In the present study, we compared the induction of heat shock proteins (HSPs) by heat and heavy metal ions in three different endothelial cell types, namely, human umbilical vein endothelial cells, human pulmonary microvascular endothelial cells, and the cell line EA.hy 926. Our results show that especially Zn2+ and Cd2+ are inducers of 70-kDa (HSP70), 60-kDa (HSP60), 32-kDa (HSP32), and 27-kDa (HSP27) HSPs. The strength of inducibility is specific for each HSP. Ni2+ and Co2+ only show an inducible effect at very high concentrations, that is, in the clearly cytotoxic range. Furthermore, we investigated the time course of HSP expression and the involvement of heat shock factor-1. Our study demonstrates that the three endothelial cell types that were under investigation show comparable stress protein expression when treated with heavy metal ions or heat shock. The expression of stress proteins may be used as an early marker for the toxic damage of cells. This damage can be an inducer of acute respiratory distress syndrome in which microvascular endothelial lesions occur early. Our study provides evidence that human umbilical vein endothelial cells or EA.hy 926 cells, which are much more easily isolated and/or cultivated than pulmonary microvascular endothelial cells, could be used as alternative cell culture systems for studies on cellular dysfunction in the lung caused by toxic substances, certainly with respect to the expression of HSPs.
- heat shock proteins
- stress response
- endothelial cell culture
many environmental changes such as heat, certain metals, toxins, and oxidative stress (anoxia, reactive oxygen metabolites) cause changes in the pattern of cellular stress protein expression (4, 9, 19). This group of proteins, referred to as heat shock proteins (HSPs), is divided into subfamilies that are characterized by their molecular mass (for reviews, see Refs. 3, 21, 39). Stress proteins have been highly conserved during evolution and are present in the entire spectrum of organisms from bacteria to human cells. In higher eukaryotes, they have been identified in many different cell types. Indeed, HSPs enable procaryotes as well as higher organisms to tolerate the different forms of stress and to survive, but they also have physiological functions and serve as molecular chaperones (3, 21, 39). Proteins of especially the 70-kDa (HSP70) and 60-kDa (HSP60) HSP families function in protein folding, assembly, and translocation between intracellular compartments (5). Other HSPs are involved in the degradation of misfolded proteins (10). Furthermore, HSPs also play a role in the immune response (33; for a review, see Ref. 24). Many HSPs are constitutively expressed, and in situations of stress, the expression pattern of many HSPs is changed. The heat shock response is regulated by a transcription factor known as heat shock factor (HSF)-1. In unstressed cells, this factor is present in a monomeric form in the cytoplasm as well as in the nucleus. With stress, HSF-1 trimerizes, and this trimerization causes a conformational change that allows HSF to bind to responsive elements in the promoters of stress-inducible genes. Furthermore, HSF becomes phosphorylated, and this phosphorylation correlates with DNA binding (14). The heat shock response is autoregulated. The synthesis of HSPs to acquire thermotolerance is “titrated” by precise mechanisms of new synthesis, accumulation, and half-life-dependent degradation of the proteins as described by Mizzen and Welch (22).
Because of their exposed location in the vasculature, endothelial cells are involved in pathological situations such as ischemia and hypoxia, viral and bacterial infections, inflammation, fever, and cancer. Changes in HSP expression by endothelial cells were demonstrated in many of these conditions (23). Other investigators (1, 37, 42) demonstrated the heat- and metal-inducible expression of HSPs in various cell types. The endothelium is a target for metal ions that are transported and distributed by the blood. Therefore, the aim of the present study was to analyze and compare heavy metal ion as well as heat shock effects in human umbilical vein endothelial cells (HUVECs), human pulmonary microvascular endothelial cells (HPMECs), and EA.hy 926 cells by measuring the levels of 27-kDa HSP (HSP27), 32-kDa HSP (HSP32), HSP60, HSP70, and 90-kDa HSP (HSP90) at both the protein and mRNA levels.
