The soluble nickel component of residual oil fly ash alters pulmonary host defense in rats

Abstract

The soluble metal fraction of residual oil fly ash (ROFA) has been shown to increase the susceptibility to infection in animal models. The goal of this study was to determine which of the primary soluble metals or metal combinations in ROFA were responsible for the increased infectivity. The soluble fraction of ROFA contained Ni, Fe, Al, and Zn. On Day 0, Sprague-Dawley rats were intratracheally (IT) instilled with NiCl₂ (55.7 μg/rat), FeSO₄ (32.7 μg/rat), Al₃(SO₄)₂ (46.6 μg/rat), or ZnCl₂ (8.69 μg/rat), or a combination of all the metals (Total Mixture). In a separate experiment, rats were instilled with metal mixtures, including the total mixture, and mixtures without Fe (Mix – No Fe), Ni (Mix - No Ni), Al (Mix – No Al), or Zn (Mix - No Zn). At Day 3, rats were instilled with 5 × 10⁴ Listeria monocytogenes. At Days 6, 8 and 10, left lungs were removed to assess bacterial clearance. Bronchoalveolar lavage (BAL) was performed on right lungs on Day 3, before infection, and on Days 6, 8 and 10 to assess lung injury and cellular activity. Prior to infection, soluble Ni and mixtures containing Ni significantly increased lung injury, inflammation, and oxidative damage to a comparable degree when compared to control. Post-infection, rats pre-treated with soluble Ni, alone or in a metal mixture, had increased bacterial lung burden on Day 6, and body weight decreased in the soluble Ni, Mix - No Fe, and Mix - No Al groups post-infection, indicating Fe and Al may act antagonistically to Ni. Ni alone and in metal mixtures increased reactive oxidants in the lung and appeared to be the most important factor in suppressing T-cell activity post-infection. Soluble Ni is likely the primary metal involved in the increased susceptibility to infection observed in rats exposed to the soluble metals of ROFA.

Keywords::

• lung
• Listeria monocytogenes
• immunotoxicology
• fly ash
• nickel compounds

Introduction

Long-term, large cohort epidemiological studies have correlated exposure to particulate matter (PM) with elevations in morbidity and mortality as indicated by increased hospital emissions for both cardiovascular and respiratory diseases (Dockery et al., 1993; Pope et al., 1995, 2004; Samet et al., 2000; Schwartz, 1994). Pulmonary effects attributed to PM exposure include, but are not limited to, increased airway reactivity and exacerbations in asthma, increased respiratory symptoms (cough, wheeze), decreased lung function, increased lung inflammation, and altered pulmonary host defense (American Thoracic Society, 1996). Pope et al. (2004) showed that the leading cause of increased morbidity and mortality in the category of pulmonary conditions was respiratory infection.

Particle composition appears to be a strong determinant in PM toxicity. An epidemiological study by Burnett et al. (2000) showed that iron (Fe), sulfate (SO₄), nickel (Ni), and zinc (Zn) in ambient PM_2.5 correlated more strongly with mortality than PM_2.5 mass, and Laden et al. (2000) were able to correlate specific metals in PM with increases in morbidity. Population studies in industrialized areas have shown that, in regions with high anthropologic PM loads, whole blood and serum levels of lymphocytes, immunoglobulin (Ig) isotypes and levels, and cytokines are significantly altered (Calderon-Garciduenas et al., 2000; Hertz-Picciotto et al., 2005; Leonardi et al., 2000; Skachkova et al., 2001). Human exposure studies have also demonstrated that PM with high metal content can alter cellular processes that are important in pulmonary responses to pathogens. Increases in the number and type of leukocytes in the lung, as well as elevations in oxidant and cytokine production by recruited phagocytes have been observed (Ghio, 2004; Ghio et al., 2000; Schaumann et al., 2004).

Animal studies have shown that PM containing metals, such as concentrated ambient air particulates (CAPs), residual oil fly ash (ROFA), copper smelter fly ash, and various forms of coal fly ash (CFA), can enhance susceptibility to infection or exacerbate pre-existing pulmonary conditions (Antonini et al., 2002; Clarke et al., 1999; Hatch et al., 1985; Kodavanti et al., 2000; Roberts et al., 2007; Zelikoff et al., 2003). Interestingly, in the study by Hatch et al. (1985), the investigators found that the ashes and particulates that were most potent at altering infectivity were the samples that were highest in soluble metal content. In addition, the various soluble metals associated with combustion-derived PM, such as Zn, Fe, Ni, and aluminum (Al), have been shown to alter susceptibility to pulmonary infection and immune responses in animal models after pulmonary exposure (Adkins et al., 1979; Cohen, 2004; Ehrlich, 1980; Hatch et al., 1981, 1985; Sherwood et al., 1981; Zelikoff et al., 2002). However, there are relatively limited, if any, studies that examine the effects of respiratory exposure to the above metals in combination with each other on susceptibility to pulmonary infection, at the levels relevant to a particular sample of PM.

Previously, our group has shown that the soluble fraction of ROFA significantly decreased the pulmonary clearance of Listeria monocytogenes in rats (Roberts et al., 2003, 2007). The soluble component of the ROFA used in our previous studies consisted primarily of Al, Fe, Ni and Zn. It is likely the inhibition of alveolar macrophage (AM) phagocytosis and killing lead to a higher bacterial burden early in the infection after treatment with the soluble fraction, and inhibition of, delay in, or alteration of the adaptive immune response resulted in the slowing of the bacterial clearance over time (Roberts et al., 2007). The purpose of the current study was to determine which metal or combination of metals, at the concentrations measured in the soluble fraction of ROFA, would best replicate the effects seen with the soluble ROFA sample. Based on the available data highlighting the inhibitory effects of Ni on immune cells involved in pulmonary host defense, and that Ni was one of the metals in highest concentration in this sample of ROFA, the hypothesis was that Ni would be one of the primary metals involved in the increase in susceptibility caused by the soluble fraction of ROFA.

