Stress proteins and glycoproteins (Review).

Proteins represent both structural and functional elements of biological organisms, however, their structural and catalytic function is directly linked to the acquisition and maintenance of a complex three-dimensional conformation. A molecular machinery to accomplish protein folding and maintenance in vivo is provided by a variety of molecular chaperones that include both heat shock proteins (Hsps), glucose-regulated proteins (Grps), and a separate class of stress glycoproteins (S-Gps). Different chaperones associate to form functional complexes (chaperone) and work coordinately to accomplish specific functions during the folding of particular proteins. In this review, we will summarize recently acquired new insights into the complexities of chaperones, the current state of S-Gps and their interactions with Hsps, and of specific chaperones that appear to be designed for the folding of cellular glycoproteins. Finally, we discuss the physiological role of chaperones by examining their function in specific cellular processes, namely tumor/host interactions and diseases associated with aberrant prion protein folding.

Interaction of heat stress glycoprotein GP50 with classical heat-shock proteins.

Cellular stress conditions are known to elevate heat-shock protein (HSP) synthesis and protein glycosylation, leading to the development of cellular thermotolerance. In the present study, we investigated the interaction of a major stress glycoprotein, GP50, with other cellular proteins during recovery from heat stress, using' mostly immunoprecipitation techniques. Parallel studies of heat-stressed CHO and M21 cells showed that both glycosylated and unglycosylated forms of GP50 interact with several members of the classical HSP families (e.g., HSP70 and HSP90) in an ATP-dependent manner. The specificity of HSP-stress glycoprotein interactions was confirmed by chemical crosslinking with a homobifunctional agent, 3,3'-dithiobis (succinimidyl propionate). Interaction of GP50 with denatured proteins was also demonstrated through binding to gelatin. Protein complexes formed between stress glycoproteins and HSPs were further characterized by gel filtration and showed an average molecular mass between 400 and 600 kDa. Overall, the consistent association of stress glycoproteins with nonglycosylated HSPs suggests a structural/functional role for protein chaperone complexes that consist of denatured proteins and the glycone/aglycone elements of cellular stress response.

Calreticulin associates with stress proteins: implications for chaperone function during heat stress.

Acute heat stress leads to the glycosylation of a "prompt" stress glycoprotein, P-SG67/64, identified as calreticulin. In the present study, we used immunoprecipitation to investigate the interactions of P-SG/calreticulin with other proteins during cellular recovery from heat stress. In heat-stressed CHO and M21 cells, both glycosylated and unglycosylated P-SGs interact with HSP90, GRP94, GRP78, and the other prompt stress glycoprotein, P-SG50, in an ATP-independent manner. Specificity of HSP-P-SG interactions was determined by chemical cross-linking with the homo-bifunctional agent DSP (3,3'-dithiobis[succinimidyl propionate]). Characterization of the cross-linked complexes involving calreticulin and heat shock proteins (HSPs) showed an average mass of 400-600 kDa by gel filtration chromatography. Overall, the consistent association of glycosylated and unglycosylated calreticulin with P-SG50 and unglycosylated HSPs suggests that P-SG/calreticulin is an active member of the cast of glycone/aglycone chaperones that cooperate to achieve cellular recovery from acute heat stress.

Stress response in a leporine renal cell model.

