Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

It is well known that exposure to asbestos causes asbestosis, lung and gastrointestinal cancer, and mesothelioma (1, 2). Because of high toxicity and carcinogenecity of asbestos, man-made mineral or vitreous fibers (MMMF or MMVF), such as glass fibers, rock wool, and titanium whisker, have been developed and are now widely used. However, health effects of those MMM(V)F have not been well investigated.

Although toxicity of asbestos has been well documented, it was not clear until now what chemical and physical characteristics are responsible for the high toxicity of asbestos fibers. Ghio and colleagues (3) and Goodglick and Kane (4) reported that transition metal ions in asbestos fibers generate oxidant stress and cause lung injury. Silanol residues on the surface of asbestos fibers may be implicated in interaction of asbestos with the cell membrane (5). The fibrous shape may confer upon the particles the high toxicity because longer glass fibers are more toxic to rat alveolar macrophages than are the shorter ones (6), and the fiber length of asbestos and MMM(V)F is proportional to in vitro toxicity in Chinese hamster ovary cells (7). It has been shown that long glass fibers are more potent than short ones in nuclear factor-B activation (8) and shorter chrysotile fibers are less fibrogenic than longer ones in rat lungs (9). On the contrary, it has been reported that the cytotoxicity and potency to generate hydrogen peroxide of shorter crocidolite were not significantly different from longer ones in thioglycolate-induced mouse peritoneal macrophages (4).

Alveolar macrophages remove particulate substances from the epithelial surface of the lung by phagocytosis. Exposure to atmospheric environment or workplace contaminants such as asbestos (10, 11) and silica (11) has been shown to stimulate alveolar macrophages to release proinflammatory cytokines and chemical mediators. Thus, phagocytosis of those particles and the following biologic responses of alveolar macrophages may be responsible for the inflammatory injury in the lung caused by those particulate substances.

Spherical titanium dioxide (S-TiO₂) has been considered inert or nuisance and has been used as control particles in toxicologic studies of silica and asbestos (12, 13). To investigate whether the shape of particles plays an important role in the cytotoxcity or phagocytosis-induced biochemical changes in macrophages, we used S-TiO₂ and fibrous TiO₂ (F-TiO₂) to minimize the chemical composition-related biologic activity of particles.

Recently, Liang and Pardee (14) developed a differential display method to search for specifically expressed messenger RNA (mRNA) in cultured cells and this method was further improved by the same group (15). We adopted this technique to seek genes that would be upregulated after phagocytosis of TiO₂ particles, especially F-TiO₂ in alveolar macrophages. In the present study we report that (1) the mRNA level of krox-20 was significantly increased in alveolar macrophages after exposure to F-TiO₂ up to 8 h, (2) the transcription of this gene was transiently upregulated after adhesion to the plastic culture dish and stimulation with lipopolysaccharide (LPS), and (3) there is a homologue to krox-20/egr-2, which has a heterogeneity in the 5' complementary DNA (cDNA) end, and transcriptions of both genes were similarly upregulated in response to phagocytosis of F-TiO₂ and adhesion to the plastic dish.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Particles

S-TiO₂ with a mean diameter of 0.059 µm was obtained from Titan Kogyo K. K. (STT-65C; Ube, Japan). F-TiO₂ with a mean length of 2.1 µm and a mean diameter of 0.14 µm (TO1) was a gift from Japan Fibrous Material Research Association (JFMRA, Tokyo, Japan). The chemical and physical characteristics of F-TiO₂ were investigated by transmission electron microscopy and X-ray fluorescence analysis (16). Crocidolite of International Union against Cancer reference samples (17) was used as the positive control in cytotoxicity assay and Northern analysis. Those particles were heated at 250°C for 2 h in an oven to remove potentially contaminating endotoxin, and were suspended in the culture medium by sonication before use. Urban dusts collected by an electric dust precipitator (B65598; Fuji Electric, Kawasaki, Japan) in the heart of Tokyo were a gift from Dr. S. Goto (National Institute of Public Health) and were autoclaved before use.

