As shown in , after a 5-week exposure to 100 ppb As in drinking water, mice infected with influenza A (H1N1) displayed a significant increase in morbidity. By day 8 postinfection (p.i.), the As-exposed mice displayed such severe morbidity (e.g., body weight decrease ≥ 20%) that those experimental groups were euthanized in compliance with institutional IACUC standards. Because of the severity of these responses, subsequent analyses focused on day 3 and day 7 p.i. In contrast, a parallel group of control mice infected with influenza but not exposed to As displayed a moderate weight loss, but then began to recover weight by day 10 p.i., with complete weight recovery by day 16 p.i. (). Exposure to As alone in the absence of viral infection did not influence weight or growth over the 5-week period, nor did anesthesia alone, with or without respiratory exposure vehicle, in either the control or As-exposed mice (data not shown). Thus, the increased morbidity was due to the combination of As in drinking water and influenza infection at an infectious dose at which mice not exposed to As recover.
Figure 1 Morbidity of influenza infection (measured by weight loss) in mice exposed to control water or water containing 100 ppb As. One experiment was conducted for days 0–16 (n = 6–8 per group), and three additional experimental repeats were (more ...)
Given that the inability to properly clear virus from the lung is positively correlated with increased risk of adverse outcomes, we examined the levels of influenza A virus in whole-lung homogenates using TCID50 determination at day 7 p.i. Relative to As-unexposed controls, the As-exposed mice exhibited a significant 10-fold increase in viral titers at this time point, correlating with their relative increase in morbidity ().
Figure 2 Viral titers for flu on day 7 p.i. in mice exposed to control water or water containing 100 ppb As. Whole-lung homogenates were assessed for viral titers by the TCID50 method. See “Materials and Methods” for experimental details. Values (more ...)
At day 7 p.i., obvious gross histologic changes in the As-exposed mice, including edema and hemorrhaging, could be observed by visual inspection of the whole lung (). We measured capillary leakage into the lungs by assaying albumin concentrations in the BALF. Albumin concentrations were significantly increased in the As-exposed versus unexposed mice at day 7 p.i. Albumin concentrations did not differ between control or As-exposed mice before infection or at day 3 p.i. (). Because an increased severity of influenza infection often correlates with a decrease in SpO2, we measured the oxygen saturation levels in infected control and As-exposed conscious mice at day 7 p.i. The As-exposed mice displayed a significant decrease in SpO2 levels compared with control mice (). The average SpO2 reading for the control mice infected with influenza was 95.2% (range, 93–97.5%), and the average SpO2 reading for As-exposed mice infected with influenza was 82.9% (range, 75–90%). Three of the six As-exposed mice had SpO2 readings considered to be dangerously low (i.e., ≤ 80%). Exposure to As in the absence of infection had no effect on peripheral oxygen saturation levels.
Figure 3 Effect of chronic exposure to As (100 ppb) in drinking water for 5 weeks followed by inoculation with influenza A. (A, B ) Gross histology of representative lungs from mice given control drinking water (Flu alone; A) and As in drinking water (Flu + As; (more ...)
We investigated cellular infiltration into the lungs at 36 hr and at day 3 and day 7 p.i. We previously reported that exposure to 100 ppb As for 5 weeks in uninfected mice did not induce changes in the total number of cells recovered from BALF, nor did it alter gross changes in lung histology (Andrew et al. 2007
; Kozul et al. 2009
). However, in the present experimental model, we observed significant differences in the total number of cells recovered postinfection from the BALF of control and As-exposed mice. At 36 hr and day 3 p.i., As-exposed mice had a significant decrease in the total number of cells recovered from the BALF compared with control mice. Conversely, at day 7 p.i., As-exposed mice had a significant increase in the number of cells in BALF () relative to controls.
Figure 4 Alteration in cell numbers at day 0, 36 hr, day 3, and day 7 p.i. in BALF of mice exposed to control water or water containing 100 ppb As (Flu + As) followed by inoculation with influenza A. (A) Number of viable nucleated cells recovered from BALF. ( (more ...)
