Sage Journals: Discover world-class research

Abstract

The semiconductor manufacturing sector plans to introduce III/V film structures (eg, gallium arsenide (GaAs), indium arsenide (InAs) onto silicon wafers due to their high electron mobility and low power consumption. Aqueous solutions generated during chemical and mechanical planarization of silicon wafers can contain a mixture of metal oxide nanoparticles (NPs) and soluble indium, gallium, and arsenic. In this work, the cytotoxicity induced by Ga- and In-based NPs (GaAs, InAs, Ga₂O₃, In₂O₃) and soluble III-V salts on human bronchial epithelial cells (16HBE14o-) was evaluated using a cell impedance real-time cell analysis (RTCA) system. The RTCA system provided inhibition data at different concentrations for multiple time points, for example, GaAs (25 mg/L) caused 60% inhibition after 8 hours of exposure and 100% growth inhibition after 24 hours. Direct testing of As(III) and As(V) demonstrated significant cytotoxicity with 50% growth inhibition concentrations after 16-hour exposure (IC₅₀) of 2.4 and 4.5 mg/L, respectively. Cell signaling with rapid rise and decrease in signal was unique to arsenic cytotoxicity, a precursor of strong cytotoxicity over the longer term. In contrast with arsenic, soluble gallium(III) and indium(III) were less toxic. Whereas the oxide NPs caused low cytotoxicity, the arsenide compounds were highly inhibitory (IC₅₀ of GaAs and InAs = 6.2 and 68 mg/L, respectively). Dissolution experiments over 7 days revealed that arsenic was fully leached from GaAs NPs, whereas only 10% of the arsenic was leached out of InAs NPs. These results indicate that the cytotoxicity of GaAs and InAs NPs is largely due to the dissolution of toxic arsenic species.

Keywords

real-time cell analysis human bronchial epithelial cells GaAs InAs nanoparticles impedance

Introduction

Gallium arsenide (GaAs) and indium arsenide (InAs) materials are being considered for semiconductor manufacturing. In comparison to silicon, these materials have direct band gap, higher electron mobility, and semi-insulating properties which help to increase electrical performance and decrease power consumption in semiconductor devices.^1

-4

Although these III/V materials have advantages, there are occupational health and environmental safety concerns due to the potential release of hazardous waste materials during wafer production via the chemical and mechanical planarization (CMP) process. The CMP uses pristine slurries such as silica (SiO₂), alumina (Al₂O₃), or ceria (CeO₂) nanoparticles (NPs) to produce smooth material surfaces.⁵ Polishing of III/V thin could generate CMP waste containing nano-sized GaAs and InAs, as well as potential by-products such as gallium and indium oxides (Ga₂O₃, In₂O₃) that are formed due to corrosion of the respective arsenides.^5,6 The CMP-derived waste streams can also contain dissolved III/V species (eg, gallium(III), indium(III), arsenic(III), and arsenic(V)) which are released into the aqueous environment during the planarization process.^5

-9 The combination of these materials in the waste stream may cause potential environmental, safety, and health problems that need to be carefully considered. In particular, there is a concern about the health risks and environmental impacts of dissolved arsenic (As), a well-known carcinogen and highly toxic metalloid,^10,11 and As-containing materials. Relatively high concentrations of As have been found in effluent streams of electronic fabrication facilities that utilize III/V materials in several countries (35-2,000 mg/L).^9,12,13 High levels of arsenide particulates are also expected in CMP waste streams,^14,15 but the concentrations found in these effluents have not been reported to date. The CMP effluents are treated at semiconductor facilities to remove regulated pollutants (eg, copper, soluble As) prior to discharge to municipal sewer systems, but current treatment systems are not designed to remove nano-sized Ga- and In arsenides or other Ga- and In-containing NPs. The limit of As in effluents from electronic manufacturing in the United States is 2.09 mg/L as the 1-day average maximum and 0.83 mg/L as the average of daily values for 30-consecutive-day maximum.¹⁶

Arsenic is well known for its toxicity and carcinogenicity¹¹ and thus is highly regulated. Gallium arsenide has also been classified as a group I carcinogen to humans.¹⁷ Exposure to As can cause different effects such as gastrointestinal distress, death, or major organ damages. There is a long history of adverse health effects by As that can target multiple organs, including the lung.¹⁸ We have shown that As compromises basic lung epithelial cell signaling and barrier function.^19

-22 The most common route for As-induced lung disease is via contaminated drinking water.^23
-25 The World Health Organization and the United States Environmental Protection Agency have established an acceptable limit of As concentration in drinking at 10 μg/L.^26,27 Since As has toxicity and carcinogenicity implications, implementation of NPs containing As in semiconductor industry can potentially cause toxic effects via different exposure routes.

In toxicity assessments, conventional in vitro cytotoxicity assays measuring fluorescence, luminescence, or absorbance are alternatives to animal tests,²⁸ but NPs can interfere with these optical measurements.^29,30 Impedance-based assays, such as the xCELLigence real-time cell analysis (RTCA) system, allow for cytotoxicity and cell viability assessment that is label-free, dynamic, and can facilitate high-throughput studies.^30

-33 The measured impedance is affected by the biological status of the cells interacting with the surface of wells in which the culture cells are growing.^30,34 There have been studies applying this system to investigate the cytotoxicity of As(V), As(III), Hg, sodium dichromate, and inorganic NPs with reliable and highly sensitive results.^34
-36

The objective of this study is to investigate the cytotoxic effects of gallium- and indium-based NPs: GaAs, InAs, Ga₂O₃, and In₂O₃ toward human bronchial epithelial cells (16HBE14o-) using impedance-based RTCA assays. We demonstrated that Ga- and In-based NPs caused cytotoxicity in 16HBE14o- cells and that this cytotoxicity could be due to corrosion of the NPs and subsequent dissolution of toxic As species.

Materials and Methods

Chemicals

Gallium(III) chloride (GaCl₃, CAS# 13450-90-3, >99.99%), sodium meta-arsenite (NaAsO₂, CAS# 7784-46-5, ≥99%), sodium arsenate dibasic heptahydrate (Na₂HAsO₄·7H₂O, CAS# 10048-95-0, ≥98%), and citric acid (C₆H₈O₇, CAS# 77–92-9) were purchased from Fisher Scientific (Pittsburgh, Pennsylvania). Indium (III) chloride tetrahydrate (InCl₃·4H₂O, CAS# 22519-64-8, >99.99%) was obtained from Strem Chemicals (Newburyport, Massachusetts). Due to their low solubility in the neutral pH range,³⁷ the stock solutions of indium(III) and gallium(III) were supplied with citrate at a molar ratio 1:3.75, metal:citrate to prevent any precipitation in the circumneutral bioassay media. Citrate, a metabolite of living cells, is a ubiquitous compound found in the natural environment.³⁸ This compound is also used as a chelating agent in metal processing industries like semiconductor manufacturing.³⁹ Therefore, it is relevant to couple citrate with In(III) and Ga(III) in this study.

