Silica-induced Chronic Inflammation Promotes Lung Carcinogenesis in the Context of an Immunosuppressive Microenvironment (2024)

Animal Model and Reagents

Eight-week-old female Balb/c mice were purchased from Harlan Laboratories (Indianapolis, IN). All mouse experiments were approved by the Ethics Committee on Animal Research of the University of Navarra.

Both NDMA (Sigma, St Louis, MO) and silica (US Silica Co, Berkeley Springs, WV) were resuspended in neutral-buffered saline and sonicated for 30 seconds. Mice were treated with oropharyngeal aspiration with a single dose of 7.5 mg/kg NDMA, 9 mg of silica, a combination of both, or saline in a final volume of 90 µl, as previously described [19]. The NDMA dose was determined as the minimal amount of carcinogen able to induce lung tumorigenesis in mice without toxic side effects (data not shown). The amount of instilled silica was chosen as a dose that triggers lung inflammation with minimal toxic side effects (data not shown).

Samples

Plasma samples and mouse lungs were obtained on months 2, 4, 6.5, 9, and 12 after treatment. Representative lung lobes were ligated, sectioned, and rapidly frozen in liquid nitrogen. The remaining material was fixed by immersion in 4% formaldehyde for 24 hours, embedded in paraffin, sectioned along the main bronchial axis, and cut at 3-µm thickness for immunohistochemical analysis.

At the time of analysis, 20-µm thickness cryostat sections of the frozen lobe were obtained and lesions were microdissected by laser capture microdissection (Leica, Wezlar, Germany). RNA and DNA from microdissected tissues were extracted using RNeasy Kit (Qiagen, Valencia, CA) and DNA Mini Kit (Qiagen), respectively.

Lesion Analysis

Tumor incidence was assessed microscopically in a representative set of six sections per mouse including all of the lung lobes. Tumor areas were calculated using the formula: A (µm²) = π x L x W, where L is the largest diameter and W is the largest diameter perpendicular to L. Tumors were classified as adenoma (solid or papillary) or adenocarcinoma, in accordance with the criteria previously published [20].

Cytokine Measurement

The concentrations of TNF-α, macrophage inflammatory protein-1α (MIP-1α), interferon gamma-induced protein 10 (IP-10), interleukin-1 beta (IL-1β), IL-6, granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-10, IL-13, interferon gamma (INF-γ), and IL-12 (p40) were determined in plasma using a mouse Milliplex MAP Kit (Millipore, Madrid, Spain) following the manufacturer's instructions.

Immunostaining

Immunostaining was performed using antibodies raised against mouse p-H2AX (1:50; Cell Signaling Technology, MA), mouse pro-SPC (1:500; Abcam, Cambridge, MA), CC10 (1:4000; Upstate, Billerica, MA), mouse Ki-67 (1:25;Neomarkers, Fremont, CA), mouse CD31 (1:40; Dianova, Hamburg, Germany), mouse cleaved caspase-3 (1:100; Cell Signaling Technology, Danvers, MA), mouse forkhead box P3 (FOXP3; 1:50; ebiosciences, Frankfurt, Germany), and Retnla (1:250; Abcam). Heat-mediated antigen retrieval was used for antibodies against Ki-67, CD31, cleaved caspase-3, FOXP3, and Retnla by microwave exposure for 15 minutes at 800 W and 15 minutes at 375 Win 10 mM sodium citrate (pH 6). Incubations with anti-CC10 and anti-Retnla antibodies were followed by incubation with the indirect avidin-biotin-peroxidase system (Dako, Barcelona, Spain). In the case of p-H2AX, pro-SPC, Ki-67, CD31, cleaved caspase-3, and FOXP3 immunostaining, Envision signal enhancement method (Dako) was used. Slides were washed in TBS and peroxidase activity was detected with 3-3′-diaminobenzidine (Dako). Slides were counterstained with Harris' hematoxylin. Omission of the primary antibodies was used as a negative control in one slide from each staining series. p-H2AX, pro-SPC, Ki-67, cleaved caspase-3, and FOXP3 immunostaining was evaluated by two independent and blinded observers and scored as the mean of the percentages of stained cells in 20 high-power fields (x400). Quantification of CD31 staining was calculated as the mean of the ratio between CD31-positive area and total area using the Analysis software (Olympus, Barcelona, Spain).

Immunohistochemistry to detect mouse CD4, CD8, CD3, CD20, myeloperoxidase, and CD68 was performed using FLEX Ready-to-Use specific antibodies (Dako) according to the manufacturer's conditions.