MATERIALS AND METHODS
Cell culture and treatments.HUVECs were isolated as previously described (13). The cells were cultured in a 1:1 mixture of Ham's F-12 medium and Iscove's modified Dulbecco's medium (GIBCO BRL) supplemented with 20% human serum, 20 U of penicillin-20 μg/ml of streptomycin solution (GIBCO BRL), and 2 mMl-glutamine (GIBCO BRL) in a humidified atmosphere containing 5% CO2 and 10% O2. Confluent HUVEC monolayers were passaged with collagenase. EA.hy 926 cells were cultured in Dulbecco's modified Eagle's medium (GIBCO BRL) supplemented with 10% fetal calf serum (GIBCO BRL), 20 U of penicillin-20 μg/ml of streptomycin solution (GIBCO BRL), 2 mM GLUTAMAX-I supplement (GIBCO BRL), and HAT supplement (5 mM hypoxanthine, 20 μM aminopterin, and 0.8 mM thymidine) (GIBCO BRL). The cells were passaged with 0.25% trypsin-EDTA.
Isolation and culture of HPMECs were performed with a modification of the method of Hewett and Murray (11). Briefly, lung tissue from pulmonary resections due to tumors was separated under sterile conditions from the overlying pleura and mechanically cut into small fragments with scissors. After removal of debris and erythrocytes through a 40-μm nylon net, the tissue was treated with dispase (1.18 U/ml at 4°C for 18 h). After filtration through a 100-μm nylon net, the tissue was treated in a volume of 4 ml with elastase (40 U), trypsin (0.05%), and EDTA (1.8 mM) for 30 min at 37°C followed by a further 100-μm net filtration. The cell clumps were then repeatedly resuspended in PBS-BSA and filtered through a 40-μm net. The cell suspension was then centrifuged for 10 min at 173 g, and the cell pellet was resuspended in medium 199 plus 20% pooled human serum. This mixed cell culture was cultivated at 37°C in a gas mixture of 5% CO2 in air for 5–7 days, but not longer, to avoid overgrowth of myofibroblastic cells. The developing monolayer was disrupted by treatment with 0.2% EDTA and 0.2% BSA for 30 min at 37°C followed by a mixture of 0.25% trypsin and 0.25% EDTA for 1 min. The positive selection of HPMECs was achieved by interacting the cell suspension with magnetic beads (1 μm diameter) coated with a mouse monoclonal antibody against human platelet endothelial cell adhesion molecule-1 (Immunotech). The subsequent pure cultures of platelet endothelial cell adhesion molecule-1-positive HPMECs were also positive for CD34, factor VIII-related antigen, CD36 (thrombospondin receptor),Ulex europeaus agglutinin-1, prostacyclin, interleukin (IL)-1, IL-6, and plasminogen activator inhibitor-1. The cells also showed uptake of 1,1′-dioctadecyl-1,3,3′,3′-tetramethylindocarbocyanine-acetylated low-density lipoprotein.
For induction experiments, cells were seeded, cultured to confluency, and then exposed to the heavy metal salts ZnCl2, NiCl2, CoCl2, and CdCl2 or to heat shock at 42.5°C for 3 h. After these treatments, the cells were washed with prewarmed (37°C) fresh medium and were then allowed to recover at 37°C for 4 h.