Gallium oxide (Ga₂O₃) nanopowder (CAS# 12024-21-4, >99.9%) was obtained from American Elements (Los Angeles, California). Indium oxide (In₂O₃) NPs (CAS# 1312-43-2, 99.995%) were purchased from SkySpring Nanomaterials, Inc. (Houston, Texas). Gallium arsenide nanopowder (GaAs) (CAS# 1303-00-0, 99.99%) and InAs nanopowder (CAS# 106097-59-0, 99.99%) were obtained from Nanoshel LLC (Wilmington, Delaware).

Particle Size Distribution and ζ Potential

The average hydrodynamic particle sizes and the particle size distribution of Ga₂O₃ and In₂O₃ NPs were measured by dynamic light scattering (DLS) using a Zetasizer Nano ZS (Malvern Instruments, Westborough, Massachusetts) with a laser wavelength of 633 nm and a scattering angle of 173°. The refractive index of the Ga₂O₃ and In₂O₃ NPs was 1.92 and 1.86, respectively. The NP concentration in the samples analyzed by DLS was 100 mg/L.

The ζ potential of the NP dispersions was also determined using the Zetasizer Nano ZS instrument. The data were determined using the Smoluchowski equation that correlates the electrophoretic mobility of particles with their ζ potential value.

Corrosion and Dissolution of III-V Particles

The dispersions of NP (500 mg/L) were prepared in aqueous medium containing 1 g/L of NaHCO₃. The pH of the dispersions was maintained at 7.85 throughout the process by supplying 1% CO₂ gas (vol/vol, based on the volume of headspace of the serum flasks) after liquid samples were collected. The flasks were shaken in an incubator at 110 rpm at 30°C. The corrosion and dissolution experiments were carried out for 7 days. Liquid samples were collected daily for analysis.

Immortalized Human Bronchial Epithelial Cell Culture

In this research, 16HBE14o- cells were obtained from California Pacific Medical Center Research Institute (San Francisco, California). 16HBE14o-, an adherent and immortalized human bronchial epithelial cell line, is often used as a model cell type to study pulmonary adsorption, transport, and permeability of environmental contaminants.^20,40,41 The cells were initially grown as previously described.^34,42 Firstly, cells were grown in tissue culture flasks coated with a collagen/fibronectin/bovine serum albumin (CFB) matrix in a controlled growth medium that contains minimum essential medium (MEM) supplemented with 10% (vol/vol) fetal bovine serum (FBS), 2 mM GlutaMAX, penicillin, and streptomycin at 37°C in a 5% CO₂ atmosphere. Next, the cells were transferred to RTCA assay plates coated with CFB and maintained with a reduced serum (5% FBS) medium.

Real-Time Cell Analysis Assay

The xCELLigence real-time cell analysis (RTCA) instrument (ACEA Biosciences, San Diego, California) allows for dynamic monitoring cytotoxicity and cell proliferation based upon impedance measurements. The RTCA instrument and the 96-well microelectronic plates (also known as E-plates 96; ACEA Biosciences) are used in this bioassay. When cells attach to the bottom of each well, they create impedance that can be detected by interdigitated gold microelectrodes integrated on the bottom of the testing place, thereby allowing measurement of the cellular status in real time.⁴³ The measured impedance is calculated and plotted as cell index (CI), which is proportional to cell biological statuses as cell number, morphology, and adhesion. The instrument is placed in a standard CO₂ cell culture incubator and interfaces via a cable with analysis and control units which are located outside the incubator.

In the RTCA assay, 150 µL of 16HBE14o- cells were plated onto 96-well E-plates at a cell density of approximately 100 000 cells/well as described above and then incubated overnight at 37°C and 5% CO₂ to verify proper attachment and cell growth. During this growth period, the impedance was monitored every 15 minutes. After the cells were plated with only medium present, a low AC voltage of 10 mV was applied, producing a small electric field between the microelectrodes that could be impeded by the presence of adherent cells and lead to the changes in measured impedance.³² As the number of cells increased over time, there would be significant changes in impedance.³² There is a proportional relationship between the impedance change and the number of cells inside each well which can provide information about the morphological and adhesive characteristics of the cells.³² After cell stabilization (15-18 hours), 50 µL of soluble species or NP stock solutions were added to each well. These stock solutions were prepared in culture medium a few minutes before addition. The impedance values were measured and recorded every 15 minutes for 24 hours. Assays were performed in quadruplicate and appropriate negative controls (eg, without soluble species or NPs) were run in parallel. Cell-free controls amended with GaAs, InAs, Ga₂O₃, or In₂O₃ NPs (300 mg/L) were tested and did not show any increase in impedance over 24 hours compared to noninoculated assays lacking NP addition.

Relative impedance is expressed as CI and presented after normalization (normalized cell index [NCI]) and was calculated for data analysis using the following equation:

{NCI}_{t} = \frac{{CI}_{t}}{{CI}_{0}},

where CI_t is the CI at any time t and CI₀ is the CI at the time of toxicant dosing (where NCI is equal to 1). The percentage inhibition was then calculated based on NCI values using the following equation:

Inhibition (%) = \frac{NCI (control) - NCI (sample)}{NCI (control) - 1} \times 100 .

When the NCI values of the samples were less than 1, inhibition values were reported as 100%. When the slope was horizontal or negative, the activity was assumed to be 0 and the inhibition value was reported as 100%. The NCI values from the 16-hour exposure time were used to develop concentration–response curves.

Uptake of NPs by 16HBE14o- Cells

Approximately 550,000 16HBE14o- cells in MEM supplemented with 5% FBS, 2 mM GlutaMAX and penicillin/streptomycin were plated onto 6-well plates coated with CFB and incubated for 24 hours. The cells were then exposed to 250 mg/L GaAs NPs for 24 hours. Finally, the cells were prepared for transmission electron microscopy (TEM) measurements after washing by Hank’s balanced salt solution. Cells were viewed in a Tecnai Spirit Bio Twin (FEI America, Hillsboro, Oregon) operated at 100 kV. Eight bit tilt images were captured via an AMT 4Mpix digital camera. Measurements were taken at eucentric height using the FEI TIA software (version 4.0).

