Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Glyoxal is a reactive -oxoaldehyde and a physiologic metabolite, formed by lipid peroxidation, oxidative degradation of glucose, and degradation of glycated proteins (1). Like other -oxoaldehydes, for example, methylglyoxal and 3-deoxyglucosone, glyoxal is capable of inducing cellular damage but may also accelerate the glycation process leading to advanced glycation end products (AGEs) which, in turn, may also be toxic for the cell. The formation of AGEs is the final result of a process characterized by addition of sugar aldehydes or ketone groups to free amino groups of proteins by a nonenzymatic reaction known as Maillard reaction (reviewed in Reference 2). Because carbonyl groups are involved in all these reactions, the term "carbonyl stress" has been suggested for this cellular stress (3).

Under conditions of normal cell metabolism, glyoxal is detoxified by the cytosolic glutathione-dependent glyoxalase system (4), which converts glyoxal to the less-reactive glycolate. There is experimental evidence that glyoxal concentration is increased in oxidative stress because of decreased availability of reduced cofactors for the enzymatic detoxification (4). Impairment of -oxoaldehyde detoxification may contribute to cytotoxicity and to the induction of apoptotic cell death by -oxoaldehydes (5).

In a previous study we reported on the in vitro formation of AGEs by human lung cells L132 after treatment with glyoxal (6). This finding may be relevant for the molecular pathogenesis of pulmonary fibrosis inasmuch as immunohistochemical evidence suggests that the increased formation of AGEs may be linked to the pathologic accumulation of extracellular matrix proteins during the development of this human disease (7). The alveolar epithelium plays an important role in this process. Previously, one major oxidatively formed AGE has been characterized as carboxymethyllysine (CML). Because Glomb and Monnier (8) have clearly shown that the presence of glyoxal leads to CML modification of proteins under physiologic conditions, the demonstration of CML formation may be used as a marker for intracellular glyoxal activity. Besides CML, characteristic AGEs formed by the reaction of glyoxal with proteins are the imidazolium crosslink GOLD (glyoxal lysin dimer) and hydroimidazolone (reviewed in Reference 9).

The human embryonic lung cell line L132 (10) has been shown to be an excellent tool for cytotoxic tests (11) and for the study of radiation-induced injury (12). In the present study we investigated the effects of glyoxal on lung epithelial cells L132 with particular emphasis on the cytotoxic (apoptotic) effects of glyoxal. Apoptosis was studied by immunostaining for the activated caspase-3 and its cleavage product by the CytoDeath Kit and by determination of the proapoptotic protein bax and the antiapoptotic proteins bcl-2 and galectin-3 (reviewed in References 13 and 14).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and Treatment

Human L132 embryonic lung cells (ATCC, Rockville, MD) were grown in Dulbecco's modified Eagle's medium/F12 medium supplemented with 5% fetal calf serum (FCS) and gentamycin (6.4 mg/ml) at 37°C in 5% CO₂. Glyoxal (Sigma, Deisenhofen, Germany) at various concentrations (20 to 400 µM) was added to the culture medium for 24 h. The viability of the lung cells was evaluated by the trypan blue exclusion test and did not exceed 7% nonviable cells in the experiments using glyoxal concentrations up to 100 µM. In additional experiments, L132 cells were maintained for 24 h in a culture medium that was supplemented with recombinant human keratinocyte growth factor (KGF) (Amgen, Inc., Thousand Oaks, CA) at a concentration of 50 ng/ml. After changing culture conditions to medium free of KGF, the cells were exposed to glyoxal as described earlier.

The experiments described later, except the immunoelectron microscopy, were repeated three times for each condition.

