearance mechanisms involved in their removal. The prediction of the in vivo dissolution kinetics of particles is potentially affected by subcellular compartments with different pHs than the lining fluids. Moreover, agglomerated airborne particles in the 1 μm range may be broken down to the size of the primary nanoparticle within these structures of the lung. Ultrafine particles may have constituents, either as chemical components or as elements adsorbed to the surface, and dissolution could release these components into the lung tissues and circulation (Bagate et al., 2004). If a toxic particle is neither soluble nor degradable in the lung, it is likely to have high durability (biopersistence) with accumulation upon sustained exposure. In addition to the known biological effects of low-soluble particulate materials, dispositional factors of NSP may also influence the effects.
Toxicology of Ultrafine Particles
Nanoscale particles, whether termed ultrafine, nano-, engineered, intentional, or incidental, pose challenges for physical, chemical, and biological characterization. A variety of biological effects have been reported, although in some cases potential hazards have been identified, generally without an assessment of a relevant or likely exposure (Donaldson et al., 2002; Borm and Kreyling, 2004; Oberdorster et al., 2005). A variety of physico-chemical effects have also been reported, such as strong interactions with other particles, either of the same size or aggregated (cohesion), and with surrounding media (adhesion). Toxic effects have been reported to include an inflammatory response with exacerbation of airways disease, cardiovascular events by hypercoagulabiliy or plaque destabilization (Pope et al., 2004), or chronic effects in animal models (Driscoll et al., 1996; Borm et al., 2004). Much of this database is from research on combustion-derived nanoscale particles and may not be generally characteristic of NSPs or may be less relevant for engineered NSPs.
Realistic exposure assessments and test procedures will be important for risk assessment of NSPs. Migration of NSPs to a distance from their application or deposition sites is possible, although the characterization of nanoscale particle pharmacokinetics is very limited, except for pharmaceutical-related products. Discrete primary NSP may have relatively low recognition and uptake by lung macrophages, and this could facilitate movement from the lungs. Exposure to aggregates (also termed “secondary particles”), which may be the more likely species of inhalation exposure, may have quite different clearance kinetics. In addition to phagocytosis (cellular ingestion of relatively large objects), endocytotic processes of pinocytosis and receptor-mediated endocytosis are likely to play a role in translocation of NSP (Gumbleton, 2001; Arredouani et al., 2004). Several authors have reported on the movement of nanoscale particles from the lung to the blood (Berry et al., 1977; Nemmar et al., 2002), and they hypothesized the potential for microthrombus formation, which has become a focus of research attention in relation to fine particulate air pollution (Nemmar et al., 2004). Oberdorster et al. (2005) reviewed studies that indicated that various NSP, including gold particles (50 nm), poliovirus (30 nm), MnO2, and carbon-spark generated NSP may be directly transported toward the brain via axonal transport, including trans-synaptic transport. Nanoscale particles may affect movement of endogenous substances (proteins, enzymes) or exhibit increased reactions with biological molecules.
The cumulative dose upon chronic exposure is dependent on exposure level and clearance pathways. Apart from macrophage clearance, dissolution and biodegradation are important components of (lung) clearance that may prevent accumulation of NSP. This is reflected in the outcomes of a chronic rat experiment, to which the tumor incidence of several durable ultrafine particles (carbon black, TiO2, diesel) and soluble (amorphous silica) ones, as well as some fine particles of the same chemical composition, were compared (Borm et al., 2004). Particles were instilled intratracheally in doses up to 60 mg. Lung tumor incidence induced by three different insoluble NSP was proportional to their surface area. The high-surface amorphous silica induced few lung neoplasms, possibly because of its high solubility, i.e., low durability in vivo. These data suggest that particle dissolution may be a useful in vitro screening method for assessment of particulate nanomaterials. In Figure 1, a schematic of the role of dissolution is illustrated by contrasting the fate of soluble versus poorly soluble particles. Dissolution may yield constituents with biologic activity, producing injury if present or accumulated in excessive amounts. For the poorly soluble nanoscale particles, a “large” surface dose (“large” equals potential to cause injury) could cause injury, presumably on the basis of surface effects. In the case of inhaled nanoscale particles, the effects may be reflected in pulmonary inflammation. The diagram also depicts potential accumulation, translocation, and distribution of nanoscale particles.
