Glutathione enhances fibroblast collagen contraction and protects keratinocytes from apoptosis in hyperglycaemic culture.

BACKGROUND: Cutaneous wound healing is relatively slow in patients with diabetes. OBJECTIVES: To test the hypothesis that this defect in healing of wounds in patients with diabetes results from dysfunction of skin fibroblasts and epidermal keratinocytes and that this dysfunction is related to disrupted intracellular glutathione (GSH) homeostasis. METHODS: We investigated the effects of esterified GSH on the contraction of fibroblasts in a fibroblast-populated collagen lattice and on keratinocyte apoptosis. RESULTS: High glucose medium (hyperglycaemia) reduced the contraction ability of fibroblasts (P < 0.05). The normalization of glucose medium concentrations for hyperglycaemic fibroblasts did not restore the contraction capacity. The percentage of apoptotic keratinocytes was statistically higher in hyperglycaemic cells (P < 0.05). GSH media concentrations ranging from 0.1 to 100 micromol L(-1) restored the ability of hyperglycaemic fibroblasts to contract the gels in a concentration-dependent manner. Primary human keratinocytes grown in hyperglycaemic medium were more susceptible to apoptosis, and treatment with esterified GSH rescued the keratinocytes from apoptosis. CONCLUSIONS: These data suggest that intracellular GSH can normalize skin cell functions disrupted by in vitro cell growth under hyperglycaemic conditions.

Endothelial barrier dysfunction caused by LPS correlates with phosphorylation of HSP27 in vivo.

Lung edema during sepsis is triggered by formation of gaps between endothelial cells followed by macrophage infiltration. Endothelial gap formation has been proposed to involve changes in the structure of the actin filament cytoskeleton. Heat shock protein 27 (HSP27) is believed to modulate actin filament dynamics or structure, in a manner dependent on its phosphorylation status. We hypothesized that HSP27 may play a role in endothelial gap formation, by affecting actin dependent events in endothelial cells. As there has been no report concerning HSP27 in lung edema in vivo, we examined induction and phosphorylation of HSP27 in lung following LPS injection, as a model of sepsis. In lung, HSP27 mainly localized in capillary endothelial cells of the alveolus, and in smooth muscle cells of pulmonary arteries. HSP27 became significantly more phosphorylated at 3 h after LPS treatment, while the distribution of HSP27 remained unchanged. Pre-treatment with anti-TNFalpha antibody, which has been shown to reduce lung injury, blocked increases in HSP27 phosphorylation at 3 h. HSP27 phosphorylation was also increased in cultured rat pulmonary arterial endothelial cells (RPAEC) by treatment with TNFalpha, LPS, or H2O2. This phosphorylation was blocked by pre-treatment with SB203580, an inhibitor of the upstream kinase, p38 MAP kinase. Increased endothelial permeability caused by H2O2 in vitro was also blocked by SB203580. The amount of actin associated with HSP27 was reduced after treatment with LPS, or H2O2. In summary, HSP27 phosphorylation temporally correlated with LPS induced pathological endothelial cell gap formation in vivo and in a cell culture model system. This is the first report of increased HSP27 phosphorylation associated with pathological lung injury in an animal model of sepsis.

HSP22, a new member of the small heat shock protein superfamily, interacts with mimic of phosphorylated HSP27 ((3D)HSP27).

Most of the members of the superfamily of mammalian small heat shock or stress proteins are abundant in muscles where they play a role in muscle function and maintenance of muscle integrity. One member of this protein superfamily, human HSP27, is rapidly phosphorylated on three serine residues (Ser(15), Ser(78), and Ser(82)) during cellular response to a number of extracellular factors. To understand better the role of HSP27, we performed a yeast two-hybrid screen of a human heart cDNA library for HSP27-interacting proteins. By using the triple aspartate mutant, a mimic of phosphorylated HSP27, as "bait" construct, a protein with a molecular mass of 21.6 kDa was identified as an HSP27-binding protein. Sequence analysis revealed that this new protein shares an overall sequence identity of 33% with human HSP27. This protein also contains the alpha-crystallin domain in its C-terminal half, a hallmark of the superfamily of small stress proteins. Thus, the new protein itself is a member of this protein superfamily, and consequently we designated it HSP22. According to the two-hybrid data, HSP22 interacts preferentially with the triple aspartate form of HSP27 as compared with wild-type HSP27. HSP22 is expressed predominantly in muscles. In vitro, HSP22 is phosphorylated by protein kinase C (at residues Ser(14) and Thr(63)) and by p44 mitogen-activated protein kinase (at residues Ser(27) and Thr(87)) but not by MAPKAPK-2.