To assess whether changes in total cell populations were accompanied by changes in the specific cell types recruited to the lung, we conducted a morphologic analysis of the cells after staining cytospin preparations. Representative cytospin preparations are shown in . The total number of macrophages () and neutrophils () were significantly decreased at the early stages of infection and increased at day 7 p.i. in As-exposed mice compared with controls, whereas lymphocytes were decreased in the As-exposed mice at day 3 p.i. but were similar to controls at day 7 p.i. (data not shown). In addition to these differences in cell number, the percentages of these different cell types within the total cells recovered from the BALF also changed, as shown in .
Figure 5 Effects of flu alone and flu plus chronic exposure to As (100 ppb) in drinking water shown at 36 hr, day 3, and day 7 p.i. in representative cytospin preparations. See “Materials and Methods” for experimental details. (A, C, E) Flu alone. (more ...)
Although the absolute number of lymphocytes was largely unaffected at day 7 p.i., when subtypes within the lymphocyte populations were analyzed by flow cytometry, CD8+ cells were significantly increased in terms of both cell percentages () and total cell numbers () in As-exposed mice relative to controls. The percentage of CD4+ T cells did not differ by As exposure (data not shown).
Figure 6 Effects of chronic exposure to As (100 ppb) in drinking water on CD8+ cells in BAL in response to influenza A. (A ) Representative fluorescence-activated cell sorting plots showing isolated BALF cells stained with fluorochrome-conjugated antibodies against (more ...)
We previously reported that As exposure decreased basal cytokine levels in the lungs of uninfected mice (Kozul et al. 2009
). In the present study we used a Bio-Plex assay to examine cytokine production in BALF at days 3 and day 7 p.i. We examined a panel of 10 cytokines: IL-1β, IL-6, MIP-1α, RANTES, MCP-1, IL-10, M-CSF, MIP-2, MIP-1β, and TNFα. With the exception of MIP-1α, cytokine production had the same general pattern. We observed no significant differences between the control and As-exposed mice at day 7 p.i. However, we observed a significant decrease in the As-exposed mice relative to controls in 9 of the 10 cytokines at day 3 p.i. (). In some cases, such as IL-1β and TNFα, cytokine production was below the limits of assay detection in the As-exposed mice. MIP-1α levels were below detection at day 3 p.i. in all mice but indicated a trending increase in the As-exposed animals relative to controls at day 7 p.i. (data not shown).
Figure 7 Effects of chronic exposure to As (100 ppb) in drinking water on cytokine and chemokine levels in BALF at day 3 and day 7 p.i. (A) RANTES. (B) MIP-2. (C) MCP-1. See “Materials and Methods” for experimental details. Values shown are mean (more ...)
The early recruitment of DCs to the lymph node is essential for the proper initiation of an immune response, and we previously reported that As exposure decreases the expression of genes involved in cell adhesion and migration (Kozul et al. 2009
). Therefore, to determine whether the migration of DCs to the lymph node was impaired by As exposure, we analyzed the MLNs for DC populations at day 3 p.i. Single-cell suspensions from MLNs were assessed for the DC markers CD11c+
. We observed a significant decrease in the total number of cells recovered from the MLNs in the As-exposed mice relative to controls at day 3 p.i. (). Flow cytometry staining for CD11c+
DCs also indicated that these DC populations were significantly decreased in MLNs of As-exposed mice compared with controls at day 3 p.i. (). We observed no differences in the intensity of the staining (data not shown).
Figure 8 Effect of chronic exposure to As (100 ppb) in drinking water on DCs. See “Materials and Methods” for experimental details. Total number of MLN cells (A), CD11c+/CD103+ cells (B), and CD11c+/B220+ cells (C); values shown are mean ± (more ...)
We then assessed effects of As on DC migration capacity, using primary bone marrow DCs isolated from control mice and mice exposed to 100 ppb As for 5 weeks (in the absence of influenza infection). Cells were cultured for 7 days in the presence of GM-CSF to produce immature DCs. Relative to controls, cells isolated from As-exposed mice had a significant decrease in migration capability toward ADP in a transwell assay (), indicating that As alone can compromise aspects of immune cell function and that this then manifests as a significantly altered innate immune response after viral infection.