Mutation Analysis

Mutations in K-RAS (codons 12 and 61), STK11 (exons 1–5, 8, and 9), and TP53 (exons 5, 6, 7, 8, and 9) genes were analyzed in DNA from microdissected malignant tissues by polymerase chain reaction (PCR) amplification and sequencing. Genes were analyzed in DNA from microdissected malignant tissues. DNA was amplified by PCR using the primers shown in Table W1. After PCR amplification, exons upstream and downstream were sequenced using a BigDye Terminator 3.1 sequencing instrument (Applied Biosystems, Carlsbad, CA); sequences were then compared to a control sequence with SeqScape v2.5 (Applied Biosystems). When a mutation was detected, an additional amplification and sequencing experiment was performed to confirm the result.

Expression Arrays

Gene expression profiles were analyzed in 16 individual lesions (9 adenomas and 7 adenocarcinomas) from NDMA-silica-treated mice and 10 individual lesions (5 adenomas and 5 adenocarcinomas) from NDMA-only-treated mice. RNA was obtained from microdissected tissue of at least eight consecutive 20-µm sections from each frozen tissue block and retrotranscribed. Gene expression was analyzed using a microarray platform containing 28,122 mouse genes (Oryzon Genomics, Barcelona, Spain) following the manufacturer's instructions. For control, cDNA from normal lung was used in all hybridizations.

Normalization of every individual array was performed using the lowest scatterplot smoother implemented in Matlab software as previously described [21]. All of the probes of the array, aside from controls, were used for normalization; the fraction of the data used for smoothing at each point was set to 5%. After calculation of the fold change values, a filtering process was applied to eliminate low expression spots. The criterion of |FC| > 2 in at least one of the samples reduced the number of probes to 33,611 (corresponding to approximately 11,200 genes).

Statistics

Differences between two groups were analyzed using the Mann-Whitney U test. To determine differences between more than two experimentally independent groups, Kruskal-Wallis H test was chosen. Chi-square test with Fisher correction was applied to study differences between mutational profiles.

Differential gene signatures in expression array experiments were identified using Linear Models for Microarray data [22]. Genes were selected as significant using a B statistic cutoff (B > 0). A threshold of |Z|O > 2.3 (P value < .01) was chosen to select genes for further analysis. R/Bioconductor [23] was used for preprocessing and statistical analysis. Functional enrichment analysis of Gene Ontology (GO) categories was conducted using Database for Annotation, Visualization and Integral Discovery [24], and the biologic interpretation was complemented through use of Ingenuity Pathway Analysis (Ingenuity Systems, Redwood City, CA), in which the database includes manually curated and fully traceable data derived from literature sources.

Silica Strongly Promotes an Inflammatory Process, DNA Damage, and Pulmonary Fibrosis in Mouse Lungs

We first performed a histologic evaluation of the inflammatory response in lungs from mice treated with silica (Figure 1, A–O). Lung chronic inflammation was confirmed 2, 4, 6.5, 9, and 12 months after treatment in all silica-treated mice irrespective of NDMA administration; moreover, time-dependent progression of inflammation was clearly observed. In particular, 2 months after silica administration, nodular silicotic granulomas accompanied by marked epithelial hyperplasia were found in central areas close to branching and terminal bronchioles in all lung lobes from all treated mice (Figure 1, A, F, and K). Four months after treatment, inflamed regions were observed in the same lung areas; however, the size of the lesions was markedly larger. Granulomas were confluent and included large acellular centers with peripheral infiltration of lymphocytes. Moderate bronchiolar hyperplasia and hyperplastic type II pneumocytes were also observed surrounding the silicotic lesions (Figure 1, B, G, and L). After 6.5 and 9 months, silicotic granulomas remained similar in size to those observed after 4 months and exhibited larger central necrotic areas. Parenchyma between silicotic nodules presented increased alveolar proteinosis and thickened alveolar septa; epithelial hyperplasia persisted in the vicinity as well as between the granulomas, and extensive areas of hyperplastic goblet cells were also observed. Interestingly, there was a marked increase of macrophages and lymphocytes in the perivascular and peribronchiolar areas in comparison to lesions at month 4 (Figure 1, C, D, H, I, M, and N). Granulomas at month 12 exhibited larger central necrotic areas. At this time point, the relative abundance of epithelial hyperplasia was reduced, whereas macrophages and lymphocytes persisted in perivascular and peribronchiolar areas (Figure 1, E, J, and O). To characterize the inflammatory infiltrate in lungs from NDMA-only-, silica-, and NDMA-silica-treated mice, the presence of CD4 and CD8 T lymphocytes, total T lymphocytes, B lymphocytes, polymorphonuclear leukocytes, and monocyte/macrophage cells was evaluated by immunohistochemistry 2, 4, and 6.5 months after treatment. All these cell populations were significantly increased in lungs from silica-treated mice irrespective of NDMA administration in all of the experimental time points analyzed (Table 1 and Figure W1). NDMA administration did not affect inflammatory structural phenotype and leukocyte infiltration in lungs from mice treated with both silica and saline. In addition to the histopathologic analysis, we measured TNF-α, MIP-1α, IP-10, IL-1β, IL-6, GM-CSF, IL-10, IL-13, INF-γ, and IL-12 (p40) plasma concentrations. According to previous findings [25], silica-treated mice presented significantly higher levels of the pro-inflammatory cytokines TNF-α, MIP-1α, and IP-10 in comparison to the non-silica-treated counterparts. No significant differences in IL-1β, IL-6, GM-CSF, IL-10, IL-13, INF-γ, and IL-12 (p40) plasma levels were found between experimental groups (Figure W2). However, we found a significant increase of mRNA expression of IL-6 and GM-CSF in lungs from NDMA-silica when compared to NDMA-only-treated mice 12 months after treatment, whereas no significant differences between both experimental groups were found for IL-1β (Figure W3).