Melanocyte-conditioned medium stimulates while melanocyte/keratinocyte contact inhibits keratinocyte proliferation.

The interaction between melanocytes and keratinocytes in epidermal tissue suggest a bidirectional interchange between these two cell types. Although keratinocytes appear to affect melanocyte function, there are no reported effects of melanocytes on keratinocytes. Using cell strains, we examined the effect of melanocytes on keratinocyte proliferation. Two conditioned medium techniques were used: one was a co-culture system, where both cell types, grown on separate surfaces shared a common volume of medium. The second was simply feeding keratinocytes melanocyte-conditioned medium. Mixed cultures (both cell types together in a monolayer) where also studied. Our results showed that melanocyte-conditioned medium and melanocytes in co-culture significantly stimulated keratinocyte proliferation as measured by bromodeoxyuridine incorporation assay. However, growth of both cell types together in culture did not affect the growth rate of either cell type. Our results showed that cultured human melanocytes produce one or more soluble factors that stimulate the growth of cultured keratinocytes.

Identification and characterization of a novel protein from Sertoli cells, PASS1, that associates with mammalian small stress protein hsp27.

hsp27 is involved in development of tolerance to stress, possibly by its involvement in molecular chaperoning, maintenance of glutathione status, and/or modulation of microfilament structure and function. We hypothesize that hsp27 function depends on specific association with other proteins. To discover proteins that associate with hsp27, we made a differentiated rat Sertoli cell cDNA expression library and screened it using the yeast two-hybrid system. We obtained a cDNA coding for a novel protein of 428 amino acids that we have named PASS1 (protein associated with small stress proteins 1). BLAST searches did not reveal major similarity of PASS1 to any known protein, but the cDNA sequence matched several mouse EST clones and shares 34% homology with a Caenorhabditis elegans genomic sequence. In vitro, bacterially expressed glutathione S-transferase-PASS1 fusion protein bound to hsp27, and hsp27 was co-immunoprecipitated with c-Myc-tagged PASS1 overexpressed in several cell lines. The region of PASS1 responsible for association with hsp27 was identified as existing predominantly between amino acids 108 and 208 of PASS1. Northern hybridization and Western blot analysis demonstrated that PASS1 is expressed in several tissues, with the highest expression occurring in testis, primarily in Sertoli cells. The presence of a 1.4-kilobase PASS1 mRNA in kidney as well as the 1. 8-kilobase mRNA seen in other tissues suggests that alternate splicing may occur in this organ. Ectopic expression of PASS1 in two cultured cell lines was observed to inhibit the ability of hsp27 to protect cells against heat shock, indicating that PASS1 does interact with hsp27 in the live cell.

Effects of glutathione depletion on oxidant-induced endothelial cell injury.