To investigate the hypothesis, male Sprague-Dawley rats were intratracheally (IT)-instilled with the soluble form of Zn, Fe, Al, or Ni at the concentrations measured in the soluble ROFA sample, either individually, or in a mixture of the four metals. In a separate experiment, rats were given an IT dose of the total metal mixture containing all four metals, or various combinations of three metals. Rats were intratracheally inoculated with L. monocytogenes 3 days after metal exposures, and bacterial clearance was monitored for up to one week post-infection.

Both innate and adaptive immune responses are necessary for clearance of L. monocytogenes from the lung. These responses are governed in part by immunomodulatory cytokines beginning with phagocytosis of the bacteria by resident macrophages, leading to a cascade of inflammatory cytokines and a burst in oxidant production to enhance macrophage killing of the bacteria. In turn, neutrophils and monocytes are recruited to the lung, and an enhanced activation of macrophages and natural killer cells occurs to aid in killing and clearing bacteria from infected cells. To monitor this portion of the innate immune response, tumor necrosis factor (TNF)-α, interleukin (IL)-6, IL-12, and nitric oxide (NO) were measured in the bronchoalveolar lavage (BAL) fluid. IL-10 was also monitored as a potential inhibitory cytokine of macrophages. In addition, total numbers and phenotypes of BAL cells were determined, and oxidant and oxidant production by macrophages was measured.

The adaptive immune response to L. monocytogenes is characterized primarily by cell-mediated immunity. The pro-inflammatory cytokines, such as IL-12 and IL-18, induce the expression of interferon (IFN)-γ by CD4⁺ T_H1 cells, CD8⁺ T-cells, and NK cells, and the presence of IL-12 and IFN-γ, in turn, drives the differentiation of naïve CD4⁺ T-cells toward the CD4⁺ T-helper-1 (T_H1) subset (Mosmann and Sad, 1996), favoring the development of the cell-mediated immune response versus a humoral immune response. T-Cell subsets will then produce auto- regulatory cytokines. IL-2 is produced primarily by CD4⁺ T_H1 cells and by naïve CD4⁺ and some CD8⁺ T-cells, and promotes growth, proliferation, and clonal expansion of T-cells. IL-4 is secreted mainly by the CD4⁺ T_H2 subset and is involved in promotion of the humoral immune response (Mosmann and Sad, 1996). The cell- mediated immune response will further promote bacterial killing through induction of oxidant production by macrophages and clearing/killing of infected cells by macrophages, NK cells, and CD8⁺ T-cells. To assess the adaptive responses after exposure to the various metals, lymphocyte phenotype in the lung was characterized and T-cell cytokines were also measured.

Methods

Animals

Male Sprague–Dawley [Hla:(SD)CVF] rats (Hilltop Laboratories, Scottdale, PA) weighing 250–300 g, ≈10-wk-old, were used for all experiments in accordance with a protocol approved by the NIOSH Animal Care and Use Committee. They were given the Teklad Global 2918 diet and tap water ad libitum, housed in a clean air and viral- and antigen-free room with restricted access in an American Association for Accreditation of Laboratory Animal Care-approved animal facility, and allowed to acclimate for one week before use. The rats were monitored and found to be free of endogenous viral pathogens, parasites, mycoplasms, Helicobacter, and CAR Bacillus.

Materials

Listeria monocytogenes (strain 10403S, serotype 1) was obtained as a gift from Dr. Rosana Schafer of the Department of Microbiology and Immunology at West Virginia University. All soluble metals were purchased from Sigma-Aldrich Co., (St. Louis, MO).

Experimental design

At Day 0, animals were pre-exposed to individual metals, metal mixtures, or PBS, (vehicle control) by IT instillation. At Day 3, the animals were inoculated via IT instillation with 5 × 10⁴L. monocytogenes. This timepoint reflects a fully-developed innate immune response where signaling toward development of an adaptive response may be occurring, and it allows enough time post-instillation to avoid measuring any response that may be due to the actual instillation of the bolus dose as a possible confounder. At Days 6, 8, and 10, the left lungs of the animals were clamped off, removed, homogenized, and the number of colony-forming units (CFUs) was assessed. The right lungs of animals were lavaged on Day 3 prior to infection, and on Days 6, 8, and 10, and the cells and the fluid were retained for analysis as described below.

Metal treatment

Stock solutions of individual soluble metals [NiCl₂ (55.7 μg/rat), FeSO₄ (32.7 μg/rat), Al₃(SO₄)₂ (46.6 μg/rat), or ZnCl₂ (8.69 μg/rat)] and soluble metal mixtures (mixtures of 3 or all of the following: soluble Fe, Ni, Al, and Zn) were prepared in deionized water and were further diluted in sterile PBS to the concentrations present in 2 mg of a residual oil fly ash (ROFA) sample that had been previously characterized (Antonini et al., 2004). The ROFA dose chosen was previously shown to induce inflammation (Antonini et al., 2002), and fell within the range of concentrations consistently used in other animal studies evaluating the pulmonary responses to ROFA (Dreher et al., 1997; Gavett et al., 1997; Kodavanti et al., 1998). The soluble fraction of ROFA was determined using inductively-coupled argon plasma, atomic emission spectroscopy and was found to consist primarily of soluble Fe, Ni, Al, and Zn at levels given in Table 1. The metal solutions were sonicated for 1 min with a Sonifier 450 Cell Disruptor (Branson Ultrasonics, Danbury, CT).

Table 1.  Concentrations for individual metals and metal combinations (μg/rat).