Indirect immunofluorescence staining.For immunofluorescence staining of HSPs and HSF-1, cells were seeded on fibronectin-coated culture slides (Nunc) and after 24 h were treated with heavy metal ions or heat shock. After the treatments, fixation was performed in 3.7% paraformaldehyde in fixation buffer (0.1 M PIPES, 1 mM EGTA, 4% polyethylene glycol 8000, and 0.1 M NaOH, pH 6.9) followed by permeabilization with 0.5% Triton X-100 in fixation buffer. After three washes, the cells were incubated in 0.1% sodium borohydride in PBS for 10 min and washed. Unspecific binding sites were blocked by incubation in 5% BSA in PBS for 30 min. The primary antibody (mouse monoclonal anti-human HSP27, rabbit polyclonal anti-human HSP32, mouse monoclonal anti-human HSP60, mouse monoclonal anti-human HSP70, mouse monoclonal anti-human HSP90, or chicken anti-human HSF-1 antibody; all 1:200 dilution; StressGen Biotechnologies) was added in 1% BSA-PBS and incubated for 1 h followed by three washes. After incubation with the secondary antibody [1:50 goat anti-mouse FITC-conjugated Ig (DAKO), 1:50 swine anti-rabbit FITC-conjugated Ig (DAKO), or 1:100 donkey anti-chicken FITC-conjugated Ig (Dianova)], the cells were washed again and counterstained with 1:50 phalloidin-TRITC (Sigma) or with rabbit anti-human fibronectin antibody (Sigma) and donkey anti-rabbit IgG-cyanine 3 (Dianova). Immunostained slides were covered with a solution consisting of 50% glycerol in PBS, 0.02% sodium azide, and 100 mg/ml of 1,4-diazabicyclo[2.2.2]octane (Sigma) and analyzed by epifluorescence microscopy (Leica DMRX).
RNA extraction and Northern blot analysis. Total RNA was extracted with an RNA extraction kit (RNeasy, Qiagen, Hilden, Germany) following the procedure described by the manufacturer. RNA samples were electrophoresed on 1.2% agarose-formaldehyde gels and blotted onto nylon membrane (Hybond-N, Amersham). To control equal loading of different RNA samples on a gel, the ethidium bromide staining of 28S and 18S rRNAs was checked before and after the samples were blotted. The blots were hybridized with digoxigenin-labeled antisense RNA probes. These were generated by in vitro transcription of cDNA probes (StressGen Biotechnologies) that were cloned into the pBluescript phagemid. Labeling and hybridization were performed with the digoxigenin RNA labeling and detection system (Boehringer Mannheim). Hybrids were visualized with a chemiluminescence-producing substrate (CSPD or CDP-Star).
Cell lysates. To prepare extracts of total cell protein, endothelial cell monolayers were washed three times with ice-cold PBS followed by the addition of radioimmunoprecipitation assay buffer (10 mM Tris ⋅ HCl, pH 7.4, 158 mM NaCl, 0.1 mM EGTA, 1% Triton X-100, 1% deoxycholate, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 10 μg/ml of aprotinin, and 10 μg/ml of leupeptin). Cell material was scraped off the flasks with a rubber policeman, transferred to a 2.0-ml vial, and allowed to lyse for an additional 30 min on ice. The samples were centrifuged for 30 min at 4°C at 14,000g, and then the supernatants were transferred to a new vial. Protein concentration was determined by the Bradford method (Coomassie Plus Protein Assay Reagent, Pierce, Rockford, IL).
Immunoblotting. To detect HSP27, HSP32, HSP60, HSP70, and HSP90, cellular protein extracts were separated by SDS-PAGE followed by Western immunoblotting. Protein samples were run on 10% SDS-polyacrylamide minigels and transferred to a nitrocellulose membrane (Schleicher and Schuell) by semidry electroblotting in a buffer containing 25 mM Tris ⋅ HCl, 190 mM glycine, 0.5% SDS, and 20% methanol. Transfer was controlled by staining the membrane with Ponceau S and the gel with Coomassie brilliant blue solution. Membranes were blocked by incubation in 5% nonfat dried milk in 20 mM Tris ⋅ HCl, pH 7.6, 137 mM NaCl, and 0.1% Tween 20 [Tris-buffered saline (TBS)-Tween 20 (TBS-T)] for 1 h at room temperature. Incubation with the primary antibody (1:2,000 mouse monoclonal anti-human HSP27, 1:1,500 rabbit polyclonal anti-human HSP32, 1:2,000 mouse monoclonal anti-human HSP60, 1:4,000 mouse monoclonal anti-human HSP70, or 1:3,000 mouse monoclonal anti-human HSP90 antibody; StressGen Biotechnologies) was performed for 60 min, and then the membranes were washed four times with TBS-T and incubated with the secondary antibody (rabbit anti-mouse or swine anti-rabbit Ig, both conjugated with horseradish peroxidase; DAKO) for 30 min. After five washes with TBS-T and one wash with TBS, development was performed with an enhanced chemiluminescence detection system (Amersham).
