The association between cardiovascular disease and exposure to airborne particulate matter (PM) is well established. Nanotechnology is already well-integrated into our daily lives, and appears to offer humankind novel benefits limited only by our creativity. While the scope of these benefits is essentially unrestricted, benefits that are specific to human health include: molecular imaging/diagnosis, drug delivery, anticancer therapy and gene therapy. However, if the unrestricted use of nanotechnology is rushed to societal integration, and the health effects of nanoparticles are not clearly identified, such technology runs the risk of causing unanticipated adverse health effects 1, and the true potential of the technology will never be fully realized 2. A fundamental first step in avoiding these possible outcomes is the realization that nanoparticles are not merely smaller variants of similarly composed fine particles. This is evidenced by the observation that nanoparticles are primarily manufactured to produce specific differences in surface reactivity, solubility or conductance.

Titanium dioxide ultrafine particles that fall within the nanoparticle size range (one dimension less than 100 nm) are commonly used as photocatalysts to clean air and water 3, as antibacterial agents on glass and steel 4, and as components of many cosmetics and sunscreens. Once in the lung, TiO₂ nanoparticles initiate pulmonary responses such as airway inflammation 5, alveolar macrophage recruitment 6, and the activation of various growth factors and chemokines 7. Ultrafine particles deposited in the lung may have the ability to translocate to systemic sites within 24 hours of deposition 8910. Similarly, within 24 hours of deposition a substantial portion of inhaled TiO₂ nanoparticles escape phagocytosis and enter the alveolar interstitium 11 and pulmonary capillaries 12.

While the pulmonary responses to nano-sized titania particles appear to be well-understood, the systemic microvascular effects of exposure to such particles are unknown. We have previously shown that pulmonary exposure to fine PM (PM_2.5, particles with a mean aerodynamic diameter of 0.1–2.5 μm) causes systemic microvascular dysfunction, most evident in the form of impaired or abolished endothelium-dependent arteriolar dilation 1314. These studies also provided evidence that this systemic effect is not necessarily the result of the inherent pulmonary toxicity of the particles, in that equivalent doses of titanium dioxide (TiO₂) and residual oil fly ash caused similar, dose-dependent degrees of microvascular dysfunction. While this indicates that larger TiO₂ particles elicit potent systemic microvascular effects, it cannot be used to conclude that ultrafine TiO₂ exposure will produce qualitatively or quantitatively similar biologic effects.

Therefore, the purpose of this study was to characterize systemic microvascular function after pulmonary exposure (via inhalation) to fine or ultrafine TiO₂ aerosols over a range of concentrations, and determine if ultrafine TiO₂ particles are inherently more toxic than larger, fine TiO₂ particles.

Table 1 lists the exposure parameters used for the current inhalation exposures, as well as the calculated and actual deposition values. The exposure parameters were used to calculate pulmonary deposition (using aerosol concentration, minute ventilation, estimated deposition fraction, and exposure duration). The actual pulmonary deposition is used to identify experimental groups in subsequent Tables and Figures. The primary count mode for the ultrafine TiO₂ aerosols was 100 nm, while it was 710 nm for the fine TiO₂ aerosols (Figure 2). When considering the whole size distribution, the former had a count geometric mean diameter of 138 nm and a geometric standard deviation of 2.2, and the latter had a count geometric mean diameter of 402 nm and a geometric standard deviation of 2.4.

The exposures selected for this study were exposures which do not alter bronchoalveolar lavage markers of pulmonary inflammation or lung damage. Consistent with the design, histopathology did not identify significant inflammation in lung sections. The principal histopathologic alterations in the lung consisted of particle accumulation within alveolar macrophages, the presence of anuclear macrophages, and an intimate association between particle-laden alveolar macrophages and the alveolar wall. Anuclear alveolar macrophages were not observed in any of the 21 sham-exposed rats. Anuclear alveolar macrophages were identified in 1 of 20 rats exposed to ultrafine TiO₂ (a rat in the 38 μg exposure group), and in 16 of the 29 rats exposed to fine TiO₂ (Figure 3A), including 1 of 6 rats in the 8 μg target exposure, 2 of 4 rats in the 20 μg target exposure, 5 of 6 rats in the 36 μg target exposure, 5 of 7 rats in the 67 μg target exposure, and 3 of 6 rats in the 90 μg target exposure. Because the macrophage cell membrane remains intact, anuclear macrophages are presumed to represent an apoptotic change.