Metal	Treatment Groups
	Saline	Fe	Ni	Al	Zn	Mix, No Fe	Mix, No Ni	Mix, No Al	Mix, No Zn	Total Mixture
Soluble Fe	0	37.2μg	0	0	0	0	37.2μg	37.2μg	37.2μg	37.2μg
Soluble Ni	0	0	55.7μg	0	0	55.7μg	0	55.7μg	55.7μg	55.7μg
Soluble Al	0	0	0	46.6μg	0	46.6μg	46.6μg	0	46.6μg	46.6μg
Soluble Zn	0	0	0	0	8.69μg	8.69μg	8.69μg	8.69μg	0	8.69μg

Rats were lightly anesthetized by an intraperitoneal injection of 0.6 ml of a 1% solution of sodium methohexital (Brevital, Eli Lilly, Indianapolis, IN), and IT instillation was performed with the individual soluble metals, soluble metal mixtures, or sterile PBS (vehicle control) at a volume of 300 μl per rat according to the method of Brain et al. (1976). Briefly, the rat was placed on a slanted board and was supported by a wire under its upper incisors. The tongue of the animal was moved aside, and the larynx was illuminated by a modified laryngoscope. Each rat was then instilled with one of the samples described above using a 20-gauge, 1.5-in animal feeding needle. The treatment groups and metal mass are listed in Table 1. There were 4–8 animals per group per timepoint.

Intratracheal bacteria inoculation

Listeria monocytogenes was cultured overnight in brain-heart infusion broth (Difco Laboratories, Detroit, MI) at 37°C in a shaking incubator. Following incubation, the bacterial concentration was determined spectrophotometrically at an optical density of 600 nm. The sample was diluted to the concentration of 5 × 10⁴L. monocytogenes in 500 μl of sterile PBS and administered IT to the rats in each treatment group 3 days post-metal sample instillation (n=4–8 per group per timepoint).

Morbidity/pulmonary clearance of Listeria

Morbidity and pulmonary clearance were monitored in all animals pre-treated with individual soluble metals, soluble metal mixtures, or saline. Animal weights were monitored over the course of the treatment period as an indicator of morbidity. Rats were euthanized with an overdose of sodium pentobarbital, and at Days 6, 8, and 10 (corresponding to Days 3, 5, and 7 post-infection) the left lungs were removed from all the rats in each treatment group. The excised tissues were suspended in 10 ml of sterile water, homogenized using a PowerGen 700 homogenizer (Fisher Scientific, Pittsburgh, PA), and cultured quantitatively on brain-heart infusion agar plates (Becton Dickinson and Co., Cockeysville, MD). The number of viable CFU was counted after an overnight incubation at 37°C.

Bronchoalveolar lavage (BAL)

BAL was performed by washing out the lungs of the rats with aliquots of PBS in order to obtain pulmonary cells for morphological and functional analysis, and the acellular BAL fluid was retained for analysis of indicators of tissue damage and cellular activity. Rats were euthanized with an overdose of sodium pentobarbital, the left lungs were clamped off, and BAL was performed on the right lungs on Day 3 prior to infection, and at Days 6, 8, and 10. The acellular fraction of the first BAL was obtained as described by Antonini et al. (2002). This concentrated aliquot was withdrawn, retained, kept separately, and was designated as the first fraction of BAL fluid. The following aliquots were 6 ml in volume, instilled once with light massaging, withdrawn, and combined until a 30 ml volume was obtained. For each animal, both fractions of BAL were centrifuged, the cell pellets were combined and resuspended in 1 ml of PBS, and the acellular fluid from the first fraction was retained for further analysis.

Analysis of albumin and lactate dehydrogenase (LDH)

The presence of albumin and LDH in the BAL fluid of all treatment groups was measured to evaluate the loss of integrity of the alveolar-capillary barrier and general cytotoxicity, respectively. Measurements of both albumin and LDH in the acellular fluid were obtained using a Cobas Mira analyzer (Roche Diagnostic Systems, Montclair, IN). Albumin was determined by spectrophotometric measurement of the reaction product of albumin with bromcresol green (628 nm) according to a method by Sigma Diagnostics (St. Louis, MO). LDH activity was quantified by detection of the oxidation of lactate coupled to the reduction of NAD⁺ at a spectrophotometric setting of 340 nm over time.

Pulmonary cell differentials and phenotyping

Cellular analysis was performed on the BAL samples of rats pre-treated with individual soluble metals, soluble metal combinations, or saline on Days 3, 6, 8, and 10. Total BAL cells were counted using a Coulter Multisizer II (Coulter Electronics, Hialeah, FL). Cell differentials were performed to determine the total number of alveolar macrophages (AM), polymorphonuclear cells (PMN), and lymphocytes. The percentage of AM, PMN, and lymphocytes were multiplied by the total number of cells to calculate the number of each cell type. To determine phenotype of the cell population in the BAL, cells were labeled with fluorescently tagged antibodies against specific cell surface markers (BD Biosciences Pharmingen, San Diego, CA). Cell types that were identified were natural killer (NK cells; CD161a⁺, CD3⁻), B-cells (CD45R⁺), and T-cells (CD3⁺). The T-cells were further differentiated as T-helper (T_H) cells (CD4⁺) or cytotoxic T-cells (CD8⁺). Using a flow cytometer (FACS Calibur, BD Biosource, San Diego, CA), the lymphocyte population of the BAL cells to be analyzed was determined by size using forward and side scattering, and the viable cells were selected by eliminating the population of dead cells that had stained positive for 7-amino-actinomycin D (7-AAD). Percentages of each cell type in BAL measured with flow cytometry were calculated and multiplied by the total number of lymphocytes determined from the cell differentials.