Immunocytochemistry

For light-microscopic detection of apoptotic markers in cultured cells we used a simple paraffin technique using a fibrin glue as embedding medium (6, 15). The 4-µm-thick paraffin sections were dewaxed, dried overnight, and treated with microwaves in 0.01 M sodium citrate buffer (16). After washing in phosphate-buffered saline (PBS), pH 7.4, the conventional immunohistochemical procedure followed using a commercially available detection system (Vectorstain Elite Kit; Vector Laboratories, Burlingame, CA). The peroxidase activity was developed with 3'-3'-diaminobenzidine and counterstaining with hemalaun was performed. The primary antibodies and their origins, dilutions, and sources are listed in Table 1. The suitability of antibodies to detect apoptosis has been shown in a previous study using in parallel the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling method with two different kits (17). The specificity of the antipentosidine antibody has already been described (18). The corresponding inhibition experiment was performed with synthesized pentosidine-bovine serum albumin (BSA) (1 mg/ml; compare Figure 1D). The specificity of the CML-recognizing antiserum (19) was verified by preabsorption with CML-human serum albumin (5 mg/ml cell culture medium). As marker of oxidative stress an antiserum to 4-hydroxynonenal (HNE) pyrrole was used to determine sites of HNE modifications in lung cells. For the specificity of the antiserum, see Reference 20. For control, the primary antibody was replaced by nonspecific rabbit or mouse immunoglobulin (Ig) G. 

View larger version (152K): [in this window] [in a new window]   Figure 1.   Paraffin sections of glyoxal-treated L132 cells (A: control; B-D: 200 µM glyoxal-treated cells; E and F: 100 µM glyoxal- treated cells. Cells in F were treated in addition with 50 ng/ml KGF). Immunohistochemical staining for LAMP-1 protein (A and B) and for pentosidine (C and D). (Note the negative immunoreactivity for pentosidine after inhibition with a pentosidine preparation in D.) E and F: Immunostaining for active caspase-3 is not present in KGF-treated cells (F ). Hematoxylin and eosin (H&E) counterstain; original magnification: ×300.

Electron Microscopy

For transmission electron microscopy, cells were fixed in 0.1 M PBS containing 4% paraformaldehyde and 0.05% glutaraldehyde for 1 h at room temperature (RT), dehydrated through a graded series of alcohols, and embedded in Epon as described previously (6). For immunoelectron microscopy, fixed L132 cells were embedded in Lowicryl HM20 (Polysciences Europe, Hamburg, Germany; for details see Reference 6). Ultrathin sections were mounted on pioloform-coated mesh nickel grids; preincubated with 10% normal goat serum in Tris-buffered saline (TBS), pH 7.6, for 45 min; and incubated overnight with the polyclonal rabbit anti-CML antiserum (dilution 1:100) at 4°C. After washings with buffer (TBS with 0.2% BSA) followed the incubation with 10-nm gold-conjugated antirabbit IgG (Biocell, Cardiff, UK), dilution 1:50. After a further rinse in buffer, the sections were stained with 2% aqueous uranyl acetate (8 min) and lead citrate (2 min). A Zeiss EM 906 transmission electron microscope (Zeiss, Oberkochen, Germany) operated at 80 kV was used for ultrastructural analysis.

Cytofluorimetric Analysis

To detect intracytoplasmatic antigens, cultured cells were fixed in 2% (wt/vol) formaldehyde in PBS for 20 min and centrifuged. The cells were then resuspended in a washing buffer consisting of 0.5% BSA in PBS. The cells were permeabilized for 20 min using 0.5% (wt/vol) saponin (Sigma) in washing buffer. Activation of caspase-3 was measured using the polyclonal antibody anti-active caspase-3 (PharMingen Europe, Hamburg, Germany) which recognizes an active human caspase-3 fragment and binds to a conformational epitope that is exposed by activation-induced cleavage or denaturation of the inactive enzyme. To detect CML-modified proteins, polyclonal rabbit anti-CML antiserum (dilution 1:1,500) was used. After incubation of cells with anti-active caspase-3 or CML-antiserum for 60 min, the cells were washed twice in PBS containing 0.5% BSA and 0.5% (wt/vol) saponin and incubated with a fluorescein isothiocyanate antimouse IgG (Sigma) for 60 min at RT. Cells were washed again and resuspended in 400 µl of PBS, and analyzed by flow cytometry (FACSCalibur; Becton Dickinson, San Jose, CA). For each analysis, 10,000 events were recorded.

