Combined effect of cigarette smoke and mineral fibers on the gene expression of cytokine mRNA.

To investigate which parameters are stimulated by mineral fibers and whether cigarette smoke enhanced a fiber-induced response, we examined the level of cytokine mRNA from alveolar macrophages (AMs) and lungs of rats exposed to mineral fibers and cigarette smoke in vivo. Male Wistar rats were given a single intratracheal instillation of 2 mg of Union Internationale Contre le Cancer chrysotile or refractory ceramic fiber (RF1). The animals then inhaled a side stream of smoke 5 days per week for 4 weeks. The expression of manganese superoxide dismutase, inducible nitric oxide synthase (iNOS), basic fibroblast growth factor (bFGF), interleukin-1[alpha] (IL-1[alpha]), interleukin-6 (IL-6), and tumor necrosis factor-[alpha] (TNF[alpha]) mRNA from lipopolysaccharide-stimulated AMs and lungs of rats exposed to mineral fibers and/or cigarette smoke were assessed using semiquantitative reverse-transcriptase polymerase chain reaction. Exposure only to cigarette smoke increased in IL-1[alpha] mRNA levels in AMs. Chrysotile stimulated the expression of IL-1[alpha], TNF[alpha], and IL-6 in AMs, and the expression of bFGF in lungs. RF1 resulted in increased expression of IL-1[alpha] and TNF[alpha] in AMs. Cigarette smoke stimulated the gene expression of iNOS in AMs and IL-6 and bFGF in lungs treated with chrysotile; IL-1[alpha] in AMs and bFGF in lungs did the same in lungs with RF1. Among these cytokines, message levels of IL-1[alpha], iNOS, and bFGF were increased in rats stimulated with mineral fibers, and the stimulating effects of mineral fibers were enhanced by cigarette smoke. Therefore, IL-1[alpha], iNOS, and bFGF would be the possible parameters of the lung remodeling induced by mineral fibers.

lsp://ehpnetl.niehs.nib.gov/docs/ll999/107p495-500morimotoabstractnml Occupational and environmental exposure to inhaled asbestos dust causes pulmonary fibrosis, bronchogenic carcinoma, and pleural mesothelioma (1). Recently, various types of man-made mineral fibers have been developed as substitutes for asbestos; the demand for these products is increasing. Some of these fibers are thought to possess the same adverse biological effects as asbestos because of their similar physiochemical properties (2). Furthermore, epidemiological and dinical studies have indicated that asbestos workers who smoked cigarettes had a higher incidence of asbestos-induced disease than those who did not (3)(4).
In asbestos-induced disease, many genes are thought to contribute to lung remodeling, such as fibrosis and carcinoma (5)(6). Asbestos fibers deposited in the lung lead to an activation of alveolar macrophages (AMs) (5). AMs and parenchymal cells release tumor necrosis factor-a (TNFa), interleukin-la (IL-la), interleukin-6 (IL-6), basic fibroblast growth factor (bFGF), and other cytokines that augment the process of inflammation (5)(6). The release of oxidants from these cells may lead to lung injury and manganese superoxide dismutase (Mn-SOD) and inducible nitric oxide synthase (iNOS) corresponding to released free radical formation (6). Some growth factors signal interstitial fibroblasts to replicate and modulate their production of connective tissue proteins (5). The accumulation of inflammatory cells and fibroblasts and the development of connective tissue matrices causes lung injury and leads to fibrosis and carcinoma (5)(6).
A reverse transcriptase-polymerase chain reaction (RT-PCR) used in this study has a number of advantages (7): a) its exquisite sensitivity gives it the ability to detect rare mRNA in small numbers of cells in a semiquantitative manner; b) unlike the protein assay system, the same methodology can be applied to analyze the expression of many genes; and c) cDNA can be used for future studies.
To estimate which parameters are more sensitive and specific for lung remodeling induced by mineral fibers, we used RT-PCR to examine cytokine mRNA levels from AMs and lungs of rats exposed to mineral fibers and/or cigarette smoke in vivo. Experimental design. Male Wistar rats (10 weeks old) were used in this study, and 30 rats were divided into six groups (saline, chrysotile, RF1, cigarette smoke, chrysotile + cigarette smoke, and RF 1 + cigarette smoke). Either saline, chrysotile suspension, or RF 1 suspension (2 mg/0.2 mL) was administered to the rats intratracheally. The rats were housed in an exposure chamber and were exposed to the smoke of 20 cigarettes Uapan Tobacco, Tokyo, Japan) for 4 hr/day, 5 days/week, for 4 weeks. The mass concentration of cigarette smoke was 10 mg/m3 and was measured gravimetrically at daily intervals by the suction of air through a glass filter. The CO concentration was 79 ppm, which is below the concentration that induces specific biological effects. The chamber volume was 100 L and the flow rate in the chamber was 50 L/min. Bronchoalveolar lavage andpreparation. After exposure, each animal was anesthetized by an intraperitoneal injection of phenobarbital. Bronchoalveolar lavage (BAL) was performed on the left lung. Cells recovered by BAL were suspended in RPMI-1640 medium, then placed on tissue culture plates and allowed to attach for 1 hr at 370C (8). Nonadherent cells were removed by one cycle of washing with RPMI, and adherent cells were supplied with an RPMI-1640 medium Articles * Morimoto et a).