Measurement of nitric oxide and reactive oxygen species

Nitric Oxide (NO) levels and reactive oxygen species (ROS) were determined by analysis of BAL from rats pre-treated with individual soluble metals, soluble metal mixtures, or saline on Days 3, 6, 8, and 10. The presence of NO in acellular BAL was determined as the accumulation of nitrite using a modified microplate assay using the Greiss reagent (Green et al., 1982) and a nitrate reductase to convert any nitrate in the sample to nitrite. Absorbance of the samples was analyzed on a SPECTRAmax TM 250 spectrophotometer (Molecular Devices Co., Sunnydale, CA) at 550 nm. The measurement of total nitrite represents the presence of both nitrate and nitrite (NO_x) in the sample.

To estimate total lung oxidant burden, luminol-dependent chemiluminescence (CL) was performed on BAL cells as a measure of the light generated by the production of ROS by AM using a Berthold LB953 luminometer (Wallace Inc., Gaithersburg, MD). Non-opsonized, insoluble zymosan (2 mg/ml), an AM stimulant, was added to the assay immediately prior to CL measurement to determine the contribution of AM to the overall production of ROS in the lungs of the rats. Measurement of CL was recorded for 15 min at 37°C, and the integral of counts per minute (cpm) per 10⁶ cells versus time was calculated.

Cytokine measurements in BAL fluid

Cytokines present in the BAL fluid of rats pre-treated with individual soluble metals, soluble metal mixtures, or saline were analyzed by enzyme-linked immunosorbent assay (ELISA) using commercially-available ELISA kits (BioSource International Inc., Camarillo, CA) to determine activity of various cell types involved in the immune response. The following cytokines were quantified: tumor necrosis factor-α (TNF-α), interleukin (IL)-2, IL-4, IL-6, IL-10, IL-12p70, IL-18, and interferon (IFN)-γ.

Statistical analysis

Results are expressed as means ± standard error of measurement (SE). Statistical analyses were carried out with the SigmaStat 3.1 statistical program (Chicago, IL). The significance of the interaction among different treatment groups for the different parameters at each time point was assessed using analysis of variance (ANOVA). The significance of difference between individual groups was analyzed using the Student-Newman-Keuls method with the criterion of significance set at p < 0.05.

Results

Morbidity and pulmonary bacterial clearance

Morbidity was monitored as the percent change in body weight after infection with L. monocytogenes in rats pre-treated with individual soluble metals (Figure 1A) or with soluble metal mixtures (Figure 1B). Body weights of rats from all treatment groups increased prior to infection from Day 0 to Day 3 and the percent change in body weight did not differ significantly among groups during this time period (data not shown). Rats treated with soluble Ni alone (Figure 1A) or in combination with other metals (Figure 1B) lost weight post-infection when compared to infected rats pre-treated with saline, with soluble Fe, Al, or Zn alone, or with the metal mixture that did not contain Ni. Although rats in all treatment groups had begun to gain weight by Day 10, there were still significant differences in body weight in the rats pre-treated with soluble Ni alone and in mixtures of metals that contained Ni, with the exception of the mixture that lacked Fe.

Figure 1.  Percent (%) change in body weight post-infection of rats that were pre-exposed to individual soluble metals (A) or to soluble metal mixtures (B) 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE (p < 0.05). Figure 1A: *significantly different from all groups; ^$significantly different from Soluble Fe and Saline groups. Figure 1B: ^asignificantly different from Saline and Mix - No Ni groups; ^bsignificantly different from Saline group; ^csignificantly different from Mix - No Ni group; ^dsignificantly different from Saline, Mix - No Ni, and Mix - No Zn groups.

Pulmonary clearance of bacteria was monitored by measuring number of bacteria colony forming units (CFU) in the left lung of rats that had been pre-treated with individual soluble metals (Table 2) or soluble metal mixtures (Table 3). Rats pre-treated with soluble Fe, Al, Zn, or a mixture of the soluble metals without Ni were able to clear bacteria form the lungs at rate equivalent to that of the saline control rats. Rats treated with soluble Ni or mixtures that contained soluble Ni had a significantly higher bacterial lung burden at the early timepoint post-infection (Day 6) when compared to saline control rats and all other groups that did not contain Ni. By Day 10, rats in all treatment groups had cleared the majority of bacteria from their lungs.

Table 2.  Bacterial clearance after pre-exposure to individual metals.

Treatment	CFUs LOG^10 Units
	Day 3 Initial Infection	Day 6	Day 8	Day 10
Saline	4.7	5.9 ± 0.10	4.3 ± 0.38	2.2 ± 0.73
Soluble Fe	4.7	5.6 ± 0.03	4.1 ± 0.12	3.1 ± 0.16
Soluble Ni	4.7	7.3 ± 0.05 ^+	5.7 ± 0.63	3.2 ± 0.22
Soluble Al	4.7	5.9 ± 0.07	4.3 ± 0.26	2.2 ± 0.80
Soluble Zn	4.7	6.1 ± 0.12	4.4 ± 0.34	2.4 ± 0.84
Total Mixture	4.7	7.4 ± 0.17 ^+	4.4 ± 0.21	2.2 ± 0.75

Number of bacterial CFUs in the left lungs of rats that were pre-exposed to individual soluble metals, the metal mixture, or saline 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE, ⁺significantly different from Saline, Soluble Fe, Soluble Al, and Soluble Zn groups (p < 0.05).

Table 3.  Bacterial clearance after pre-exposure to metal mixtures.

