Interaction between 4-hydroxy-2,3-alkenals and the platelet-derived growth factor-beta receptor. Reduced tyrosine phosphorylation and downstream signaling in hepatic stellate cells.

Hepatic stellate cells (HSC) undergo activation toward myofibroblast-like cells during early stages of liver injury associated with fibrogenesis. Platelet-derived growth factor (PDGF), particularly its BB isoform, has been identified as the most potent mitogen for HSC. 4-Hydroxy-2,3-nonenal and related 4-hydroxy-2, 3-alkenals (HAKs) have been suggested to modulate the process of HSC activation. In this study we investigated the relationship between HAKs and PDGF receptor activation in human HSC. By employing noncytotoxic concentrations (10(-6) m) of HAKs, we observed a significant inhibition of PDGF-BB-dependent DNA synthesis. HAKs inhibited relevant pathways of PDGF-BB-dependent mitogenic signaling, including autophosphorylation of PDGF receptor (PDGF-R) beta subunits and activation of phosphatidylinositol 3-kinase and extracellular regulated kinases 1/2. Inhibition of DNA synthesis was reversible, and recovery of PDGF-mediated mitogenic signaling occurred within 24-48 h and was associated with HAKs-induced up-regulation of PDGF-R beta gene expression. 4-Hydroxy-2,3-nonenal, used as a model HAK, inhibited the intrinsic tyrosine kinase activity associated with the PDGF-R beta subunit, whereas binding of PDGF to its receptor was unaffected. This study identifies a novel regulatory mechanism of reactive aldehydes on PDGF receptor signaling and biologic actions, which may be relevant in several pathophysiological conditions, including liver fibrosis.

Up-regulation of genes encoding for PDGF A-and B-chains as well as for PDGF receptor (PDGF-R) ␣and ␤-subunits has been shown to be involved in the development of chronic human inflammatory fibroproliferative diseases. These include atherosclerosis (4), liver fibrosis (5,6), mesangial proliferative glomerulonephritis (7,8), scleroderma (9), rheumatoid arthritis (10), and idiopathic pulmonary fibrosis (11). In these clinical settings one of the most relevant effect of PDGF is the stimulation of proliferation of target cells of mesenchymal origin. In the liver, hepatic stellate cells are major effector cells involved in the development and progression of liver fibrosis and are particularly responsive to PDGF (5,6). In addition, overexpression of PDGF isoforms and their cognate receptors has been shown during active fibrogenesis in human liver (12).
PDGF mitogenic signaling involves the binding of PDGF isoforms (AA, AB, or BB) to specific high affinity receptor ␣ (or type A) and ␤ (or type B) subunits (3,13). The binding is followed by the dimerization of PDGF receptors, activation of the PDGF-R intrinsic tyrosine kinase, and consequent autophosphorylation of the receptors on critical tyrosine residues (3). Several downstream signal transduction molecules have been shown to bind to different autophosphorylation sites in PDGF-R, including phosphatidylinositol 3-kinase (PI 3-K), phospholipase C-␥, the Src family of tyrosine kinases, the tyrosine phosphatase SHP-2, GTPase-activating protein for Ras, as well as other adaptor molecules (14). PDGF mitogenic signaling has been shown to require activation of phospholipase C-␥ and PI 3-K (15). In addition, a major role has been demonstrated for Ras activation, which is followed by a kinase cascade including Raf-1, MEK, and extracellular-signal regulated kinase (ERK) (14,16). ERK activation is followed by nuclear translocation and phosphorylation of several transcription factors, including activator protein-1, Elk-1, and SAP, and, at least in some cells, increased expression of c-fos (6,17,18).
We have investigated whether HNE and other HAKs may be able to affect PDGF mitogenic signaling in cultured HSC, as a paradigm of mesenchymal cells involved in organ fibrosis. Following chronic liver tissue injury, HSC activate and transdifferentiate into highly proliferative myofibroblast-like cells that account for deposition of extracellular matrix leading to hepatic fibrosis and cirrhosis (5,6). In this context, activated HSC have been shown to be highly sensitive to the effects of both ROI and HNE (33, 40 -44). The results of the present study indicate that HAKs specifically block PDGF-dependent HSC DNA synthesis by inhibiting tyrosine phosphorylation of the PDGF receptor and its downstream signaling when used at low, nontoxic concentrations.