Western Blots

Cytosolic and nuclear proteins were separated and prepared as described by Andrews and Faller (21) and the protein concentrations were determined according to Bradford using the Bio-Rad protein assay reagent (22). Cytosolic and nuclear extracts were separated by sodium dodecyl sulfate (SDS) polyacrylamide (7.5%) gel electrophoresis. Proteins were transferred to nitrocellulose by semidry electroblotting (transfer buffer: 48 mM Tris, 39 mM glycine, 0.0375% SDS, 20% [wt/vol] methanol). After blotting, the nitrocellulose membranes were blocked with NET buffer (150 mM NaCl, 50 mM Tris/HCl [pH 7.4], 5 mM ethylenediaminetetraacetic acid, 0.05% Triton X-100, and 0.25% gelatine) and incubated with the CML antibody (diluted 1:8,000 in NET) overnight at 4°C. After washing with NET, the membranes were incubated with horseradish peroxidase-conjugated antirabbit or antigoat IgG for 1 h at RT. Visualization of immunocomplexes was performed by enhanced chemiluminescence as described by Bouaboula and colleagues (23).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Immunocytochemical Detection of the AGE Products CML and Pentosidine and the Lipid Peroxidation Product HNE in L132 Cells

In fetal epithelial lung cells we observed an increasing percentage of cells immunoreactive to the CML-specific antibody with increasing glyoxal concentrations ranging from 0 to 400 µM (Figures 2A-2D; compare also the fluorescence-activated cell sorter [FACS] data given later). At the light-microscopic level, in contrast to the dose-dependent increase of CML, first positive reactions with the antibodies against pentosidine (another nonglyoxal-derived AGE), and HNE (a product of lipid peroxidation), were found at glyoxal concentrations of 200 µM (Figure 1C, HNE immunostaining not shown) and remained positive at higher doses of glyoxal treatment (not shown). The technique of paraffin embedding and cutting of the cells allowed a precise detection of the antigen in different compartments of the cell. Positive immunostaining for CML was seen in the cytoplasm, at the cell membrane, and in a large number of cells also in the nuclei (Figure 2). Immunoelectron microscopy confirmed this staining of the entire cell body (Figure 5).

View larger version (145K): [in this window] [in a new window]   Figure 2.   Paraffin sections of glyoxal-treated L132 cells (A, E, I: control; B, F, J: 100 µM; C, G, K: 200 µM; D, H, L: 400 µM glyoxal). Immunohistochemical demonstration of CML (A-D), caspase-3 (E-H), and caspase-cleavage product CytoDeath (I- L). H&E counterstain; original magnification: ×300.

View larger version (144K): [in this window] [in a new window]   Figure 5.   Lowicryl HM 20-embedded L132 cells. Immunoelectron microscopic demonstration of CML in untreated (A) and glyoxal-treated (200 µM) (B) cells. Note the strong labeling of cytoplasmic and nuclear components (B). Original magnification: ×19,400.

Assessment of Apoptosis

Only occasional apoptotic cells were seen in sections of untreated L132 cells stained with antibodies against active caspase-3 and the cytokeratin cleavage product (Figures 2E and 2I). In cells treated with increasing glyoxal concentrations, an increasing number of cells immunopositive with antibodies specific for caspase-3 (Figures 2F-2H) and CytoDeath (Figures 2J-2L) were observed, indicating that in these cells apoptosis had been induced. To evaluate the activation pathway of apoptosis the expression of pro- and antiapoptotic molecules was studied. The proapoptotic bax was enhanced with increasing glyoxal concentrations (Figures 4A-4D). In contrast, staining of molecules with antiapoptotic effects such as bcl-2 and galectin-3 was most intense in untreated cells and decreased with increasing glyoxal concentrations (Figures 4E-4H and 4I-4L).

View larger version (165K): [in this window] [in a new window]   Figure 4.   Paraffin sections of glyoxal-treated L132 cells (A, E, I: control; B, F, J: 100 µM; C, G, K: 200 µM; D, H, L: 400 µM glyoxal). Immunohistochemical demonstration of bax (A-D), galectin-3 (E-H), and bcl-2 (I- L). H&E counterstain; original magnification: ×300.