Assessing durability and dissolution have been important for the development and regulation of man-made vitreous fibers (MMVF), with markedly reduced potential for pulmonary toxicity. However, procedures for definition and measurement of nanoscale particle dissolution are needed to allow evaluation and implementation of such a method. Theoretical and practical aspects of NSP dissolution and measurement are discussed in the following sections.
MATERIALS SCIENCE ASPECTS
Before describing the dissolution process in more detail, it is important to differentiate nanosized particulate materials from nanosized molecules. As already mentioned, nanoparticles are generally defined as particles that have at least one dimension in the 1–100 nm size range. Though this definition is accurate by contemporary standards, questions arise as to what differentiates a molecule in solution from a particle—many NSPs such as fullerenes, dendrimers, and protein molecules are composed of a single molecule but embody particle-like characteristics; whereas others, such as polyethylene glycol (PEG) molecules, arguably may not. One reasonable assessment is that if a molecule contains segments, or an internal core that is insoluble (has an energetically unfavorable interaction with the solvent promoting phase separation), then that molecule as a whole should be classified as a particle. Because insolubility restricts a molecule's conformational and thermal freedom, it can be argued that it contributes to a more particle-like behavior. In many cases, NSPs such as metal oxide and polymeric nanoparticles, are composed of multiple molecular constituents that exist as relatively rigid structures independent of the solvent present. In contrast to solvated molecular species, these molecules closely interact with each other and are tightly packed together via intermolecular binding forces (e.g., metal binding, van der Waal's forces, ionic binding) that are often heterodesmic. During the dissolution process, these intermolecular bonds are broken via a process that results in a lower free energy of the system. It should be noted that particles and aggregates of particles can also be broken down to smaller particles via milling and/or dispersion—these phenomena are distinct from dissolution in that they predominantly result in the generation of smaller particles, rather than solvated molecules, and typically require relatively large amount of mechanical work to reduce the particle size. The change in free energy involved in breaking apart particulate constituents or increasing surface area may be an important consideration for assessing dissolution of a species. However, issues surrounding differentiating spontaneous dispersion of single molecule nanoparticles versus classical dissolution phenomena are yet to be resolved. More definition is needed in this area. For the purpose of this review, dissolution is defined as described in the next section.
Dissolution
Dissolution is the dynamic process by which a particle goes into the solution phase to form a homogeneous mixture. For a material to dissolve, its constituent molecules must have some solubility in the local environment. For certain classes of NSPs, as discussed above, the size of the component molecule may be on the nanoscale. For example, a single C-60 is approximately 1 nm and a hemoglobin molecule is approximately 6 nm. (It should be noted that further breakdown of these NSPs results in degradation rather than dissolution.) Material solubility is highly dependent on solvent properties (e.g., pH, ionic strength, constituent solvated molecules, and concentration), and therefore may vary with the region of exposure in vivo. Both particle dissolution kinetics and solubility are size dependent, and nanoparticulate materials are often expected to dissolve more quickly and to a greater extent (i.e., higher equilibrium solubility) than macroscopic particles of the same material. Because of this, the dissolution of nanoparticulate materials can govern their persistence in vivo and should be considered an important factor when interpreting biological responses to NSP exposures. Considering the difficulty in predicting dissolution behavior in complex systems (e.g., in vivo), experimental toxicologists need to be aware of fundamental aspects of the dissolution process as well as other variables that can affect the phenomenon. Such knowledge will be useful in designing more meaningful experimental protocols, characterizing systems, and interpreting data from particle exposure studies.
During particle dissolution, constituent molecules of the dissolving solid migrate from the surface to the bulk solution through a diffusion layer (Fig. 2). The diffusion layer represents a region between the particle surface and bulk solution environment that is densely populated by solvated molecules (e.g., ions, solute molecules, biomolecules, etc.) and is demarcated by the shear plane that separates the stagnant fluid layer that travels with the particle, from the bulk solution phase. The driving force for dissolution depends on the materials solubility within a given environment as well as the concentration gradient between the particle surface and the bulk solution phase. Hence, the kinetics of dissolution of soluble materials is surface area dependent. Nanoparticulate materials are anticipated to dissolve faster than larger sized materials of the same mass by surface area considerations alone—although other factors such as surface curvature/roughness also play a role. The dissolution process slows as the concentration in the bulk solution phase reaches equilibrium solubility levels.