Open in a separate window

Figure 1

Effects of oropharyngeal administration of silica on mouse lungs. The left panel shows the histologic analysis of lung silica-induced chronic inflammation in sections stained with hematoxylin and eosin. Silica alone-treated mice were sacrificed at months 2 (n = 4), 4 (n = 4), 6.5 (n = 4), 9 (n = 4), and 12 (n = 4) after treatment. Lungs were harvested and all the lung lobes were analyzed (one field per lobe). (A–E) Silicotic granulomas at different experimental points following silica treatment. Insets show the details of the granulomatous proliferations. (F–O) Alveolar (F–J) and bronchiolar (K–O) hyperplasia in areas close to the granulomas at 2, 4, 6.5, 9, and 12 months after silica treatment. Alveolar proteinosis (J), hyperplastic goblet cells (N), and perivascular, peribronchiolar inflammatory reaction (asterisk, O) and lung fibrotic lesions (data not shown) are common features in this model. Images were obtained at the same magnification. The right panel includes representative examples of p-H2AX staining in lungs from NDMA-only- and NDMA-silica-treated mice. Samples were considered positive when six or more cells presented positive immunoreactivity in at least two of three fields analyzed (x40).

To further evaluate the effects of silica instillation in mouse lungs, we analyzed the phosphorylation of histone H2AX to indicate the level of silica-mediated DNA damage present within lung epithelial cells. One hundred percent of lungs from NDMA-silica-treated mice presented positive p-H2AX staining in hyperplastic areas from bronchiolar epithelium (n = 6; Figure 1, right panel). This percentage was reduced over time to 50% (month 4; n = 8) and 33% (month 6.5; n = 3) and disappeared after month 9 (n = 18). Interestingly, p-H2AX was only observed in hyperplastic tissue areas near to granulomatous areas. None of the lungs from NDMA-only-treated (n = 28) or saline-treated (n = 20) mice were positive for p-H2AX. These results indicate that silica promotes DNA damage in lungs. Another important consequence of silica instillation was, as previously described [19], the development of fibrotic areas extending to the peripheral and distal regions of the lung in all of the experimental time points (data not shown). As previously described [18], we did not detect adenomas nor adenocarcinomas in mice treated with silica alone.

Chronic Inflammation Increases the Incidence and Multiplicity of NDMA-Induced Lung Tumors

We next analyzed whether silica-mediated chronic lung inflammation influences the carcinogenic effects of NDMA. We examined the preneoplastic and neoplastic lesions in mice treated with saline, silica, a poorly tumorigenic dose of NDMA, and combination of the same dose of NDMA and silica. Silica-treated mice developed chronic inflammation; however, no lung adenomas and adenocarcinomas were detected at any time during the experiment. However, 75% of mice treated with NDMA exhibited adenomas after 6.5 months. Premalignant adenomas were detected earlier in mice treated with the combination of silica and NDMA compared with NDMA-only-treated mice (4 months after treatment vs 6.5 months, respectively). Moreover, the incidence and multiplicity of these adenomas were markedly higher in all of the experimental time points in the combined treatment group compared with NDMA-only treatment (Figure 2A). Interestingly, mice treated with the combination of silica and NDMA also demonstrated a three-fold increase in both incidence and multiplicity of lung malignant adenocarcinomas compared with mice treated with NDMA only (Figure 2B). The ratios of adenocarcinoma to adenoma were statistically equivalent: 0.20 ± 0.3 in the combined treatment and 0.12 ± 0.27 in NDMA only (P = .36). Taken together, these data suggest that silica-mediated local chronic inflammation significantly reduces the latency period for the appearance of preneoplastic lung lesions and strongly promotes lung carcinogenesis.

Open in a separate window

Figure 2

Incidence and multiplicity of adenomas and lung tumors in mice treated with NDMA-only or NDMA-silica. (A) Evaluation of adenomatous structures at different experimental time points. (B) Quantitation of adenocarcinomas at month 12 after treatment. NDMA-only-treated mice were analyzed at months 4 (n = 4), 6.5 (n = 4), 9 (n = 4), and 12 (n = 19) after treatment. NDMA-silica-treated mice were analyzed at months 4 (n = 8), 6.5 (n = 8), 9 (n = 8), and 12 (n = 21) after treatment. Incidence rate was calculated as the percentage of mice with at least one lung lesion. Multiplicity is represented as the average of the number of lesions found in mouse lungs ± SEM. P values were calculated using the Kruskal-Wallis test in adenomas, whereas the test chosen for the analysis of adenocarcinomas was the Mann-Whitney U test.