Treatment	CFUs LOG^10 Units
	Day 3 Initial Infection	Day 6	Day 8	Day 10
	4.8 ± 0.01	6.1 ± 0.09	4.3 ± 0.31	3.0 ± 0.18
Mix - No Fe	4.8 ± 0.01	7.2 ± 0.14^a	4.7 ± 0.20	3.1 ± 0.17
Mix - No Ni	4.8 ± 0.01	6.1 ± 0.08	4.0 ± 0.14	2.2 ± 0.52
Mix - No Al	4.8 ± 0.01	7.2 ± 0.08^a	5.3 ± 0.48	3.2 ± 0.59
Mix - No Zn	4.8 ± 0.01	7.1 ± 0.09^a	4.8 ± 0.30	3.0 ± 0.13
Total Mixture	4.8 ± 0.01	6.9 ± 0.19^a	4.3 ± 0.16	3.3 ± 0.23

Number of bacterial CFUs in the left lungs of rats that were pre-exposed to soluble metal mixtures, the total metal mixture, or saline 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE; ^asignificantly different from Saline and Mix - No Ni groups (p < 0.05).

Biochemical analysis of BAL fluid: LDH and albumin

The presence of the intracellular enzyme LDH in the BAL fluid of rats was used as a marker for general cellular cytotoxicity (Figure 2A and B), and the presence of albumin in the BAL fluid served as an indicator of the breakdown of the air-blood barrier (Figure 2C and 2D). On Day 3, prior to infection, rats treated with soluble Ni alone or mixtures of metals containing soluble Ni had significant elevations in both LDH and albumin when compared to saline control, Fe, Al, or Zn alone (Figure 2A and 2C), and when compared to the mixture without Ni (Figures 2B and 2D). The increases observed in the metal mixtures containing Ni did not differ from that of Ni alone. The same pattern of increased injury was also present post-infection on Days 6 and 8. By Day 10, there were no significant differences in LDH among treatment groups, and only a slight, but significant, increase in albumin was present in the soluble Ni group (Figure 2B).

Figure 2.  Lactate dehydrogenase (LDH) (A and B) and albumin (C and D) in the BAL fluid of rats that were pre-exposed to individual soluble metals or to soluble metal mixtures 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE (p < 0.05). Figures 2A and 2C: *significantly different from all groups; ⁺significantly different from Saline, Soluble Fe, Soluble Al, and Soluble Zn groups. Figures 2B and 2D:^asignificantly different from Saline and Mix - No Ni groups (p < 0.05).

BAL cellular profiles

Total number of AM and PMN were calculated from cell differentials for both the soluble metals (Figures 3A and 3C) and the metal mixtures (Figures 3B and D). On Day 3, prior to infection, increases in PMN numbers were observed in all groups that contained Ni, and were significantly elevated in the total mixture group and the mixture without Zn when compared to saline control. There were also slight, but significant, increases in AM number in the total mixture group and the mixture without Fe. Post-infection, AM numbers were increased in all groups that contained Ni, including the rats treated with soluble Ni alone, when compared to saline, individual soluble metals excluding Ni (Fe, Al, Zn), or metal mixtures that did not contain Ni (Mix, No Ni) at all timepoints post-infection, and PMN were increased primarily at the earlier timepoints post-infection (Days 6 and 8).

Figure 3.  Totals AM (A and B) and total PMN (C and D) present in the BAL fluid of rats that were pre-exposed to individual soluble metals or to soluble metal mixtures 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE (p < 0.05). Figures 3A and 3C: ⁺significantly different from Saline, Soluble Fe, Soluble Al, and Soluble Zn groups; ^#significantly different from Saline group; ^$significantly different from Soluble Fe and Saline groups; ^&significantly different from Saline, Soluble Fe, and Soluble Al groups. Figures 3B and 3D: ^asignificantly different from Saline and Mix - No Ni groups; ^csignificantly different from Mix - No Ni group; ^esignificantly different from Saline, Mix – No Fe, and Mix – No Ni groups.

To assess the adaptive immune response, flow cytometry, in conjunction with cell differentials, was used to determine the total number of lymphocytes (B-cells, T-cells, and CD8⁺ and CD4⁺ T-cell subsets) and NK cells for rats pre-treated with individual soluble metals (Table 4) or soluble metal mixtures (Table 5). Prior to infection, there were no significant differences in NK cell number, lymphocyte number, or lymphocyte populations among treatment groups (data not shown). Total number of lymphocytes and NK cells increased in all groups over the course of the infection (data not shown) peaking on Day 10, with no significant differences in number or cell type until Day 10. On Day 10, rats pre-treated with soluble Ni had a significant increase in T-cells and both T-cell subsets when compared to all other groups, and a slight, but not significant, increase in B-cells (Table 4). Rats pre-treated with metal mixtures containing Ni had a significant increase in the B-cell population when compared to the saline and Mix-No Ni groups. There was also a large increase in T-cells, both CD4⁺ and CD8⁺, for the groups exposed to metal mixtures containing Ni (Table 5).

Table 4.  BAL cell phenotype on Day 10 after pre-exposure to individual metals as determined by flow cytometry (3 10⁴ Cells).

Treatment	NK Cells	B-Cells	Total T-Cells	CD4^+ T-Cells	CD8^+ T-Cells
Saline	7.95 ± 3.30	8.62 ± 1.98	34.7 ± 10.3	21.7 ± 5.97	13.0 ± 4.38
Soluble Fe	5.23 ± 1.67	12.9 ± 4.04	35.7 ± 14.1	19.0 ± 7.49	16.6 ± 7.45
Soluble Ni	5.66 ± 2.28	19.6 ± 3.61	191 ± 7.30 ^*	85.6 ± 20.5 ^*	105 ± 53.1 ^*
Soluble Al	3.69 ± 0.590	10.1 ± 3.47	31.0 ± 2.68	19.0 ± 1.93	12.0 ± 0.90
Soluble Zn	3.88 ± 1.03	6.93 ± 0.450	25.5 ± 3.76	16.8 ± 2.95	8.69 ± 1.07
Total Mixture	1.62 ± 0.049	9.83 ± 5.47	37.3 ± 18.6	23.1 ± 13.0	14.2 ± 5.63

Note: Values are means ± SE, ^*significantly different from all groups (p < 0.05).

