4-Hydroxy-2-nonenal-modified glyceraldehyde-3-phosphate dehydrogenase is degraded by cathepsin G

Degradation of oxidized or oxidatively modified proteins is an essential part of the antioxidant defenses of cells. 4-Hydroxy-2-nonenal (HNE), a major reactive aldehyde formed by lipid peroxidation, causes many types of cellular damage. It has been reported that HNE-modified proteins are degraded by the ubiquitin–proteasome pathway or, in some cases, by the lysosomal pathway. However, our previous studies using U937 cells showed that HNE-modified glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is degraded by an enzyme that is sensitive to a serine protease inhibitor, diisopropyl fluorophosphate (DFP), but not a proteasome inhibitor, MG-132, and that its degradation is not catalyzed in the acidic pH range where lysosomal enzymes are active. In the present study, we purified an HNE-modified GAPDH-degrading enzyme from a U937 cell extract to a final active fraction containing two proteins of 28 kDa (P28) and 27 kDa (P27) that became labeled with [³H]DFP. Using peptide mass fingerprinting and a specific antibody, P28 and P27 were both identified as cathepsin G. The degradation activity was inhibited by cathepsin G inhibitors. Furthermore, a cell extract from U937 cells transfected with a cathepsin G-specific siRNA hardly degraded HNE-modified GAPDH. These results suggest that cathepsin G plays a role in the degradation of HNE-modified GAPDH.     

Degradation of HNE-modified proteins--possible role of ubiquitin

4-Hydroxynonenal (HNE) is a lipid peroxidation product that is able to modify proteins. HNE-modified proteins are degraded to a considerable extend by the proteasomal system. It is unclear whether the recognition of HNE-modified proteins is mediated by ubiquitin, or whether the ubiquitin-independent proteasomal pathway is involved. In this study we demonstrate that HNE-modified GAPDH is preferentially ubiquitinated in vitro. In an attempt to demonstrate the formation of poly-ubiquitinated HNE-modified proteins in living cells we explored E36 fibroblasts. A clear rise in HNE-protein modification could be demonstrated after HNE treatment of the cells. Using inhibitors, we could show that the ubiquitin-dependent, ubiquitin-independent, and the lysosomal pathways affect the presence of HNE-modified proteins. We conclude that, although several proteolytic pathways exist for the degradation of HNE-modified proteins, there is the possibility of involvement of ubiquitin-dependent degradation.     

Degradation of glyceraldehyde-3-phosphate dehydrogenase triggered by 4-hydroxy-2-nonenal and 4-hydroxy-2-hexenal

Lipid peroxidation products such as 4-hydroxy-2-nonenal (HNE) may be responsible for various pathophysiological events under oxidative stress, since they injure cellular components such as proteins and DNA. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which is a key enzyme of glycolysis and has been reported to be a multifunctional enzyme, is one of the enzymes inhibited by HNE. Previous studies showed that GAPDH is degraded when incubated with acetylleucine chloromethyl ketone (ALCK), resulting in the liberation of a 23-kDa fragment. In this study, we examined whether GAPDH incubated with HNE or other aldehydes of lipid peroxidation products are degraded similarly to that with ALCK. The U937 cell extract was incubated with these aldehydes at 37 degrees C and analyzed by Western blotting using anti-GAPDH antibodies. Incubation with HNE or 4-hydroxy-2-hexenal (HHE) decreased GAPDH activity and GAPDH protein level, and increased the 23-kDa fragment, in time- and dose-dependent manners, but that with other aldehydes did not. Gel filtration using the Superose 6 showed that the GAPDH-degrading activity was eluted in higher molecular fractions than proteasome activity. The enzyme activity was detected at the basic range of pH and inhibited by serine protease inhibitors, diisopropyl fluorophosphate and phenylmethylsulfonyl fluoride, but not by other protease inhibitors including a proteasome inhibitor, MG-132, and a tripeptidyl peptidase II (TPP II) inhibitor, AAF-CMK. These results suggest that GAPDH modified by HNE and HHE is degraded by a giant serine protease, releasing the 23-kDa fragment, not by proteasome or TPP II.     