Lesions in Mice Treated with NDMA and the Combination of NDMA with Silica Exhibit Similar Histologic and Molecular Features

Despite an increase in lung cancer incidence and multiplicity in mice treated with the combination of NDMA and silica, we did not observe significant histologic differences between the lesions present in the two experimental groups (Figure 3, A–H). Immunohistochemical analysis of the expression of CC10, an airway Clara cell marker, and pro-SPC, a type II pneumocyte marker, revealed that adenomas and adenocarcinomas from NDMA-only- and NDMA-silica-treated mice were consistently positive for pro-SPC and negative for CC10, whereas non-malignant normal and hyperplastic bronchiolar epithelium exhibited positive staining for CCL10, and only hyperplastic alveoli were positive for pro-SCP (Figure 3, I–P). These data suggest that tumors are potentially derived from hyperplastic type II cells or less committed cells following the type II differentiation pathway. Furthermore, we found no statistical differences in the average size of adenomas (P = .405) or adenocarcinomas (P = .256) between both experimental groups. Immunohistochemical analysis of Ki-67 and CD31 revealed that there was no increase in the proliferative capacity and in the vascular density of adenomas from NDMA-silica when compared to NDMA-only-treated mice. However, a statistically significant increase of active caspase-3 immunostaining was found in adenomas from NDMA-silica-treated mice when compared to those from NDMA-only-treated mice (P < .001), suggesting that silica promotes apoptosis within adenomas (Figure W4).

Open in a separate window

Figure 3

Histopathologic features of lung lesions from NDMA and NDMA-silica-treated mice. All lesions classified as adenomas and adenocarcinomas in Figure 2 were analyzed. (A–H) Hematoxylin and eosin staining of bronchiolar (A) and alveolar (B) epithelium, a solid adenoma (C), and a papillary adenocarcinoma (D) from NDMA-treated mice and bronchiolar hyperplasia (E), alveolar epithelium (F), a solid adenoma (G), and papillary adenocarcinoma (H) from NDMA-silica-treated mice. (I–P) Immunohistochemical analysis of pro-SPC and CC10 in tumors and adenomas from NDMA-silica- and NDMA-treated mice. Slides were counterstained with Harris' hematoxylin. All adenomas (I, M) and adenocarcinomas (J, N) exhibited positive staining for pro-SPC in both NDMA-silica- and NDMA-treated mice. Adenomas (K, O) and adenocarcinomas (L, P) were negative for CC10 staining in both experimental groups, whereas normal bronchiolar epithelial cells close to the tumors (*) exhibited CC10 positive staining. Scale bar, 100 µm.

Silica-mediated Chronic Inflammation Modifies the K-RAS Mutation Profile in NDMA-Induced Lung Tumors

We next evaluated the presence of somatic mutations in oncogenes and tumor suppressor genes associated with lung carcinogenesis in humans and mice: K-RAS, STK11, and p53. In particular, we analyzed activating mutations in K-RAS (codons 12 and 61) and inactivating mutations in TP53 (exons 5–9) and STK11 (exons 1–5, 8, and 9). This analysis was performed on DNA from lung adenocarcinomas of mice treated with NDMA-silica (n = 10) or NDMA only (n = 5). Of the 15 adenocarcinomas studied, 13 exhibited mutation in K-RAS, either in codon 12 or codon 61. Importantly, the mutational pattern was different between the NDMA-only and NDMA-silica-induced lesions. Specifically, the K-RAS mutations in tumors that originated following NDMA-silica treatment was primarily G-to-A transition (G12D) in codon 12, whereas A-to-G transition (Q61R) in codon 61 was the most frequent alteration in adenocarcinomas from mice treated with NDMA only (P = .031; Figure 4). None of the adenocarcinomas analyzed presented somatic mutations in p53 and STK11 hotspots.

Open in a separate window

Figure 4

K-RAS mutational analysis of adenocarcinomas from the NDMA-silica- and NDMA-only-treated mice. (A) Total number of adenocarcinomas with either K-RAS codon 12 or 61 mutations in NDMA-silica- and NDMA-treated mice. (B) Representative electropherograms showing G-to-A transition in K-RAS codon 12 and A-to-G transition in codon 61. Differences between groups were analyzed using the Fisher exact test.