Table 5.  BAL cell phenotype on Day 10 after pre-exposure to metal combinations as determined by flow cytometry (3 10⁴ Cells).

Treatment	NK Cells	B-Cells	Total T-Cells	CD4^+ T-Cells	CD8^+ T-Cells
Saline	7.29 ± 1.12	9.50 ± 1.36	65.4 ± 12.8	41.0 ± 9.90	24.4 ± 4.11
Mix – No Fe	9.21 ± 2.20	20.1 ± 2.31^a	110 ± 15.0^a	57.5 ± 9.07	52.9 ± 8.93
Mix – No Ni	6.01 ± 0.815	8.06 ± 0.975	49.0 ± 9.68	27.3 ± 5.72	21.7 ± 6.28
Mix – No Al	8.06 ± 1.68	21.2 ± 3.53^a	161 ± 52.6^a	81.8 ± 21.6^a	78.7 ± 32.0
Mix – No Zn	9.12 ± 1.83	21.1 ± 4.37^a	123 ± 22.3^a	77.8 ± 12.6^a	45.2 ± 9.91
Total Mixture	4.99 ± 1.17	18.8 ± 2.74^a	119 ± 27.5^a	65.6 ± 23.8	52.9 ± 13.0

Note: Values are means ± SE, ^asignificantly different from Saline and Mix – No Ni groups (p < 0.05).

Production of reactive oxygen and nitrogen species

Reactive oxygen and nitrogen species (ROS and RNS, respectively) are known to act as antimicrobial agents; however, large increases in production of these species can lead to oxidative stress. The presence of nitric oxide was measured as total nitrite (NO_x) in the BAL fluid of rats pre-exposed to individual soluble metals (Figure 4A) or soluble metal mixtures (Figure 4B). On Day 3, prior to infection, levels of NO_x in the BAL were significantly elevated in rats exposed to the total mixture of soluble metals and to soluble Ni alone when compared to the other individual soluble metal groups and to saline control rats. There were also increases in all metal mixtures that contained soluble nickel. After infection, on Day 6, all groups that contained soluble Ni, including soluble Ni alone, had a significantly greater increase in NO levels in the BAL when compared to saline, soluble Fe, Al, and Zn alone, as well as when compared to the metal mixture without Ni. There were no significant differences on Days 8 and 10 in BAL NO levels.

Figure 4.  Nitrate and nitrite (NO_x) in the BAL fluid (BAL) of rats that were pre-exposed to individual soluble metals (A) or to soluble metal mixtures (B) 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE (p < 0.05). Figure 4A: *significantly different from all groups; ⁺significantly different from Saline, Soluble Fe, Soluble Al, and Soluble Zn groups. Figure 4B: ^asignificantly different from Saline and Mix - No Ni groups; ^csignificantly different from Mix - No Ni group.

The production of ROS by AM was assessed by measuring luminol-dependent chemiluminescence after zymosan stimulation in AM recovered from rats exposed to individual soluble metals (Figure 5A) or soluble metal combinations (Figure 5B) followed by infection. On Day 3, prior to infection, the oxidative potential of AM was elevated in the rats exposed to the total mixture of soluble metals when compared to control. After infection, on Day 6, there was an increase in ROS production by AM in all groups. However, at all timepoints post-infection there were significant elevations in AM ROS production in rats pre-exposed to mixtures containing soluble Ni, as well as in rats pre-treated with soluble Ni alone. Rats treated with the mixture of metals without Ni, or with the individual soluble metals, Fe, Al, or Zn, followed by infection did not differ from saline-treated rats.

Figure 5.  BAL cellular chemiluminescence (CL) depicted in total counts per 15 min for total BAL AM after stimulation with zymosan in rats that were pre-exposed to individual soluble metals (A) or to soluble metal mixtures (B) 3 days prior to intratracheal inoculation with L. monocytogenes. Values are means ± SE (p < 0.05). Figure 5A: *significantly different from all groups; ⁺significantly different from Saline, Soluble Fe, Soluble Al, and Soluble Zn groups. Figure 5B: ^asignificantly different from Saline and Mix - No Ni groups; ^bsignificantly different from Saline group; ^csignificantly different from Mix - No Ni group; ^fsignificantly different from all groups (p < 0.05).

BAL cytokine analysis

Cytokines present in the BAL were evaluated as indicators of cell specific activity in response to soluble metals followed by bacterial infection. Cytokines involved in the inflammatory response, in early modulation of adaptive immune responses, and T-cell cytokines were measured on Day 3 (Figure 6A and 6B), prior to infection, after exposure to individual soluble metals (Figure 6A) or soluble metal mixtures (Figure 6B). There were no significant differences among groups in levels of TNFα, IL-6, IL-12p70, IL-18, IL-2, or IL-4 in the BAL fluid prior to infection. However, there was significantly less IL-10, a cytokine involved in inhibition of both AM activation and T-cell-mediated immunity, in all the soluble metal mixture groups when compared to saline treated rats. On Day 6, 3 days post-infection, IL-6 (indicative of an acute phase response) was elevated in rats pre-exposed to soluble Ni alone and the total mixture of soluble metals (Figure 6C), as well as in mixtures that contained soluble Ni (Figure 6D), when compared to saline control. IL-12p70 and IL-18 are two cytokines that are involved in development and activation of T_H1 cells (a subset of CD4⁺ T-cells), cytotoxic T-cells (CD8⁺), and NK cells. IL12p70 was significantly elevated in the rats pre-treated with soluble Ni alone.