Role of Surface Morphology and Curvature
Different sized and shaped particles of the same material at equivalent surface area dosages and starting free (dissolved) solute concentrations undergo dissolution at different rates (Tang et al., 2004). This phenomenon has been explained through alterations in diffusion layer properties (Tinke et al., 2005), as well as variations in equilibrium solubility (Iler, 1979).
Diffusion layer thickness has been experimentally found to decrease with particle size leading to faster transport of solvated molecules to bulk solution and hence more rapid dissolution (Tinke et al., 2005). This consideration can also be extended to surface morphological features, and may provide explanations for the preferential dissolution of convex versus concave surface structures. The overall hydrodynamic shape of a particle defines its surrounding shear plane boundary; therefore smaller particle surface features can have varying local diffusion layer thicknesses. Intuitively, we can conclude that convex features will have thinner diffusion layers because they extend from the surface toward the shear plane, whereas concave features will have thicker diffusion layers because they are recessed on the surface—leading to relatively faster and slower dissolution rates, respectively.
Though there is some basic understanding of trends in particle dissolution with size and surface curvature, toxicologists should be aware that particles of different size and/or shape are often manufactured/generated through different routes. The route of manufacture/generation often leaves fingerprints of imperfections (e.g., localized stress, structural defects, and dislocations), contaminants (e.g., dispersant molecules and residual molecules from the synthesis process), and surface features (e.g., dominant crystal orientation, roughness, passivation layers, and morphology) that can further modify dissolution behavior (Adamson et al., 1997). Actual particle surfaces tend to be complex and imperfect in many ways and are often poorly characterized. To further complicate the matter, for small NSPs (1–10 nm) quantum mechanical effects and cooperative events from a limited number of atoms may alter various physical properties (Adamson et al., 1997). Our current understanding of nanoscale systems is inadequate to predict the complex behavior of these materials. However, in the literature there are reports of phenomena unique to small NSPs. The size-dependant optical properties have been well established for a variety of semiconductor materials, e.g., quantum dots (Alivisatos, 1996a, b). More recently, small NSP crystallites have been shown to change crystalline structure in different solvent environments, which may have a impact on dissolution phenomena (Zhang et al., 2003).
Impact of Agglomeration
In addition to the complications described above, nanoparticles have a tendency to agglomerate, or form multiple single particle clusters, in biological fluids. The formation of agglomerates can hinder dissolution (increasing persistence) by reducing the average equilibrium solubility of the particle system and by introducing kinetic hindrance to the diffusion process. Experimental toxicologists should be aware of the potential role of agglomerates in dissolution, because the state of agglomeration is affected in large part by the mode of administration (i.e., internasal/intertracheal inhalation, insufflation, or instillation). The probability for agglomeration increases with increased localized dose and for doses given over shorter time frames.
When agglomerates form a neck between two or more particles, this creates an area of negative surface curvature (Fig. 3) (Israelachvili, 1992). Areas of negative curvature have lower theoretical equilibrium solubility, as previously discussed. Under equilibrium conditions nucleation is predicted to occur at this interface, which can result in fusion of the agglomerate and a reduction in the total particle surface area. This process is most likely to occur toward the center of rather large aggregates in vivo because saturated and quasi-equilibrium solution conditions are more likely to occur in the agglomerate interior due to restricted diffusion to the bulk solution phase. Because the hydrodynamic shear plane is defined around the agglomerate, the formation of a clump or cluster of particles greatly increases the overall diffusion layer thickness. This significantly hinders the diffusion efficiency of solute molecules from interior particles to the bulk solution. The extent of this hindrance process will ultimately be related to the aggregate volume and packing factor, with larger, more densely packed aggregates exhibiting slower dissolution.
Impact of Adsorbed Species on Dissolution
The adsorption of molecules and ions from solution can also have a profound effect on dissolution probability and kinetics. They can serve to catalyze or passivate dissolution in addition to modifying diffusion layer characteristics (Adamson et al., 1997; Fukushi and Sato, 2005). The adsorption of molecules at the particle solid–liquid interface generally results in a lower interfacial tension and adds an additional molecular diffusion barrier, resulting in lower and slower solubility. However, certain classes of molecules may specifically promote or demote the dissolution process; e.g., chelating agents are known to modify the solubility of various materials (Fredd and Fogler, 1998; Gao et al., 2004; Perry et al., 2005). Solubility and dissolution are enhanced when organic or ionic molecules are able to form complexes with surface molecules that collectively are soluble. In contrast, complexes that result in a less soluble, “stickier” construct will tend to hinder dissolution. The adsorption of molecules from solution is a dynamic process, even at “equilibrium” conditions. Adsorbed ions and molecules are constantly exchanged at the interface driven by thermal motion and their affinity to the interface. More solvated molecules undergo faster and more energetically favorable exchange processes and therefore are able to enhance dissolution by essentially sequestering and directly removing surface ions (Iler, 1979).