To assess the integrity of lysosomal membranes in our experimental conditions, a monoclonal antibody specific for the lysosomal membrane protein LAMP-1 was included in immunocytochemical studies. There were no remarkable alterations of immunoreactivity for LAMP-1 detectable in both untreated and glyoxal-treated cells (Figures 1A and 1B).

Treatment of cells with KGF, an inhibitor of apoptosis as demonstrated in several in vivo and in vitro experiments (24, 25), resulted in strong reduction in the expression of the caspase-3 immunoreactivity (Figures 1E and 1F). This effect could be achieved only at low concentrations (50 to 100 µM) of glyoxal.

Figure 3 illustrates the key changes in the ultrastructure of L132 cells occuring in the majority of cells at concentrations of 200 µM glyoxal pretreatment: formation of membrane blebs, fragmentation of nuclei, and condensation of chromatin. Using an antiserum with high affinity to protein-bound CML (26), immunoreactivity for this AGE product was detected by immunoelectron microscopy at the cell membrane and in the cytosol as well as in the nucleus of epithelial lung cells (Figure 5).

View larger version (128K): [in this window] [in a new window]   Figure 3.   Effect of glyoxal on apoptosis in L132 cells. Electron microscopy features of L132 cells cultured alone (A) or in the presence of 200 µM glyoxal for 24 h (B). Compared with the untreated cells (A), after glyoxal treatment L132 cells show apoptotic nuclei with condensed chromatin, abnormal cytoplasmic vesicles, and blebbing of the cell membrane (B). Original magnification: ×4,170.

To confirm the immunocytochemically found nuclear localization of CML-modified proteins, nuclear extracts from glyoxal-treated cells were isolated and CML-modified proteins were analyzed by Western blotting. As shown in Figure 6, distinct proteins ranging from < 39 up to 220 kD showed increased CML staining with increasing glyoxal concentrations. To detect whether very low glyoxal concentrations also cause CML modification, cells were also incubated with 20 µM glyoxal. Even at this concentration an increase of CML modification was found when compared with control. It is noteworthy that nuclear extracts previously not exposed to exogeneous glyoxal also showed faint CML staining of distinct proteins, particularly in the range of 66 to 160 kD.

View larger version (87K): [in this window] [in a new window]   Figure 6.   Effect of increasing glyoxal concentrations on CML modification of nuclear proteins in L132 cells. Cells were incubated with 0 to 200 µM glyoxal for 24 h. Nuclear extracts were prepared and Western blots with CML antibody were performed as described in MATERIALS AND METHODS.

The glyoxal-induced formation of CML-modified proteins and parallel expression of caspase-3 was studied by FACS analysis. As shown in Figure 7, elevation of glyoxal concentration substantially enhanced the number of CML-positive cells. Even at low concentrations of 50 µM glyoxal, significant cellular CML modification was found. Similar to the CML modification shown in Figure 7, activated caspase-3 was dose-dependently enhanced with increasing glyoxal concentrations (Figure 8).

View larger version (30K): [in this window] [in a new window]   Figure 7.   Flow cytometric analysis of CML in L132 cells incubated with glyoxal. Relative CML concentrations in the cells were quantified by flow cytometry. Cells were treated for 24 h with (black area) or without (white area) 50 µM (A), 100 µM (B), 150 µM (C), or 200 µM (D) glyoxal.

View larger version (25K): [in this window] [in a new window]   Figure 8.   Flow cytometric analysis of active caspase-3 expression in L132 cells incubated with glyoxal. Cells were untreated (A) or treated for 24 h with 50 µM (B), 100 µM (C), 150 µM (D), or 200 µM (E) glyoxal. The vertical line indicates the mean fluorescence intensity of active caspase-3 in untreated cells (A).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