The influence of differential processing of procathepsin H on its aminopeptidase activity, secretion and subcellular localization in human cell lines

Cathepsin H is a unique member of the cysteine cathepsins that acts primarily as an aminopeptidase. Like other cysteine cathepsins, it is synthesized as an inactive precursor and activated by proteolytic removal of its propeptide. Here we demonstrate that, in human cells, the processing of the propeptide is an autocatalytic, multistep process proceeding from an inactive 41kDa pro-form, through a 30kDa intermediate form, to the 28kDa mature form. Tyr87P and Gly90P were identified as the two major endopeptidase cleavage sites, converting the 30kDa form into the mature 28kDa form. The level of processing differs significantly in different human cell lines. In monocyte-derived macrophages U937 and prostate cancer cells PC-3, the 28kDa form is predominant, whereas in osteoblasts HOS the processing from the 30kDa form to the 28kDa form is significantly lower. The aminopeptidase activity of the enzyme and its subcellular localization are independent of the product, however the 30kDa form was not secreted in HOS cells. The activity of the resulting cathepsin H in U937 cells was significantly lower than that in HOS cells, presumably due to the high levels of endogenous cysteine protease inhibitor cystatin F present specifically in this cell line. These results provide an insight into the dependence of human cathepsin H processing and regulation on cell type.     

Cathepsin D is one of the major enzymes involved in intracellular degradation of AGE-modified proteins

Abstract

Oxidized and cross-linked modified proteins are known to accumulate in ageing. Little is known about whether the accumulation of proteins modified by advanced glycation end products (AGEs) is due to an affected intracellular degradation. Therefore, this study was designed to determine whether the intracellular enzymes cathepsin B, cathepsin D and the 20S proteasome are able to degrade AGE-modified proteins in vitro. It shows that AGE-modified albumin is degraded by cathepsin D, while cathepsin B was less effective in the degradation of aldehyde-modified albumin and the 20S proteasome was completely unable to degrade them. Mouse primary embryonic fibroblasts isolated from a cathepsin D knockout animals were found to have an extensive intracellular AGE-accumulation, mainly in lysosomes, and a reduction of AGE-modified protein degradation compared to cells isolated from wild type animals. In summary, it can be assumed that cathepsin D plays a significant role in the removal of AGE-modified proteins.     

Proteolysis of C3 on U937 cell plasma membranes. Purification of cathepsin G

Covalent binding of C3 fragments to U937 cell membranes involved a cell surface-associated proteolytic activity. Two proteases able to cleave C3 were purified from U937 plasma membranes. Purification involved solubilization of the membranes and ion exchange chromatography. One of the purified proteases was identified as elastase, based upon a substrate specificity for benzyloxycarbonylalanine-o-nitrophenyl ester and complete inhibition by elastatinal and methoxysuccinyl-alanyl-alanyl-prolyl-valyl-chloromethyl-ketone. The other protease (m.w. 28,000) is cathepsin G, as deduced from the amino acid composition, the amino-terminal sequence, and the substrate specificity for succinyl-alanyl-alanyl-phenylalanine-p-nitroanilide. These two lysosomal proteases are present on the U937 cell surface, as confirmed by immunofluorescence analysis. Plasma membrane elastase and cathepsin G from U937 cells cleave C3 into C3a- and C3b-like fragments; further incubation leads to C3c- and C3dg-like fragments, as judged from SDS-PAGE analysis of the digests. Sequencing of the C3b-like fragment purified by reverse phase chromatography indicates that initial cleavage of C3 by purified cathepsin G occurs at two positions in the amino-terminal part of the alpha-chain, at a Arg-Ser bond located between residues 748 and 749 and at a Leu-Asp bond between residues 751 and 752. These proteases are, thus, able to generate, on the U937 surface, active fragments of C3, which are likely to be involved in cell-protein and cell-cell interactions.     