In biological systems, a vast variety of proteins, enzymes, polysaccarides, and other molecules are present that are known to adsorb to interfaces (e.g., opsonization) in a matter of seconds (Ratner, 1996). Difficulties, however, are encountered when one tries to rationalize what particular molecules are interacting with the surface at any given time. In addition, components from the biological system may react with surface components on the particle, such as described for ascorbate and iron on the silica surface (Fenoglio et al., 2000). Exposures to particles by inhalation typically result in the triggering of a number biochemical pathways that ultimately result in a series of modified solution conditions. For instance, particle exposures have been shown to lead to the upregulation of genes that promote fibronectin and collagen (potential “sticky” particle-coating molecules) expression in the lung epithelium (Kuwahara et al., 1994). At the same time the infiltration of neutrophils and macrophages through chemotaxis may result in exposure to various enzymes and acidic molecules or, perhaps, particle uptake, but overall an interaction with a completely different environment (Renwick et al., 2004). In addition to these considerations, the adsorption and denaturing of biomolecules to interfaces is apparently also a size-dependent phenomenon—adding further complexity to the nanoparticle dissolution process (Borm and Kreyling, 2004; Vertegel et al., 2004). Tighter, more packed biomolecule layers are typically found on larger particles. Moreover, the formation of passivation layers (e.g., oxide films), and the presence of impurities, stress loci, and dislocations can also modify the adsorption process, as well as the dissolution phenomena in vivo.
Analytical Aspects
Particle dissolution studies can be performed through the evaluation of particulate properties, such as particle size or number, concentration, or weight, by assay of the concentration of dissolved particulate (solvated molecular species) generated, or through a combination of both. For toxicological interpretation, it may be relevant to try to simulate more than one compartment within the body or in the environment. In some cases, sinks and issues of non-steady-state thermodynamics may also influence the fate of the particle. Monitoring of both particulate and molecular parameters ultimately enables the completion of a mass balance. Choice of technique to monitor particle dissolution will be dependent on factors such as the desired precision and/or accuracy, the particle properties (composition, morphology, phase, size, and shape), the analytes, the chemical make-up of the dissolution fluid when attempting to simulate biological systems, and the desired cycle time of an analysis. As a result, each measurement technique will have a discrete range of applicability. The discussion of dissolution technologies will be limited to the study of particles that have been dispersed in a fluid that may or may not contain additives intended to simulate a biological fluid, thus enabling measurements under well-controlled or ideal conditions. When analyzing particle size distribution of liquid-borne particulates or the concentration of dissolved species, the response of the analysis system to either form or to separate solid-phase or solubilized molecular species, must be clearly understood. For instance, ultraviolet-visible (UV-VIS) spectroscopy has been used extensively to determine the concentration of molecular species in solution during drug dissolution studies. In the UV-VIS spectrophotometric analysis of submicron particle dispersions, modeling has been developed and implemented as a software package that enables the output of a spectrophotometer to be converted into particle size distribution information (Alba, 2004). The results of such an analysis are dependent on the scattering from the particulate present and will be greatly influenced by the presence of absorbing molecular species as well. Whether determining solubilized molecular species concentration or dispersed particle size with UV-VIS spectrophotometry and possibly other chemical assay techniques, one must fully understand how the simultaneous presence of undissolved particles and dissolved molecular species affect the measurement. Sampling the particle or solvent may affect dissolution kinetics. If sampling volumes are not relatively small, and if dissolution kinetics is not relatively slow, effects of sampling should be considered.
Under well-controlled conditions, filters may be used to remove particulates that have been found to interfere with the measurement of solubilized molecular species during particle dissolution. Filters with high efficiency ratings at nanoparticle sizes are available from several filter manufacturers.
Monitoring Dissolution via Chemical Assay
Chemical assays of molecular species in solution have been performed using a wide variety of analytical techniques. The applicability of any given technique will be dependent on the particle's chemical composition and/or the concentration of soluble species that one is trying to detect. The concentration of molecular species from organic particle dissolution can be measured using instrumental techniques (Settle, 1997) such as fluorescence spectroscopy, UV-VIS spectroscopy, liquid chromatography, and electrochemical techniques, to name a few. The limit of detection and quantitation for each technique will be dependent on the detection scheme used in the chromatographic techniques and on the chromophore or electroactive species being monitored.