Cell Isolation and Culture-Human HSC were isolated from a wedge section of normal human liver tissue unsuitable for transplantation by collagenase-Pronase digestion and centrifugation on stractan gradiens. Procedures for cell isolation, culture, and characterization have been described extensively elsewhere (45,46). Cells were cultured in complete medium supplemented with 20% fetal bovine serum. Data presented herein were obtained using three separate cell lines used between passages 4 and 7. As already reported (45), at these stages of culture, human HSC showed transmission electron microscopy features of "myofibroblast-like cells," thus indicating complete transition to their activated phenotype.
In all the experiments employing HNE and other HAKs, HSC were incubated in serum-free/insulin-free (SFIF) medium for 24 h before the addition of these compounds. Unless otherwise stated, HAKs were always added to culture medium 15 min before the addition of PDGF-BB (10 ng/ml).
Measurement of DNA Synthesis-Confluent HSC in 24-well dishes were washed with phosphate-buffered saline and incubated in SFIF medium for 24 h. The cells were then incubated with HAKs for varying periods of time and then exposed to PDGF-BB (10 ng/ml) for an additional 24 h; DNA synthesis was measured as the incorporation of [ 3 H]thymidine, as described previously (47).
Binding Studies-Confluent cells in 24-well dishes were incubated in SFIF for 24 h and then incubated for 15 min in the presence or absence of 1 M HNE. The procedure for the binding assay was exactly as previously reported (48). Scatchard analysis of the binding data was performed using Ligand software.
Preparation of Cell Lysates-Confluent and 24 h serum-starved cells were treated with the appropriate conditions and then quickly placed on ice and washed with ice-cold phosphate-buffered saline, pH 7.4. Cell lysates were obtained as described previously (49). HSC monolayers were lysed in immunoprecipitation assay buffer consisting in Tris-HCl buffer, pH 7.4, containing 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 1 mM sodium orthovanadate, 1 mM phenylmethylsulphonyl fluoride, and 0.05% aprotinin (w/v). Insoluble proteins were discarded by high speed centrifugation at 4°C. Protein concentration in the supernatant was measured in triplicate using a commercially available assay (Bio-Rad).
Western Blot Analysis-Proteins were separated by SDS-PAGE according to Laemmli (51) and transferred electrophoretically to nitrocellulose membranes (Hybond-C extra; Amersham Pharmacia Biotech). The membranes were blocked overnight at 4°C for unspecific binding using 5% (w/v) nonfat dry milk in 50 mM Tris-HCl, pH 7.4, containing 200 mM NaCl, 0.05% (v/v) Tween 20 (TBS-Tween). The blots were incubated with appropriated primary antibodies, followed by incubation with peroxidase-conjugated anti-mouse or anti-rabbit immunoglobulins (as secondary antibodies) in TBS-Tween containing 1% (w/v) nonfat dry milk. Immunoblots were developed with the ECL plus reagents from Amersham Pharmacia Biotech according to manufacturer's instructions.
ERK Assay-ERK activity was detected using myelin basic protein as substrate. Briefly, proteins from cell lysates were immunoprecipitated with polyclonal anti-ERK antibodies (Santa Cruz Biotechnology, Santa Cruz, CA). After washing, immunobeads were incubated for 30 min at 30°C in reaction buffer consisting in 20 mM Hepes, pH 7.4, 1 mM sodium orthovanadate, 1 mM MgCl 2 , 1 mM dithiothreitol, 1 mM ATP, 1 Ci of [␥-32 P]ATP, and 0.4 mg of myelin basic protein. Reaction was stopped by addition of Laemmli buffer and then subjected to 15% SDS-PAGE. The gel was stained with Coomassie Blue, dried, and autoradiographed; bands corresponding to myelin basic protein were identified and evaluated by laser densitometry.