Ischemia-reperfusion produces edema in vivo by disrupting endothelial cell junctional integrity. A cultured rat pulmonary artery endothelial cell (RPAEC) model was used to analyze the effects of oxidants and ischemic plasma in vitro. RPAEC cultures were treated with ischemic human plasma from transverse rectus abdominis musculocutaneous (TRAM) flaps following mastectomy or with an equal quantity of nonischemic plasma taken peripherally. Endothelial cells treated with ischemic plasma rounded and formed gaps within 5 min, then ruffled and blebbed after 10 min. Cultures treated with human nonischemic plasma had no gross morphological changes. Additionally, cultures treated with human ischemic plasma demonstrated an increase in diffusion rate of 125I-albumin across monolayers while monolayers treated with human nonischemic plasma had no increase in diffusion rate. RPAEC monolayers were treated with malic acid diethyl ester (DEM) or L-buthionine-[S, R]-sulfoximine (BSO) to decrease cellular stores of glutathione before exposure to oxidant stress. Cultures depleted of cellular glutathione stores were significantly (P < 0.05) more susceptible to 50 microM H2O2 than controls, as determined by an increase in diffusion rate of 125I-albumin across monolayers. To determine if ischemic plasma effects were mediated by oxidants, cultures were depleted of glutathione by DEM or BSO pretreatment before exposure to plasma from the ischemic hind limbs of Sprague-Dawley rats. Glutathione-depleted RPAEC monolayers were significantly (P < 0.05) and substantially (2-3 X) more susceptible to the effects of ischemic plasma than were cultures with normal glutathione levels. Glutathione depletion had no effect on cultures treated with an equal amount of nonischemic plasma from sham-operated rats. These data strongly suggest that ischemic plasma in the absence of any cellular component are able to induce an oxidant injury in endothelial cells and thereby compromise junctional integrity.

Variation in expression of hsp27 messenger ribonucleic acid during the cycle of the seminiferous epithelium and co-localization of hsp27 and microfilaments in Sertoli cells of the rat.

The purpose of these studies was to define expression of hsp27 mRNA during the cycle of the seminiferous epithelium and to determine the distribution of hsp27 protein in the rat testis. To study hsp27 mRNA expression, a rat hsp27 cDNA was isolated and sequenced (GenBank no. M86389). The cDNA was used in Northern blot analysis to estimate the relative levels of hsp27 mRNA in rat seminiferous tubule segments selected for different stages of the cycle of the seminiferous epithelium. The level of hsp27 mRNA was low during stages IX-XII of the cycle of the seminiferous epithelium. Approximately 15-fold higher levels of hsp27 mRNA were expressed during stages II-VI, with intermediate levels being expressed during stages XIII-I and VII-VIII. No effect of FSH on hsp27 mRNA expression was detected in cultured Sertoli cells, suggesting that hsp27 synthesis in Sertoli cells is not directly regulated by FSH. The distribution of hsp27 was also studied by use of immunofluorescence in frozen sections of rat testis, in isolated seminiferous tubules, in primary cultures of Sertoli cells isolated from 19-26-day-old rats, in peritubular myoid cells from 26-day-old rats, and in several cell lines. Microfilaments were localized in similar preparations by using rhodamine-phalloidin or BODIPY-phallicidin (Molecular Probes, Eugene, OR). The hsp27 was co-localized with micro-filaments in Sertoli cells from 20-day-old and older rats, but not in Sertoli cells from younger rats. In other cell types, hsp27 was diffusely distributed throughout the cytoplasm. These results demonstrate that hsp27 expression varies with the cycle of the seminiferous epithelium and provide the first direct morphological evidence that hsp27 is associated with microfilaments in a normal, intact tissue. They also suggest that Sertoli cell micro-filaments, by virtue of their associated hsp27, may be different in composition and function from microfilaments of peritubular myoid cells and many other cell types.

TNF-alpha potentiates oxidant and reperfusion-induced endothelial cell injury.

Pulmonary edema following reperfusion is a major clinical problem. Changes in endothelial cell shape induced by oxidant injury may account for immediate capillary leakage associated with reperfusion injury. In these experiments we examined the role of tumor necrosis factor-alpha (TNF-alpha) in acute endothelial cell injury following ischemia-reperfusion. Sprague-Dawley rats were treated with a neutralizing antisera directed against TNF-alpha prior to production of distal ischemia. These rats demonstrated a significant reduction (P < 0.05) in acute lung edema in response to 4 hr of ischemia and 30 min of reperfusion when compared to rats undergoing the same procedure without antisera treatment. An in vitro model was developed to determine if TNF-alpha had a direct effect on endothelial cell response to ischemia-reperfusion. The effects of TNF-alpha and oxidant stress on the integrity of cultured endothelial cell monolayers was measured. Rat pulmonary artery endothelial cell monolayers reacted in vitro to oxidant stress by an increase in permeability. The cells changed shape and an increase in diffusion of 125I-albumin across cell monolayers resulted when these cells were exposed to 50 microM hydrogen peroxide (H2O2) or plasma from the ischemic hind limb of a Sprague-Dawley rat (50 microliters/ml). Pretreatment of cultured cells with low levels of recombinant mouse TNF-alpha significantly affected both the cell shape change and the increase in permeability (P < 0.05). Increased permeability of cell monolayers in vitro was not due to cell lysis as determined by media lactate dehydrogenase levels. The effect appeared to be due to cellular rounding and contraction seen using video time lapse microscopy. These data suggest a direct effect of TNF-alpha on endothelial cells, whereby the cells are rendered more susceptible to oxidant injury accompanying reperfusion.