Techniques for monitoring inorganic molecular species include x-ray fluorescence spectroscopy, inductively coupled plasma spectroscopy, neutron activation analysis, and ion chromatography, among others to name a few. Once again, the technique to employ and the ultimate detection limit will be dependent on the analyte of interest.
A sample preparation besides filtration, i.e., extraction, concentration, dilution, may be required in order to complete the assay.
Monitoring Dissolution via Particle Characterization
When it comes to the particle size analysis of nanoparticles suspended in an aqueous fluid, there are numerous techniques available. The goal of performing a particle size analysis would be to monitor the particle size distribution as a function of time to obtain a particle dissolution profile. Chromatographic techniques such as field flow fractionation (Giddings, 1993; Chen et al., 2005) and hydrodynamic chromatography (Small, 1974; DosRamos, 1991; Tribe et al., 2003) enable the measurement of complex size distributions with relatively high resolution in the <10 to >100 nanometer size range. Moreover, during a field flow fractionation and hydrodynamic chromatography analysis, particulate and solubilized molecular species are separated, enabling one to determine the concentration of both. Another particle size technique that is fractionation based is photosedimentometry (Weiner, 1991, Fitzpatrick, 1998). The lower size limit of the photosedimentometer can be limited by the density difference between the particulate of interest and the analysis fluid. At low density differences, the sedimentation of <100-nm-sized particles requires long analysis times, and when this occurs Brownian motion can detrimentally affect the fractionation and results.
Light-scattering techniques have also been used to monitor size and number concentration of particles present in a fluid. Static scattering techniques have been applied to particle characterization in a wide variety of forms including UV-VIS spectroscopy (Alba, 2004), turbidimetric (Tucker, 2004), and photon migration (Dali et al., 2005). The lower size and concentration limits of detection for each of these techniques should be evaluated for each particle system of interest. The ability to monitor changes in size and concentration at 10 nm may be limited for most of these techniques. Dynamic light scattering (Allen, 1992) has been applied extensively in the nanometer size range. However, its ability to accurately characterize complex size distributions is somewhat limited.
Finally, microscopy techniques can be used to monitor changes in particle size and particle concentration. Recently, advances in visible-light-optical microscopy have enabled visualization of particles and/or the light scattered by particles with diameters less than 100 nm using a visible light microscope. Darkfield illumination techniques now commercially available may be capable of enabling visualization of the particle dissolution process using a visible light microscope (Aetos Technologies, Inc., Auburn, AL, USA; Nanosight Ltd., Wiltshire, UK). Plasmon resonant particles (Schultz, 2003) also have the unique characteristic of being “ultra-bright” when evaluated in light-scattering experiments thus enabling viewing with a light microscope.
State of Aggregation and Agglomeration
Aggregates are defined as a secondary particle composed of primary particles that are strongly bonded together (fused, sintered) and acting as a unit. An agglomerate is defined as a secondary particle which, upon handling, breaks into aggregates. The degree of aggregation and agglomeration may affect movement of the particle in external media (for example, aggregation would influence aerodynamic diameter) as well as internally. Agglomerates and aggregates act as large particles with internal porosity in their aerodynamic properties. It is difficult to determine the degree to which an agglomerate will break apart and to calculate an aggregate size, as this depends on the mechanical stresses to which the agglomerate is subjected. It is unlikely that an aggregated secondary particle will itself de-aggregate to become a collection of separate primary particles (Rödelsperger et al., 2003), but there are no recognized test standards to follow on this matter. A recent study by Cherukuri et al. (2004) combined the use of fluorescence microscopy and near-infrared excitation to show that morphological aggregates of carbon nanotubes in macrophages were behaving chemically as simple primary carbon nanotubes. This suggests that physical properties of NSP may help to characterize the fate of NSP in the body.