In addition to the particle accumulation, macrophages containing particles were frequently intimately associated with the alveolar wall (Figures 3 and 4), a location, suggestive of, but not diagnostic of, macrophage activation. This change was difficult to quantify because it was dependent upon identifying the macrophages which were intimately associated with the alveolar wall. These macrophages were best demonstrated using cross-polarization to demonstrate the titanium particles (Figures 3B and 4B). Fine TiO₂ particles were distinctly birefringent in cross-polarized light and could be identified in both intracellular and extracellular locations, while ultrafine TiO₂ particles were faintly birefringent in polarized light and were only identified within macrophages (Figures 4B). Since the size of ultrafine TiO₂ particles should be below the size visible in the light microscope, the observed ultrafine TiO₂ particles are presumed to represent agglomerates of ultrafine particles. Consistent with the ultrafine particles only being visible when agglomerated, ultrafine particles were not observed in sham-exposed rats or rats receiving the lowest exposure, but were seen in 1 of 4 rats at 6 μg target exposure, 1 of 4 rats at the 10 μg target exposure, 2 of 4 rats at the 19 μg target exposure, and all rats receiving the 38 μg target exposure. Conversely, particles were seen in the lungs of all rats exposed to fine TiO₂ at any dose.

At the time of intravital experiments, rats in the 38 μg ultrafine TiO₂ group were slightly older than rats in all other experimental groups (Table 2). Despite this subtle age difference in one group, mean arterial pressure was not different among all experimental groups. Resting diameter, passive diameter and resting tone were not affected by aerosol dose, particle size or exposure method (Table 3).

Consistent with our previous findings in rats exposed to fine TiO₂ via intratracheal instillation 1314, inhalation of fine TiO₂ impaired arteriolar dilation in response to A23187 infusion in a dose-dependent manner (Figure 5). Although significantly compromised, arteriolar dilation was still present after exposure to as much as 90 μg of fine TiO₂. The no-effect dose was determined to be 8 μg. At this dose, arteriolar dilation at each ejection pressure was identical to that observed in the sham-control group.

Inhalation of ultrafine TiO₂ impaired arteriolar dilation in response to A23187 infusion in a dose-dependent manner (Figure 6). This was most evident at the 38 μg dose, in which not only was dilation completely abolished, but significant arteriolar constriction resulted during A23187 infusion. The no-effect dose was determined to be 4 μg, in which arteriolar responsiveness was not different from responses observed in sham-controls.

Because aerosol concentration and exposure time were manipulated to obtain an array of pulmonary depositions, the possibility existed that such subtle manipulations could themselves contribute to the resultant microvascular dysfunction. To address this possibility, additional exposures were performed in which exposure time (2–8 hrs) and aerosol concentrations (3–12 mg/m³) were manipulated to produce identical calculated pulmonary depositions. Three groups of rats displayed identical levels of microvascular dysfunction after exposure to 30 μg ultrafine TiO₂via different conditions (Figure 7). This indicates that manipulation of exposure time or aerosol concentration within the parameters used in the current study neither attenuates nor augments the resultant microvascular dysfunction (i.e., the response is dependent on the time × concentration product of exposure).

Because actual pulmonary mass deposition was determined, three pair-wise comparisons of arteriolar responsiveness can be made between rats exposed to fine and ultrafine TiO₂ (Figure 8). At a deposition of 8–10 μg, fine TiO₂ produced no significant microvascular effects, whereas ultrafine TiO₂ significantly impaired arteriolar dilations at 20 and 40 psi A23187 by 53% and 75%, respectively (Top Panel). At a deposition of 19–20 μg, fine TiO₂ showed a trend towards an impaired dilation, while ultrafine TiO₂ produced arteriolar constrictions that were significantly different from the responses in both the sham-control and fine TiO₂ groups (Middle Panel). At a deposition of 36–38 μg, effects similar to the 19–20 μg lung burden occurred but the intensity of arteriolar constriction after exposure to ultrafine TiO₂ was more pronounced (Bottom Panel). This suggests that at similar pulmonary burdens, ultrafine particles produce greater systemic microvascular dysfunction than fine particles.

This is the first study to directly identify an impaired vasodilator capacity in the systemic microcirculation after nanoparticle inhalation. A second novel observation in the current report is that for the same pulmonary mass deposition, nanoparticles produce significantly greater systemic microvascular dysfunction than their larger, fine counterparts of the same composition.