Immunoprecipitation. For immunodetection of HSF-1 in cell extracts, the protein was immunoprecipitated before SDS-PAGE and Western blot analysis. Therefore, the cells were lysed (see Cell lysates) in lysis buffer containing 10 mM Tris ⋅ HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% deoxycholate, 1 mM phenylmethylsulfonyl fluoride, 10 μg/ml of leupeptin and aprotinin, 10 mM NaF, 1 mM Na3VO4, and 1 μM calyculin A. The extract was precleared with 50% protein A-Sepharose in lysis buffer for 15 min at 4°C. Sepharose was collected by centrifugation, and the supernatant was transferred to a new vial. After protein determination, 400 μg of protein of each sample were taken off and PBS was added to give a volume of 500 μl. One microliter of rabbit polyclonal anti-human HSF-1 antibody (StressGen Biotechnologies) was added and incubated overnight at 4°C with rotation. Fifty microliters of protein A-Sepharose were added, and incubation was performed for 2 h at 4°C. The Sepharose was washed three times with lysis buffer and twice with 10 mM Tris ⋅ HCl, pH 7.5, and immunoprecipitated protein was eluted by boiling the sepharose in 50 μl of Laemmli sample buffer for 5 min at 95°C. Twenty-five microliters of the supernatant were separated on a 7.5% SDS-polyacrylamide gel and blotted onto nitrocellulose membrane (seeImmunoblotting) for immunoblot analysis. The membrane was blocked with 2% BSA in 10 mM Tris ⋅ HCl, pH 7.5, 150 mM NaCl, 0.05% Tween 20, and 10 mM NaF for 1 h followed by incubation with 1:5,000 rabbit polyclonal anti-human HSF-1 antibody (StressGen Biotechnologies) for 60 min. The membrane was washed and incubated with the second antibody (swine anti-rabbit HRP-conjugated Ig; DAKO) for 30 min. Immunoreactive protein was detected with the enhanced chemiluminescence system (Amersham).
Induction of HSPs in endothelial cells by heavy metals. We investigated the expression of HSP27, HSP32, HSP60, HSP70, and HSP90 in HUVECs treated with heavy metal ions. Immunofluorescence staining of HUVECs resulted, in the case of HSP32 and HSP60, in weak signals (data not shown). On the contrary, HSP27 was strongly expressed in the cytoplasm of nonstressed cells but not in the nucleus (Fig.1 A). The expression was not equal in all cells. HSP90 was expressed constitutively in the cytoplasm as well as in the nucleus, and its distribution appeared granular (Fig.1 B). HSP70 was hardly detectable in noninduced endothelial cells, but it could be shown by immunofluorescence staining that the amount of this protein was significantly augmented in the nucleus in Zn2+- and Cd2+-treated as well as in heat-shocked cells (Fig. 1, C andD). A similar effect was observed concerning HSP32 expression (data not shown).
Western blot analysis confirmed our data from immunofluorescence staining. We found that there was almost no expression of HSP70 in quiescent cells, but the protein is strongly induced by Zn2+ and Cd2+ (Fig.2). Zn2+ was most inductive in a concentration of 1 mM, whereas Cd2+ induced HSP70 at much lower concentrations. A 100 μM concentration of Cd2+ was still inductive. Ni2+ and Co2+ had no influence on HSP70 expression, even at the high concentrations of 1 or 2 mM. HSP60 was induced by Zn2+ (1 mM) and, to a lesser extent, by Cd2+ (Fig. 2). The induction of HSP60 was much weaker than that of HSP70. HSP32 was not detectable in untreated HUVECs by Western blot analysis, but there was a strong induction of HSP32 by Zn2+ (1 mM) and Cd2+ (0.1 mM and 0.5 mM). Furthermore, there was a weak but clear expression of HSP32 after treatment with Co2+ at 1.5 mM (Fig. 2). The expression of HSP90 was not significantly enhanced in HUVECs by Ni2+, Zn2+, or Cd2+ (Fig.3).