Phosphatidylinositol 3-Kinase Assay-PI 3-K was measured as the recruitment of PI 3-K activity in anti-phosphotyrosine immunoprecipitates, as described previously (49). Briefly, proteins from different experimental conditions were immunoprecipitated using anti-phosphotyrosine antibodies. After washing, the immunobeads were resuspendend in 50 l of 20 mM Tris-HCl, pH 7.4, 100 mM NaCl, and 0.5 mM EGTA. One microliter of 20 mg/ml phosphatidylinositol was added, and the samples were then incubated at 25°C for 10 min. One microliter of 1 mM MgCl 2 and 10 Ci of [␥-32 P]ATP (3000 Ci/mmol) were then added simultaneously, and the incubation was continued for 10 more minutes. The reaction was stopped by addition of 150 l of chloroform, methanol, 37% HCl (10:20:0.2). Samples were extracted with chloroform and dried. Radioactive lipids were separated by thin layer chromatography using chloroform, methanol, 30% ammonium hydroxide, water (46:41: 5:8). After drying, the plates were autoradiographed. The radioactive spots were then scraped and counted in a ␤-counter.
Assay of Akt Activity-This assay was carried out essentially as described by Sandirasegarane et al. (50). One hundred micrograms of protein was immunoprecipitated using anti-Akt antibodies and protein G-agarose. The immunobeads were washed three times with washing buffer (20 mM Hepes, pH 7.5, 40 mM NaCl, 50 mM NaF, 1 mM EDTA, 1 mM EGTA, 0.5% Nonidet P-40, 20 mM ␤-glycerophosphate, 0.5 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 10 g/ml pepstatin, and 10 g/ml aprotinin). The assay was performed resuspending the beads in kinase buffer (50 mM Hepes, pH 7.5, 100 mM NaCl, 10 mM MgCl 2 , 10 mM MnCl 2 , 10 mM ␤-glycerophosphate, and 0.5 mM sodium orthovanadate) in the presence of 1 M protein kinase A inhibitor peptide, 50 M unlabeled ATP, and 6 Ci of [␥-32 P]ATP, using exogenous histone H2B (1.5 g/assay tube) as the substrate and incubating for 20 min at room temperature. The proteins in the samples were resolved by 12% SDS-PAGE, and the gel was stained with Coomassie Blue and subjected to autoradiography.
Assay of PDGF-R ␤ Tyrosine Kinase Activity-Cell lysates were obtained from serum-deprived HSC exposed to the following conditions: sample A, untreated cells; sample B, 1 M HNE for 15 min; sample C, PDGF for 10 min; and sample D, HNE for 15 min followed by PDGF for 10 min. One aliquot of the samples B and D and two aliquots of samples A and C were subjected to immunoprecipitation with anti PDGF-R ␤ subunit antibodies (49). The immunoprecipitated proteins were washed twice in lysis buffer and once in 50 mM Tris-HCl, pH 7.4, containing 1 mM sodium orthovanadate and finally resuspended in 50 mM Hepes at pH 7.4. At this point, 1 M HNE was added to one aliquot of samples A and C for 15 min to establish the effects of HNE on purified PDGF-R ␤ subunits in vitro. The other aliquots of samples A and C as well as samples B and D received no additions. At the end of this 15-min incubation, MnCl 2 (final concentration, 10 mM) and [␥-32 P]ATP (20 Ci) were added to all samples, which were incubated at 30°C for 15 min. The beads were then washed once with immunoprecipitation assay buffer (see before) and once with 50 mM Tris-HCl, pH 7.4. Finally, the beads were resuspended in 20 l of Laemmli's sample buffer, boiled for 5 min, and separated by a 7.5% SDS-PAGE.
RNA Purification and RNase Protection Assay-Total RNA from cultured HSC was isolated by the guanidine isothiocyanate-phenolchlorophorm method (52). Total RNA of the samples was calculated spectrophotometrically by the absorbance at 260 nm. Purity of RNA was determined from the 260/280 nm absorbance ratio. The level of PDGF-R ␤ subunit mRNA was determined by the RNase protection assay that was performed essentially as extensively reported elsewhere (12, 53). The previously described cDNA fragments encoding for human PDGF-R ␤ subunit mRNA and subcloned in pGEM1 were employed in this assay. The gene product glyceraldehyde-phosphate dehydrogenase (a 110-base pair cDNA fragment subcloned in the HindIII site of pT7.2) was used as housekeeping gene. After the linearization of the plasmid with the restriction endonuclease T7, RNA polymerase (Promega) was used to run off antisense 32 P-labeled RNA probes. The newly synthesized RNA probes (specific activity, approximately 1 ϫ 10 9 cpm/g) were hybridized overnight to 10 g of total RNA from individual samples at 55°C. After hybridization, the reaction was treated with RNases A (50 g/ml) and T1 (2 g/ml) for 30 min at room temperature and for an additional 30 min at 37°C in 10 mM Tris-HCl, pH 7.5, 5 mM EDTA, and 300 mM NaCl. Protected fragments were analyzed by electrophoresis through a 7.0 M urea, 6% acrylamide gel and then exposed to Kodak X-Omat film at Ϫ80°C for 72 h.