The parameters that change as primary particles aggregate and agglomerate are the hydrodynamic size (becomes larger), the number concentration (becomes smaller), and possibly the specific surface area. The agglomerated particles will behave as a population of larger particles, and the size distribution, specific surface area, dissolution rate, or any other parameters that can be used in toxicological assessment can be measured on the agglomerated system, but it must be understood that the results cannot be fully predictive of mobile, discrete NSP. It may be possible to assess the state of agglomeration by analyzing electron micrographs of the particles; however, sample preparation often involves a dispersion step followed by evaporation, and the mechanical stresses may break up agglomerates. Any non-imaging techniques will provide a size distribution but no direct measurement of the state of agglomeration. The state of agglomeration can be assessed with prior knowledge of the size distribution of the primary particles.
All of the size characterization techniques outlined earlier can be used to measure the size distribution of agglomerated and non-agglomerated particles. Because the assumption of sphericity is used in all particle-sizing technologies, with the exception of imaging techniques, it is expected that the data generated for non-spherical particles may be somewhat dependent on the particular technique used. In addition, when sedimentation techniques are used, the density of the particles needs to be known for accurate measurement, as Brownian motion can affect hydrodynamic flow at the smallest measurable size. Furthermore, if the particles under measurement consist of distinct populations having different densities, the size distributions generated may lack accuracy.
DISCUSSION AND FUTURE DIRECTIONS
Dissolution assessment has been successfully applied to macro-sized materials such as man-made mineral fibers that are both slightly soluble and insoluble (Fubini et al., 1998). In a very strict sense, the final form of any soluble solid material is a nanoscale particle that becomes smaller than the classical critical nucleus size. It is also a formalism that all materials that precipitate from a fluid pass from the original soluble species through the critical nucleus size up and through the nanosize, until they attain the macro-sizes that allow sedimentation, filtration, or centrifugation. However, the current interest in nanoscale materials is for what are traditionally considered insoluble materials. Furthermore, the use of dissolution data when applied to toxicological investigations will focus on degrees of insolubility. The difference between a persistent, durable, solid nanoscale substance and a less durable, “biosoluble” nanoscale substance can be quite small and will require discerning analysis.
Amorphous silica was used earlier in this article as an example of a soluble nanoscale material that caused few neoplasms following intratracheal instillation, relative to more durable NSP (Borm et al., 2004). Amorphous silica has a solubility ranging from 2.00 mmol l−1 at around pH 7 and 25°C to 4.5 mmol l−l at the around pH 7 and 45°C (Vogelsberger et al., 1995) which are discretely different than cristobalite at 0.45 mmol l−l and quartz at 0.09 mmol l−l (Iler, 1979). With silicas, in addition to dissolution, surface reactivity has also been important in determining biological reactivity (Donaldson and Borm, 1998; Fubini et al., 2004; Schins et al., 2002). In Warheit's review of pulmonary responses to different forms of silica (2001), cristobalite was indicated to produce the greatest measure of lung injury; quartz produced intermediate effects, and amorphous silica produced minimal pulmonary effects. In terms of analytical technique, small differences in dissolution exist among these polymorphs of silica, and dissolution, in turn, influences pulmonary effects through the concept of persistence.
Synthetic amorphous silica has also been the subject of dissolution testing using a simulated biological medium (Roelofs and Vogelsberger, 2004) at 37°C and pH near 7. The authors report a maximum solubility of 2.3–2.7 mmol l−l, depending on the source of the commercial silica samples and on primary particle size. These values are consistent with the same authors' earlier study in water. Roelof and Vogelsberger also confirmed the fact that silica dissolution rate is more rapid than the reverse precipitation rate. In dissolution studies, this means that the solution becomes supersaturated, and the experimentalist would need to wait a considerable time to reach equilibrium. Hence, silica dissolves both in water and in simulated biological systems beyond the equilibrium concentration value.
Isolated NSP or agglomerates located in the alveoli will not necessarily attain an equilibrium state in terms of dissolution. Thermodynamically, the alveoli represent an open system where pulmonary fluids circulate and can be replenished. Some constituents of these fluids have ion-complexation/-sequestration properties that enhance dissolution even further. The subcompartmental kinetic rates of dissolution can therefore dominate over equilibrium values for solubility. This is likely the situation observed with synthetic amorphous silica particles, where the tendency to supersaturate aids in rapidly dissolving and being eliminated from the lung. Other substances may have different properties. Their dissolution may be inhibited by adsorbed biological and tramp species, or hydrous surface layers may act to limit dissolution. All of these factors, the equilibrium solubility value, the kinetic rates of dissolution and precipitation, and the possibility of enhancement or inhibition due to hydrated surface layers and adsorbed species may be relevant when evaluating a particle's dissolution in the context of pulmonary persistence. Nevertheless, for man-made vitreous fibers, mathematical modeling of dissolution and subsequent biological responses in the rat lung has successfully been done (Tran et al., 2003). Nanoscale particles may be subject to endogenous breakdown mechanisms that are available for proteins, nucleic acids, or general xenobiotics. However, few studies are available on the biotransformation of NSP. Studies with different water-soluble fullerenes have shown that minor alterations in the fullerene structure have a large effect on the cytotoxicity in two different human cell lines (Sayes et al., 2004). This suggests that biotransformation of molecular components or NSP itself maybe a relevant mechanism for the prevention of cytotoxicity.