Processing of the human transferrin receptor at distinct positions within the stalk region by neutrophil elastase and cathepsin G

The ectodomain of the human transferrin receptor (TfR) is released as soluble TfR into the blood by cleavage within a stalk. The major cleavage site is located C-terminally of Arg-100; alternative cleavage sites are also present. Since the cleavage process is still unclear, we looked for proteases involved in TfR ectodomain release. In the supernatant of U937 histiocytic cells we detected alternatively cleaved TfR (at Glu-110). In membrane fractions of these cells we identified two distinct proteolytic activities responsible for TfR cleavage within the stalk at either Val-108 or Lys-95. Both activities could be inhibited by serine protease inhibitors, but not by inhibitors of any other class of proteases. Protein purification yielded a 28 kDa protein that generated the Val-108 terminus. The protease activity could be ascribed to neutrophil elastase according to the substrate specificity determined by amino acid substitution analysis of synthetic peptides, an inhibitor profile, the size of the protease and the use of specific antibodies. The results of analogous experiments suggest that the second activity is represented by another serine protease, cathepsin G. Thus, membrane-associated forms of neutrophil elastase and cathepsin G may be involved in alternative TfR shedding in U937 cells.     

Proteasome involvement in the degradation of the G(q) family of Galpha subunits

Metabolically unstable proteins are involved in a multitude of regulatory networks, including those that control cell signaling, the cell cycle and in many responses to physiological stress. In the present study, we have determined the stability and characterized the degradation process of some members of the G(q) class of heterotrimeric G proteins. Pulse-chase experiments in HEK293 cells indicated a rapid turnover of endogenously expressed Galpha(q) and overexpressed Galpha(q) and Galpha(16) subunits. Pretreatment with proteasome inhibitors attenuated the degradation of both G alpha subunits. In contrast, pretreatment of cells with inhibitors of lysosomal proteases and nonproteasomal cysteine proteases had very little effect on the stability of the proteins. Significantly, the turnover of these proteins is not affected by transient activation of their associated receptors. Fractionation studies showed that the rates of Galpha(q) and Galpha16 degradation are accelerated in the cytosol. In fact, we show that a mutant Galpha(q) which lacks its palmitoyl modification site, and which is localized almost entirely in the cytoplasm, has a marked increase in the rate of degradation. Taken together, these results suggest that the G(q) class proteins are degraded through the proteasome pathway and that cellular localization and/or other protein interactions determine their stability.     

Protein oxidation and degradation during cellular senescence of human BJ fibroblasts: part I--effects of proliferative senescence.

Oxidized and cross-linked proteins tend to accumulate in aging cells. Declining activity of proteolytic enzymes, particularly the proteasome, has been proposed as a possible explanation for this phenomenon, and direct inhibition of the proteasome by oxidized and cross-linked proteins has been demonstrated in vitro. We have further examined this hypothesis during both proliferative senescence (this paper) and postmitotic senescence (see the accompanying paper, ref 1 ) of human BJ fibroblasts. During proliferative senescence, we found a marked decline in all proteasome activities (trypsin-like activity, chymotrypsin-like activity, and peptidyl-glutamyl-hydrolyzing activity) and in lysosomal cathepsin activity. Despite the loss of proteasome activity, there was no concomitant change in cellular levels of actual proteasome protein (immunoassays) or in the steady-state levels of mRNAs for essential proteasome subunits. The decline in proteasome activities and lysosomal cathepsin activities was accompanied by dramatic increases in the accumulation of oxidized and cross-linked proteins. Furthermore, as proliferation stage increased, cells exhibited a decreasing ability to degrade the oxidatively damaged proteins generated by an acute, experimentally applied oxidative stress. Thus, oxidized and cross-linked proteins accumulated rapidly in cells of higher proliferation stages. Our data are consistent with the hypothesis that proteasome is progressively inhibited by small accumulations of oxidized and cross-linked proteins during proliferative senescence until late proliferation stages, when so much proteasome activity has been lost that oxidized proteins accumulate at ever-increasing rates. Lysosomes attempt to deal with the accumulating oxidized and cross-linked proteins, but declining lysosomal cathepsin activity apparently limits their effectiveness. This hypothesis, which may explain the progressive intracellular accumulation of oxidized and cross-linked proteins in aging, is further explored during postmitotic senescence in the accompanying paper (1).     