Analysis of Cytotoxicity-HAKs were tested for cytotoxicity on confluent and 24 h serum-starved HSC. HSC cultured on 24-well plates were exposed to increasing concentrations of test molecules, and cytotoxicity was evaluated at the end of a 24-h incubation by means of trypan blue exclusion dye test.
Data Analysis-Data presented as bar graphs are the means (Ϯ S.E.) of three independent experiments. Autoradiograms and luminograms are representative of at least three experiments with similar results. Statistical analysis was performed by Student's t test (p Յ 0.05 was considered significant).

Effects of HAKs on Basal and PDGF-dependent DNA Synthesis in Relation to
Their Cytotoxic Effect-We tested HAKs of different chain length, namely HOE, HNE, and HUE, on basal and PDGF-dependent DNA synthesis. HAKs exerted a significant inhibitory effect on DNA synthesis of unstimulated cells when added at concentrations of 100 M or higher, whereas lower doses (range 0.1-10 M) had no effect (Fig. 1, A-C). In contrast, HAK concentrations as low as 1 M, chosen because this low dose has been found to elicit other biological responses in HSC (33,40,41), completely blocked PDGF-dependent incorporation of [ 3 H]thymidine in cultured HSC (Fig. 1D). All three aldehydes had comparable effects on DNA synthesis (Fig. 1).
We next investigated whether the observed effects of HAKs on DNA synthesis were related to a cytotoxic action. Cultured HSC were exposed to increasing concentrations of HAKs (HOE, HNE, and HUE) for 24 h, and cell viability was measured by trypan blue exclusion (Table I). Concentrations of HAKs in the 0.1-10 M range were devoid of any cytotoxic effects (Table I), whereas exposure of HSC to concentrations of 100 M or higher was associated with a marked reduction in cell viability (Table  I). Again, no significant differences were observed in the cytotoxic action of the three different aldehydes. Therefore, although the reduction in basal [ 3 H]thymidine incorporation by high doses of HAKs was likely to depend on a cytotoxic effect, the inhibition of PDGF-induced DNA synthesis observed at doses of 1 M cannot be ascribed to this action (compare Table  I and Fig. 1). On the basis of these results, we decided to concentrate our efforts to the understanding of the mechanisms of action of low dose HAKs on PDGF-mediated DNA synthesis.
HAKs Inhibit PDGF-␤ Receptor Autophosphorylation and Downstream Signaling-We first analyzed the effects of HAKs on PDGF-␤ receptor autophosphorylation and the resulting activation of downstream pathways, such as ERK, PI 3-K, and Akt. HSC were pretreated for 15 min with 1 M HOE, HNE, or HUE and then exposed to PDGF-BB for 10 min. In the absence of HAKs, PDGF induced a marked increase in tyrosine phosphorylation of the PDGF-␤ receptor ( Fig. 2A). Pretreatment with HAKs at a concentration of 1 M resulted in a marked inhibition of receptor autophosphorylation. In a similar fashion, the effects of PDGF on activation of ERK and PI 3-K were markedly reduced in cells preincubated with HOE, HNE, or HUE (Fig. 2, B and C). Once again, the three HAKs exerted a similar degree of inhibition on PDGF signaling. To confirm the inhibition of PI 3-K, we tested the effects of HNE, as a model HAK, on the activation of Akt, which lies downstream of PI 3-K (54). In agreement with the data on PI 3-K, HNE reduced the activation of Akt induced by PDGF (Fig. 2D).
Inhibition of PDGF-BB-dependent DNA Synthesis by HAKs Is Reversible-Because the inhibitory effects of HAKS on PDGFdependent [ 3 H]thymidine incorporation was not due to cytotoxicity, we performed experiments to investigate whether the inhibitory effect on DNA synthesis was a transient or irreversible event. HSC were treated with 1 M HAKs for different periods of time and then exposed to PDGF. Although the effects

4-Hydroxy-2,3-alkenals and PDGF Receptor Signaling
of HAKs after a 6-h exposure were comparable with those at 15 min, we observed a recovery of the effects of PDGF on DNA synthesis at 48 h (Fig. 3).