Nanoparticle inhalation attenuated systemic arteriolar dilation in a dose-dependent manner (Figure 6). This was most evident at a pulmonary deposition of 38 μg, in which arterioles constricted in response to intraluminal A23187 infusion. Arteriolar constriction during A23187 infusion is inconsistent with a healthy endothelium because under these conditions A23187 normally interacts only with endothelial cells to increase intracellular Ca²⁺ concentration and subsequently stimulate nitric oxide production 33. Altered endothelial integrity and/or permeability could allow luminal A23187 to interact with smooth muscle cells, thereby increasing intracellular Ca²⁺ concentration 34 and ultimately stimulate vasoconstriction 35. Alternatively, endothelial integrity may be unaltered, but some aspect of particle exposure offsets the prevailing balance of vasoconstrictors and vasodilators. Indeed, A23187 has the potential to stimulate endothelial production of vasoconstrictor prostanoids 3637. Future studies must determine if endothelial integrity has been compromised and/or the production of vasoconstrictor prostanoids has been altered by nanoparticle exposure. Of equal importance, future investigations must characterize the ability of extrapulmonary nanoparticles to alter the bioavailability of vasoactive metabolites.

The loss of microvascular vasodilator capacity can be a profound influence on normal homeostasis in any organ. In its most basic sense, this microvascular impairment would decrease tissue perfusion, and therefore, compromise function 38. Our current techniques and findings focus upon vasomotor function in single arterioles after pulmonary nanoparticle exposure. To fully appreciate the net effect of alterations in vascular regulation, total tissue or organ blood flow must be studied. These approaches can be challenging, and as such, other hemodynamic variables that relate to tissue or organ blood flow can offer considerable insight into the systemic microvascular consequences that follow particle exposure. Mills et al. 39 showed that diesel exhaust inhalation attenuates forearm blood flow responsiveness to vasodilators. Further, pulmonary exposure to diesel particles in the ultrafine range potentiates myocardial ischemia in patients with pre-existing coronary heart disease 40. Urch et al. 41 have reported acute increases in diastolic blood pressure within 2 hrs of particle exposure. Changes in diastolic blood pressure are achieved primarily by alterations within the resistance vasculature. Rundell et al. 42 reported a deficit in hemoglobin reoxygenation following arterial occlusion. Taken together, these findings are highly suggestive of larger disturbances in microvascular blood flow regulation after particle exposure.

Brook et al. 43 initially identified a subtle vasoconstriction of the brachial artery in humans that inhaled fine particle matter. Recently, Rundell et al. 42 reported that this conduit artery constricts similarly after ultrafine particle inhalation. Because the smaller particles used by Rundell et al. did not appear to be associated with a more robust vasoconstriction in these studies (than those used by Brook et al.), one might conclude that particle size does not dictate resultant vascular dysfunction. However, it is possible that this experimental model lacks sufficient sensitivity to reveal such differences, or that fundamental differences in particle composition, exposure protocols, and/or experimental group profiles prevent a meaningful comparison between the two studies.

Ultrafine TiO₂ has been shown to cause significantly greater airway inflammation than fine TiO₂ 5. Based upon the count geometric mean diameter (Figure 2), nano-sized titania aerosols attenuate systemic endothelium-dependent arteriolar dilation to a greater degree than fine TiO₂ aerosols (Figures 5, 6 and 8) at similar lung burdens (Table 1). This observation is most evident at lung burdens of 36 μg and 38 μg for fine and ultrafine particles, respectively. In this case, nano-sized titania aerosol exposure was consistently associated with arteriolar constriction, whereas fine TiO₂ inhalation was consistently associated with an impaired arteriolar dilation (Figure 8, bottom panel). This suggests that given the diverse nature of particle size, shape and chemistry, the deposited particle mass may not be the ideal dose metric.