In conclusion, dissolution may be among the critical steps for some nanoscale materials in determining fate in the environment and within the body. Translocation of insoluble particles into cells is likely to be dependent on processes of endocytosis (mainly pinocytosis and phagocytosis or receptor-mediated endocytosis). Similarly, movement of particles from live cells may occur largely by exocytosis. Diffusional movement of nanoparticles through cell membranes is likely to be limited under normal conditions (Heckel et al., 2004; Meiring et al., 2005), although more research in this area may provide a better understanding of nanoscale particle translocation processes. There is potential for accumulation of (insoluble) NSP in the body. Effects would be dependent on the dose, the properties of the nanoscale material, and the location of accumulation. Dissolution, even over the time frame of weeks or months, may significantly enhance clearance of nanoscale materials. Characterizing the state of the component molecule or particle, and understanding the processes of translocation, dissolution, and particle receptor interactions are key challenges to defining the toxicology of NSPs.
Measuring fiber dissolution in vivo (Bellmann et al., 1986; Warheit et al., 1994) and in vitro (Jaurand, 1994; Zoitos et al., 1997) has been done to assess their biopersistence and predict potential chronic effects. Both morphological biopersistence (measuring fibers with microscopic techniques) and dissolution rate (kdis, in units of mass per surface area of particle) have been used. Biological fluid simulants at defined pH are used in the in vitro assessment. These principles may be usefully applied to nanoscale materials, although substantial technical challenges exist. Following the morphological transformation of a particle into single molecules will require resolution at the nanometer scale. Measuring the appearance of the chemical in the surrounding fluid will require analytical sensitivity and resolution at extraordinarily low concentrations to evaluate nanoparticles at low dissolution rates.
The description of material science concepts and analytical approaches to dissolution at the nano scale is intended to provide a useful basis for developing relevant dissolution assay(s) for nanoscale particles and addressing the challenges of how this may relate to the in vivo situation.
The authors gratefully acknowledge the ILSI-HESI organization, especially M. P. Holsapple, for support of the ILSI-HESI Nanomaterials EHS Committee and Dissolution group. We thank James E. Gibson for review of the manuscript as part of the ILSI-HESI review process.
References
Adamson, A. W., Gast, A. P., and NetLibrary Inc. (
Arredouani, M., Yang, Z., Ning, Y., Qin, G., Soininen, R., Tryggvason, K., and Kobzik, L. (
Alba, F. (
Alivisatos, A. P. (
Alivisatos, A. P. (
Bagate, K., Meiring, J. J., Gerlofs-Nijland, M. E., Vincent, R., Cassee, F. R., and Borm, P. J. (
Bellmann, B., Konig, H., Muhle, H., and Pott, F. (
Berry, J. P., Arnoux, B., Stanislas, G., Galle P., Chretien, J. (
Borm, P. J. A. (
Borm, P. A., and Kreyling, W. (
Borm, P. J. A., Schins, R. P. F., and Albrecht, C. A. (
Chen, B., Jiang, H., Zhu, Y., Cammers, A., and Selegue, J. P. (
Cherukuri, P, Bachilo, S. M., Litovsky, S. H., and Weisman, R. B. (
Colvin, V. (
Dali, S. S., Rasmussen, J. C., Huang, Y., Roy, R., and Sevick-Muraca, E. M. (