Because the reduction of DNA synthesis was associated with inhibition of PDGF receptor autophosphorylation, we analyzed the ability of PDGF to phosphorylate its receptor 24 or 48 h following exposure to HAKs. Our results indicate that a complete recovery of PDGF-␤ receptor autophosphorylation was detectable only when HSC were exposed to PDGF 48 h after HAKS (Fig. 4).
HAKs Up-regulate the Synthesis of PDGF-␤ Receptor during Recovery-To investigate whether the time-dependent recovery of HSC sensitivity to PDGF-stimulated DNA synthesis is associated with an up-regulation of PDGF-␤ receptor gene expression, we performed RNase protection assay and Western blot analysis. A single exposure of HSC to 1 M HAKs was followed by a marked up-regulation of PDGF-␤ receptor mRNA expression (Fig. 5). This effect was associated with increased expression at the protein level that was barely detectable after 24 h and markedly evident at 48 h (Fig. 6). Transforming growth factor ␤1 was used as a positive control for up-regulation of PDGF-␤ receptor protein synthesis in human HSC (48).
HNE Inhibits Intrinsic PDGF␤ Receptor Tyrosine Kinase Activity-We next tested whether the actions of HAKs occurred at the level of ligand-receptor binding or at that of receptor activation. HNE (1 M) did not significantly affect the binding of the radiolabeled ligand to its receptor, as evaluated by Scatchard plot analysis (Fig. 7).
The effects of HNE on tyrosine kinase activity of PDGF-␤ receptor were then investigated. In HSC sequentially exposed to 1 M HNE and to PDGF, we observed a reduction of the intrinsic tyrosine kinase activity of the immunoprecipitated PDGF-␤ receptor, in agreement with the already reported inhibition of autophosphorylation (Fig. 8, lanes 1-4). To establish whether the action of HNE occurred directly on the receptor, we obtained total extracts from untreated HSC and from HSC treated for 10 min with PDGF; these extracts were immunoprecipitated with anti-phosphotyrosine antibodies and then exposed in vitro (Fig. 8, lanes 5 and 6) to HNE before the assay for tyrosine kinase activity. Also in these conditions HNE clearly inhibited PDGF-␤ receptor intrinsic tyrosine kinase activity, indicating a direct inhibition of HAKS on this enzymatic activity.
The Effects of HAKs Are Specific for the PDGF-R ␤ Subunit-To test whether the effects of HAKs are specific for the PDGF-R ␤ subunit or may extend to other receptor tyrosine kinases, we tested the effects of preincubation with HNE, as a model HAK, on autophosphorylation of the PDGF-R ␣ subunit and of the EGF-R. In contrast to the marked inhibitory effects of HNE on PDGF-R ␤ subunit phosphorylation, the activation FIG. 2. HAKs inhibit major pathways of PDGF-BB-dependent signaling. HSC deprived of serum for 24 h were incubated with HAKs at 1 M concentration 15 min before exposure to PDGF-BB (10 ng/ml) for 10 min. Cell lysates were obtained and processed as described under "Experimental Procedures." A, PDGF-R ␤ subunits were immunoprecipitated (IP ) and analyzed by immunoblotting using antibodies against phosphotyrosine (P-Tyr, upper panel) or against the PDGF-R ␤ subunit (lower panel). B, ERK was immunoprecipitated and an immune complex kinase assay was performed using myelin basic protein as a substrate. C, cell lysates were immunoprecipitated with anti-phosphotyrosine antibodies and PI 3-K activity associated with the immunobeads was measured using phosphatidylinositol as a substrate. D, cell lysates were immunoprecipitated with anti-Akt antibodies and an immune complex kinase assay was performed using histone 2B as a substrate.