Analytical Methods

The concentration of dissolved As, Ga, and In in liquid samples was determined by inductively coupled plasma optical emission spectrometry (ICP-OES model 5100; Agilent Technologies, Santa Clara, California) at a wavelength of 188.980 nm (for As), 294.363 nm (for Ga), and 230.606 nm (for In). The limit of detection for all 3 metals was 50 μg/L. Prior to ICP analysis, liquid samples were centrifuged (12,000 rpm for 10 minutes) and filtered (through 25-nm membrane filter). The filtered samples were then diluted to proper concentrations with 2% HNO₃ for analysis. Arsenic speciation (As(III) and As(V)) was measured using high-performance liquid chromatography (HPLC) for pretreatment and species separation followed by inductively coupled plasma mass spectrometer (ICP-MS; Agilent 7500a) for detection. The system is an Agilent 1100 HPLC (Agilent Technologies, Inc) with a reverse-phase C18 column (Prodigy 3u ODS(3), 150 × 4.60 mm; Phenomenex, Torrance, California) maintained at 50°C. The mobile phase (pH 5.85) contained 4.7 mM tetrabutylammonium hydroxide, 2 mM malonic acid, and 4% (vol/vol) methanol at a flow rate of 1.2 mL/min. The detection limit for this instrument was 0.1 µg/L for As speciation and 0.01 µg/L for As totals. All other analyses, including pH, were performed according to standard methods.

Results and Discussion

Physiochemical Characterization of NPs

Figure 1 shows the TEMs of the different nanomaterials investigated in this study. The micrographs show that the particle morphology of the nanomaterials is diverse and includes rod-like, nearly spherical, and amorphous particles. The average particle size values determined by TEM are reported by manufacturers as 20 to 70 nm (In₂O₃ NPs), 500 nm (Ga₂O₃ NPs), 80 to 100 nm (InAs NPs), and 80 to 100 nm (GaAs NPs).

Figure 1.

Transmission electron microscope (TEM) images of Ga₂O₃ (A), In₂O₃ (B), gallium arsenide (GaAs) (C), and indium arsenide (InAs) (D) nanoparticles provided by the respective nanomaterial manufacturers.

Characterization of NPs in the medium of study is important, since the formation of large aggregates may influence and change their potential toxicity. Dynamic light scattering was utilized to evaluate the particle size distribution in bioassay medium. The average hydrodynamic diameter includes the particle with water or other molecules adsorbed on its surface, and therefore, it is generally higher than the primary size determined for the dry particles by electron microscopy. The results show that the average hydrodynamic particle sizes of the NPs in bioassay medium were much higher than their primary sizes. The average particle size determined for Ga₂O₃ and In₂O₃ ranged from 626 to 654 nm. Additionally, the ζ potential of the Ga₂O₃ and In₂O₃ dispersions was low, averaging 15.1 and 1.3 mV, respectively. Colloidal dispersions exhibiting ζ potential values higher than +30 mV or lower than −30 mV are considered stable, while particles with low surface charge tend to aggregate due to limited electrostatic repulsion. Since the isoelectric point of Ga₂O₃ and In₂O₃ NPs dispersions in pure water (8.4-9.2)^44,45 is relatively close to the pH of the bioassay media (ca. 7.0), the observed particle aggregation in the bioassay medium is not surprising. In addition, the high ionic strength of the culture media, which contains high concentrations of NaHCO₃ (2.2 g/L), NaCl (6.8 g/L), and other salts, is also expected to contribute to compress the electrical double layer and, consequently, facilitate particle agglomeration. It is well known that many NP dispersions are destabilized in complex aqueous media⁴⁶ containing high levels of salts and/or organic compounds,¹² resulting in aggregates that have effective sizes that are orders of magnitude higher compared to that of the primary particles. The arsenide NP dispersions were highly unstable in culture medium and formed large aggregates that settled readily and could not be assessed for particle size distribution and ζ potential using the Zetasizer Nano ZS instrument utilized in this study to characterize the NP dispersions.

Cytotoxicity of Ga- and In-Based NPs

The CMP process uses pristine slurries such as silica (SiO₂), alumina (Al₂O₃), or ceria (CeO₂) NPs to produce smooth material surfaces.⁵ When new materials are introduced onto the wafer, CMP could generate waste containing GaAs and InAs NPs as well as potential by-products such as Ga₂O₃ and In₂O₃ NPs. We first evaluated the potential cytotoxicity of GaAs and InAs NPs to human bronchial epithelial cells using the xCELLigence RTCA system. 16HBE14o- cells were exposed to GaAs and InAs NPs at concentrations ranging from 0 to 250 mg/L and the NCI was recorded for 24 hours (Figure 2A and B). The GaAs NPs caused strong inhibition to the target cells. Increasing concentration of GaAs NPs resulted in a high initial increase in NCI within the first 4 hours of exposure, followed by a loss of NCI over the 24-hour period (Figure 2A). At the highest tested concentration of GaAs NPs (125 mg/L), the initial NCI increase was rapid, peaking near 2 hours, followed by a drastic decrease in NCI values until there was no impedance within 10 hours of exposure. These changes are consistent with GaAs-induced cellular signaling (ie, increased NCI) followed by cytotoxicity (ie, loss of NCI). The RTCA measurements following InAs NPs resulted in a similar concentration-dependent short-term signaling and long-term cytotoxicity, albeit with a significantly reduced response (Figure 2B). A 16-hour exposure time point was chosen to compare toxicity and develop a concentration response (Figure 2C and D). With GaAs NPs, a growth inhibition of 95% was observed at 25 mg/L and 100% inhibition at 125 mg/L. The 50% growth inhibition concentrations, IC₅₀ value, after 16-hour exposure was 6.2 mg/L (Table 1). For InAs NPs, a concentration of 125 mg/L resulted in ca. 60% inhibition, and full inhibition was only reached at the highest concentration tested (250 mg/L). The calculated 16-hour IC₅₀ for InAs NPs (68.0 mg/L; Table 1) was 10-fold higher than that observed for GaAs NPs. Similar cytotoxicity toward BEAS-2B cells (an alternative human bronchial cell line) has been reported for GaAs NPs (<10 nm) using single time point (24 hours) fluorescent assay (IC₅₀ = 12.5 mg/L).¹⁵ In the same study, InAs NPs (20-200 nm) displayed limited cytotoxicity toward BEAS-2B cells (<10% cell death at 98.7 mg/L).¹⁵ The use of the RTCA in the current experiments allowed for an examination of physiological changes (including cytotoxicity as an end point) over time. The RTCA system provided the percentage inhibition toward the 16HBE14o- cells at different concentrations at multiple time points such as after 8, 16, and 24 hours of exposure (Figure 3A and B). This is a significant difference from previous work in which cytotoxicity was reported at a single time point.¹⁵ Such analysis shows that early (ie, 1-4 hours) signaling, as noted in the rapid increase in NCI, is predictive of a later cytotoxicity response, recorded here as a loss of NCI and previously in single time point fluorescence assays.¹⁵

Figure 2.