Materials and Methods
containing 100% fetal bovine serum. They were adjusted to a final concentration of 1 x 105/mL and stimulated with 10 pg/mL lipopolysaccharide (Sigma, St. Louis, MO). The AiMs were cultured for 2 and 6 hr on cell culture plates at 37°C in the CO2 incubator. After incubation, mRNA in AMs was extracted using a Quick Prep kit (Pharmacia Biotech, Uppsala, Sweden). Total RNA from the lung was prepared in the presence of guanidium thiocyanate (9).
Preparation of RNA, cDNA synthesis, and PCR. Total RNA (0.5 pg) was used for the synthesis of single-strand cDNA using Moloney murine leukemia virus-derived reverse transcriptase (Perkin Elmer, Norwalk, CT). An equal amount of cDNA from each sample, standardized to give identical signals on gel following amplification with 3-actin primer, was amplified by specific primers for each gene (Table 1) (10)(11)(12)(13)(14)(15)(16). The amplification was performed using a Thermocycler (Astech, Japan) under the following conditions: denaturation at 94°C for 45 sec, annealing at 60°C for 45 sec, and extension at 72°C for 2 min for cytokines and 1-actin genes.
Detection of the fragments amplified by PCR was made by electrophoresis on a 2% agarose gel. PCR products were resolved using gel electrophoresis and visualized by ethidium bromide staining. The gel was photographed with Polaroid Type 665 positive/negative film (Polaroid Corp., Cambridge, MA) under ultraviolet light at identical exposure and development times. The bands from the positive film were scanned, and the density of each PCR product was measured using National Institute of Health image 1.55 software (provided by Wane Rasband, National Institutes of Health, Bethesda, MD). The ratio of each specific gene product to the f-actin product was used for the analysis. Quantitative analysis. To quantify the transcriptional level of cytokine mRNA, the following analysis was used: IL-1a samples were coamplified with 3-actinspecific primers as an internal standard ( Figure 1). Aliquots containing cDNA were subjected to 26, 29, 32, 35, or 38 cycles of amplification under identical conditions as in protocols. The amounts of density recovered from the excised bands were plotted as a function of the number of cycles. The rates of amplification were exponential between 29 and 35 cycles for both templates. The number of thermocycles used allowed quantitation without saturation. The data were normalized to represent equivalent RNA loading based on the density of 1-actin at the appropriate cycle of both genes (7).
We compared the data of 2-and 6-hr incubation of cytokines in AMs and presented the one that showed more gene expression.
Statistical analysis. Values were expressed as the mean + 1 SD. The difference between values was assessed using a Mann-Whitney U-test.
Histopathology. Lungs were fixed in 10% buffered formalin. Embedded samples were then sectioned and stained with hematoxylin and eosin.