4-Hydroxy-2,3-alkenals and PDGF Receptor Signaling
of the EGF-R was not affected, and autophosphorylation of the PDGF-R ␣ subunit was only modestly inhibited (Fig. 9). DISCUSSION Recent data suggest that ROI as well as HNE and related HAKs may act as mediators of pathophysiological effects of oxidative injury (4, 19 -21, 35-39). In chronic fibrogenic diseases, where PDGF has been described to play a relevant role as a mitogenic factor, oxidative stress-related molecules are indeed generated and have been suggested to be able to modulate the proliferative response of key myofibroblast-like target cells such as, for example, activated HSC in the case of liver fibrosis (43,44). In this study we report that noncytotoxic doses of HAKs (1 M), which can be easily reached in vivo in conditions of mild to moderate oxidative stress (20), affect PDGFinduced thymidine incorporation of HSC in primary culture. We also showed that the molecular basis for this effect is the inhibition of the intrinsic tyrosine kinase activity of PDGF-R ␤ subunits and, as a consequence, the reduction of ERK, and PI 3-K activity, two pathways known to play a key role in medi-ating the mitogenic effect of PDGF in HSC. Remarkably, the inhibitory effects of HAKs were specific for the PDGF-R ␤ subunit, because only a minor inhibition was observed on the PDGF-R ␣ subunit activation, and no changes were detected on EGF-R autophosphorylation.
It is well established that HNE and related HAKs are potent alkylating agents able to react as nucleophilic compounds in Michael-type reaction to form adducts with defined amino acid residues. In particular, these compounds interact with sulfydryl groups of cysteine or with the ⑀-amino group of lysine and with the imidazole ring nitrogen of histidine (20,38,39,55,56). Histidine as well as cysteine residues are indeed numerous in the structure of both the extracellular and intracellular domains of PDGF-R ␤ subunit, particularly in the two tyrosine kinase domains (57). Because of these characteristics, the PDGF receptor is likely to be a target for the action of HAKs, which can potentially affect either PDGF binding to its receptor and consequent dimerization or the intrinsic tyrosine kinase activity. The results of the present study suggest that intracellular tyrosine kinase domains are potential targets for HAKs action because HNE, as a model HAK, did not affect the binding of radiolabeled PDGF to its receptors. The ability of HNE to interfere with tyrosine kinase activity of PDGF-␤ receptor is indicated by three lines of evidence: 1) the reduced steady state tyrosine phosphorylation as shown by Western blot analysis; 2) the reduced tyrosine kinase activity shown in immune complex kinase assay of the PDGF-␤ receptor from cells pretreated with HNE; and 3) the inhibition of tyrosine kinase activity caused by addition of HNE directly to the immunoprecipitated PDGF receptor before the assay for tyrosine kinase activity. The possibility that this effect is mediated by the interaction of HNE with sulfydryl groups and/or hystidine residues is supported by previous observations. Depletion of intracellular GSH correlates with a strong inhibition of PDGF-R ␤ subunit autophosphorylation in NIH-3T3 murine fibroblasts (58). Moreover, GSH alone, given extracellularly, was able to increase significantly PDGF-R ␤ subunit autophosphorylation possibly by acting as a reducing agent (58). Similar considerations may be proposed for the possible relevance of HNE interactions with histidine residues. In fact, the signal transduction pathway associated with muscarinic acetylcholine and metabotrobe glutamate receptors was inhibited as a consequence of direct interaction of the aldehyde with histidine residues within a G␣ q/11 protein involved in the signal transduction cascade (59). In this context, it is interesting to note that interaction of HNE or other HAKs with target proteins not always results in an inhibition of an enzymatic activity. The activity of several proteins involved in signal transduction, such as adenylate cyclase (60) Confluent cells were incubated for 24 h in a serum-free and insulin free Iscove's medium (SFIF medium) and then exposed to different HAKs (concentration, 1 M) for 6 h. Total RNA and RNase protection assay were performed as described in details under "Experimental Procedures." Equal loading of samples was checked by using a probe for glyceraldehyde-phosphate dehydrogenase (GAPDH) as a housekeeping gene.
FIG. 6. Western blot analysis of PDGF-R ␤ subunit synthesis 24 and 48 h after treatment with individual HAKs. Confluent cells were incubated for 24 h in a serum-free and insulin free Iscove's medium (SFIF medium) and then exposed to different HAKs (concentration, 1 M) and to transforming growth factor ␤1 (TGF␤1, 10 ng/ml), with the latter used as a positive control for PDGF-R ␤ subunit synthesis. Cell lysates and Western blot analysis were performed as described in details under "Experimental Procedures." 4-Hydroxy-2,3-alkenals and PDGF Receptor Signaling kinase (33,34), has been shown to be stimulated by HNE. This suggests that the overall effect of the exposure of defined proteins to HAKs depends on the changes in protein function and/or conformation resulting from the interaction of the aldehydes with critical amino acid residues.