Processing and transport of lysosomal enzymes in human monocyte line U937

Precursors of cathepsin D and beta-hexosaminidase synthesized in the U937 monocyte line are processed to mature forms with similar kinetics as in fibroblasts. In U937 cells the processing of the precursor of the beta-chain of beta-hexosaminidase, however, results in a larger fragment that resembles a processing intermediate in fibroblasts. This difference is explained by differences in the equipment of the cells with proteinases, since cross-feeding of the precursors to the cells results in a processing characteristic for the recipient cell type. In sucrose gradients the precursors are found partly in a low- and partly in a high-density region. Mature polypeptides and activity of lysosomal enzymes fractionate mainly in the higher density region. In U937 cells the transport and maturation of endogenous lysosomal enzymes are less sensitive to bases (NH4Cl, chloroquine, tilorone) and to antibody against the mannose 6-phosphate specific receptors than in fibroblasts. A small portion of enzymes released from U937 cells contains the markers recognized by the mannose-6-phosphate specific receptors. U937 cells express these receptors and utilize them for transport of endogenous and exogenous lysosomal enzymes. It appears, however, that a fraction of lysosomal enzymes is transported in U937 cells independent of the mannose-6-phosphate-specific receptors.     

The proteasomal system and HNE-modified proteins

Metabolic processes and environmental conditions cause the constant formation of oxidizing species over the lifetime of cells and organisms. This leads to a continuous oxidation of intracellular components, including lipids, DNA and proteins. During the extensively studied process of lipid peroxidation, several reactive low-molecular weight products are formed, including reactive aldehydes as 4-hydroxynonenal (HNE). These aldehydic lipid peroxidation products in turn are able to modify proteins. The degradation of oxidized and oxidatively modified proteins is an essential part of the oxidant defenses of cells. The major proteolytic system responsible for the removal of oxidized cytosolic and nuclear proteins is the proteasomal system. The proteasomal system by itself is a multicomponent system responsible for the degradation of the majority of intracellular proteins. It has been shown that some, mildly cross-linked, HNE-modified proteins are preferentially degraded by the proteasome, but extensive modification with this cross-linking aldehyde leads to the formation of protein aggregates, that can actually inhibit the proteasome. This review summarizes our knowledge of the interactions between lipid peroxidation products, proteins, and the proteasomal system.     

Synthetic fibril peptide promotes clearance of scrapie prion protein by lysosomal degradation

Transmissible spongiform encephalopathies are infectious and neurodegenerative disorders that cause neural deposition of aggregates of the disease-associated form of PrP(Sc). PrP(Sc) reproduces by recruiting and converting the cellular PrP(C), and ScN2a cells support PrP(Sc) propagation. We found that incubation of ScN2a cells with a fibril peptide named P9, which comprises an intrinsic sequence of residues 167-184 of mouse PrP(C), significantly reduced the amount of PrP(Sc) in 24 hr. P9 did not affect the rates of synthesis and degradation of PrP(C). Interestingly, immunofluorescence analysis showed that the incubation of ScN2a cells with P9 induced colocalization of the accumulation of PrP with cathepsin D-positive compartments, whereas the accumulation of PrP in the cells without P9 colocalized mainly with lysosomal associated membrane proteins (LAMP)-1-positive compartments but rarely with cathepsin D-positive compartments in perinuclear regions. Lysosomal enzyme inhibitors attenuated the anti-PrP(Sc) activity; however, a proteasome inhibitor did not impair P9 activity. In addition, P9 neither promoted the ubiquitination of cellular proteins nor caused the accumulation of LC3-II, a biochemical marker of autophagy. These results indicate that P9 promotes PrP(Sc) redistribution from late endosomes to lysosomes, thereby attaining PrP(Sc) degradation.     