Results
Bronchoalveolar lavage. An increased number of total cells was recovered from rats exposed to mineral fibers and/or cigarette smoke than from saline-exposed rats (cigarette 6.4 ± 3.4 x 105 cells; chrysotile 5.6 + 1.9 x 105 cells; RFI 5.5 ± 1.9 x 105 cells; chrysotile + cigarette smoke 5.7 + 1.5 x 105 cells; RF1 + cigarette smoke 9.1 + 1.6 x 105 cells; control 4.4 ± 1.2 x 105 cells; Table 2). AMs accounted for >95% of the cells in all groups.  Expression of IL-la, TNFa, and IL-6 mRNA. In comparison with the control (saline) group, levels of H lu mRNA in AMs increased significantly in groups exposed to cigarette smoke, chrysotile, chrysotile + cigarette smoke, RF 1, and RF 1 + cigarette smoke (Figure 2A). IL-lu mRNA in the group exposed to RF 1 + cigarette smoke was maximally expressed in all groups, but it was not significant compared to cigarette smokeor RFI-exposed group.
Elevation in the gene expression of IL-l t mRNA was not observed in rat lungs exposed to mineral fibers anid/or cigarette smoke ( Figure 2B). A significant increase in TNFo mRNA in AMs was obtained in chrysotileor  ,r 6, June 999 * Environmental Health Perspectives RF1-exposed rats as compared to control rats, but not in cigarette smoke-exposed rats ( Figure  3A). Mineral fibers and cigarette smoke did not have combined effects on message levels of TNFa in AMs. TNFa mRNA in the lung was increased in chrysotile-exposed rats. Similar to chrysotile, RF 1 tended to increase in the gene expression of TNFa mRNA; however, there were no significant differences in the levels of TNFa mRNA when compared to control rats ( Figure 3B). Chrysotile-only exposure induced increases in IL-6 mRNA levels in AMs ( Figure 4A). Elevation of IL-6 mRNA was not increased in rat AMs exposed to RF1 and/or cigarette smoke. Exposure to cigarette smoke or mineral fibers did not induce an increase in IL-6 levels in the lung; however, exposure to chrysotile and cigarette smoke had a synergistic effect on the expression of IL-6 mRNA ( Figure 4B). Expression of iNOS and Mn-SOD mRNVA. In rats exposed to cigarette smoke or mineral fibers, mRNA levels for iNOS in AMs were not increased ( Figure 5A). On the other hand, mRNA levels for iNOS Cigarette smoke-exposed El M _ Mineralfiber-exposed among AMs from rats treated with mineral fibers and cigarette smoke increased significantly when compared with control rats.
Levels of iNOS mRNA declined markedly in cigarette smoke-exposed rat lungs ( Figure  5B). Exposure to mineral fibers tended to induce an increase in iNOS mRNA; however, differences between mineral fibers and control were not significant. *p<4.05 and **p<0.01 significantly different from a control (saline-exposed) group.
M Cigarette smoke-exposed M Mineral fiber-exposed *p<0.05 significantly different from a control (saline-exposed) group. Results are expressed in ratio of bFGF to f-actin (mean ± standard error). *p<0.05 and **p<0.01 significantly different from a control (saline-exposed) group.
No significant changes in levels of Mn-SOD mRNA in AMs and lungs exposed to mineral fibers and/or cigarette smoke was observed ( Figure 6).
Expression of bFGF mRNA. The message of bFGF in AMs was not observed in all groups, although the cycle of PCR was up to 40 ( Figure 7A). Chrysotile induced an increase in bFGF mRNA in the lungs, but cigarette smoke or RF1 did not. The combined effect of mineral fibers and cigarette smoke on the message level of bFGF was significant ( Figure 7B).
Pathologicalfindings. Foreign body giant cells and mononuclear cells around bronchioles were observed in groups exposed to chrysotile and chrysotile + cigarette smoke (Figure 8). In the other four groups, definite pathological changes were not observed.

Discussion
We set the recovery time after the intratracheal instillation of mineral fibers at 4 weeks. Unlike the inhalation findings, some reports showed that nonspecific responses at the acute phase were seen after the intratracheal injection. Some investigators (17,18) Volume 107, Number 6 June 1999 Environmental Health Perspectives   Figure 9. Ethidium bromide staining of PCR products separated in 2% agarose gel. Abbre' basic fibroblast growth factor; IL-la, interleukin-lac; IL-6, interleukin-6; M, DNA marker (OX1 SOD, manganese superoxide dismutase; iNOS, inducible nitric oxide synthase; RCF, refra fiber; TNFa, tumor necrosis factor a. showed that saline injection in BAL caused a neutrophil recruitment by the lower respiratory tract in 72 hr, and that an increase in neutrophils and AMs recovered from bronchoalveolar lavage fluid were seen in the first week after injecting water alone. We previously reported that inhalation for one day of cigarette smoke transiently enhanced the ability of mineral fibers to stimulate AMs to produce TNF in rats (151, and later reported that no specific changes were fo in rats with cigarette smoke int weeks (20). To avoid the initial acute response, we exposed rat smoke for 4 weeks in this exp( time line for gene expression of AMs and lungs following ini exposure is now under investigal Many previous reports exan level or gene expression of proi cytokines and growth factors that were stimulated by mineral fibers in lungs or in AMs iNOS (696 bp) (5,6,(19)(20)(21). However, the mineral fibers l-Actin (357 bp) used for experimental studies are not always the same among laboratories. The physio-iNOS chemical properties of mineral fibers affect P-Actin biological response in cells and lungs. For instance, the size of mineral fibers causes different responses even if they are the same fibers. We previously reported a difference in the production of TNFa in AMs treated Mn-SOD (383 bp) with long and short mineral fibers (22). Numerous in vitro studies have also shown 1B-Actin (357 bp) that long fibers induced greater cytotoxicities than short fibers in vitro (2,23,24). In this experiment each cytokine was examined tin-SO (383bp from the same samples using RT-PCR under the identical experimental process, and it is useful to estimate the sensitivity and selectivity of parameters in lung remodeling induced by mineral fibers. In this experiment, these parameters were examined on bFGF(260bp) the following points: whether the gene expression of cytokine mRNA is induced by mineral fibers and whether mRNA levels for cytokines are additively or synergistically increased by combined exposure to cigarette IL-la (623 bp) smoke and mineral fibers.