An important observation provided by the present study relates to the general relevance of oxidative stress-related molecules in the modulation of the proliferative response of target cells of mesenchymal origin, such as HSC. Inhibition of DNA synthesis by HAKs is not surprising, because reduced cell growth by these compounds has been extensively reported (20,38,39). However, this report shows for the first time that HNE (and related HAKs) can inhibit DNA synthesis by directly interfering with the signal transduction of a growth factor receptor. The inhibitory effect displayed by 1 M HAKs on PDGF-stimulated DNA synthesis was reversible, because PDGF-induced thymidine incorporation was recovered after 48 h. The fact that the inhibition of the actions PDGF was transient is not surprising, because we have previously shown that in human HSC HAKs rapidly disappear from the culture medium either by forming aldehyde-protein adducts or via cell metabolism (33). The recovery of thymidine uptake in response to PDGF was associated with increased abundance of both PDGF-R ␤ subunit mRNA and the relative protein, although while the message was already up-regulated 6 h after exposure to HAKs, increased protein expression was observed not earlier than after 48 h. The reason for this dissociation in the time course is presently unclear. Induction of gene expression by HAKs has already been reported in several cells for different target genes (38,39). Along these lines, studies performed in FIG. 7. Specific binding of [ 125 I]P-DGF-BB to human HSC in monolayer. Triplicate wells of confluent HSC were preincubated for 15 min in the absence (E) or in the presence (q) of 1 M HNE, used as a model HAK and then incubated with [ 125 I]PDGF-BB at the concentration shown for 2 h at 4°C. Nonspecific binding was determined and was subtracted from the total binding to determine the specific binding. The ordinate indicates bound activity. Values shown are fmol bound/10 5 . Scatchard plot analysis of the [ 125 I]PDGF-BB binding data, representative of three experiments, is shown in the inset.
FIG. 8. Immune complex kinase assay for PDGF-R ␤ subunit intrinsic tyrosine kinase activity. Confluent cells were serum-deprived for 24 h and then either left untreated (lanes 1 and 5) or exposed to 1 M HNE (lane 2), 10 ng/ml PDGF-BB for 10 min (lanes 3 and 6), or HNE 15 min before adding PDGF-BB (lane 4). PDGF-R ␤ subunits were immunoprecipitated and after washing, and the immunobeads were resuspended in 50 mM Hepes, pH 7.4. At this point, samples shown in lanes 5 and 6 were treated in vitro with 1 M HNE for 15 min, whereas the others were left untreated. At the end of this incubation, the kinase assay was started by adding MnCl 2 and radiolabeled ATP, as described under "Experimental Procedures." The samples were then resolved by SDS-PAGE and autoradiographed.
FIG. 9. The effects of HNE are specific for the PDGF-R ␤ subunit. Serum-deprived HSC were incubated in the presence or absence of 1 M HNE for 15 min and then exposed to 10 ng/ml PDGF-BB (A), 100 ng/ml EGF (B), or 50 ng/ml PDGF-AA (C), as indicated. Samples were immunoprecipitated (IP) with antibodies against the PDGF-R ␤ subunit (A), the EGF-R (B), or the PDGF-R ␣ subunit (C) and analyzed by immunoblotting using antiphosphotyrosine (P-Tyr) antibodies (A-C, upper panels) or with antibodies against the PDGF-R ␤ subunit (A, lower panel), the EGF-R (B, lower panel), or the PDGF-R ␣ subunit (C, lower panel).

4-Hydroxy-2,3-alkenals and PDGF Receptor Signaling
human HSC have shown that 1 M HNE up-regulates gene expression of key molecules involved in the liver tissue repair process such as monocyte chemotactic protein-1 (41) and procollagen type I (40). Therefore, the biological action of HAKs may result in multiple effects within the complex network of events occurring in the process of fibrogenesis following reiterated liver tissue injury, the resulting overall biological effect being influenced by the actual HAKs concentration, the target cell(s) available, and the presence of growth factors and other mediators in the microenvironment as well as, as recently pointed out, by a decrease in antioxidant enzymatic defenses versus HAKs in activated hepatic stellate cells (62).