Silica-induced Chronic Inflammation Exhibits a Pro-inflammatory Gene Expression Profile

To further characterize the effects of silica-mediated chronic inflammation in lung carcinogenesis, we performed microarray analysis. Minimal differences in gene expression were found in adenocarcinomas from NDMA-silica- and NDMA-only-treated mice (Table W2). However, marked gene expression differences were observed between NDMA-silica and NDMA-only adenomas. In particular, 36 genes were upregulated and 13 downregulated in adenomas from mice treated with NDMA-silica compared to mice treated with NDMA only (Figure 5A and Table 2). Interestingly, a substantial group of the most relevant upregulated genes in the NDMA-silica model, such as Saa3, Retnla, Irg1, Serpina3n, Adam8, and Lta, plays a key role in inflammation. To confirm this finding at the protein level, we chose one of these genes, Retnla, upon commercial availability of commercial anti-mouse specific protein antibody. Retnla expression was analyzed by immunohistochemistry (Figure 5B). Most of the adenomas (17 of 23) from NDMA-silica-treated mice exhibited positive staining for Retnla, whereas adenomas (11 of 16) from NDMA-only-treated mice were negative (P = .011). Functional analysis of the genes differentially expressed in adenomas and adenocarcinomas from both experimental groups demonstrated differences in the enriched GO categories. In adenomas, the presence of silica-mediated inflammation affected GO categories associated with inflammation and immune response, whereas enriched categories in adenocarcinomas were unrelated to these processes (Table W3).

Open in a separate window

Figure 5

Differentially expressed genes after a comparison of gene expression profiles of adenomas from NDMA and NDMA-silica-treated mice. (A) Matrix expression of the representative signature for the set of genes differentially expressed (B > 0; |log FC| > 1.5). Each line represents the gene expression profile of a mixture of RNA from at least two different adenomas (total n = 14). Red and green represent upregulated and downregulated genes, respectively, when compared to the corresponding normal lung tissue expression profile. (B) Retnla negative staining in an NDMA adenoma (left panel) and Retnla positive staining in an NDMA-silica adenoma located close to silica-induced lung inflammation (right panel). Retnla staining was performed in adenomas from 10 mice treated with NDMA and 16 mice treated with NDMA-silica. Differences between groups were analyzed using the Fisher exact test.

Silica-induced Chronic Inflammation Promotes an Immunosuppressive Microenvironment in Adenomas

Malignant transformation may be associated with an immunosuppressive microenvironment. Using microarray expression data, we analyzed whether silica-induced chronic inflammation generates an immunosuppressive microenvironment in adenomas and adenocarcinomas in NDMA-treated mice (Figure 6). Genes associated with immunosuppression, such as PD-1, transforming growth factor-β1 (TGF-β1), lymphocyte-activation gene 3 (LAG3), monocyte chemotactic protein 1 (MCP-1), or FOXP3, presented higher expression levels in adenomas from NDMA-silica-treated mice compared to mice treated with NDMA only (Figure 6A). No differences were observed in the expression of these genes among adenocarcinomas from both experimental groups (data not shown). Additionally, we assessed the presence of FOXP3-positive cells within adenomas from mice treated with NDMA-silica (n = 15) and NDMA only (n = 15) using immunohistochemistry. All FOXP3-positive cells analyzed presented lymphocyte morphology. Positive FOXP3-positive lymphocytes were counted separately in the lesions and in the immediate peripheral lung parenchyma tissue (the “capsule” of the lesion). We did not observe FOXP3- positive lymphocytes within adenomas from NDMA-only-treated mice, whereas 7.1% of adenomas from NDMA-silica-treated mice exhibited at least one FOXP3-positive lymphocyte (Figure 6B). We analyzed the FOXP3-positive cells present at the periphery of adenomas; one hundred percent of adenomas from NDMA-silica-treated mice exhibited at least one FOXP3-positive peripheral lymphocyte, whereas only 16.5% of adenomas generated following NDMA-only treatment exhibited at least one FOXP3-positive lymphocyte in the peritumoral area (Figure 6B).

Open in a separate window

Figure 6

Expression of different immunomodulators in adenomas from NDMA-silica- and NDMA-treated mice. (A) Signature of immunosuppressive genes expressed in adenomas from NDMA-silica-treated mice compared to NDMA-treated mice. Each line represents the gene expression profile of a mixture of RNA from at least two different adenomas (total n = 14). Upregulated and downregulated genes are shown in red and green, respectively. (B) Percentage of lesions with FOXP3-positive cells within adenomas and adenoma peripheral areas (“capsule”) from NDMA-silica-treated (n = 15) and NDMA-treated (n = 15) mice. (C) Percentage of FOXP3-positive lymphocytes to total lymphocytes counted on lymphoid clusters located in non-injured lung parenchyma areas from NDMA-silica- and NDMA-treated mice. Values are represented as the means ± SEM. (D) FOXP3 immunohistochemical staining of lymphoid clusters from NDMA- and NDMA-silica-treated mice. The P value was calculated using the Mann-Whitney U test.