Changes in HSP mRNA expression were studied by Northern blot analysis. Endothelial cells were treated with Cd2+, Zn2+, Ni2+, and Co2+ in three different concentrations for 3 h followed by a further incubation with fresh medium without metal ions for an additional 4 h. The results are shown in Fig. 4. HSP70 mRNA was increased by Cd2+ (0.5 mM) and Zn2+ (1 and 0.5 mM) and further by Ni2+ at an extremely high concentration of 10 mM. HSP60 mRNA was only augmented by 1 mM Zn2+. HSP27 was induced on the transcriptional level by Cd2+ and Zn2+ but not by Ni2+ or Co2+.
In addition to HUVECs, we analyzed HSP expression in HPMECS. The results were very similar to those obtained from HUVECs concerning the induction of HSP70, HSP60, and HSP32 (Fig.5 A). Zn2+ and Cd2+ were inductive, whereas Ni2+ was not. In agreement with the effects seen for HUVECs, we detected the strongest induction of HSP60 by Zn2+ at a 1 mM concentration. HSP32 was also induced by Zn2+ (1 mM) and Cd2+ (0.1 and 0.5 mM) in HPMECs. Furthermore, we investigated HSP70 expression in EA.hy 926 cells as shown in Fig. 5 B. HSP70 was induced by Zn2+ and Cd2+ as demonstrated for HUVECs and HPMECs. Unlike the latter two endothelial types, EA.hy 926 cells gave a clearly detectable constitutive expression of HSP70 (Figs. 2 and5, A andB). On the mRNA level, our results with EA.hy 926 cells were very similar to those obtained with HUVECs (data not shown).
HSP induction by heat shock. We analyzed the expression of HSP70 in HUVECs during the time course of heat shock. A strong induction of HSP70 transcripts was found as shown in Fig.6 A. The mRNA level reached a maximum after 2 h and then decreased. The protein level was augmented continuously during 6 h of heat shock, beginning between 60 and 90 min (Fig. 6 B), so the increase in the protein level was delayed compared with the mRNA level.
Although the level of HSP27 mRNA was increased by heat shock (Fig.7 A), HSP27 protein decreased in the early phase of heat shock (Fig.7 B). Compared with HSP70 mRNA expression, the level of HSP27 mRNA increased more slowly and declined later.
As shown in Fig. 5 A, we detected a weak induction of HSP60 by heat shock, but this effect could not be verified in all experiments. We did not detect an increase in HSP60 mRNA after heat shock (data not shown). Furthermore, we could not measure any heat shock effect on HSP32 expression (Fig.5 A) or on the synthesis of HSP90 (Fig. 3).
Analysis of HSF-1 in HUVECs. We analyzed the activation of HSF-1 in HUVECs during heat shock by immunoprecipitation followed by immunoblot analysis. The electrophoretic mobility of HSF-1 in SDS-polyacrylamide gels is due to its phosphorylation status as demonstrated by an earlier study (31). Therefore, we analyzed the behavior of HSF-1 in HUVECs from heat-shocked and untreated cells. We show that HSF-1 from cells that were incubated at 42.5°C for up to 60 min migrated more slowly than HSF-1 from cells incubated at 37°C (Fig.8). Phosphorylation of HSF-1 declined again to the basic level when the cells were heat shocked for 3 or 8 h. Furthermore, we studied the localization of HSF-1 in endothelial cells by immunofluorescence staining. We observed a translocation of HSF-1 from the cytoplasm into the nucleus in endothelial cells 30 min after transfer to a 42.5°C incubator (data not shown).