P-Actin (357 bp)
Mineral fibers increased in gene expression of TNFa, IL-1, IL-6, and iNOS IL-la mRNA in AMs and bFGF mRNA in P-Actin lungs. These proinflammatory cytokine mRNA or proteins have been detected in the AMs or lungs of patients with idiopathic pulmonary fibrosis (25) and in animal lungs with pulmonary fibrosis (26).
VNAFa(295 bp) Blackford et al. (27) reported that intratracheal instillation of silica up regulated the 3-Actin gene expression of iNOS mRNA in AMs.

TNFa
These data were accordance with our data. However, mRNA expression of Mn-SOD was not increased in AMs and lung tissue from rats exposed to mineral fibers. Quinlan et al. (28) reported that the mes-1L-6 (614 bp) sage level of Mn-SOD in rat lungs inhaling ,B-Actin (357 bp) crocidolite peaked at 3-9 days and decreased toward control level thereafter.  In our experiments, levels of Mn-SOD 3-Actin mRNA might return from peak to normal because recovery time was 4 weeks. viations: bFGF, Cigarette smoke is thought to increase c4/HaeIIl); Mnthe incidence of asbestos-induced disease (3,4). It is generally accepted that smoking is associated with a higher risk of carcinound for AMs ma of the lung (3); however, its association ialation for 4 with asbestosis is controversial, and a 1 influence of majority of studies showed a positive inters to cigarette action between smoking and asbestos (4).

eriment. The
Studies suggest that smoking and asbestos f cytokines in act in combination in the development of itial asbestos fibrosis. In our experiments, cigarette tion. smoke and mineral fibers have combined nined protein effects of gene expression of IL-ia, iNOS, inflammatory bFGF, and IL-6 mRNA (Figure 9). There Environmental Health Perspectives * Volume 107, Number  have been few experimental studies to investigate the combined effects of cigarette smoke and mineral fibers on the gene expression of proinflammatory cytokines. We previously reported that the cigarette smoke and mineral fibers had synergistic effects on matrix metalloproteinases (MMPs) in rat lungs (29). IL-1 up regulates the production of MMPs (30), and the increase in gene expression of MMPs in rats lungs exposed by two agents may be affected by IL-1. Jackson et al. (31) reported that cigarette smoke and asbestos increased DNA damage synergistically and suggested that this synergism might involve hydroxyl radical production. Free radical formation including nitric oxide radical and peroxynitrite anion through iNOS (2Z) might be related to lung injury induced by asbestos and cigarette smoke. The gene expression of proinflammatory cytokines, iNOS, and bFGF might be representative cytokines of inflammation, pulmonary damage by oxidants, and a neovascularization, respectively (4,27,32) Therefore, a change in gene expression of these factors, which play a key role in the inflammatory process, suggested that these factors might contribute to lung remodeling induced by cigarette smoke and mineral fibers. The lungs in rats exposed to chrysotile or chrysotile + cigarette smoke showed inflammatory cells around the bronchioles without a fibrotic form. These pathological findings may support that two agents have a combined effect on cytokines which are related to inflammation or lung injury. Taken together, IL-lot, iNOS, and bFGF mRNA were up regulated by exposure to mineral fibers, and were affected by exposure to mineral fibers and cigarette smoke in combination. Therefore, these factors were thought to be possible parameters for a risk assessment of mineral fibers.

Summary
We examined cytokine mRNA levels from AMs and the lungs of rats exposed to mineral fibers and/or cigarette smoke in vivo.
Message levels of IL-1l, iNOS, and bFGF were increased by exposure to mineral fibers and enhanced by combined exposure to mineral fibers and cigarette smoke. These data suggested that IL-loc, iNOS, and bFGF would be the possible parameters of the lung remodeling induced by mineral fibers.