It is well established that renal proximal tubule (RPT) cells grown under standard in vitro conditions attenuate many of their in vivo properties and functions. Thus, the study of renal stress response mechanisms requires an appropriate cell culture model. In the present study, we compared the heat stress (10 min, 45 degrees C) response of freshly isolated RPT cells with that of RPT cells grown in vitro for 6 days under two different culture conditions: (1) SHAKE conditions, where oxygen levels and physiological functions are maintained via continuous media motion [Nowak G, Schnellmann RG: Am J Physiol 1996;271:C2072-2080] and (2) STILL conditions, involving standard cell culture which leads to partial hypoxia and a marked reduction in physiological functions. The freshly isolated RPT cells progressively synthesized heat shock proteins (HSPs) and stress glycoproteins (SGs) during a 3-hour culture period in vitro. Under these conditions, heat stress did not further increase HSP and SG synthesis. In RPT cells grown under SHAKE conditions, HSP70 synthesis was detected 1 h after heat stress and decreased below detection by 3 h. In contrast, the uptake of radiolabeled mannose into (glycoprotein) GP62 (Mr 62,000), GP50, and GP38 was observed in control SHAKE cultures and was not further increased after heat stress. These results are consistent with immunohistochemistry studies, where similar changes in HSP70 and GP50 expression were noted. RPT cells grown under STILL conditions showed both increased synthesis of HSP70 and increased glycosylation of GP62, GP50, and GP38 as early as 1 h after heat stress, but in contrast to SHAKE conditions, this heat-induced stress response further intensified at 3 h after heat stress. By 7 h after heating, HSP synthesis returned to control levels, while glycosylation of GP62 and GP50 remained elevated. Based on our results, we conclude that freshly isolated RPT cells exhibit a stress response that may be caused by acute cell isolation/culture stress. While this stress response unfolds, freshly isolated RPT cells appear unable to respond to additional heat stress. RPT cells grown under SHAKE and STILL conditions exhibit high rates of SG glycosylation, especially that of GP62, possibly reflecting a 'stress' condition associated with growth on plastic substrate. Concurrently, RPT cells from STILL cultures show a higher capacity for responding to acute heat stress than SHAKE cultures, evidenced by the transiently increased HSP synthetic rates. The interpretation of the renal stress response capacity, therefore, must be linked to a specific culture condition.

Intracellular distribution of stress glycoproteins in a heat-resistant cell model expressing human HSP70.

Heat stress results in the cellular accumulation of heat-shock proteins (HSP) and increased protein glycosylation. Among known stress glycoproteins, GP50 and GP62 are associated with the expression of thermotolerance. In the present study, we characterized subcellular localization and redistribution of GP62 and GP50 in a rodent cell line, M21, before and after cellular heat-stress. M21 cells are heat-resistant cells that overexpress human HSP70 and also have concomitantly high GP62 levels. Cellular fractionation by differential centrifugation showed that GP62 and GP50 was present in each subcellular fraction. However, each stress glycoprotein exhibited a characteristic kinetic pattern of redistribution during cellular recovery after heat stress. For example, glycosylated GP62 was seen predominantly in the mitochondria before heat-stress. Immediately after heat-stress, its presence in the mitochondrial fraction was dramatically reduced, while it increased in lysosomes, microsomes and cytosol. By 1 h after heat stress, it had largely disappeared from microsomal and cytosolic fractions. After 24 h, all subcellular fractions showed only trace amounts of residual GP62. By comparison, GP50 was also highest in the mitochondrial fraction before heat-stress, redistributed like GP62 immediately after heat stress, but remained relatively unchanged thereafter. In contrast to GP62, GP50 showed little redistribution during 24 h after heat-stress and remained at high concentrations in all cell fractions, including microsomes. Distribution of GP50 and GP62 before and after heat stress, based on differential centrifugation, was consistent with immunolocalization data. Following heat stress, both GP50 and GP62 showed a partial overlap in distribution with HSP70. The above results indicate that each stress glycoprotein has a specific subcellular location, both before and after heat stress. The presence of GP62 in virtually all cell fractions is consistent with a multifunctional role for GP62 in the cellular stress response.

Intracellular distribution of heat-induced stress glycoproteins.

Cellular heat stress results in elevated heat-shock protein (HSP) synthesis and in thermotolerance development. Recently, we demonstrated that protein glycosylation is also an integral part of the stress response with the identification of two major stress glycoproteins, GP50, associated with thermotolerance, and P-SG67, the "prompt" stress glycoprotein induced immediately during acute heat stress. In the present study, we characterized the subcellular location and redistribution of these proteins during the cellular injury and recovery phase. In unheated and heated CHO cells, both stress glycoproteins were present in each subcellular fraction isolated by differential centrifugation. However, the subcellular redistribution in the course of cellular recovery after heat stress was specific for each stress glycoprotein. GP50 was present in all subcellular fractions before heat stress, but showed relatively little redistribution after heat stress. By 24 h of recovery following stress, GP50 showed partial depletion from lysosomes and microsomes, and was mainly present in the mitochondria. Glycosylated P-SG67 was redistributed in a more complex fashion. It was seen predominantly in the lysosomes and microsomes immediately following heat-stress, but after 6 h of recovery following heat stress, it largely disappeared from the microsomes and was present mainly in the cytosol. By 24 h of recovery following heat stress, it was found predominantly in the nucleus-rich fraction and mitochondria. The localization of GP50 and P-SG67 by subcellular fractionation is consistent with immunolocalization studies and contrasts with the translocation of HSP70 after heat stress from cytosol to nuclei and nucleoli. These results reflect a characteristic distribution for each stress glycoprotein; their presence in virtually all subcellular fractions suggests multifunctional roles for the various stress glycoproteins in the cellular heat stress response.