We demonstrated that the stress proteins HSP70, HSP60, HSP32, and HSP27 are inducible in human endothelial cells by heat shock and heavy metal ions. In particular, regarding the effects of metals, we show that HSP70 is induced especially by Zn2+ and Cd2+ at the mRNA and protein levels, with Cd2+ still being inductive at lower concentrations. A significant effect of Ni2+ and Co2+ even at high concentrations of 1 or 2 mM on HSP70 expression was not detectable either in Western or in Northern blot analysis. Only treatment of the cells with a 10 mM concentration of Ni2+, which obviously represents marked stress and induced morphological changes such as cytoplasmic retraction, induced elevated levels of HSP70 mRNA.
Besides Zn2+ and Cd2+, arsenite has been described as a very strong inducer of HSP. It was shown by Brown and Rush (2) and Ribeiro et al. (28) that an injection of sodium arsenite into rabbits and rats induces the synthesis of proteins of the HSP70 family.
HSP60 was much more weakly inducible in HUVECs by heavy metal ions than HSP70. We observed the strongest induction by Zn2+ at a 1 mM concentration in both protein and transcript analyses. As for HSP70, the expression of HSP27 was induced by Zn2+ and Cd2+, whereas Co2+ or Ni2+ did not have a significant influence as demonstrated by Northern blot analysis.
HSP32, also referred to as heme oxygenase-1 (HO-1), was first described for its enzymatic function in heme metabolism, that is, the cleavage of heme with the generation of biliverdin, carbon monoxide, and iron (35). Its role in the defense against toxic substrates like sodium arsenite and cadmium chloride in HeLa and HL60 cells was described by Taketani et al. (34). Shibahara et al. (32) demonstrated the heat inducibility of HO-1 in rat glioma cells. Furthermore, HO-1 plays a role in an antioxidant defense mechanism as in ischemic rat brain where the protein is upregulated (25). Several other studies analyzed HSP32 expression in endothelial cells. The protein is involved in the regulation of inflammatory processes because it is induced in human endothelial cells by IL-1 and tumor necrosis factor-α (36). Wagener et al. (38) demonstrated the induction of HSP32 in HUVECs after exposure to tumor necrosis factor-α or heme. Another function of HSP32 is its involvement in nitric oxide-mediated changes of the cellular balance between heme and iron and between nonheme and iron in vascular endothelial cells (41). In our study, we have shown that the synthesis of HSP32 is enhanced in human endothelial cells by Zn2+ and Cd2+ and, to a lesser extent, by Co2+.
There was no augmentation of HSP90 in HUVECs after treatment with Zn2+, Cd2+, Ni2+, or Co2+. Bauman et al. (1) described the induction of HSP90 in rat hepatocytes by Cd2+ and Zn2+, but the induction was only twofold or less. HSP90 is a very abundant protein in mammalian cells and is known to interact with tyrosine kinases, e.g., pp60src (26), steroid hormone receptors (16), and the eukaryotic initiation factor-2α kinase (30), which is involved in regulating cellular protein synthesis.
Furthermore, we analyzed the inducibility of endothelial stress proteins by heat shock at 42.5°C. HSP70 expression was augmented after heat shock in all three endothelial cell types investigated. A time-course study showed that HSP70 mRNA is first increased and then declines, whereas protein expression is maintained much longer. When we analyzed the expression of HSP27, we found that this protein is regulated in a different way in heat-shocked endothelial cells than HSP70. We detected a slight increase in HSP27 mRNA, but the protein level decreased during the first 90 min of heat shock treatment. It must be concluded that the level of HSP27 is not only dependent on the presence of mRNA but that protein stability and protein degradation are important regulatory mechanisms. HSP27 has been implicated in the organization of actin filaments, and it was demonstrated that this function is accompanied by its phosphorylation (17, 18). In addition, it was shown by Huot et al. (12) that oxidative stress mediated by H2O2induces in HUVECs the reorganization of actin by HSP27, which is phosphorylated via the p38 mitogen-activated protein (MAP) kinase pathway. Li et al. (20) demonstrated phosphorylation of HSP27 in HUVECs treated with shear stress. Freshney et al. (7) described the identification of new kinases that are involved in the phosphorylation of HSP27. These are not MAP kinases, but they could be related to MAP kinase cascades. These data suggest that phosphorylation may be the main regulatory mechanism for HSP27 activity and not for the protein level. We conclude that in our study HSP27 activity can be enhanced despite the heat-induced decrease of the protein level. Further experiments have to elucidate these HSP27-regulating mechanisms.