DosRamos, J. G. (
Donaldson, K., and Borm, P. J. A. (
Donaldson, K., Brown, D., Clouter, A., Duffin, R., Macnee, W., Renwick, L., Tran, L., and Stone, V. (
Driscoll, K. E., Carter, J. M., Howard, B. W., Hassenbein, D. G., Pepelko, W., Baggs, R. B., and Oberdorster, G. (
Fenoglio, I., Martra, G., Coluccia, S., and Fubini, B. (
Fitzpatrick, S. T. (
Fredd, C. N., and Fogler, H. S. (
Fubini, B., Aust, A. E., Bolton, R. E., Borm, P. J. A., Bruch, J., Ciapetti, G., Donaldson, K., Elias, Z., Gold, J., Jaurand, M. C., et al. (
Fubini, B., Fenoglio, I., Ceschino, R., Ghiazza, M., Martra, G., Tomatis, M., Borm, P., Schins, R. P. F., and Bruch, J. (
Fukushi, K., and Sato, T. (
Gao, H., Tan, T., and Wang, D. (
Giddings, J. C. (
Gumbleton, M. (
Heckel, K., Kiefmann, R., Dorger, M., Stoeckelhuber, M., and Goetz, A. E. (
Iler, R. K. (
Jaurand, M. C. (
Kuwahara, M., Verma, K., Ando, T., Hemenway, D. R., and Kagan, E. (
Macinnis, I. N., and Brantley, S. L. (
Meiring, J. J., Borm, P. J. A., Bagate, K., Semmler, M., Seitz, J., Takenaka, S., and Kreyling, W. G. (
Nemmar, A., Hoet, P. H., Vanquickenborne, B., Dinsdale, D., Thoemer, M., Hoylaerts, M. F., Vanbilloen, H., Mortelsmans, L., and Nemery, B. (
Nemmar, A., Hoylaerts, M. F., Hoet, P. H. M., and Nemery, B. (
Perry, T. D., Duckworth, O. W., Kendall, T. A., Martin, S. T., and Mitchell, R. (
Pope, C. A. 3rd, Burnett, R. T., Thurston, G. D., Thun, M. J., Calle, E. E., Krewski, D., and Godleski, J. J. (
Ratner, B. D. (
Renwick, L. C., Brown, D., Clouter, A., and Donaldson, K. (
Rödelsperger, K., Podhorsky, S., Brückel, B., Dahmann, D., Hartfiel, G. D., and Woitowitz, H. J. (
Roelofs, F., and Vogelsberger, W. (
Sayes, C. M., Fortner, J. D., Guo, W., Lyon, D., Boyd, A. M., Ausman, K. D., Tao, Y. J., Sitharaman, B., Wilson, L. J., Hughes, J. B., et al. (
Schins, R. P. F., Duffin, R., Höhr, D., Knaapen, A. M., Shi, T., Weishaupt, C., Stone, V., Donaldson, K., and Borm, P. J. A. (
Schultz, D. A. (
Settle, F. (editor). (
Small, H. (
Tang, R., Orme, C. A., and Nancollas, G. H. (
Tinke, A. P., Vanhoutte, K., De Maesschalck, R., Verheyen, S., and De Winter, H. (
Tran, C. L., Jones, A. D., Miller, B. G., and Donaldson, K. (
Tribe, K., Meehan, E., and McConville, J. (
Tucker, C. J. (
Vertegel, A. A., Siegel, R. W., and Dordick, J. S. (
Vogelsberger, W., Mittelbach, T., Seidel, A. (
Wang, J., and Flanagan, D. R. (
Warheit, D. B. (
Warheit, D. B., Hartsky, M. A., McHugh, T. A., and Kellar, K. A. (
Weiner, B. B. (
Zhang, H., Gilbert, B., Huang, F., and Banfield, J. F. (
Zoitos, B. K., Meringo, A. D., Rouyer, E., Thelohan, S., Bauer, J., Law, B., Boymel, P. M., Olson, J. R., Christensen, V. R., Guldberg, M., et al. (
Author notes
*Centre of Expertise in Life Sciences, Zuyd University, Heerlen, the Netherlands; †Aerosil & Silanes, Degussa Corporation, Piscataway, New Jersey 08855, USA; ‡Toxicology and Environmental Research and Consulting, The Dow Chemical Company, Midland, Michigan 48674, USA; §Particle Engineering Research Center, University of Florida, Gainesville, Florida 32611, USA; ¶Institut für Toxikologie, Bayer HealthCare, Wuppertal, Germany; ‖ILSI Health and Environmental Sciences Institute, Washington, DC 20005, USA; |||Analytical Sciences, The Dow Chemical Company, Freeport, Texas 77541, USA; and ‖‖Analytical Sciences, The Dow Chemical Company, Midland, Michigan 48674, USA
Comments