Collection of Alveolar Macrophages

Specific pathogen-free, male Sprague-Dawley rats, 7 to 9 wk old (Clea Japan, Tokyo, Japan), were used. They were allowed free access to commercial chow and distilled water in a clean air-conditioned room (23 ± 1°C, relative humidity 55 ± 10%). The rats were anesthetized with intraperitoneal injection of sodium pentobarbital (50 mg/kg body weight) and killed by exsanguination from abdominal aorta. The lungs were lavaged eight times with endotoxin-free saline (Otsuka Pharm. Co., Naruto, Japan) and the lavageate was centrifuged at 400 × g for 5 min at 4°C. The pellet was washed in RPMI 1640 containing 100 U/ml penicillin and 100 µg/ml streptomycin and supplemented with 10% heat-inactivated fetal bovine serum (FBS), and resuspended in fresh RPMI 1640 at 1.0 or 1.2 × 10⁶ viable cells/ml. The viability of the lavaged cells was more than 95% as determined by the ability of trypan blue exclusion. More than 97% of the lavaged cells were alveolar macrophages, as determined by differential cell counting using Diff-Quik staining.

Cytotoxicity Assay

A 100-µl sample of cell suspension (1.0 × 10⁶ viable cells/ml) was aliquotted into a 96-well culture dish (Costar, Cambridge, Mass). After 0.5 h of culture the monolayer was washed three times with RPMI 1640 culture medium without FBS and without phenol red. S-TiO₂, F-TiO₂, or crocidolite samples suspended in serum-free RPMI 1640 were added to each well to final concentrations of 0.0625, 0.125, 0.25, 0.5, and 1 mg/ml. The cells were further cultured for 20 h and the cytotoxicity of those particles was measured with the Lactate Dehydrogenase (LDH)-Cytotoxic Test (Wako Pure Chem., Osaka, Japan) according to manufacturer's instructions. Briefly, a 50-µl sample of the conditioned medium was transferred to a V-bottomed 96-well dish. After centrifugation at 500 × g for 5 min, the supernatant was collected. Triton X-100 was added both to the remaining cellular pellet in the V-bottomed dish and to the original dish to a final concentration of 0.1%. The activity of LDH was measured colorimetrically using a microplate reader (CS9300; Shimadzu, Kyoto, Japan) and the cytotoxicity was calculated according to the following equation:

Differential Display

Alveolar macrophages were precultured in the culture dish (60 mm diameter, Costar) for 20 h at 1.2 × 10⁶ cells/ml. S-TiO₂ and F-TiO₂ were added to the final concentration of 100 µg/ml and the cells were further cultured for 3 h. Total RNA was prepared from control and S- and F-TiO₂-exposed alveolar macrophages using TRIZOL (GIBCO BRL, Rockville, MD) according to the manufacturer's instructions. The total RNA was subjected to DNase I (GIBCO BRL) digestion in the presence of RNase inhibitor (GIBCO BRL) at room temperature for 30 min. The RNA was precipitated again and dissolved in diethyl pyrocarbonate-treated water (5 µl/10⁶ cells). The estimated total RNA concentration was 0.2 µg/µl according to the preliminary experiments. A 2-µl sample of RNA solution was subjected to reverse transcriptase/polymerase chain reaction (RT-PCR) using a GeneAmp RNA PCR kit (Perkin Elmer, Foster City, CA) with a minor modification. Briefly, single-stranded cDNA (scDNA) were prepared from the RNA sample using Moloney murine leukemia virus RT and anchor primers (T₁₂VA, C, G, and T, where V is the equimolar mixture of A, C, G, and T; Operon, Alameda, CA) and the cDNA was amplified by PCR with an equivolume mixture of Amplitaq (Perkin-Elmer) and GeneTaq (Nippon Gene, Toyama, Japan) in the presence of [-³³P]deoxyadenosine triphosphate. Primers used for the PCR reaction were the anchor primers described earlier and 10 mer primers (OP-26-01 to -10, Operon). The PCR products were resolved by polyacrylamide gel electrophoresis (PAGE). After drying the gel, an X-ray film was exposed to the gel overnight at 80°C.

cDNA Cloning

Because the gene of interest seemed to be amplified with anchor primer (T₁₂VA) alone in differential display-PCR, as shown later, the corresponding cDNA was extracted from the gel and reamplified by PCR using only T₁₂VA primer. The PCR product was TA-cloned into a pCR2.1-TOPO vector (Invitrogen, Carlsbad, CA) and the insert was sequenced using the Thermo Sequenase Fluorescent Labelled Primer Cycle Sequencing kit with 7-deaza- deoxyguanidine triphosphate (Amersham, Buckinghamshire, UK).