Cytotoxicity response of human bronchial epithelial cells (16HBE14o-) exposed to increasing concentrations (mg/L) of different nanoparticles (NPs) as a function of time: gallium arsenide (GaAs) NPs (A) and indium arsenide (InAs) NPs (B). Lines represent average values of 4 replicates. Inhibitory effects (%) of GaAs NPs (C) and InAs NPs (D) on human bronchial epithelial cells after 16 hours of exposure as a function of concentration.

Figure 3.

Inhibitory effects (%) of gallium arsenide (GaAs) nanoparticles (NPs), indium arsenide (InAs) NPs, Ga₂O₃ NPs, and In₂O₃ NPs on human bronchial epithelial cells at different concentrations (mg/L) as a function of time. The label shown next to each curve indicate the nanoparticle concentration in the bioassays (in mg/L).

Table 1.

Concentrations of Soluble and Nano-Sized III/V Chemicals Tested Causing 50% Inhibition (IC₅₀) After 16 Hours of Exposure in the Impedance-Based RTCA Bioassay.

Tested chemical	IC₅₀ (mg/L)
Soluble species
As(III)	2.7
As(V)	4.5
Ga(III)^a	155.0
In(III)^a	ND^b
Nanomaterials
GaAs	6.2
InAs	68.0
Ga₂O₃	260.0
In₂O₃	ND^c

Abbreviations: ND, not detected; RTCA, real-time cell analysis.

^a Ga(III) and In(III) were complexed with citrate at molar ratio as 1:1.31 (Ga:Cit) and 1:3.75 (In:Cit).

^b The highest In(III) concentration tested (30 mg/L) only caused 30% inhibition.

^c The highest In₂O₃ concentration tested (300 mg/L) caused 30% inhibition.

In addition to the arsenide NPs derived from the III-V materials, Ga₂O₃ and In₂O₃ NPs can also be formed as potential byproducts of the CMP process of the semiconductor fabrication plant.^5,6 Both Ga₂O₃ and In₂O₃ NPs were found to be relatively nontoxic to the 16HBE14o- target cells. Unlike the As-containing NPs above, the cytotoxicity traces for Ga₂O₃ and In₂O₃ NPs showed significant decreases only at the highest concentrations tested (30 and 300 mg/L for Ga₂O₃ NPs and 300 mg/L for In₂O₃ NPs; Figure 4A and B). Similarly, at a concentration of 300 mg/L, Ga₂O₃ and In₂O₃ NPs only caused 56% and 30% growth inhibition, respectively, after 16 hours of exposure. Data in this current study are in agreement with single time point (24 hours) viability assays (MTT assay) where In₂O₃ (>200 nm) has been shown to have minimal effects on BEAS-2B cells at 50 mg/L and 50% reduction in viability at 1,000 mg/L.⁴⁷ A direct comparison of the Ga₂O₃ and In₂O₃ NPs RTCA traces with the GaAs and InAs NPs RTCA traces shows a loss of early signaling upon addition of the oxidized compounds and a distinct loss of cytotoxicity.

Figure 4.

Cytotoxicity response of human bronchial epithelial cells (16HBE14o-) exposed to increasing concentrations (mg/L) of different NPs as a function of time: Ga₂O₃ nanoparticles (NPs) (A) and In₂O₃ NPs (B). Lines represent average values of 4 replicates. Inhibitory effects (%) of Ga₂O₃ NPs (C) and In₂O₃ NPs (D) on human bronchial epithelial cells after 16 hours of exposure as a function of concentration.

Corrosion and Dissolution of GaAs and InAs NPs

Both GaAs and InAs particles have been shown previously to be susceptible to corrosion in aqueous medium with concomitant release of soluble As species.⁴⁸ Therefore, the heightened signaling and subsequent cytotoxicity observed in GaAs and InAs NPs compared to Ga₂O₃ and In₂O₃ NPs reported above could be caused by the NPs themselves or by toxic species released into solution due to NP corrosion. The concentrations of soluble As, Ga, and In detected in the medium at the end of the cytotoxicity assays (ie, 24 hours after NP addition) are listed in Table 2. The results indicate that GaAs NPs underwent rapid dissolution in the bioassay medium, leading to the release of high concentrations of toxic soluble As. Soluble As was also detected in the samples originally spiked with InAs NPs, albeit at much lower concentrations. Arsenic speciation analysis revealed that As(III) was the dominant As species, accounting for 81.2% to 93.3% of the total concentration of soluble As, depending on the assay.

Table 2.

Soluble As, Ga, and In Detected in the RTCA Cytotoxicity Bioassay Medium at the End of the Incubation (24 Hours) in Assays Spiked With Different Concentrations of InAs and GaAs.^a

Nanomaterial	As (mg/L)	Ga^b (mg/L)	In^c (mg/L)	As (%)	Ga (%)	In (%)
GaAs NPs
25 mg/L	11.8	0.062	ND	91.20	0.05	-
125 mg/L	48.8	0.273	ND	75.41	0.23	-
250 mg/L	83.5	0.648	ND	64.52	0.54	-
InAs NPs
25 mg/L	2.6	ND	ND	26.2	-	ND
125 mg/L	3.2	ND	ND	6.39	-	ND
250 mg/L	6.7	ND	ND	6.80	-	ND

Abbreviations: GaAs, gallium arsenide; InAs, indium arsenide; ND, not detected; RTCA, real-time cell analysis.

^a The As, Ga and In detected is expressed as concentration (mg/L) and percentage (based on the total element concentration present in the nanoparticles at time = 0). The limits of detection for Ga and In in the inductively coupled plasma optical emission spectrometry system were 30 and 20 μg/L, respectively.

Dissolution experiments were performed with GaAs and InAs NPs over 7 days to better characterize the potential release of soluble As in aqueous medium under circumneutral pH conditions commonly found in natural waters (Figure 5). In aqueous media, the arsenide NPs underwent significant dissolution. Of the initial 500 mg/L of GaAs NPs (80-100 nm) incubated at the start of the experiment, almost all of the As were leached out (259 mg As/l, corresponding to 100%) after 7 days. Speciation results showed that 50% of total measured As was As(III) and 11% was As(V). Under similar testing conditions, only 20 mg As/l (≈10% of the As) was leached out of InAs NPs. Speciation results of InAs NPs study indicated that about 68% of total measured As was As(III) and 7% was As(V). The results of dissolution and speciation studies indicate that the potential cytotoxic effects of GaAs NPs and InAs NPs can come from the toxic soluble As species. In the current study, the concentration of soluble Ga and In detected both in the cytotoxicity bioassays (Table 2) and the corrosion tests (Figure 5) was very low. For example, the dissolved concentrations of Ga and In recorded in the corrosion tests on day 7 were only 0.03 and 12.3 mg/L, corresponding for 0.01% and 4.1% of the respective metal initially present in the GaAs and InAs NPs. Higher concentrations of soluble Ga(III) and In(III) would be expected based on the concentrations of dissolved As determined. In the corrosion tests (day 7), the molar ratio of dissolved As:Ga was over 8,000 and the molar As:In ratio was 2.5. These ratio values were higher than the theoretical ratio of 1:1 in the solid phase NPs, suggesting that a large fraction of Ga and In released by corrosion had been oxidized to insoluble products. It is well known that Ga(III) and In(III) are sparingly soluble (0.01-1.0 mg/L of soluble Ga, <0.1 mg/L of soluble In) in water at pH values ranging 6.0 to 8.0 and ambient temperature.³⁷ In addition, the 16-hour IC₅₀ values of As(III) and As(V) at 2.7 and 4.5 mg/L (Table 1), respectively, were very similar to that of GaAs NPs at 6.2 mg/L. When combining the results of corrosion and dissolution experiments of GaAs and InAs NPs with the IC₅₀ values of the cytotoxicity tests, it is evident that the release of toxic soluble As species significantly contributes to the observed cytotoxic effects to human bronchial epithelial cells.