Finally, we investigated whether silica-induced chronic inflammation increases the number of FOXP3-positive lymphocytes in distant lung tissue. We analyzed the presence of FOXP3-positive lymphocytes in lymphoid clusters located in non-injured lung areas distant from adenomas. Twelve percent of NDMA-silica lymphocytes stained positive for FOXP3 in these distant clusters, whereas only 2.7% were positive in the NDMA treatment (Figure 6, C and D). Taken together, these results demonstrate that silica-mediated chronic inflammation promotes the development of an immunosuppressive microenvironment.

Supplementary Figures and Tables:

¹This work was supported by the UTE project CIMA and the Spanish Ministries of Health/Economy and Competitivity (ISCIII: RTICC RD12/0036/0040, FIS-PI 04/2128, and 10/00166). No potential conflicts of interest declared.

²This article refers to supplementary materials, which are designated by Tables W1 to W3 and Figures W1 to W4 and are available online at www.neoplasia.com.

1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J Clin. 2009;59:225–249. [PubMed] [Google Scholar]

2. Boffetta P. Epidemiology of environmental and occupational cancer. Oncogene. 2004;23:6392–6403. [PubMed] [Google Scholar]

3. Donaldson K, Poland CA. Inhaled nanoparticles and lung cancer—what we can learn from conventional particle toxicology. Swiss Med Wkly. 2012;142:w13547. [PubMed] [Google Scholar]

4. Houghton AM, Mouded M, Shapiro SD. Common origins of lung cancer and COPD. Nat Med. 2008;14:1023–1024. [PubMed] [Google Scholar]

5. Xu J, Yin Z, Gao W, Liu L, Wang R, Huang P, Yin Y, Liu P, Yu R, Shu Y. Meta-analysis on the association between nonsteroidal anti-inflammatory drug use and lung cancer risk. Clin Lung Cancer. 2012;13:44–51. [PubMed] [Google Scholar]

6. Dannenberg AJ, Subbaramaiah K. Targeting cyclooxygenase-2 in human neoplasia: rationale and promise. Cancer Cell. 2003;4:431–436. [PubMed] [Google Scholar]

7. de Visser KE, Eichten A, Coussens LM. Paradoxical roles of the immune system during cancer development. Nat Rev Cancer. 2006;6:24–37. [PubMed] [Google Scholar]

8. Wistuba II, Gazdar AF. Lung cancer preneoplasia. Annu Rev Pathol. 2006;1:331–348. [PubMed] [Google Scholar]

9. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–674. [PubMed] [Google Scholar]

10. Du C, Wang Y. The immunoregulatory mechanisms of carcinoma for its survival and development. J Exp Clin Cancer Res. 2009;30:12. [PMC free article] [PubMed] [Google Scholar]

11. Melief CJ. Cancer immunotherapy by dendritic cells. Immunity. 2008;29:372–383. [PubMed] [Google Scholar]

12. Sinha P, Clements VK, Bunt SK, Albelda SM, Ostrand-Rosenberg S. Cross-talk between myeloid-derived suppressor cells and macrophages subverts tumor immunity toward a type 2 response. J Immunol. 2007;179:977–983. [PubMed] [Google Scholar]

13. Brahmer JR, Tykodi SS, Chow LQ, Hwu WJ, Topalian SL, Hwu P, Drake CG, Camacho LH, Kauh J, Odunsi K, et al. Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med. 2012;366:2455–2465. [PMC free article] [PubMed] [Google Scholar]

14. Topalian SL, Hodi FS, Brahmer JR, Gettinger SN, Smith DC, McDermott DF, Powderly JD, Carvajal RD, Sosman JA, Atkins MB, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443–2454. [PMC free article] [PubMed] [Google Scholar]

15. Farago AF, Snyder EL, Jacks T. SnapShot: Lung cancer models. Cell. 2012;149:246–246.e1. [PubMed] [Google Scholar]

17. Leung CC, Yu IT, Chen W. Silicosis. Lancet. 2012;379:2008–2018. [PubMed] [Google Scholar]

18. Castranova V, Wallace WE, Vallyathan V. Silica and Silica-Induced Lung Diseases. Boca Raton, FL: Informa Healthcare, CRC; 1995. [Google Scholar]

19. Lakatos HF, Burgess HA, Thatcher TH, Redonnet MR, Hernady E, Williams JP, Sime PJ. Oropharyngeal aspiration of a silica suspension produces a superior model of silicosis in the mouse when compared to intratracheal instillation. Exp Lung Res. 2006;32:181–199. [PMC free article] [PubMed] [Google Scholar]

20. Nikitin AY, Alcaraz A, Anver MR, Bronson RT, Cardiff RD, Dixon D, Fraire AE, Gabrielson EW, Gunning WT, Haines DC. Classification of proliferative pulmonary lesions of the mouse: recommendations of the mouse models of human cancers consortium. Cancer Res. 2004;64:2307–2316. [PubMed] [Google Scholar]