Northern Analysis

Alveolar macrophages (1.2 × 10⁶ viable cells/ml) were precultured for 20 h in RPMI 1640 medium (10% FBS) and exposed to 100 µg/ml of S-TiO₂, urban dusts, and crocidolite for 3 h or F-TiO₂ for 0.5, 1.5, 3, and 8 h. In adhesion study, freshly lavaged alveolar macrophages were washed and resuspended in RPMI 1640 medium at 1.0 × 10⁶ viable cells/ml and a 2-ml sample of the cell suspension was aliquotted into untreated or poly (2-hydroxyethyl methacrylate) (poly-HEMA)-treated culture dishes (35 mm diameter; Costar). It has been shown that adhesion of alveolar macrophages to the culture dish was inhibited when the dish was treated with poly-HEMA (18). The poly-HEMA-treated dish was prepared by evaporating 2 ml of 10 mg/ml poly-HEMA (Sigma, St. Louis, MO) solution in 95% ethanol in the culture dish (35 mm diameter). The precultured alveolar macrophages were also stimulated with 2 µg/ml LPS (phenol extract, Escherichia coli 0127:B8; Sigma) for 0.5, 1, 3, 5, and 8 h in the poly-HEMA-treated culture dish to aviod adhesion to the plastic culture dish.

Total RNA was extracted using TRIZOL as described earlier. The RNA was resolved in formaldehyde-denatured agarose gels (1%) and blotted onto a nylon membrane (Hybond-N; Amersham) using a vacuum blotter (Model 785; Bio-Rad, Hercules, CA). The blot was prehybridized in ExpressHyb (Clontech, Palo Alto, CA) at 65°C for 90 min, and hybridized with a [³²P]-labeled cDNA probe for the detection of krox-20/egr-2 mRNA. The cDNA probe (492 base pairs [bp]) for the detection of krox-20/ egr-2 mRNA was obtained by digestion of pBluescript II SK+ (Stratagene, La Jolla, CA) with a full-length krox-20/egr-2 cDNA at a multicloning site (see below) with XhoI. The cDNA of RT-PCR product amplified with rat -actin primers (520 bp, BioSource, Camarillo, CA) was TA-cloned into pCR2.1-TOPO (Invitrogen) and the EcoRI-digested fragment was used for the detection of -actin mRNA. The -actin cDNA insert was sequenced before the blotting and the sequence was confirmed to be identical to the registered one in The National Center for Biotechnology Information (NCBI) library (accession number J00691). The probes were labeled using the Rediprime DNA labeling system (Amersham) with [a-³²P]deoxycytidine triphosphate. The blots with the krox-20/egr-2-XhoI probe were stripped in 0.5% sodium dodecyl sulfate solution at 90°C for 10 min and reprobed with [³²P]-labeled -actin cDNA to normalize krox-20/egr-2 mRNA with a housekeeping gene. The radioactivity on the membranes was analyzed and quantitated using a bioimage analyzer (BAS 2000; Fuji, Tokyo, Japan).

Rapid Amplification of cDNA Ends (RACE) and Heterogeneity of krox-20/egr-2

Because cDNA of interest obtained from differential display- PCR was not corresponding to the 3' end of krox-20/egr-2, 5' and 3' rapid amplification of cDNA ends (RACE) reactions were carried out to obtain a full length of this gene and to analyze the cDNA end for any heterogeneity. Gene-specific primers were constructed using the sequence of the differential display-PCR product (5'-CCT GTA ACA CTG CCC ACA TCA CAC A-3' for 5' RACE and 5'-TTC TCC GAG TTC TGA ACC TTT GGG A-3' for 3' RACE). RACE reactions were performed using a Marathon cDNA Amplification kit (Clontech). 5' and 3' RACE products were TA-cloned into pCR2.1-TOPO. The inserts in pCR2.1-TOPO were digested again with BamHI and XhoI and the fragments were cloned into pBluescript II SK+ for DNA sequencing. 5' and 3' RACE products were digested with EcoT22I separately, and the digested cDNAs were ligated using T4 DNA ligase (Takara, Otsu, Japan). The ligated cDNA containing a full-length cDNA of krox-20/egr-2 and krox-20H1 (see subsequent text) was cloned into pBluescript II SK+.