Figure 5.

Time course of dissolution of gallium arsenide (GaAs) nanoparticles (NPs) (80-100 nm) (▪) and indium arsenide (InAs) NPs (80-100 nm) (•) in 1 g/L NaHCO₃ at pH 7.8. The initial NP concentration was 500 mg/L.

Cytotoxicity of Soluble As, In, and Ga Species

It has been suggested that the toxicity of GaAs NPs, and to a lesser extent that of InAs NPs, is due to the As that dissociates in solution.¹⁵ To test for such a speciation-driven toxicity, we performed RTCA toxicity assays with 16HBE14o- cells and As(III), As(V), and the relatively soluble Ga(III)-citrate and In(III)-citrate. As(III) displayed a distinct pattern of cytotoxicity over the 24-hour experiment. At the highest As(III) concentration tested (5 mg/L), an RTCA pattern similar to cytotoxic concentrations of GaAs NPs was observed; there was a rapid development of increasing NCI (1-3 hours) followed by a dramatic decrease in NCI that continued throughout the 24-hour experiment (Figure 6A). The cytotoxic effect of As(III) was concentration dependent and concentrations as low as 1 mg/L caused cytotoxicity (Figure 6A). The IC₅₀ value after 16 hours of exposure (2.7 mg/L) was even lower than that observed for GaAs NPs (6.2 mg/L; Table 1). Additionally, data in this current study indicated that as the exposure time increased, the inhibitory effect increased (Figure 7). It is evident that even at low concentrations, As(III) is significantly cytotoxic.

Figure 6.

Cytotoxicity response of human bronchial epithelial cells (16HBE14o-) exposed to increasing concentrations (mg/L) of As(III) (A) and As(V) (B) as a function of time. Lines represent average values of 4 replicates. Inhibitory effects (%) of As(III) (C) and As(V) (D) on human bronchial epithelial cells after 16 hours of exposure as a function of concentration.

Figure 7.

Cytotoxicity response of human bronchial epithelial cells (16HBE14o-) exposed to increasing concentrations (mg/L) of Ga(III)-citrate (A) and In(III)-citrate (B) as a function of time. Lines represent average values of 4 replicates. Inhibitory effects (%) of Ga(III)-citrate (C) and In(III)-citrate (D) on human bronchial epithelial cells after 16 hours of exposure as a function of concentration.

Next, the cytotoxicity of As(V) was tested. At the highest tested concentration of As(V) (30 mg/L), the NCI initially increased in values during the first 3 hours, followed by a dramatic decrease in the NCI and eventual loss of signal (Figure 6B). Also similar to As(III) traces, lower concentrations of As(V) resulted in reduced cellular signaling and a reduction in overall impedance loss, indicative of less cytotoxicity. The dose-related inhibition after 16-hour exposure is shown in Figure 6D in which 100% growth inhibition was observed at 30 mg/L. The IC₅₀ value after 16-hour exposure was 4.5 mg/L (Table 1). Similarly, evaluation of the cytotoxicity caused by different As(V) concentrations after 8, 16, and 24 hours of exposure indicated that the inhibitory impact increased with increasing exposure time (Figure 7B).

The evidence suggests that As(III) has a major impact on human bronchial epithelial cells. Considering various human cell lines, the IC₅₀ of As(III) in the literature ranges from 0.14 to 12.2 mg/L.^15,25,49 The distinctive pattern of the rising and decreasing NCI was reported once before and discussed by Xing et al with NIH 3T3 cells (a standard fibroblast cell line) treated with As(III).²⁸ In their study, it was observed that NIH 3T3 cells fused together and formed larger multinuclear cell bodies during the first 3 hours of exposure to As(III).²⁸ However, after 8 hours of exposure to As(III), the multinuclear cells began to dissociate into smaller cells, which led to the decrease in NCI.²⁸ This phenomenon of the NCI rising and decreasing is unique to As since it does not occur with other metals.²⁸ In this current study, As(III) did not lead to any cell detachment or cell death at lower tested concentrations that is consistent with the findings of Xing et al²⁸ who did not observe an early increase in the NCI at a lower As(III) dose (0.094 mg/L). Even though the cell line in this study is different from the other studies, the results are within the range of reported values and prove that As(III) can lead to acute toxic effects to human lung cells at relatively low concentrations. The other As species behaved similarly.

Results in the current study and previous work showed that As(V) is cytotoxic to human cells. Considering different human cell lines, the 24-hour IC₅₀ values reported for As(V) in the literature range from 3.8 to 97.6 mg/L.^15,50,51 The observed pattern of the rising and decreasing NCI when human lung cells were exposed to As(V) was also reported before by Xiao et al and Moe et al with the fibroblastic V79 cell line and the human lung carcinoma cell line A549 exposed to As(V), respectively.^50,51 In both of these earlier studies, it was observed that at the highest tested concentrations of 1.4 and 279 mg As(V)/l, the NCI values increased during the first 5 to 6 hours, then slowly decreased below the NCI values of the control group.^50,51 Similar to As(III), this phenomenon of the NCI rising and decreasing is unique to As since it does not occur with other metals.²⁸ Compared to As(III), in this current study, As(V) did not lead to any cell detachment or cell death at the lowest tested concentrations. This finding is consistent with the previous findings of Xiao et al and Moe et al who did not observe the early increase in the NCI at low As(V) doses (3-7 mg/L).^50,51 The results in this current study are within the previously reported IC₅₀ range in the literature, suggesting that As(V) significantly affects human epithelial lung cells.