Figure 6.  Cytokines involved in inflammation and the adaptive immune responses measured in the BAL fluid 3 days after intratracheal instillation with individual soluble metals (A) or soluble metal mixtures (B). Cytokines measured in the BAL fluid on Day 6 (3 days post-infection with L. monocytogenes) after intratracheal instillation with soluble metals (C) or soluble metal mixtures (D). Values are means ± SE (p < 0.05). Figures 6A and 6C: *significantly different from all groups; ⁺significantly different from Saline, Soluble Fe, Soluble Al, and Soluble Zn groups. Figures 6B and 6D: ^asignificantly different from Saline and Mix - No Ni groups; ^bsignificantly different from Saline group; ^csignificantly different from Mix - No Ni group; ^esignificantly different from Saline, Mix – No Fe, and Mix – No Ni groups; ^gsignificantly different from Saline, Mix - No Fe, and Mix - No Al groups.

In addition, IL-18, a cytokine involved in initiation and activation of T_H1 cells, was significantly elevated on Day 8 in rats exposed to soluble Ni alone prior to infection when compared to all other individual soluble metals and saline control (data not shown), and a similar trend was apparent in all metal mixtures containing soluble Ni. On Day 6, T-cell cytokines IL-2 (a promoter of T-cell development and T_H1 cell activation) and IL-4 (a promoter of T_H2 cell activation and a humoral immune response) were significantly lower in rats pre-exposed to soluble Ni alone or mixtures that contained soluble Ni when compared to other individual soluble metals, the metal mixture without soluble Ni, and saline control. There were also lower levels of IL-2 in the group exposed soluble Ni alone at Days 8 and 10 as well (data not shown).

Discussion

The purpose of this study was to determine which soluble metal or metal combination was responsible for the alterations in pulmonary host defense and increased susceptibility observed after exposure to a soluble ROFA sample (Roberts et al., 2007). Intratracheal (IT) instillation of the artificial metal mixture used in the study decreased pulmonary clearance of bacteria, increased lung injury and inflammation, and altered cellular profiles in a similar fashion, although not to the same degree, as was observed with the soluble fraction of ROFA. The four metals used in the artificial mixture (Al, Fe, Ni, and Zn) could account for the majority of the effects observed with the soluble ROFA sample at the early timepoints post-infection (Day 6); however, the infection at the later time points was not as severe as that observed with the soluble fraction of ROFA.

The major findings of this study are that primarily soluble Ni, alone or in combination with soluble Fe, Zn, or Al, at the concentrations present in the soluble sample of ROFA, accounted for an increased susceptibility to infection and had the greatest effect on lung injury and inflammation. The increased levels of BAL NO_x, cellular ROS, and IL-6 in all groups that contained Ni on Day 6 were also comparable to that observed in rats treated with the soluble fraction of ROFA. Lymphocyte profiles in rats pre-treated with soluble Ni, alone or in combination with other metals, also followed the same pattern as the soluble ROFA sample, with increased T-cells infiltrating the lungs by Day 10. Although a large difference in bacterial lung burden was not observed at the later timepoints post-infection, as was observed when the soluble fraction of ROFA was administered, there still appeared to be an effect on the T-cell population. A reduction in T-cell cytokine production (IL-2 and IL-4) was observed in groups that were treated with soluble Ni, indicating that there may be adverse effects on the development of the adaptive immune response, and that could be attributed, at least in part, to the soluble Ni content.

The findings concerning the role of Ni in injury and lung inflammation are consistent with those of Dreher et al. (1997) and Kodavanti et al. (1998), where the investigators found that soluble Ni at concentrations present in ROFA produced a more severe inflammation than when in combination with soluble Fe or V. In the ROFA sample used in our study (Roberts et al., 2007), there was minimal amounts of V present in the soluble fraction; therefore, soluble V was not tested in the current study. Instead, in the present study, oxidant production and alterations in IL-10 production prior to infection correlated best with total soluble metal content. Also, treatment with Ni alone had potential for inducing increased ROS and NO, although not to the degree of the Total Metal mixture. It is possible that the existing inflammation, excess injury, and increased oxidative stress in the groups that contained soluble Ni may have predisposed rats in those groups to a greater susceptibility to intracellular bacterial infection.

Bacterial clearance from the lungs of rats pre-treated with the Total Mixture was significantly reduced on Day 6 when compared to all other groups that did not contain Ni, and body weight was also significantly reduced at the early time point post-infection in this group as well. This result was duplicated only with Soluble Ni alone, or in combination with the other metals. Interestingly, rats treated with Ni alone appeared to have a more severe infection than the Total Mixture, as indicated by significant loss of body weight and increased lung bacterial burdens out to Day 8. Also, the metal combination that produced the greatest increase in susceptibility to infection was the group that contained Ni, but lacked Al, and the group with the second worst infection lacked Fe. In addition, rats treated with Fe alone had the fastest clearance rate on Day 6. These findings suggest that soluble Al and Fe may act antagonistically to Ni in the immune response to L. monocytogenes.

The method of IT instillation of the soluble metals bypasses the upper respiratory tract (nasal cavity, pharynx, and upper trachea) and results in different deposition and clearance patterns from that of the more physiological exposure route of inhalation. Despite these differences, Hatch et al. (1981) have shown that the physical technique of IT instillation does not alter infectivity, and there was an agreement between dose-response curves generated by more toxic salts when comparing instillation versus inhalation. In addition, other Investigators that have examined the effects of soluble Ni on infectivity after inhalation are in agreement with the findings of this study. Adkins et al. (1979) found that inhalation of nickel chloride (NiCl₂) or nickel sulfate (NiSO₄) enhanced mortality in mice infected with Streptococcus pyogenes. Zelikoff et al. (2002) showed that inhalation of NiCl₂ or iron chloride (FeCl₂) resulted in increased infectivity in rats that had been previously IT-infected with S. pneumoniae. Inhalation of NiCl₂ was found to inhibit the clearance of the bacteria from the lungs, whereas FeCl₂ facilitated an increase in overall bacterial number.