RT-PCR with krox-20/egr-2- and krox-20H1- Specific Primers

To confirm that both krox-20/egr-2 and krox-20H1 are expressed in F-TiO₂-exposed and adherent alveolar macrophages, mRNA levels of those two genes were examined by RT-PCR using a GeneAmp RNA PCR kit (Perkin-Elmer) and gene-specific primers. The following primers were used for PCR reactions: upstream 5'-CGA GGG GAC ACA CTG ACT G-3' and downstream 5'-GAA GAG GCT GTG GTT GAA GC-3', for krox-20/egr-2; upstream 5'-CAC CCA CTT ACC CAT TCT GG-3' and downstream 5'-GTG CAC ACG CTC TCT CTC AC-3' for krox-20H1. Alveolar macrophages were precultured for 20 h in poly-HEMA- treated culture dishes. The cells were transferred to an untreated culture dish and allowed to adhere for 1 h or exposed to 100 µg/ ml F-TiO₂ for 3 h. After total RNA extraction scDNA was synthesized using an oligo d(T)₁₆ primer and Moloney murine leukemia virus transcriptase (42°C, 15 min). The annealing/extension and denaturing temperatures for PCR reaction were 60 and 95°C, respectively. PCR cycles were 23 and 28 for krox-20/egr-2 and krox-20H1, respectively. The experiment was duplicated using alveolar macrophages obtained from different animals.

Statistics

Each cytotoxicity value represents the mean ± standard error of the mean (SEM) of four determinations with duplicate replicates. In the Northern analysis, using different particles, alveolar macrophages were obtained separately from four different groups of rats and each value represents the mean ± SEM of four groups. Statistical analysis was performed using two-way analysis of variance followed by Bonferroni's post hoc comparison. A probability value of less than 0.05 was accepted as indicative of a significant difference.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Figure 1 indicates that the cytotoxicity of F-TiO₂ to alveolar macrophages was significantly higher than that of S-TiO₂. The median lethal concentration (LC₅₀) values of crocidolite and F-TiO₂ were calculated to be 147 and 556 µg/ml, respectively, using GraphPad PRISM (San Diego, CA). Although the cytotoxicity of crocidolite was significantly higher than that of F-TiO₂, the toxicity of F-TiO₂ was comparable to that of crocodolite at higher concentrations. At 3 h after in vitro exposure, almost all alveolar macrophages phagocytosed both S- and F-TiO₂ and those particles were present in the cytosolic compartment of the cells (Figure 2).

View larger version (21K): [in this window] [in a new window]   Figure 1.   Cytotoxicity of S-TiO2 (open circles), F-TiO2 (filled circles), and crocidolite (open triangles) in rat alveolar macrophages. Alveolar macrophages were suspended in RPMI 1640 medium (10% FBS) at 1.0 × 106 viable cells/ml. After 0.5 h of preculture, nonadherent cells were removed by washing and the particle suspensions were added to the wells at final concentrations of 62.5, 125, 250, 500, and 1,000 µg/ml. The cells were further cultured for 20 h, and the viability was assessed by leakage of LDH into the culture medium. Data are presented as means ± SEM of four experiments with duplicate determinations. Cytotoxicities of S- and F-TiO2 and crocidolite were significantly different from each other (P < 0.05).

View larger version (87K): [in this window] [in a new window]   Figure 2.   Photomicrographs of TiO2-loaded rat alveolar macrophages. Alveolar macrophages were precultured in RPMI 1640 medium (10% FBS) for 20 h. The cells were further cultured in the presence of 100 µg/ml S-TiO2 (B) or F-TiO2 (C) or without particles (A, control) for 3 h. The cells were detached from the culture dish by pipetting and the suspended cells were cytocentrifuged and stained with Diff-Quik.

To investigate what genes were expressed in those phagocytosing alveolar macrophages, total RNA was extracted from those cells and the transcripts were resolved by PAGE using a differential display-PCR technique. In Figure 3 the transcripts (marked by an arrow) were increased in TiO₂-exposed alveolar macrophages, especially in F-TiO₂-exposed cells. It is noteworthy that the cDNA was amplified by differential display-PCR regardless of the 10 mer primers, suggesting that the gene was amplified by an anchor primer alone or that both ends of this cDNA fragment have the same sequence. Comparison of the sequence of this cDNA with those in the NCBI database revealed that this gene was krox-20/egr-2.