Data presented in the current study indicate that on an equivalent mass basis, ultrafine TiO₂ was approximately one order of magnitude more potent than fine TiO₂ in causing systemic microvascular dysfunction. Oberdorster 44 has proposed that particle surface area may be the more appropriate dose metric for pulmonary effects of ultrafine particles than mass. Therefore, we also analyzed the results on an equivalent surface area of particles deposited in the lungs. BET analysis, as described by Brunauer, Emmett and Teller 45, was used to determine that the surface area of ultrafine TiO₂ was 48.08 m²/g while fine TiO₂ was 2.34 m²/g; i.e., ultrafine TiO₂ had approximately 20 times more surface area per unit mass than fine TiO₂. Therefore, if one normalized the systemic microvascular response to equivalent total particle surface area, the fine TiO₂ would be more potent than the ultrafine TiO₂. This conclusion is likely the result of an over estimation of the total ultrafine surface area delivered to the lungs since the BET method measures the gas absorptive surface area of the primary particles, rather than the actual physical surface of the aerosolized agglomerates. Indeed, Shvedova et al. 46 reported that the pulmonary response to ultrafine carbon black was significantly increased upon improved nanoparticle dispersion.

It is important to note that actual pulmonary particle deposition was significantly less than the calculated deposition (Table 1). This may be due to a number of reasons. First, it is very likely that particle clearance occurs in the 24 hrs after exposure, and particularly during the 720 minute exposures that took place over three days. Second, biological heterogeneity among rats may contribute to differences in the calculated vs. the actual particle depositions. In this case, subtle differences in minute volumes or deposition fractions among rats could contribute to the divergent measurements. However, despite manipulation of aerosol concentration (3–12 mg/m³) or exposure time (2–8 hrs), the intensity of microvascular dysfunction was not different among these groups 24 hrs after exposure (Figure 7). This suggests that the ranges of variables used to obtain our target pulmonary loads are not responsible for observed biologic effects in either the lung or the systemic microcirculation.

The opportunity to directly compare the microvascular responses to two different exposure methods presented itself in this study. When arteriolar responses between rats exposed to similar doses of fine TiO₂ (90 μg via inhalation vs. 100 μg via intratracheal instillation 14) are compared, no significant differences are apparent in the resultant microvascular dysfunction (data not shown). This suggests that the method of particle introduction to the lungs does not artificially induce systemic microvascular dysfunction.

The physiological relevance of intratracheal instillation is frequently questioned. Intratracheal instillation is a commonly used exposure method because it is rapid, economical and consistent across multiple animals. Further, intratracheal instillation can produce a relatively uniform pulmonary particle distribution if exposure doses are kept low 47. However, in regards to the true in vivo environment, intratracheal instillation bypasses many important physiological processes. Hence, the "gold standard" of particle exposure remains via inhalation.

In the time since we first reported that systemic microvascular dysfunction follows PM exposure 13, nanotechnology products have become considerably imbedded in most every aspect of daily life. It is also clearly apparent now that the benefits of nanotechnology are far reaching. Unfortunately, not only are the precise mechanisms through which inhaled particles exert systemic biologic effects unknown, but also the scope of biologic targets and/or end points is greater when nanoparticles are considered. By definition, this later observation may be due largely because of particle size, but surface chemistry of nanoparticles must also be taken into account in future studies. The ultrafine particles used in the current study represent manufactured particles in the "nano" size range, and after pulmonary deposition, they exerted a robust biologic effect. It is unclear at this time what the most appropriate dose metric is for nanoparticles, but it is apparent from our studies and many others that mass does not consistently appear to be the ideal metric. Independent of the appropriate dose metric is our consistent observation that pulmonary particle exposure initiates systemic microvascular dysfunction, and this observation is now extended to nanoparticles.

A23187 – Calcium ionophore. ADO – Adenosine. ANOVA – Analysis of variance. ICP-AES – Inductively-coupled plasma-atomic emission spectroscopy. D_c – Control diameter. D_pass – Passive diameter. D_ss – Steady state diameter. i.p. – Intraperitoneal. N – Number of animals. n – Number of arterioles. PM – Particulate matter. TiO₂ – Titanium dioxide

The author(s) declare that they have no competing interests.

TRN performed intravital microscopy experiments. DWP performed ICP-AES experiments. AFH performed pulmonary histology. JLC, BTC and DGF performed inhalation exposures. TRN and VC conceived and designed the study. All authors read and approved the final manuscript.

The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health.

Research described in this article was conducted under contract to the Health Effects Institute (HEI), an organization jointly funded by the United States Environmental Protection Agency (EPA) (Assistance Award No. R-82811201) and certain motor vehicle and engine manufacturers. The contents of this article do not necessarily reflect the views of HEI, or its sponsors, nor do they necessarily reflect the views and policies of the EPA or motor vehicle and engine manufacturers.

Health Effects Institute Award #4730 and NIEHS ES015022 (TRN).