Partial homology of stress glycoprotein GP62 with HSP70.

Thermotolerance and heat resistance are often associated with elevated levels of heat shock proteins (HSPs) and a selective increase in protein glycosylation. In the present study, we have characterized heat stress-induced protein glycosylation in M21 cells, derived from the rat fibroblast line, Rat-1. M21 cells are characterized by constitutive overexpression of human HSP70 gene and show increased heat resistance without loss of its normal capacity for thermotolerance development after heat conditioning (Li et al., 1991, Proc. Natl. Acad. Sci. USA 88, 1681-1685). The data presented here show that the elevated heat resistance in these cells is associated not only with the constitutive overexpression of human HSP70, but also with increased glycosylation of a major stress glycoprotein, GP62 (Mr of 62,000). We further purified GP62 by sequential preparative isoelectric focusing and two-dimensional isoelectric focusing/SDS-polyacrylamide gel electrophoresis. The purified protein was digested and partially characterized by microsequencing of two peptide fragments, comprising of 14-15 amino acids each. These fragments had a 100% sequence homology with HSP70 and a 71-100% sequence homology with HSC70 from various species. Western blotting using both HSP70 and HSC70 antibodies showed positive reactivity of GP62 with HSP70. Affinity characterizations showed strong binding of GP62 to wheat germ agglutinin and concanavalin A, consistent with the presence of both alpha-D-mannosyl/glucosyl and N-acetyl-beta-D-glucosylaminyl/glucosamine oligomer residues in GP62. These data confirm the glycosylated status of GP62 and indicate that GP62 is a heat stress-induced glycoprotein with partial homology to HSP70.

Protein glycosylation in a heat-resistant rat fibroblast cell model expressing human HSP70.

Thermotolerance and heat resistance are frequently associated with elevated levels of heat shock proteins (HSPs). Elevated heat resistance is also found to be associated with the overexpression of high levels of HSP70, as seen in M21 cells, derived from the Rat-1 line. In the present study, we report that M21 cells also feature an increase in general protein glycosylation and specific expression of the stress glycoprotein, GP62, both of which correlate with cellular heat resistance. The expression of GP50, a major stress glycoprotein in cell lines such as CHO, however, did not correlate with cellular heat resistance in M21 cells. Protein glycosylation that occurs during acute heat stress ("prompt" glycosylation) was associated with the glycosylation of a major "prompt" stress glycoprotein, P-SG64 (M(r) of 64,000), that was identified by immunoblotting as a glycosylated form of calreticulin. The higher level of protein glycosylation in M21 cells correlated well with increased D-[2-3H]mannose incorporation into precursor pools of dolichyl phosphomannose and dolichyl pyrophosphoryl oligosaccharides and into glycoproteins. Thus, heat resistance in M21 cells is associated not only with expression of high levels of HSP70, but also with a concomitant increase in protein glycosylation. These data support the hypothesis that stress-induced protein glycosylation is a component of cellular stress response, either in association with HSPs or as an independent mechanism.

Prompt glycosylation of calreticulin is independent of Ca2+ homeostasis.

Selective glycosylation of "prompt" stress glycoproteins (P-SG), mainly P-SG67 and P-SG64 (M(r) of 64,000, pI = 5.1), occurs immediately during acute heat-stress. In the present study, P-SG64 was purified by sequential gel filtration, anion-exchange, affinity chromatography, and two-dimensional isoelectric focusing/SDS-PAGE. Purified P-SG64 was further characterized by microsequencing of a peptide fragment, PT-61, which showed a 100% sequence homology with calreticulin, suggesting that P-SG64 is identical to calreticulin. PT-61 also showed 55%, 58% and 63% sequence homologies with calnexin, HIV-1 gp120 and HIV-2 envelope polyprotein, respectively. 45Ca2+ overlay studies confirmed Ca(2+)-binding of P-SG64. P-SG67 was also recently identified as calreticulin (8), which suggests that CHO cells either have two isoforms of calreticulin or express variable states of calreticulin glycosylation during acute heat stress. The role of intracellular Ca2+ ([Ca2+]i) during heat-induced "prompt" glycosylation was also examined and indicated an 8-fold increase in [Ca2+]i. Chelation of this increased cytoplasmic Ca2+ by BAPTA reduced glycosylation of P-SG67/P-SG64/calreticulin only by approximately 20%. This observation suggests that altered [Ca2+]i homeostasis is not directly linked to calreticulin glycosylation, instead, heat-induced calreticulin glycosylation is a Ca(2+)-independent effect.