dT diaphorase: increased enzyme activity and mRNA expression in oxidant stress of skin.

DT diaphorase is a flavoprotein that enzymatically transfers two electrons from quinones as intermediate substrates and has been reported to increase its activity in the liver after exposure to toxicants. In this series of experiments, we tested the hypothesis that DT diaphorase also increases its activity after exposure to oxidants following gradient ischemia in skin. Using dorsal rat flaps, oxidant stress was induced immediately or during a 7-day period of preconditioning as a bipedicle flap before the distal attachment was divided. DT diaphorase activity (delta Abs/min/100 g) or expression of message was measured during the period of preconditioning to determine the relationship between skin survival, enzyme activity, and expression of message. There was 4.7 +/- 0.8 cm of skin necrosis in the distal end of acute flaps while the preconditioned flaps had no skin necrosis after the distal attachment was divided. In the acute flaps, the DT diaphorase activity was equal throughout the flap for the first 6 hr. After 24 hr of ischemia, the DT diaphorase activity was significantly higher in the proximal end of the flap (1.83 +/- 0.21 delta Abs/min/100 g) than that in the distal end (0.005 +/- 0.01 delta Abs/min/100 g), which was significant (P < 0.05). In the preconditioned flaps, enzyme activity did not increase but there was as 50-fold increase in DT diaphorase activity at the distal end 24 hr after they were divided (P < 0.05). Maximal enzyme induction of DT diaphorase activity occurred after 4 days of preconditioning and correlated with the maximal expression of mRNA. These studies provide the first evidence that DT diaphorase enzyme activity is inducible after oxidant stress. The data also suggests that DT activity remains elevated for at least 6 hr of ischemia and may be a potential source of anti-oxidant activity in ischemic skin.

The low molecular weight heat shock protein (hsp27) in rat Sertoli cells: evidence for identity of hsp27 with a germ cell-responsive phosphoprotein.

The low mol wt heat shock protein (hsp27) exists as three forms. hsp27a is the most basic and is not phosphorylated. The other two forms are phosphorylated, with hsp27b of intermediate isoelectric point and hsp27c as the most acidic form. We hypothesized that a Sertoli cell protein, previously discovered as being responsive to germ cells and referred to as GC1, is hsp27c. To investigate this hypothesis, cultured Sertoli cells were treated with CdCl2 or were heat shocked and then labeled with H3(32)PO4, [35S]methionine, or 3H-labeled amino acids. Sertoli cell proteins were then prepared for two-dimensional polyacrylamide gel electrophoresis and autoradiography or fluorography. The 27K protein did not incorporate [35S] methionine. After CdCl2 or heat shock treatments, hsp27b and hsp27c (GC1) showed increasing 32P incorporation with time. To test if this result was due to increased synthesis of hsp27 Sertoli cell proteins were labeled with 3H-labeled amino acids. Heat shock and CdCl2 resulted in time-dependent increases in 3H label incorporation into all three forms of hsp27. Shortly after heat shock, a greater proportion of hsp27b and hsp27c was observed in the pellet fraction of Sertoli cell homogenates, with more of these hsp27 forms being found in the supernatant fraction after 4 h recovery from heat shock. Treatment of CdCl2-exposed Sertoli cells with germ cells increased 32P labeling of hsp27c (GC1). The location on two-dimensional gels of the most acidic form of hsp27, hsp27c, was exactly coincident with the location of GC1. Taken together, the results support the conclusion that GC1 in Sertoli cells is hsp27c.

Synthesis of transferrin and transferrin mRNA in bovine Sertoli cells in culture and in vivo: sequence of partial cDNA clone for bovine transferrin.