View larger version (42K): [in this window] [in a new window]   Figure 3.   Differential display of gene expression in control and S- and F-TiO2-exposed alveolar macrophages. Rat alveolar macrophages were precultured for 20 h and further cultured in RPMI 1640 medium (10% FBS) with 100 µg/ml of either S- or F-TiO2 for 3 h or without treatment (control). The experiment was performed in duplicate using alveolar macrophages obtained from different groups of animals. Only the result of differential display- PCR with T12VA as an anchor primer and five of the 10 mer primers described later is shown, where V is an equimolar mixture of A, C, and G. Lanes 1-6: TAC AAC GAG G, lanes 7-12: TGG ATT GGT C, lanes 13-18: CTT TCT ACC C, lanes 19-24: TTT TGG CTC C, and lanes 25-30: GGA ACC AAT C. Lanes 1, 7, 13, 19, and 25: Control-1; lanes 4, 10, 16, 22, and 28: Control-2; lanes 2, 8, 14, 20, and 26: S-TiO2-1; lanes 5, 11, 17, 23, and 29: S-TiO2-2; lanes 3, 9, 15, 21, and 27: F-TiO2-1; and lanes 6, 12, 18, 24, and 30: F-TiO2-2. The arrow indicates a differentially expressed gene.

To exclude a possibility that krox-20/egr-2 was false-positive in differential display, Northern blot analysis was performed using a XhoI fragment of krox-20/egr-2. Total RNA was extracted from two different batches of alveolar macrophages that had been exposed to S- and F-TiO₂ for 3 h and the transcription of krox-20/egr-2 was quantitated (Figure 4). mRNA levels of krox-20/egr-2 in S- and F-TiO₂-exposed alveolar macrophages were higher than the control value by 3.7- and 17.6-fold, respectively. The time course of changes in krox-20/egr-2 mRNA level suggested that the transcription of this gene was increased up to 8 h after exposure to F-TiO₂ in alveolar macrophages (Figure 5). The upregulation of krox-20/egr-2 gene was also examined using other particulate substances. Crocidolite, another fibrous particle, significantly increased krox-20/egr-2 level compared with the control and S-TiO₂ groups. However, the mRNA level of crocidolite-exposed alveolar macrophages was less than that of F-TiO₂-exposed alveolar macrophages (Figure 6).

View larger version (13K): [in this window] [in a new window]   Figure 4.   Northern analysis of krox-20/egr-2 gene expression in rat alveolar macrophages after exposure to S- and F-TiO2. Alveolar macrophages obtained from two different groups of rats were precultured in RPMI 1640 (10% FBS) for 20 h and exposed to S- or F-TiO2 (100 µg/ml) for 3 h before RNA extraction. The bar graph shows the mean relative transcription of krox-20/egr-2 to that of -actin (n = 2).

View larger version (14K): [in this window] [in a new window]   Figure 5.   Northern analysis of krox-20/egr-2 gene expression in rat alveolar macrophages after exposure to F-TiO2. Alveolar macrophages were precultured in RPMI 1640 (10% FBS) for 20 h and exposed to F-TiO2 (100 µg/ml) for 0.5, 1.5, 3, and 8 h. The bar graph shows relative transcription of krox-20/egr-2 to that of -actin.

View larger version (35K): [in this window] [in a new window]   Figure 6.   Northern analysis of krox-20/egr-2 gene expression in rat alveolar macrophages after exposure to S-TiO2, urban dusts, F-TiO2, and crocidolite. Alveolar macrophages were precultured in RPMI 1640 (10% FBS) for 20 h and exposed to 100 µg/ml of each sample for 3 h. The bar graph shows the means ± SEM of relative transcription of krox-20/egr-2 to that of -actin (n = 4). *Significantly different from all other groups. # Significantly different from control and S-TiO2-exposed groups.