In addition to soluble As species, the potential cytotoxicity of soluble Ga(III)-citrate and In(III)-citrate to 16HBE14o- cells was studied. Both Ga(III)-citrate and In(III)-citrate were distinctly less toxic than the inorganic As species (Figure 7A and B). The NCI values from samples exposed to Ga(III)-citrate were only slightly lower than that of control group at 250 mg/L, while at lower concentrations, no significant decrease was observed (Figure 7A). In Figure 7B, approximately 60% inhibition was recorded at this highest tested Ga(III)-citrate concentration (250 mg/L). Additionally, at 3 different time points of exposure (8, 16, and 24 hours), Ga(III)-citrate was moderately cytotoxic at the concentrations of 250 and 500 mg/L (Figure 8C). The IC₅₀ value after 16-hour exposure was 155 mg/L of Ga(III)-citrate (Table 1). Like Ga(III)-citrate, In(III)-citrate was not very toxic as evidenced by the NCI values from samples exposed to different concentrations of In(III)-citrate being were very close to that of the control after 24-hour exposure (Figure 7B). A similar trend of noncytotoxicity of Ga(III)-citrate and In(III)-citrate was reported in previous studies using different human cell lines.^15,52 The results in this current study demonstrated that Ga(III)-citrate and In(III)-citrate were less cytotoxic to target cells compared to soluble As species.

Figure 8.

Inhibitory effects (%) of As(III) (A), As(V) (B), Ga(III)-citrate (C), and In(III)-citrate (D) on human bronchial epithelial cells at different concentrations (mg/L) as a function of time. Panel A: (▬) 0.03, (*) 0.1, (•) 0.3, (▴) 1.0, (▪) 3.0, and (♦) 5.0. Panel B: (×) 0.1, (+) 0.3, (▬) 0.6, (*) 1.0, (ˆ) 3.0, (▵) 6.0, (□) 10.0, and (◊) 30.0. Panel C: (+) 2.5, (*) 5.0, (•) 25.0, (▴) 50.0, (▪) 250.0, and (♦) 500.0. Panel D: (*) 1.0, (ˆ) 3.0, (▵) 10.0, (□) 30.0, and (◊) 100.0.

Nanoparticle Uptake by 16HBE14o- Cells

Due to their small size, NPs more likely enter mammalian cells and can potentially contribute to toxic effects compared to larger size particles with the same chemical composition. In this work, GaAs NPs were shown to be taken up by human bronchial epithelial cells which can contribute to the potential cytotoxic effects. The GaAs NPs were taken up and accumulated in Golgi body of the cells (Figure 9), which could affect its biological function and eventually affect the overall activity of human lung cells. Additionally, the potential release of As(III) from GaAs NPs could increase the inhibitory effects. The finding in this study agrees with one previous study in which GaAs NPs were taken up into endosomal compartments of the BEAS-2B cells.¹⁵ The researchers of that previous study also reported that the GaAs NPs likely resided in lysosomes, and because of the low pH (4.5) inside lysosomes, they observed an increase in GaAs particle dissolution (compared to neutral pH).¹⁵

Figure 9.

Transmission electron microscopy images of gallium arsenide (GaAs) nanoparticles (NPs) in 16HBE14o- human bronchial epithelial cells. Panel A shows that GaAs NPs were accumulated in the Golgi body of the cells. Panel B shows a close-up image of the GaAs particles (average diameter 25 nm) inside the Golgi body.

Potential Mechanisms of Species Cytotoxicity

Corrosion and speciation studies suggest that the species primarily responsible for the cytotoxicity of GaAs and InAs NPs is As(III). The literature provides several plausible mechanisms for As(III) toxicity. The toxic effect of As(III) has been attributed to oxidative stress due to the generation of reactive oxygen species (ROS).⁵³ Oxidative stress develops when ROS are formed and react with cellular constituents including thiols, sulfides, and lipids.⁵⁴ These oxidants deplete glutathione, and the redox status of the cell can be affected causing stressful toxic situation.⁵⁴ In addition, As(III) increases heat shock or stress proteins in cultured human cells.⁵⁵ In addition to As(III), As(V) is also released from GaAs and InAs NPs, which can add to the observed cytotoxicity.

According to previous studies, one of the main mechanisms of toxicity of As(V) relates to the chemical similarities of As(V) and phosphate. As(V) has the identical structure and acid dissociation constants as phosphate, so As(V) can substitute phosphate in many biochemical reactions.^56,57 When As(V) substitutes for phosphate, the methylation metabolism can convert As(V) into compounds with different properties from phosphate.⁵⁷ Additionally, As(V) reacts with adenosine diphosphate (ADP) to form a mixed arsenate–phosphate bond in molecules like adenosine triphosphate (ATP).⁵⁷ This bond can be hydrolyzed easily and the mixed arsenate–phosphate molecule can dissociate quickly without generating an ATP.^57,58 As(V) can also replace phosphate in the sodium pump and the anion exchange transport system of the human red blood cell.⁵⁹

In this study, Ga(III)-citrate was found to cause only low to moderate toxicity at low concentrations (16-hour IC₅₀= 166 mg Ga/l). Previous studies have shown that Ga(III) can compete with Fe(III) in cellular systems. One recent study reported that the interference of Ga(III) on Fe homeostasis can cause disruption of ribonucleotide reductase and mitochondrial function as well as the regulation of transferrin receptor and ferritin.⁶⁰ Ga(III) shares certain properties with Fe(III), for instance, Ga(III) has an octahedral ionic radius of 0.620 Å and a tetrahedral ionic radius of 0.47 Å, while those of Fe(III) are 0.645 and 0.49 Å.⁶⁰ Because of these similar properties, Ga(III) can compete with Fe(III) for binding to several redox enzymes and eventually interfere with some essential biological reactions.⁶¹

Conclusions

Both GaAs and InAs NPs caused noteworthy acute toxicity to 16HBE14o- human bronchial epithelial cells due to the release of inorganic arsenic species, As(III) and As(V), via corrosion and dissolution. On the other hand, Ga(III) and In(III) were only moderately toxic. Oxide NPs of Ga₂O₃ and In₂O₃ were not cytotoxic and were also not subject to dissolution. The results of this study indicate that the discharge of GaAs and InAs NPs can potentially be hazardous to humans if semiconductor manufacturing sector plans to introduce these materials.

Footnotes

Author Contributions

Chi H. Nguyen contributed to conception and design, contributed to acquisition, analysis, and interpretation, and drafted the manuscript. Chao Zeng contributed to conception and contributed to acquisition and analysis. Scott Boitano, Jim A. Field, and Reyes Sierra-Alvarez contributed to conception and design and contributed to acquisition, analysis, and interpretation. All authors critically revised the manuscript, gave final approval, and agree to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Semiconductor Research Corporation (SRC) Engineering Research Center for Environmental Benign Semiconductor Manufacturing [Award 425.052] and by the National Science Foundation (NSF-CBET/GOALI) [Award 1507446].