There are studies from other laboratories that analyzed the hyperthermic effects in endothelial cells. Rinaldo et al. (29) showed that hyperthermia induces HSP70 in bovine pulmonary artery endothelial cells, whereas lipopolysaccharide does not. Another effect of heat on endothelial cells (HUVECs) is the induction of a hyperfibrinolytic state by enhancement of plasminogen activator and plasminogen activator receptor (8). Furthermore, other modulators of stress proteins in endothelial cells have been described. Thus oxidized low-density lipoprotein induces HSP70 as demonstrated with HUVECs and EA.hy 926 cells (43). HSP70 is suggested to play a critical role in atherosclerosis because it was found together with 65-kDa HSP in atherosclerotic plaques (40).
Interestingly, the expression of HSPs was correlated with cellular aging and senescence. Piotrowicz et al. (27) demonstrated that the overexpression of HSP27 in arterial endothelial cells stimulates cell growth and accelerates senescence.
The expression of HSP mRNA is regulated by the inducible transcription factor HSF-1. We could confirm the translocation of HSF-1 from the cytoplasm into the nucleus by immunofluorescence staining. Furthermore, the phosphorylation status of HSF-1 is enhanced in the early phase of heat shock, and this correlates with the induction of HSP70 mRNA.
In the present study, we investigated the induction of HSPs in HUVECs as well as in microvascular endothelial cells that were isolated from the human lung (HPMECs) and in the cell line EA.hy 926, which was established by fusion of HUVECs with the epithelial cell line A549 (6). We show that these different endothelial cell types behave very similarly with regard to the induction of HSPs under stress. For further studies, the cell line EA.hy 926 may be used as a model of endothelial cells for in vitro studies, although the constitutive expression of HSP70 in untreated cells does differ from the situation in primary isolated endothelial cells, which under unstimulated conditions do not express HSP70. The marked similarity of reaction for HUVECs and HPMECs provides evidence that macro- and microvascular endothelial cells (ECs) have no major difference in their HSP response to metal ion treatment. This is of relevance for research on toxic damage in the lung, which can lead to the severe condition of acute respiratory distress syndrome. The latter is known to involve impairment of the pulmonary microcirculation (15). Research activity in the pathogenesis of pulmonary endothelial dysfunction is hampered by the fact that only very few groups are able to obtain and cultivate HPMECs. Our results suggest that it may be valid to use the much more widely available HUVEC model in such studies.
Regarding HSP induction by metal ions, our results clearly show that not all metals may increase the production of stress proteins, and if they do, their effects can be stronger or weaker in comparison to each other. Because some stress proteins are already activated at lower metal concentrations, as demonstrated for Cd2+, the heat shock response is not only a marker for irreversible cytotoxicity (as occurs at higher concentrations) but also for low-level toxicity. Furthermore, the incubation of endothelial cells at 42.5°C did not induce all HSPs under investigation depending on the specific function of each stress protein.
We gratefully acknowledge the kind gift of the EA.hy 926 cell line from Dr. Cora-Jean S. Edgell (University of North Carolina at Chapel Hill).
Address for reprint requests and other correspondence: C. J. Kirkpatrick, Institute of Pathology, Johannes Gutenberg Univ., Langenbeckstr. 1, D-55101 Mainz, Germany (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 1999 the American Physiological Society