Because preculture seemed to be important to obtain low transcription level of krox20/egr-2 in control alveolar macrophages, we next investigated whether transcription of this gene was related to adhesion to the culture dish. Freshly isolated alveolar macrophages were washed and resuspended in the culture medium and aliquotted into the untreated or poly-HEMA-treated culture dishes (35 mm diameter; Costar). The mRNA level of krox-20/egr-2 was highest just after adhesion to the plastic dish (0.5 h after aliquotting) and decreased gradually (Figure 7). The adhesion-dependent transcription level of krox-20/egr-2 was much lower when alveolar macrophages were cultured in the poly-HEMA-treated culture dish, confirming that the transient transcription of krox20/egr-2 was related to the nonspecific adhesion to the plastic surface. We also investigated the time-course profile of krox-20/egr-2 mRNA transcription in LPS-exposed alveolar macrophages to study the expression of this gene in response to the typical inflammatory stimulus. As shown in Figure 8, the mRNA level increased transiently, decreased to the control level after 3 h, and then increased again after 8 h.

View larger version (17K): [in this window] [in a new window]   Figure 7.   Northern analysis of krox-20/egr-2 gene expression in rat alveolar macrophages after adhesion to the plastic culture dish. Alveolar macrophages were cultured in RPMI 1640 (10% FBS) for 0.5, 1.5, 3, 8, and 20 h in untreated culture dishes (filled columns) or cultured for 3 h in a poly-HEMA-coated culture dish (open column). The bar graph shows relative transcription of krox-20/egr-2 to that of -actin.

View larger version (21K): [in this window] [in a new window]   Figure 8.   Northern analysis of krox-20/egr-2 gene expression in rat alveolar macrophages after exposure to LPS. Alveolar macrophages were precultured in RPMI 1640 (10% FBS) for 20 h in a poly-HEMA-treated culture dish. The cells were exposed to 2 µg/ml LPS for 0.5, 1.5, 3, 5, and 8 h. The bar graph shows relative transcription of krox-20/egr-2 to that of -actin. The open column indicates the control value (without LPS.)

While cloning a full length krox-20/egr-2 with RACE, we found that this gene had heterogeneity at the upstream region. We cloned a different type of krox-20 into pBluescript SK+ with a new sequence at the 5' cDNA end (krox-20H1). To study whether the transcription of those two types of krox-20 genes were differentially regulated after adhesion and phagocytosis of F-TiO₂, specific primers for each krox-20 were prepared and PCR was performed using those primers. Both krox-20/egr-2 and krox-20H1 transcriptions were similarly upregulated after either adhesion to the plastic dish or phagocytosis of F-TiO₂, as shown in Figure 9.

View larger version (15K): [in this window] [in a new window]   View larger version (40K): [in this window] [in a new window]   Figure 9.   Heterogeneity of 5' cDNA end of krox-20. 5' RACE indicated that there are two different types of krox-20: krox-20/ egr-2 and krox-20H1, with an unreported sequence at 5' cDNA end (A). Striped areas, ORF; lightly shaded areas, amplified in DD-PCR; darker shaded area, different sequence; and arrows indicate amplification by PCR. PCR was performed to determine whether mRNA levels of those two different types of krox-20 were increased in both adherent and F-TiO2-exposed alveolar macrophages. The sizes of PCR products deduced from the sequences of krox-20/egr-2 and krox-20H1 are 724 and 399 bp, respectively. (B) Lane 1, molecular size marker; lanes 2 and 3, control alveolar macrophages; lanes 4 and 5, adherent alveolar macrophages; lanes 6 and 7, F-TiO2-exposed alveolar macrophages.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Although cytotoxicity of both asbestos and MMMF or MMVF has been well documented, it is not clear whether fibrous forms confer on those particles a potency to adversely interact with the cells. The present study addressed whether fibrous mineral particles are more cytotoxic to alveolar macrophages compared with fine and spherical ones, and what genes are expressed in response to phagocytosis of mineral fibers of different shapes using S- and F-TiO₂ particles. Because S-TiO₂ particles are considered to be nuisance or inert, we used those particles to extract only shape-dependent biologic changes.

Figure 1 shows that F-TiO₂ is more cytotoxic than S-TiO₂ to alveolar macrophages. Several lines of evidence suggest that longer asbestos and vitreous fibers are more cytotoxic than shorter ones (6, 12). The present study further confirms that the fibrous shape confers on TiO₂ particles more potent cytotoxicity. It should be noted that the cytotoxicity of F-TiO₂ was comparable to that of crocidolite at higher concentrations, suggesting that attention should be paid to the biologic effects of some MMMF or MMVF. However, the question of why the shape of particles affects the cytotoxicity remains to be answered. One plausible explanation is that when macrophages phagocytose long and fibrous particles, the cell membrane may be injured more seriously compared with fine and spherical particles, although interleukin-1-converting enzyme (ICE) family-mediated apoptosis in asbestos-exposed macrophages (19) cannot be explained directly by the injury of cell membrane.