ORCID iD

Chi H. Nguyen

Reyes Sierra-Alvarez

References

Dick

Caroff

Bolinsson

, et al. Control of III-V nanowire crystal structure by growth parameter tuning. Semicond Sci Tech. 2010;25(2):024009.

Tomioka

Ikejiri

Tanaka

, et al. Selective-area growth of III-V nanowires and their applications. J Mater Res. 2011;26(17):2127–2141.

Vurgaftman

Meyer

Ram-Mohan

. Band parameters for III-V compound semiconductors and their alloys. J Appl Phys. 2001;89(11):5815–5875.

Yamaguchi

Nishimura

K-I

Sasaki

, et al. Novel materials for high-efficiency III–V multi-junction solar cells. Sol Energy. 2008;82(2):173–180.

Brightup

Goorsky

. Chemical-mechanical polishing for III-V wafer bonding applications: polishing, roughness, and an abrasive-free polishing model. ECS Transactions. 2010;33(4):383–389.

Matovu

Ong

Leunissen

LHA

Krishman

Babu

. Fundamental investigation of chemical mechanical polishing of GaAs in silica dispersions: material removal and arsenic trihydride formation pathways. ECS J Solid State Sc. 2013;2(11):432–439.

Shepard

Brenner

. An occupational exposure assessment for engineered nanoparticles used in semiconductor fabrication. Ann Occup Hyg. 2014;58(2):251–265.

Speed

Westerhoff

Sierra-Alvarez

, et al. Physical, chemical, and in vitro toxicological characterization of nanoparticles in chemical mechanical planarization suspensions used in the semiconductor industry: towards environmental health and safety assessments. Environ Sci: Nano. 2015;2(3):227–244.

Torrance

Keenan

Hursthouse

Stirling

. Measurement of arsenic and gallium content of gallium arsenide semiconductor waste streams by ICP-MS. J Environ Sci Heal A. 2010;45(4):471–475.

10.

Jain

Ali

. Arsenic: occurrence, toxicity and speciation techniques. Water Res. 2000;34(17):4304–4312.

11.

Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological profile for arsenic. In: ATSDR, U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA; 2007.

12.

Huang

Pan

Lee

Yen

. Treatment of high-level arsenic-containing wastewater by fluidized bed crystallization process. J Chem Technol Biot. 2007;82(3):289–294.

13.

Luo

Pan

. Arsenic removal of high-arsenic wastewater from gallium arsenide semiconductor production by enhanced two-stage treatment. Desalin Water Treat. 2015;55(5):1285–1292.

14.

Osborne

Lin

Jiang

, et al. Differential effect of micron- versus nanoscale III–V particulates and ionic species on the zebrafish gut. Environl Sci: Nano. 2017;4(11):1350–1364.

15.

Jiang

Lin

Chang

, et al. Implications of the differential toxicological effects of III–V ionic and particulate materials for hazard assessment of semiconductor slurries. ACS Nano. 2015;9(12):12011–12025.

16.

United States Environmental Protection Agency. Effluent Guidelines and Standards. Part 469. Electrical and electronic components point source category. Effluent limitations representing the degree of effluent reduction attainable by the application of the best practicable control technology currently available (BPT). Title 40 e-C.F.R. x 469.24. 2015. Office of the Federal Register and the Government Publishing Office. Federal Register. 1998;48(24155690):55690–55705.

17.

Flora

Dwivedi

. A toxicochemical review of gallium arsenide. Defence Sci J. 2012;62(2):95–104.

18.

Hughes

Beck

Chen

Lewis

Thomas

. Arsenic exposure and toxicology: a historical perspective. Toxicol Sci. 2011;123(2):305–332.

19.

Olsen

Liguori

Zong

Lantz

Burgess

Boitano

. Arsenic upregulates MMP-9 and inhibits wound repair in human airway epithelial cells. Am J of Physiol-Lung C. 2008;295(2):L293–L302.

20.

Sherwood

Lantz

Boitano

. Chronic arsenic exposure in nanomolar concentrations compromises wound response and intercellular signaling in airway epithelial cells. Toxicol Sci. 2013;132(1):222–234.

21.

Sherwood

Lantz

Burgess

Boitano

. Arsenic alters ATP-dependent Ca² ⁺ signaling in human airway epithelial cell wound response. Toxicol Sci. 2011;121(1):191–206.

22.

Sherwood

Liguori

Olsen

Lantz

Burgess

Boitano

. Arsenic compromises conducting airway epithelial barrier properties in primary mouse and immortalized human cell cultures. PLoS One. 2013;8(12):e82970.

23.

Ferreccio

González

Milosavjlevic

Marshall

Sancha

Smith

. Lung cancer and arsenic concentrations in drinking water in Chile. Epidemiology. 2000;11(6):673–679.

24.

Smith

Lingas

Rahman

. Contamination of drinking-water by arsenic in Bangladesh: a public health emergency. B World Health Organ. 2000;78(9):1093–1103.

25.

Xie

Huang

Martin

Wise

. Arsenic is cytotoxic and genotoxic to primary human lung cells. Mutat Res Genet Toxicol Environ Mutagen. 2014;760:33–41.

26.

World Health Organization. Guidelines for Drinking-Water Quality. 2nd ed. Vol. 1. Recommendations. France: World Health Organization; 1993.

27.

United States Environmental Protection Agency. National primary drinking water regulations: arsenic and clarifications to compliance and new source contaminants monitoring final. In: United States Environmental Protection Agency, ed; 2001:6976.

28.

Xing

Zhu

Jackson

, et al. Dynamic monitoring of cytotoxicity on microelectronic sensors. Chem Res Toxicol. 2005;18(2):154–161.

29.

Fisichella

Dabboue

Bhattacharyya

, et al. Mesoporous silica nanoparticles enhance MTT formazan exocytosis in HeLa cells and astrocytes. Toxicol in Vitro. 2009;23(4):697–703.

30.

Wang

Abassi

. The xCELLigence system for real-time and label-free monitoring of cell viability. Methods Mol Biol. 2011;740:33–43.

31.

Moe

McGuigan

X-F

Dabek-Zlotorzynska

Gabos

. Cell-electronic sensing of cellular responses to micro- and nanoparticles for environmental applications. In: Meyers

, ed. Encyclopedia of Analytical Chemistry. Hoboken, NJ: Wiley; 2014.

32.

Abassi

Zhang

, et al. Kinetic cell-based morphological screening: prediction of mechanism of compound action and off-target effects. Chem Biol. 2009;16(7):712–723.

33.

Moe

Gabos

. Real-time cell-microelectronic sensing of nanoparticle-induced cytotoxic effects. Anal Chim Acta. 2013;789:83–90.

34.