Proteasome activation by poly-ADP-ribose-polymerase in human myelomonocytic cells after oxidative stress

Cytotoxic action of a variety of antitumor drugs generate oxidatively modified proteins that are predominantly metabolized via the proteasome. In the present study, a differentiation-retrodifferentiation cell system was exposed to oxidative stress by hydrogen peroxide treatment. Thus, the activity of the nuclear proteasome in proliferating human U937 leukemic cells increased by 2.5-fold after hydrogen peroxide treatment. In contrast, growth-arrested differentiated U937 cells demonstrated 40% less constitutive proteasomal activity, which was not inducible after hydrogen peroxide exposure. After a retrodifferentiation process, however, in which differentiated U937 cells resume autonomous growth again, the proteasomal activity was indistinguishable from that in U937 control cells, both constitutively and after induction of oxidative stress. Moreover, cells of TUR, a differentiation-resistant U937 subclone, expressed an elevated constitutive proteasomal activity that increased by 2.5-fold after oxidative stress. Immunoblot analysis revealed that these differences in proteasomal activities did not correlate with proteasome protein expression but with protein levels of the nuclear enzyme poly-ADP-ribose-polymerase (PARP). Further studies using specific PARP inhibitors revealed that the noninducible proteasome activity in differentiated U937 cells was PARP independent, whereas the increased activity level in oxidatively stressed TUR cells was downregulated upon PARP inhibition. Immunoprecipitation experiments demonstrated a protein-protein interaction of the functional active PARP with the proteasome in correlation with the proteasome activity. Similar results were obtained by analyzing protein carbonyls after oxidative stress. Taken together, these data suggest that proliferating, rather than growth-arrested, cells metabolize oxidatively damaged nuclear proteins via the proteasome by expressing high levels of PARP.     

An Active 32-kDa Cathepsin L Is Secreted Directly from HT 1080 Fibrosarcoma Cells and Not via Lysosomal Exocytosis

Cathepsin L [EC 3.4.22.15] is secreted via lysosomal exocytosis by several types of cancer cells, including prostate and breast cancer cells. We previously reported that human cultured fibrosarcoma (HT 1080) cells secrete cathepsin L into the medium; this secreted cathepsin is 10-times more active than intracellular cathepsin. This increased activity was attributed to the presence of a 32-kDa cathepsin L in the medium. The aim of this study was to examine how this active 32-kDa cathepsin L is secreted into the medium. To this end, we compared the secreted active 32-kDa cathepsin L with lysosomal cathepsin L by using a novel gelatin zymography technique that employs leupeptin. We also examined the glycosylation and phosphorylation status of the proteins by using the enzymes endoglycosidase H [EC 3.2.1.96] and alkaline phosphatase [EC 3.1.3.1]. Strong active bands corresponding to the 32-kDa and 34-kDa cathepsin L forms were detected in the medium and lysosomes, respectively. The cell extract exhibited strong active bands for both forms. Moreover, both forms were adsorbed onto a concanavalin A-agarose column. The core protein domain of both forms had the same molecular mass of 30 kDa. The 32-kDa cathepsin L was phosphorylated, while the 34-kDa lysosomal form was dephosphorylated, perhaps because of the lysosomal marker enzyme, acid phosphatase. These results suggest that the active 32-kDa form does not enter the lysosomes. In conclusion, our results indicate that the active 32-kDa cathepsin L is secreted directly from the HT 1080 cells and not via lysosomal exocytosis.     