21. Yang YH, Dudoit S, Luu P, Speed TP. Normalization for cDNA microarray data. In: Bittner ML, Chen Y, Dorsel AN, Dougherty ER, editors. Microarrays: Optical Technologies and Informatics. Bellingham, WA: SPIE Press; 2001. Vol. 4266 of Proceedings of SPIE. [Google Scholar]

22. Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol. 2004;3:46–69. Article 3. [PubMed] [Google Scholar]

23. Gentleman R, Carey V, Huber W, Irizarry R, Dudoit S. Bioinformatics and computational biology solutions using R and bioconductor. New York, NY: Springer; 2005. pp. 49–69. [Google Scholar]

24. Huangda W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4:44–57. [PubMed] [Google Scholar]

25. Pryhuber GS, Huyck HL, Baggs R, Oberdorster G, Finkelstein JN. Induction of chemokines by low-dose intratracheal silica is reduced in TNFR I (p55) null mice. Toxicol Sci. 2003;72:150–157. [PMC free article] [PubMed] [Google Scholar]

26. Matsuzaki H, Maeda M, Lee S, Nishimura Y, Kumagai-Takei N, Hayashi H, Yamamoto S, Hatayama T, Kojima Y, Tabata R, et al. Asbestosinduced cellular and molecular alteration of immunocompetent cells and their relationship with chronic inflammation and carcinogenesis. J Biomed Biotechnol. 2012;2012:492608. [PMC free article] [PubMed] [Google Scholar]

27. Driscoll KE, Lindenschmidt RC, Maurer JK, Higgins JM, Ridder G. Pulmonary response to silica or titanium dioxide: inflammatory cells, alveolar macrophage-derived cytokines, and histopathology. Am J Respir Cell Mol Biol. 1990;2:381–390. [PubMed] [Google Scholar]

28. Malkinson AM, Koski KM, Evans WA, Festing MF. Butylated hydroxytoluene exposure is necessary to induce lung tumors in BALB mice treated with 3-methylcholanthrene. Cancer Res. 1997;57:2832–2834. [PubMed] [Google Scholar]

29. Bojan F, Nagy A, Herman K. Effect of butylated hydroxytoluene and paraquat on urethan tumorigenesis in mouse lung. Bull Environ Contam Toxicol. 1978;20:573–576. [PubMed] [Google Scholar]

30. Barta P, Van Pelt C, Men T, Dickey BF, Lotan R, Moghaddam SJ. Enhancement of lung tumorigenesis in a Gprc5a Knockout mouse by chronic extrinsic airway inflammation. Mol Cancer. 2012;11:4. [PMC free article] [PubMed] [Google Scholar]

31. Takahashi H, Ogata H, Nishigaki R, Broide DH, Karin M. Tobacco smoke promotes lung tumorigenesis by triggering IKKβ- and JNK1-dependent inflammation. Cancer Cell. 2010;17:89–97. [PMC free article] [PubMed] [Google Scholar]

32. Mass MJ, Jeffers AJ, Ross JA, Nelson G, Galati AJ, Stoner GD, Nesnow S. Ki-ras oncogene mutations in tumors and DNA adducts formed by benz[j]aceanthrylene and benzo[a]pyrene in the lungs of strain A/J mice. Mol Carcinog. 1993;8:186–192. [PubMed] [Google Scholar]

33. You M, Candrian U, Maronpot RR, Stoner GD, Anderson MW. Activation of the Ki-ras protooncogene in spontaneously occurring and chemically induced lung tumors of the strain A mouse. Proc Natl Acad Sci USA. 1989;86:3070–3074. [PMC free article] [PubMed] [Google Scholar]

34. Nesnow S, Ross JA, Nelson G, Wilson K, Roop BC, Jeffers AJ, Galati AJ, Stoner GD, Sangaiah R, Gold A, et al. Cyclopenta[cd]pyrene-induced tumorigenicity, Ki-ras codon 12 mutations and DNA adducts in strain A/J mouse lung. Carcinogenesis. 1994;15:601–606. [PubMed] [Google Scholar]

35. Nuzum EO, Malkinson AM, Beer DG. Specific Ki-ras codon 61 mutations may determine the development of urethan-induced mouse lung adenomas or adenocarcinomas. Mol Carcinog. 1990;3:287–295. [PubMed] [Google Scholar]

36. Wessner LL, Fan M, Schaeffer DO, McEntee MF, Miller MS. Mouse lung tumors exhibit specific Ki-ras mutations following transplacental exposure to 3-methylcholanthrene. Carcinogenesis. 1996;17:1519–1526. [PubMed] [Google Scholar]

37. Ji H, Houghton AM, Mariani TJ, Perera S, Kim CB, Padera R, Tonon G, McNamara K, Marconcini LA, Hezel A, et al. K-ras activation generates an inflammatory response in lung tumors. Oncogene. 2006;25:2105–2112. [PubMed] [Google Scholar]