Otero-Gonzalez

Sierra-Alvarez

Boitano

Field

. Application and validation of an impedance-based real time cell analyzer to measure the toxicity of nanoparticles impacting human bronchial epithelial cells. Environ Sci Technol. 2012;46(18):10271–10278.

35.

Ponti

Ceriotti

Munaro

, et al. Comparison of impedance-based sensors for cell adhesion monitoring and in vitro methods for detecting cytotoxicity induced by chemicals. Altern Lab Anim. 2006;34(5):515–525.

36.

Ceriotti

Ponti

Broggi

, et al. Real-time assessment of cytotoxicity by impedance measurement on a 96-well plate. Sensors Actuat B-Chem. 2007;123(2):769–778.

37.

Wood

Samson

. The aqueous geochemistry of gallium, germanium, indium and scandium. Ore Geol Rev. 2006;28(1):57–102.

38.

Hoyt

Gewanter

. Citrate. In: de Oude

, ed. Detergents. Berlin, Germany: Springer; 1992:229–242.

39.

Thomas

Beswick

Basnakova

Moller

Macaskie

. Growth of naturally occurring microbial isolates in metal–citrate medium and bioremediation of metal–citrate wastes. J Chem Technol Biotechnol. 2000;75(3):187–195.

40.

Ehrhardt

Kneuer

Laue

Schaefer

Kim

K-J

, Lehr C-M. 16HBE14o- human bronchial epithelial cell layers express P-glycoprotein, lung resistance-related protein, and caveolin-1. Pharm Res. 2003;20(4):545–551.

41.

Forbes

Shah

Martin

Lansley

. The human bronchial epithelial cell line 16HBE14o- as a model system of the airways for studying drug transport. Int J Pharm. 2003;257(1-2):161–167.

42.

Zeng

Nguyen

Boitano

Field

Shadman

Sierra-Alvarez

. Cerium dioxide (CeO₂) nanoparticles decrease arsenite (As(III)) cytotoxicity to 16HBE14o- human bronchial epithelial cells. Environ Res. 2018;164:452–458.

43.

Solly

Wang

Strulovici

Zheng

. Application of real-time cell electronic sensing (RT-CES) technology to cell-based assays. Assay Drug Dev Technol. 2004;2(4):363–372.

44.

Kosmulski

. Pristine points of zero charge of gallium and indium oxides. J Colloid Interf Sci. 2001;238(1):225–227.

45.

Kosmulski

. Isoelectric points and points of zero charge of metal (hydr)oxides: 50 years after Parks’ review. Adv Colloid Interfac. 2016;238:1–61.

46.

Banerjee

. Current status of slurries and cleans for CMP of III-V device fabrications—a critical review. ECS Transactions. 2014;61:43–54.

47.

Badding

Stefaniak

Fix

Cummings

Leonard

. Cytotoxicity and characterization of particles collected from an indium–tin oxide production facility. J Toxicol Env Health Part A. 2014;77(20):1193–1209.

48.

Nguyen

Field

Sierra-Alvarez

. Microbial toxicity of gallium- and indium-based oxide and arsenide nanoparticles. J Environ Sci Health A Tox Hazard Subst Environ Eng. 2020;55(2),168–178.

49.

Styblo

Del Razo

Vega

, et al. Comparative toxicity of trivalent and pentavalent inorganic and methylated arsenicals in rat and human cells. Arch Toxicol. 2000;74(6):289–299.

50.

Moe

Peng

, et al. Comparative cytotoxicity of fourteen trivalent and pentavalent arsenic species determined using real-time cell sensing. J Environ Sci. 2016;49:113–124.

51.

Xiao

Lachance

Sunahara

Luong

JHT

. Assessment of cytotoxicity using electric cell−substrate impedance sensing: concentration and time response function approach. Anal Chem. 2002;74(22):5748–5753.

52.

Rzhepishevska

Ekstrand-Hammarström

Popp

, et al. The antibacterial activity of Ga³⁺ is influenced by ligand complexation as well as the bacterial carbon source. Antimicrob Agents Ch. 2011;55(12):5568–5580.

53.

Qiu

Chen

, et al. Reactive oxygen species contribute to arsenic-induced EZH2 phosphorylation in human bronchial epithelial cells and lung cancer cells. Toxicol Appl Pharm. 2014;276(3):165–170.

54.

Hughes

. Arsenic toxicity and potential mechanisms of action. Toxicol Lett. 2002;133(1):1–16.

55.

Keyse

Tyrrell

. Heme oxygenase is the major 32-kDa stress protein induced in human skin fibroblasts by UVA radiation, hydrogen peroxide, and sodium arsenite. P Natl Acad Sci USA. 1989;86(1):99–103.

56.

Dixon

HBF

. The biochemical action of arsonic acids especially as phosphate analogues. In: Sykes

, ed. Advances in Inorganic Chemistry. Cambridge, MA: Academic Press; 1996:191–227.

57.

Carter

Aposhian

Gandolfi

. The metabolism of inorganic arsenic oxides, gallium arsenide, and arsine: a toxicochemical review . Toxicol Appl Pharm. 2003;193(3):309–334.

58.

Goyer

Clarkson

. Toxic effects of metals. In: Klaasen

, ed. Casarett and Doullis Toxicology: The Basic Science of Poisons. New York, NY: McGraw-Hill; 2001:827–834.

59.

Kenney

Kaplan

. Arsenate substitutes for phosphate in the human red cell sodium pump and anion exchanger. J Biol Chem. 1988;263(17):7954–7960.

60.

Chitambar

. Gallium and its competing roles with iron in biological systems. Biochim Biophys Acta. 2016;1863(8):2044–2053.

61.

Antunes

LCS

Imperi

Minandri

Visca

. In vitro and in vivo antimicrobial activities of gallium nitrate against multidrug-resistant Acinetobacter baumannii . Antimicrob Agents Chemother. 2012;56(11):5961–5970.

Cytotoxicity Assessment of Gallium- and Indium-Based Nanoparticles Toward Human Bronchial Epithelial Cells Using an Impedance-Based Real-Time Cell Analyzer

Abstract

Keywords

Introduction

Materials and Methods

Chemicals

Particle Size Distribution and ζ Potential

Corrosion and Dissolution of III-V Particles

Immortalized Human Bronchial Epithelial Cell Culture

Real-Time Cell Analysis Assay

Uptake of NPs by 16HBE14o- Cells

Analytical Methods

Results and Discussion

Physiochemical Characterization of NPs

Cytotoxicity of Ga- and In-Based NPs

Corrosion and Dissolution of GaAs and InAs NPs

Cytotoxicity of Soluble As, In, and Ga Species

Nanoparticle Uptake by 16HBE14o- Cells

Potential Mechanisms of Species Cytotoxicity

Conclusions

Footnotes

Author Contributions

Declaration of Conflicting Interests

Funding

ORCID iD

References