Differential display has been used to seek for differentially expressed novel genes in mammalian cells. Using this technique, we found a gene that was slightly and greatly transcribed in response to phagocytosis of S-TiO₂ and F-TiO₂, respectively. In Figure 2, this gene was amplified regardless of the 10 mer primers, suggesting that the PCR amplification proceeded with an anchor primer alone. As expected, this PCR product was not corresponding to the 3' end of cDNA. This gene, krox-20/egr-2, is known to be expressed in the development of the hind brain (20, 21) and hippocampus after long-term potentiation (22). It is also known that transcription of this gene is increased when cells proliferate in the presence of growth factors (23, 24). To our knowledge, the present study is the first that reports upregulation of krox-20/egr-2 gene transcription in response to either phagocytosis or adhesion. However, further investigation is required to reveal the function of Krox-20/Egr-2 in phagocytosing or adhering alveolar macrophages. Although krox-20/egr-2 transcription was significantly increased after exposure to crocidolite, the mRNA level of crocidolite-exposed alveolar macrophages was less than that of F-TiO₂- exposed cells, suggesting that the expression of this gene is not directly associated with cell death.

The transcription of krox-20/egr-2 was transiently increased in alveolar macrophages in response to nonspecific adhesion to the plastic culture dish with a maximum at 0.5 h after plating. When alveolar macrophages were cultured in a poly-HEMA-treated dish, the transcription of krox-20/egr-2 was significantly suppressed (Figure7). In contrast to nonspecific adhesion, krox-20/egr-2 mRNA increased up to 8 h when alveolar macrophages were exposed to F-TiO₂. It is known that rat alveolar macrophages adhere to the plastic culture dish quickly (< 0.5 h) and adherence force decreases thereafter (25). Probably greater adhesion to the foreign substrate or having foreign material of relatively large size in their bodies may be important for the transcription of this gene. There are two phases in phagocytosis. The first phase is adhesion and extension of pseudopods along the particle surface, and the second is enclosing those attached particles into phagosomes. It has been shown that different molecules, such as phosphatidylinositol 3 kinase and type VI myosin, are involved in each phase (26). The difference in time course of krox-20/ egr-2 gene expression in between phagocytosis of F-TiO₂ and adhesion to the plastic dish may be related to the assembly of different molecules in those two phases.

It is interesting that there are two phases in krox-20/ egr-2 gene expression in the LPS-exposed alveolar macrophages. When alveolar macrophages were stimulated with LPS in suspension culture (in the poly-HEMA-treated culture dish), the cell started to aggregate after 3 to 5 h. Thus, it is plausible that the second phase of the krox-20/ egr-2 gene expression is due to cell-to-cell interaction.

In cloning a full length of krox-20/egr-2 using 5' and 3' RACE, we found a homologue to krox-20/egr-2 with a different and unreported sequence at the 5' cDNA end (krox-20H1). PCR primers were constructed to amplify krox-20/ egr-2 and krox-20H1 mRNA separately. The transcription of krox-20/egr-2 and krox-20H1 appeared to be increased in a similar way in alveolar macrophages after either exposure to F-TiO₂ or adhesion to the plastic culture dish.

It has been shown that macrophages adhere to the plastic dish using scavenger receptor (27) and ₂-integrin (28, 29). Recently, macrophage receptor with collagenous structure, one of the scavenger receptors, has been shown to play a pivotal role in binding of unopsonized environmental particles, including TiO₂ (30). Although a functional role of Krox-20/Egr-2 in phagocytosing macrophages is not yet clear, signals from those receptors, including phosphorylation of proteins (18), may be implicated in transcription of krox-20/egr-2 and krox-20H1 in alveolar macrophages.

In summary, transcription of krox-20/egr-2 was greatly increased after exposure to fibrous TiO₂ particles in rat alveolar macrophages. The transcription of this gene was also increased transiently after adhesion to the plastic dish. A new gene, krox-20H1, was also upregulated in either phagocytosis or adhesion in alveolar macrophages. The increased level of these genes are probably associated with signals triggered by interaction with foreign substances and adhesion.