Interleukin 1 and tumour necrosis factor increase phosphorylation of the small heat shock protein. Effects in fibroblasts, Hep G2 and U937 cells

Interleukin 1 alpha and tumour necrosis factor-alpha stimulated phosphorylation of three 27 kDa phosphoproteins in MRC-5 fibroblasts which was sustained for up to 2 h after adding the cytokines. All three phosphoproteins were immunoprecipitated by a specific antiserum to the small mammalian heat shock protein, hsp 27. The three phosphoproteins from stimulated or control cells contained phosphoserine but not phosphothreonine or phosphotyrosine. Similar increases in phosphorylation of immunoprecipitable 27 kDa proteins were seen in U937 cells stimulated by TNF alpha and Hep G2 cells stimulated by IL1 alpha.     

Purification and characterization of human lysosomal membrane glycoproteins

Two human cell lysosomal membrane glycoproteins of approximately 120 kDa, hLAMP-1 and hLAMP-2, were identified by use of monoclonal antibodies prepared against U937 myelomonocytic leukemia cells or blood mononuclear cells. The two glycoproteins were purified by antibody affinity chromatography and each was found to be a major constituent of human spleen cells, representing approximately 0.05% of the total detergent-extractable protein. Both molecules were highly glycosylated, being synthesized as polypeptides of 40 to 45 kDa and cotranslationally modified by the addition of Asn-linked oligosaccharides. NH2-terminal sequence analysis indicated that each was approximately 50% identical to the corresponding mLAMP-1 or mLAMP-2 of mouse cells. Electron microscopic studies of human blood monocytes, HL-60, and U937 cells demonstrated that the principal location of these glycoproteins was intracellular, in vacuoles and lysosomal structures but not in the peroxidase-positive granules of monocytes. Transport of the proteins between organelles was evidenced by their marked accumulation in the membranes of phagolysosomes. A fraction of each glycoprotein was also detected on the plasma membrane of U937 and HL-60 cells but not on a variety of other tissue culture cells. This cell-surface expression may be differentiation related, since the proteins were not detected in the plasma membrane of normal blood monocytes and their expression on U937 and HL-60 cells was reduced when the cells were treated with differentiating agents. Cell-surface expression of both glycoproteins was markedly increased in blood monocytes but not in U937 cells after exposure to the lysosomotropic reagent methylamine HCl, indicating differences in LAMP-associated membrane flow in these cell types.     

Phosphorylation-dependent degradation of the cyclin-dependent kinase inhibitor p27.

The p27(Kip1) protein associates with G1-specific cyclin-CDK complexes and inhibits their catalytic activity. p27(Kip1) is regulated at various levels, including translation, degradation by the ubiquitin/proteasome pathway and non-covalent sequestration. Here, we describe point mutants of p27 deficient in their interaction with either cyclins (p27(c-)), CDKs (p27(k-)) or both (p27(ck-)), and demonstrate that each contact is critical for kinase inhibition and induction of G1 arrest. Through its intact cyclin contact, p27(k-) associated with active cyclin E-CDK2 and, unlike wild type p27, p27(c-) or p27(ck-), was efficiently phosphorylated by CDK2 on a conserved C-terminal CDK target site (TPKK). Retrovirally expressed p27(k-) was rapidly degraded through the proteasome in Rat1 cells, but was stabilized by secondary mutation of the TPKK site to VPKK. In this experimental setting, exogenous wild-type p27 formed inactive ternary complexes with cellular cyclin E-CDK2, was not degraded through the proteasome, and was not further stabilized by the VPKK mutation. p27(ck-), which was not recruited to cyclin E-CDK2, also remained stable in vivo. Thus, selective degradation of p27(k-) depended upon association with active cyclin E-CDK2 and subsequent phosphorylation. Altogether, these data show that p27 must be phosphorylated by CDK2 on the TPKK site in order to be degraded by the proteasome. We propose that cellular p27 must also exist transiently in a cyclin-bound non-inhibitory conformation in vivo.