Techniques were developed for generating enriched cultures of bovine Sertoli cells and indifferent supporting cells (immature Sertoli cells). The [35S]methionine and [35S]sulfate-labeled proteins secreted by cultured cells were analyzed by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) and fluorography. The electrophoretic pattern of the major Sertoli cell-secreted proteins was distinct from that of the major proteins secreted by cultured peritubular cells (the predominant contaminating cell type). Five major polypeptides ranging in molecular mass from 22 kDa to 77 kDa were resolved by 2D-PAGE in reducing conditions and were assigned numbers for reference purposes. Polypeptides 1 and 2 appeared to be analogous to two rat Sertoli cell-secreted proteins, sulfated glycoprotein-1 and sulfated glycoprotein-2, because of similar molecular mass, isoelectric point, subunit composition, sulfation, and sialation characteristics. Transferrin was detected in conditioned medium by immunoprecipitation using an antibody to bovine serum transferrin. Cultured Sertoli cells isolated from prepubertal bulls secreted higher levels of transferrin than did cells isolated from infant bulls. An 850 bp cDNA corresponding to the 3' portion of bovine transferrin mRNA was cloned and sequenced. Transferrin message was shown to be present in testicular tissue isolated from infant and prepubertal bulls and it increased as bulls matured. Levels of testicular transferrin mRNA were subsequently shown to correlate with daily sperm production in yearling beef bulls.

Seminal transferrin and spermatogenic capability in the bull.

The objective of this study was to determine the relationship between seminal transferrin and sperm output in ejaculates from mature dairy bulls. Caudal sperm reserves in mature Holstein bulls (n = 15) were depleted by 8 successive ejaculations during a 50-70-min period (depletion phase). Bulls were then ejaculated 6 times per week for a period of 4 weeks (6X phase). Weekly sperm output (WSO) and weekly transferrin output (WTfO) were the sums of sperm and transferrin levels in 6 ejaculates taken in 1 week of the study. Mean WSO ranged from 20.7 billion to 39.6 billion and mean WTfO ranged from 334 micrograms to 1872 micrograms among the bulls. Regression analysis of sperm and transferrin levels in ejaculates collected during the depletion phase indicated that approximately 40% of seminal transferrin was not related to sperm output and probably was from accessory fluids. A relationship between total seminal transferrin and total sperm in ejaculate was observed (p less than 0.01, r = 0.54). This relationship was stronger when the transferrin was corrected for accessory fluid contribution (p less than 0.01, r = 0.65). The relationship between WSO and WTfO corrected for accessory fluid transferrin contribution (cWTfO) was significant (p less than 0.01, r = 0.64). The relationship between WSO and cWTfO can be interpreted to reflect the relationship between actual testicular sperm production and transferrin from testicular or epididymal origin.

Isolation and characterization of Sertoli cell plasma membranes and associated plasminogen activator activity.

Plasma membranes were isolated from the cultured Sertoli cells of 20-day-old rat testes by differential centrifugation and sucrose density fractionation. The distribution and purity of subcellular components was determined by marker enzyme analysis of gradient fractions. The plasma membrane fraction showed an enrichment in two plasma membrane marker enzymes, 5'-nucleotidase and ouabain-sensitive Na+/K+-ATPase-specific activities, of 9- and 23-fold, respectively. Forty-two percent and 52% of the total cellular 5'-nucleotidase and ouabain-sensitive Na+/K+-ATPase activities, respectively, were found in the membrane fraction. The protein yield of plasma membrane was approximately 6% of the total cellular protein. Two-dimensional polyacrylamide gel electrophoresis was used to compare [35S] methionine- and [3H] glucosamine-labeled membrane proteins. The incorporation of [35S] methionine and [3H] glucosamine was increased in several proteins when the cultured Sertoli cells were treated with follicle-stimulating hormone, insulin, retinol, and testosterone. Isolated Sertoli cell membranes contained a membrane-associated form of plasminogen activator. Analysis of this plasminogen activator demonstrated that the membrane-associated enzyme existed primarily as a single 38,000-40,000-Mr form.