Heat shock-induced prompt glycosylation. Identification of P-SG67 as calreticulin.

Acute heat shock initiates the phenomenon of "prompt" glycosylation, which is characterized by selective glycosylation of specific cellular proteins called the prompt stress glycoproteins (P-SG). Prompt glycosylation rapidly occurs even during short heating periods, e.g. 10 min at 45 degrees C, and is not affected by the presence of cycloheximide (Henle, K. J., Kaushal, G. P., Nagle, W. A., and Nolen, G. T. (1993) Exp. Cell Res. 207, 245-251). The major P-SG in Chinese hamster ovary cells, P-SG67, was characterized by an M(r) of 67,000 and a pI = 5.1. In the present study, we purified P-SG67 by sequential gel filtration, anion exchange, affinity chromatography with concanavalin A-Sepharose, and two-dimensional isoelectric focusing/SDS-polyacrylamide gel electrophoresis. The purified protein was digested and partially characterized by microsequencing of three major peptide fragments. The fragments, comprising a total of 46 amino acid residues, had an almost 100% sequence homology with calreticulin and partial homology with calnexin. Calcium binding studies with 45Ca2+ overlay confirmed that P-SG67 is a Ca(2+)-binding protein. These observations support the notion that P-SG67 is identical to calreticulin and that the glycosylation status of calreticulin can respond to environmental stress conditions.

Platelet phosphoinositide turnover in streptozotocin-induced diabetes.

Increased platelet aggregation and secretion in response to various agonists has been described in both diabetic humans and animals. Alterations in the platelet membrane fatty acid composition of phospholipids and changes in the prostacyclin and thromboxane formation could only partly explain the altered platelet function in diabetes. In the present study, we have examined the role of phosphoinositide turnover in the diabetic platelet function. We report alterations in 2-[3H] myo-inositol uptake, phosphoinositide turnover, inositol phosphate and diacylglycerol (DAG) formation, phosphoinositide mass, and phospholipase C activity in platelets obtained from streptozotocin (STZ)-induced diabetic rats. There was a significant increase in the 2-[3H] myo-inositol uptake in washed platelets from diabetic rats. Basal incorporation of 2-[3H] myo-inositol into phosphatidylinositol 4,5-bisphosphate (PIP2), phosphatidylinositol 4-phosphate (PIP) or phosphatidylinositol (PI) in platelets obtained from diabetic rats was, however, not affected. Thrombin stimulation of platelets from diabetic rats induced an increase in the hydrolysis of [32P]PIP2 but indicated no change in the hydrolysis of [32P]PIP and [32P]PI as compared to their basal levels. Thrombin-induced formation of [3H]inositol phosphates was significantly increased in both diabetic as well as in control platelets as compared to their basal levels. This formation of [3H]inositol phosphates in diabetic platelets was greater than controls at all time intervals studied. Similarly, there was an increase in the release of DAG after thrombin stimulation in the diabetic platelets.(ABSTRACT TRUNCATED AT 250 WORDS)

Modifications of platelet phospholipid fatty acid composition in streptozocin-induced diabetic rats.