38. Okudela K, Woo T, Kitamura H. KRAS gene mutations in lung cancer: particulars established and issues unresolved. Pathol Int. 2010;60:651–660. [PubMed] [Google Scholar]

39. Hegi ME, Soderkvist P, Foley JF, Schoonhoven R, Swenberg JA, Kari F, Maronpot R, Anderson MW, Wiseman RW. Characterization of p53 mutations in methylene chloride-induced lung tumors from B6C3F1 mice. Carcinogenesis. 1993;14:803–810. [PubMed] [Google Scholar]

41. Biswas S, Guix M, Rinehart C, Dugger TC, Chytil A, Moses HL, Freeman ML, Arteaga CL. Inhibition of TGF-β with neutralizing antibodies prevents radiation-induced acceleration of metastatic cancer progression. J Clin Invest. 2007;117:1305–1313. [PMC free article] [PubMed] [Google Scholar]

42. Qian BZ, Li J, Zhang H, Kitamura T, Zhang J, Campion LR, Kaiser EA, Snyder LA, Pollard JW. CCL2 recruits inflammatory monocytes to facilitate breast-tumour metastasis. Nature. 2011;475:222–225. [PMC free article] [PubMed] [Google Scholar]

43. Fridlender ZG, Buchlis G, Kapoor V, Cheng G, Sun J, Singhal S, Crisanti MC, Wang LC, Heitjan D, Snyder LA, et al. CCL2 blockade augments cancer immunotherapy. Cancer Res. 2010;70:109–118. [PMC free article] [PubMed] [Google Scholar]

44. Ni L, Ma CJ, Zhang Y, Nandakumar S, Zhang CL, Wu XY, Borthwick T, Hamati A, Chen XY, Kumaraguru U, et al. PD-1 modulates regulatory T cells and suppresses T-cell responses in HCV-associated lymphoma. Immunol Cell Biol. 2011;89:535–539. [PMC free article] [PubMed] [Google Scholar]

45. Marie JC, Letterio JJ, Gavin M, Rudensky AY. TGF-β1 maintains suppressor function and Foxp3 expression in CD4⁺CD25⁺ regulatory T cells. J Exp Med. 2005;201:1061–1067. [PMC free article] [PubMed] [Google Scholar]

46. Hsieh CS, Lee HM, Lio CW. Selection of regulatory T cells in the thymus. Nat Rev Immunol. 2012;12:157–167. [PubMed] [Google Scholar]

47. Lu LF, Rudensky A. Molecular orchestration of differentiation and function of regulatory T cells. Genes Dev. 2009;23:1270–1282. [PMC free article] [PubMed] [Google Scholar]

48. Quezada SA, Peggs KS, Simpson TR, Allison JP. Shifting the equilibrium in cancer immunoediting: from tumor tolerance to eradication. Immunol Rev. 2012;241:104–118. [PMC free article] [PubMed] [Google Scholar]

49. Koyama K, Kagamu H, Miura S, Hiura T, Miyabayashi T, Itoh R, Kuriyama H, Tanaka H, Tanaka J, Yoshizawa H, et al. Reciprocal CD4⁺ T-cell balance of effector CD62L^low CD4⁺ and CD62L^highCD25⁺ CD4⁺ regulatory T cells in small cell lung cancer reflects disease stage. Clin Cancer Res. 2008;14:6770–6779. [PubMed] [Google Scholar]

50. Ju S, Qiu H, Zhou X, Zhu B, Lv X, Huang X, Li J, Zhang Y, Liu L, Ge Y, et al. CD13⁺CD4⁺CD25^hi regulatory T cells exhibit higher suppressive function and increase with tumor stage in non-small cell lung cancer patients. Cell Cycle. 2009;8:2578–2585. [PubMed] [Google Scholar]

51. Huang CT, Workman CJ, Flies D, Pan X, Marson AL, Zhou G, Hipkiss EL, Ravi S, Kowalski J, Levitsky HI, et al. Role of LAG-3 in regulatory T cells. Immunity. 2004;21:503–513. [PubMed] [Google Scholar]

53. Lo Re S, Lecocq M, Uwambayinema F, Yakoub Y, Delos M, Demoulin JB, Lucas S, Sparwasser T, Renauld JC, Lison D, et al. Platelet-derived growth factor-producing CD4⁺ Foxp3⁺ regulatory T lymphocytes promote lung fibrosis. Am J Respir Crit Care Med. 2011;184:1270–1281. [PubMed] [Google Scholar]

54. Granville CA, Memmott RM, Balogh A, Mariotti J, Kawabata S, Han W, Lopiccolo J, Foley J, Liewehr DJ, Steinberg SM, et al. A central role for Foxp3+ regulatory T cells in K-Ras-driven lung tumorigenesis. PLoS One. 2009;4:e5061. [PMC free article] [PubMed] [Google Scholar]