As described in a previous paper, glyoxal is a suitable metabolite able to induce the in vitro formation of AGEs in epithelial lung cell line L132 (6). Knowing that -oxoaldehydes such as methylglyoxal and 3-deoxyglucosone have apoptotic effects (5), we speculated that glyoxal or one of its AGEse.g., CMLmay induce apoptosis in living cells. Our study showed the presence of AGEs in L132 cells and a dose-dependent induction of apoptosis. The presence of CML modifications in all cellular compartments indicates that glyoxal or glyoxal-derived modifications of cellular proteins directly contributed to the apoptotic cell death in the present experiment. The exact mechanisms, however, by which either one of these metabolites exerts its apoptotic activity remains unknown. Because we were cultivating the cells in FCS-containing medium, preformed AGEs may also contribute synergistically to the metabolic changes. Results from a preliminary experiment have indicated that CML-modified proteins are formed in cell culture medium upon incubation with 50 to 200 µM glyoxal for 24 h inasmuch as CML antibody reactivity was blocked after incubation with the glyoxal-treated medium for 1 h (Kasper, unpublished data). It is noteworthy that 0.1 to 1-µM glyoxal has been found in plasma of normal and diabetic subjects (27) whereas intracellular concentrations of -oxoaldehydes may be much higher, e.g., 300 µM for methylglyoxal (28).

From our experiments we cannot distinguish whether glyoxal itself or glyoxal-derived protein modifications are initiating the apoptosis. Because it has recently been shown that CML-modified proteins may bind to the receptor for AGEs (RAGE) and activate signal transduction leading to nuclear factor (NF)-B activation (2), it may well be that glyoxal-induced CML modification rather than glyoxal may initiate and/or propagate cellular apoptosis.

Pulmonary fibrosis is characterized by a complex process involving alveolitis, fibroblast proliferation, and abnormal accumulation and deposition of interstitial collagen. This process could be the result of DNA damage and apoptosis. There is experimental evidence that Fas ligand and Fas antigen are upregulated in fibrosing lung disease (29). Other investigators describe a potential role of bax and bcl-2 expression in diffuse alveolar damage (30). Fibroblasts isolated from fibrotic lung produce factors capable of inducing apoptosis and necrosis of alveolar epithelial cells in vitro, and in fibrotic lung apoptotic alveolar epithelial cells are located immediately adjacent to activated myofibroblasts (31). Possible pathogenetic factors implicated in the development of pulmonary fibrosis are AGE-modified proteins, which accumulate in pulmonary fibrosis (7). They induce the generation of reactive oxygen intermediates and the upregulation of proinflammatory cytokines (32). Reactive oxygen intermediates are involved in the formation of early glycation adducts such as CML (33). Interaction of AGEs with specific cell-surface receptors (for example, RAGE) on macrophages and probably also on certain epithelial cells induces oxidative stress, which results in activation of the transcription factor NF-B and subsequent gene expression, which might be relevant for chronic diseases (reviewed in Reference 2). The involvement of oxidative stress in induction of apoptosis by glyoxal is supported by the detection of lipid peroxidation products such as HNE and the formation of other nonglyoxal-derived AGEs such as pentosidine in our study.

Finally, glyoxal-triggered apoptosis was also suppressed by KGF. In our in vitro model, KGF could have reduced the AGE-induced apoptosis by decreasing the expression of the activated caspase-3 enzyme. KGF has been used successfully to prevent alveolar damage induced by oxygen-induced injury in mice (25). Further, treatment with KGF protects epithelial cells against tumor necrosis factor-- or Actinomycin D-induced apoptosis (24). KGF also protects type II epithelial cells cultured from hyperoxic rats against oxygen-induced DNA damage and subsequent apoptosis (34). Inasmuch as apoptotic cell death plays an essential role in the development of pulmonary fibrosis, KGF may prevent AGE-induced lung injury by limiting the oxidative stress.

In conclusion, our data show that treatment of L132 lung cells with glyoxal induces apoptosis. At the ultrastructural level, L132 cells showed dramatic and characteristic changes in nuclear shape and organization, which is an early and relatively unequivocal hallmark of apoptosis (35). In addition, pro- and antiapoptotic markers were increased or decreased, respectively, with increasing glyoxal concentrations. Further, increasing glyoxal concentrations cause an enhanced formation of AGE products, particularly CML modification. The experimental induction of apoptosis in L132 cells described here will provide a suitable system to study the role of AGEs for apoptotic processes in alveolar epithelial cells in vitro. The usage of in vitro glyoxal-modified extracellular matrix proteins or the glycation of more lung-specific products such as surfactant protein are the necessary steps to improve this model in the future.