Increased thromboxane A2 (TXA2) generation by platelets has been reported both in diabetic patients and streptozocin-induced diabetic rats. This increase is in contrast to the decreased prostacyclin (PGI2) synthesis by endothelial cells in diabetes. An imbalance in the ratio of TXA2/PGI2 has been implicated in increased platelet aggregation and a high incidence of vascular disease in human diabetes. The mechanism for this imbalance, however, remains elusive. In a previous study from our laboratory, we reported unchanged arachidonic acid levels in platelet membrane phospholipids of 3-week diabetic rats, but a decreased arachidonic acid level in platelet membrane phospholipids of 6-week diabetic rats. In the present communication, we report the role of enzymes that are involved in remodeling arachidonic acid levels of platelet membrane phospholipids in both 3- and 6-week diabetic rats. No alterations were observed in the activities of arachidonoyl-CoA synthetase, acyl-CoA: lysophosphatidylcholine acyltransferase, or phospholipase A2 in platelets from both 3- and 6-week diabetic rats. However, both increased uptake and incorporation of [14C]arachidonic acid into platelets were observed in the diabetic platelet-rich plasma. In conclusion, increased TXA2 formation in diabetic platelets is not due to alterations in the activities of enzymes involved in the incorporation into or release of arachidonate from the diabetic platelet membrane phospholipid, but may be due to increased efficiency of uptake, incorporation or possibly redistribution of this fatty acid among phospholipid classes in diabetic platelets.

Decreased incorporation of long-chain fatty acids into erythrocyte phospholipids of STZ-D rats.

We studied the mechanisms for the altered fatty acid composition in erythrocytes (RBCs) derived from streptozocin-induced diabetic (STZ-D) rats. After 3-wk duration of diabetes, blood glucose, plasma triglyceride, and plasma free-fatty acid levels were all significantly increased. In the diabetic platelet-poor plasma (PPP), the most significant increases in free fatty acids were stearate, linoleate, eicosatrienoate (n-6), and docosahexaenoate (n-3). Fatty acid composition of RBC phospholipids was also altered, with significant decreases in arachidonate, docosatetraenoate (n-6), and docosapentaenoate (n-6) and increases in linoleate and docosahexaenoate. Insulin treatment of the diabetic rats resulted in normalization of docosapentaenoate, arachidonate, and linoleate levels in RBC phospholipids but not of docosahexaenoate or docosatetraenoate levels. The incorporation of [5,6,8,9,11,12,14,15-3H]arachidonate into diabetic RBC phospholipids was significantly decreased compared with the corresponding control RBC, regardless of the incubation medium used, which was the PPP derived either from the control or diabetic rats. Therefore, the decreased incorporation of [5,6,8,9,11,12,14,15-3H]arachidonate into diabetic RBC phospholipids was independent of the altered lipid composition of the PPP incubation media. Furthermore, the decreased incorporation was not specific for arachidonate, because the incorporation of other long-chain fatty acids such as [9,10-3H]oleate, [1-14C]palmitate, [2-14C]eicosatrienoate (n-6), and [1-14C]linoleate into RBC phospholipids was also comparably decreased. More important, the decreased fatty acid incorporations were reversed by insulin treatment of the diabetic rat.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of ingestion of thermally oxidized edible oils on plasma lipids, lipoproteins and postheparin lipolytic activity of rats.

Acute (after 4 hr) and short-term (after 7 days) effects of ingesting heated and unheated groundnut, coconut and safflower oils on plasma lipids, lipoproteins and postheparin lipopolytic activity (PHLA) were studied in rats. All heated oils were characterized by increases in carbonyl value, peroxide value and free fatty acid (FFA) content, except heated coconut oil which showed a decrease in FFA content. Heating procedure also did not alter to an appreciable extent their fatty acid compositions. Acute and short-term effects of feeding heated and unheated oils showed no significant differences in rat plasma levels of total cholesterol (TC), total triglycerides, total phospholipids, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol + very-low-density lipoprotein cholesterol, TC/HDL-C ratio and PHLA. Inspite of certain changes in some of the indices of thermal alteration of these heated oils, consumption of such heated oils by rats did not have any significant effect on various plasma parameters in these animals.

Effect of cod liver oil supplementation on plasma lipids, lipoproteins, lipase activity and platelet aggregation in normotensive and hypertensive volunteers.

No significant change in plasma levels of total cholesterol (TC), triglycerides, phospholipids, very-low-density lipoprotein cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol (HDL-C), lipase activity and TC/HDL-C ratio could be observed in both normotensive and hypertensive individuals after cod liver oil supplementation. Measure of platelet aggregation rates did not also show any significant change after cod liver oil ingestion in both normotensive and hypertensive individuals. The results suggest that supplementation of normal diets with 600 mg cod liver oil per day for 50 days neither affects plasma lipids, lipoproteins and lipase activity nor affects platelet aggregation in both normotensive and hypertensive individuals.