The reduction in bacterial clearance at the early timepoint post-infection likely indicates that innate immune responders, such as resident AM or NK cells, are being inhibited by soluble Ni. The immune system as a target after soluble Ni treatment has been well documented, resulting in reduction in AM phagocytosis (ATSDR, 2005; Adkins et al., 1979; NTP, 1996a, b, and c), NK cell cytotoxicity (Goutet et al., 2000; Smialowicz, 1985), and lymphoid hyperplasia in the lung associated lymph nodes (NTP, 1996a,b,c). Numerous studies have shown that soluble Ni is capable of suppressing AM function (Camner et al., 1984; Goutet et al., 2000; Graham et al., 1975; Sunderman et al., 1989). In fact, Adkins et al. (1979) demonstrated that decreased phagocytosis by AM in response to inhalation of soluble Ni may be partly responsible for increased susceptibility to respiratory infection with S. pyogenes. In addition, an in vitro study by Graham et al., (1975) showed that NiCl₂ directly inhibited the phagocytic ability of AM. It is also likely that the soluble Ni in the current study reduced the ability of resident AM to effectively kill or clear the L. monocytogenes infection early after infection, as evidenced by higher bacterial lung burdens early after infection.

The pro-inflammatory cytokine cascade, which includes TNFα, IL-6, and IL-12, is a critical step in the innate immune response to L. monocytogenes and leads to a progression that results in initiation of cell-mediated immunity and further activation of innate responders (Seder and Gazzinelli, 1999). The elevation in IL-6 on Day 6 in rats pre-exposed to soluble Ni alone or in a metal combination may be indicative of increased lung injury and an exacerbated acute-phase response due to a delay in, or the suppression of, the early innate immune response that lead to the increased bacterial burden on Day 6.

These cytokines are also known to induce production of ROS and NO (antimicrobial agents) by AM and PMN, either directly or through the induction of IFNγ production (Billiau, 1996). Elevations in NO and ROS in the same groups at that time point suggests that soluble Ni induces an excessive oxidative burst in the lungs of rats post-infection. The increase in lung oxidant burden in these rats is likely due to the increased influx of PMN and monocytes into the lungs. In the current study, the effects of the elevation in oxidant burden in the lungs of rats pre-treated with soluble Ni are not clear. High levels of oxidants have been shown to have adverse effects on adaptive immune responses, such as altered T-cell growth, cytokine production, and survival (Blesson et al., 2002; Brant and Fabisiak, 2008; Guan et al., 2007; Hoffman et al., 2002, van der Veen et al., 2000); however, the current study does not provide enough direct evidence to draw a firm conclusion regarding this potential mechanism of toxicity.

In the current study, although T-cell numbers in the lung did not differ on Day 6, T-cell activity appeared to be inhibited, particularly at the early time points post-infection, as indicated by reduced levels of IL-2 and IL-4, cytokines involved in T-cell growth, differentiation, and proliferation. This response was similar, although not as persistent, as that observed after infection in rats pre-treated with the soluble fraction of ROFA in our previous study (Roberts et al., 2007). The findings concerning the inhibitory effects of soluble Ni on T-cells have also been observed by others. For example, Zelikoff et al. (2002) found that respiratory exposure to soluble Ni, at concentrations present in ambient air particulates, in uninfected rats altered the ability of splenic lymphocytes to proliferate, and T-cells appeared to be more sensitive to NiCl₂ than B-cells. Graham et al. (1978) also found that inhalation of NiCl₂ in mice diminished the antibody-producing capabilities of splenic lymphocytes.

Despite indications of T-cell suppression early after infection in the current study, on Day 10 there is a significant increase in the T-cell population in the lungs in all group containing Ni or after exposure to soluble Ni alone, attributable to both CD8⁺ and CD4⁺ T-cells. The reason for this is unclear. A similar response was observed after exposure to the soluble ROFA fraction followed by infection (Roberts et al., 2007). In that study, T-cell cytokine production remained low throughout the course of the infection, suggesting that despite increase T-cell populations in the lung, adverse effects of soluble metals on T-cells were still present.

Soluble Ni has been shown to induce lymphoid hyperplasia in the lung associated lymph nodes (NTP, 1996a, b and c), but it is not clear in either of the studies if this is the mechanism for the population increase, or if it is simply a heightened response due to increased bacterial burdens early on. Differences between the two studies may be due to complex interactions between the primary soluble metals and other soluble components present in ROFA. Also, the metal solutions as they exist in this study do not completely replicate the original solution.

In summary, ROFA is a complex mixture of soluble metals, sulfates, acids, fuel contaminants, and other unknown components combined with an insoluble, particulate carbon core. Our data indicate that soluble Ni, at the concentration present in a particular ROFA sample (Roberts et al., 2007), is likely responsible for increased lung injury and inflammation, decreased bacterial clearance, altered AM function, and adverse effects on T-cells that were observed after exposure to the soluble metals of a ROFA sample. Importantly, the soluble metal component, the concentration of those metals, and their bioavailability may vary and are dependent on the fuel source, the combustion conditions in the power plant (including temperature, pressure changes, gas velocities, and duct work), and efficiency of emission control devices (Burckle, 1977; El-Mogazi et al., 1988). In addition, depending on which soluble metals are present in a sample, metal-metal interactions may alter pulmonary immune responses and be the ultimate determinant as to whether a particular air pollutant will induce susceptibility to infection.

Acknowledgements

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper