Coupling endoplasmic reticulum stress to the cell-death program: a novel HSP90-independent role for the small chaperone protein p23.

The endoplasmic reticulum (ER) is the principal organelle for the biosynthesis of proteins, steroids and many lipids, and is highly sensitive to alterations in its environment. Perturbation of Ca(2+) homeostasis, elevated secretory protein synthesis, deprivation of glucose or other sugars, altered glycosylation and/or the accumulation of misfolded proteins may all result in ER stress, and prolonged ER stress triggers cell death. Studies from multiple laboratories have identified the roles of several ER stress-induced cell-death modulators and effectors through the use of biochemical, pharmacological and genetic tools. In the present work, we describe the role of p23, a small chaperone protein, in preventing ER stress-induced cell death. p23 is a highly conserved chaperone protein that modulates HSP90 activity and is also a component of the steroid receptors. p23 is cleaved during ER stress-induced cell death; this cleavage, which occurs close to the carboxy-terminus, requires caspase-3 and/or caspase-7, but not caspase-8. Blockage of the caspase cleavage site of p23 was associated with decreased cell death induced by ER stress. Immunodepletion of p23 or inhibition of p23 expression by siRNA resulted in enhancement of ER stress-induced cell death. While p23 co-immunoprecipitated with the BH3-only protein PUMA (p53-upregulated modulator of apoptosis) in untreated cells, prolonged ER stress disrupted this interaction. The results define a protective role for p23, and provide further support for a model in which ER stress is coupled to the mitochondrial intrinsic apoptotic pathway through the activities of BH3 family proteins.

A novel translational repressor mRNA is edited extensively in livers containing tumors caused by the transgene expression of the apoB mRNA-editing enzyme.

Transgene expression of the apolipoprotein B mRNA-editing enzyme (APOBEC-1) causes dysplasia and carcinoma in mouse and rabbit livers. Using a modified differential display technique, we identified a novel mRNA (NAT1 for novel APOBEC-1 target no. 1) that is extensively edited at multiple sites in these livers. The aberrant editing alters encoded amino acids, creates stop codons, and results in markedly reduced levels of the NAT1 protein in transgenic mouse livers. NAT1 is expressed ubiquitously and is extraordinarily conserved among species. It has homology to the carboxy-terminal portion of the eukaryotic translation initiation factor (eIF) 4G that binds eIF4A and eIF4E to form eIF4F. NAT1 binds eIF4A but not eIF4E and inhibits both cap-dependent and cap-independent translation. NAT1 is likely to be a fundamental translational repressor, and its aberrant editing could contribute to the potent oncogenesis induced by overexpression of APOBEC-1.

Hyperediting of multiple cytidines of apolipoprotein B mRNA by APOBEC-1 requires auxiliary protein(s) but not a mooring sequence motif.

An RNA-binding cytidine deaminase (APOBEC-1) and unidentified auxiliary protein(s) are required for apolipoprotein (apo) B mRNA editing. A sequence motif on apoB mRNA ("mooring sequence," nucleotides 6671-6681) is obligatory for the editing of cytidine 6666 (C6666), the only cytidine on apoB mRNA converted to uridine in normal animals. Transgenic animals with hepatic overexpression of APOBEC-1 develop liver tumors, and other non-apoB mRNAs are edited, suggesting a loss of the normally precise specificity. In this study, we examined apoB mRNA from these transgenic animals to determine if cytidines aside from C6666 are edited. Multiple cytidines downstream from C6666 in apoB mRNA were edited extensively by the overexpressed APOBEC-1. This pathophysiological "hyperediting" could be mimicked in vitro by incubating a synthetic apoB RNA substrate with the transgenic mouse liver extracts. Multiple cytidines in the synthetic apoB RNA were edited by recombinant APOBEC-1 but only with supplementation of the auxiliary protein(s). Mutations in the mooring sequence markedly decreased the normal editing of C6666 but, surprisingly, increased the hyperediting of downstream cytidines. Furthermore, cytidines in an apoB RNA substrate lacking the mooring sequence were also edited in vitro. These results indicate that the hyperediting of apoB mRNA by overexpressed APOBEC-1 depends upon auxiliary protein(s) but is independent of the mooring sequence motif. These results suggest that hyperediting may represent the first step in a two-step recognition model for normal apoB mRNA editing.

Cloning and mutagenesis of the rabbit ApoB mRNA editing protein. A zinc motif is essential for catalytic activity, and noncatalytic auxiliary factor(s) of the editing complex are widely distributed.

Apolipoprotein (apo) B mRNA editing is the specific deamination of cytidine (nucleotide 6666) to uridine in apoB mRNA. We isolated a full-length cDNA clone encoding the rabbit apoB mRNA editing protein (REPR), a subunit of the editing complex. Rabbit REPR is analogous to a rat enterocyte 27-kDa protein that has been shown to have cytidine deaminase activity. Like rat REPR, rabbit REPR edited synthetic apoB RNA when mixed with chicken enterocyte extract. Surprisingly, the REPR also acquired editing activity when mixed with extracts from various organs of the rabbit (liver, gallbladder, stomach, intestine, adrenals, thyroid, testes, spleen, kidney, and lung) or the chicken (kidney and liver). In contrast, the rabbit REPR mRNA was found only in the small and large intestine. Thus, the auxiliary protein(s) of the apoB mRNA editing complex, which are essential for editing activity, exist in organs devoid of significant apoB mRNA editing or apoB synthesis. REPR requires zinc for its catalytic activity. We mutated putative zinc-coordinating residues (His61, Cys93, Cys96) and 2 additional residues (Glu63, Pro92) of the rabbit REPR that are conserved in other cytidine or deoxycytidylate deaminases and in rat REPR. The wild-type and mutant REPR cDNAs each produced 28-kDa proteins when transcribed and translated in vitro. Compared with the wild-type editing activity, the mutations of His61-->Ala, Glu63-->Ala, Cys93-->Ala, and Cys96-->Ala abolished or greatly reduced editing activity, whereas the mutations of His61-->Cys (which also can coordinate zinc) and Pro92-->Ala had a lesser effect. These results indicate that His61, Cys93, and Cys96 are essential for editing activity, probably because they coordinate zinc, whereas Glu63 also is essential, because it may be involved in the deaminase reaction. In addition, the widespread distribution of the auxiliary factor(s) portends their involvement in other RNA editing reactions.

The mechanism for apo-B mRNA editing is deamination.

Apolipoprotein (apo-) B mRNA editing at nucleotide 6666 converts cytidine to uridine, transforming the codon for glutamine-2153 to a termination codon. To investigate this editing mechanism, [a-32P] and [5-3H] CTP were incorporated into synthetic apo-B RNA. After the substrate had been edited extensively in vitro by a partially purified editing extract, the edited base was isolated and analyzed for radioactivity. The uridine-6666 resulting from the editing reaction had the same ratio of 3H to 32P as did the cytidine-6666, demonstrating that deamination rather than base exchange or nucleotide replacement is the mechanism for apo-B mRNA editing.

Characterization of apolipoprotein B mRNA editing from rabbit intestine.

Apolipoprotein (apo) B-48 is generated by a unique physiological process. Cytidine 6,666 of the apo B primary transcript is posttranscriptionally converted to a uridine by an RNA editing mechanism that transforms the codon for glutamine 2,153 to a termination codon. The editing reaction can be duplicated in a cell-free extract. In this study, the apo B-48 mRNA editing activity derived from partially purified extracts of rabbit enterocytes was characterized. The optimum conditions for the editing reaction were determined to be a salt concentration of 0.125-0.150 M NaCl or KCl, a pH of 8-8.5, and a temperature of 30 degrees C. The reaction rate was linear up to 45 minutes and was proportional to the editing extract concentration. No metal ion cofactors, DNA or RNA cofactors, or energy requirements were identified. At optimum conditions, the reaction followed Michaelis-Menten kinetics, with a Km of 0.4 nM for the rabbit RNA substrate. In addition, the reaction rate was enhanced by the addition of 25 micrograms/ml heparin or 40% glycerol. The characteristics of the editing reaction suggest that it is catalyzed by a nucleotide sequence-specific cytidine deaminase that is either a single enzyme or a multimeric protein.

Apolipoprotein B mRNA editing. Direct determination of the edited base and occurrence in non-apolipoprotein B-producing cell lines.

In humans, apolipoprotein (apo) B48 is synthesized in the intestine as an obligatory constituent of chylomicrons. Apolipoprotein B48 is identical to the amino-terminal 2152 amino acids (240 kDa) of apoB100 and is translated from an edited apoB mRNA in which codon 2153 has been converted from glutamine (CAA) to what is recognized as a premature stop codon (UAA). To determine whether the apoB mRNA editing in fact converts cytosine 6666 in codon 2153 to uracil, we incubated a synthetic apoB RNA containing 32P-labeled cytosines in an in vitro editing system prepared from rabbit enterocytes. The in vitro edited RNA was purified and digested to nucleoside 5'-monophosphates, which were analyzed on two-dimensional thin-layer chromatography. We found that the edited base co-migrated with authentic uridine 5'-monophosphate. Thus, cytosine 6666 is converted to uracil, most likely by a nucleotide-specific cytosine deaminase. To determine whether apoB mRNA editing occurs in cell lines that do not synthesize apoB, we stably transfected a high expression vector containing 354 base pairs of apoB sequence into 18 different cell lines. We found apoB mRNA editing activity in five osteosarcoma cell lines and one epidermoid cell line, none of which synthesizes any detectable apoB. Thus, apoB mRNA editing occurs in cell lines that do not synthesize apoB, which suggests that mRNA editing may be a common biological phenomenon in eukaryotic cells.

Apolipoprotein B48 RNA editing in chimeric apolipoprotein EB mRNA.

Apolipoprotein (apo) B occurs in two forms, apoB100 (512 kDa) and apoB48 (240 kDa); both are derived from the same gene. A novel mechanism involving editing of the apoB mRNA causes the formation of apoB48; the first base of codon 2153 is changed from cytosine to uracil, converting a glutamine codon to a premature stop codon. To identify the apoB mRNA sequence elements recognized by the apoB mRNA editing mechanism, two apoB cDNA fragments (354 and 63 base pairs) with codon 2153 near their centers were inserted into a high expression vector of another secreted apolipoprotein, apoE. The resulting vectors, pHEB-354 and -63, were transfected into Chinese hamster ovary cells, HepG2 cells, and apoB48-producing CaCo-2 cells. The secreted chimeric apolipoproteins (apoEB354 and apoEB63) were analyzed for premature truncation, and the mRNA was analyzed for the presence of an edited base. The pHEB-354 construct produced a truncated protein only in CaCo-2 cells, whereas pHEB-63 produced no truncated protein in any of the three cell types. The mRNA was converted to cDNA and amplified by the polymerase chain reaction technique. Differential hybridization of the polymerase chain reaction products with CAA (Gln) and TAA (Stop) specific probes detected an edited base only in cDNA from CaCo-2 cells transfected with pHEB-354, in agreement with the protein analysis. We conclude that the nucleotide sequence of the apoB cDNA insert in pHEB-354 contains sufficient information to be edited in CaCo-2 cells. In these cells, a cryptic polyadenylation site was activated in the edited pHEB-354 mRNA. As a result, CaCo-2 cells transfected with pHEB-354 produced a short, edited pHEB-354 mRNA and a long, unedited pHEB-354 mRNA. Chinese hamster ovary cells transfected with pHEB-354 or CaCo-2 cells transfected with pHEB-63 produced only a full length transcript. Amplification of the pHEB-354 cDNA using 3'-primers upstream and downstream of the poly(A) addition site and hybridization with the TAA probe confirmed these results. This unusual mRNA editing apparently occurs before polyadenylation, probably in the nucleus.

Structural relationship of human apolipoprotein B48 to apolipoprotein B100.

Although the complete amino acid sequence of human apolipoprotein (apo) B100 is known (4536 amino acids), the structure of apo B48 has not been defined. The objective of our study was to define the structure of apo B48 and its relationship to apo B100. Antibodies were produced against 22 synthetic peptides corresponding to sequences in human apo B100. The levels of immunoreactivity of the antipeptides to apo B100 and apo B48 were used to define the structural relationship between these two species of apo B. Six antibodies from sequences in the amino-terminal half of apo B100, including antipeptide 2110-2129, bound to both apo B100 and apo B48. 15 other apo B-specific antipeptides from sequences carboxyl-terminal to residue 2152 bound to apo B100, but not to apo B48. Immunoblots of cyanogen bromide digests of apo B100 and apo B48 with antipeptides 2068-2091 and 2110-2129 detected a 16-KD fragment (residues 2016-2151) in the apo B100 digest and a fragment of identical size in the apo B48 digest. Because apo B48 appears to contain the apo B100 cyanogen bromide fragment 2016-2151 and because an antiserum specific for the peptide 2152-2168 does not bind to apo B48, we conclude that apo B48 represents the amino-terminal 47% of apo B100 and that the carboxyl terminus of apo B48 is in the vicinity of residue 2151 of apo B100.

Regulation of beta-adrenoceptor number and subtype in 3T3-L1 preadipocytes by sodium butyrate.

In mouse 3T3-L1 preadipocytes, the glucocorticoid dexamethasone has been shown to promote a switch in beta-adrenoceptor subtype expression from beta 1 to beta 2 and to increase the total number of beta-adrenoceptors. The present study demonstrates that sodium butyrate also modulates beta-adrenoceptor expression in these cells. Incubation of preadipocytes with 2-10 mM butyrate for 24-48 h promoted a dose- and time-dependent switch in beta-adrenoceptor subtype from a near equal mixture of beta 1 and beta 2 to greater than 85% beta 2 and caused an approximate doubling of the receptor number. beta-Adrenoceptors were assayed in membranes prepared from 3T3-L1 cells using the radiolabeled antagonist [125I]iodocyanopindolol and the beta 2-selective antagonist ICI 118.551. Other short chain acids were not as effective as butyrate in promoting changes in beta-adrenoceptor expression. Cycloheximide (1.0 microgram/ml) inhibited the effects of butyrate on both beta-adrenoceptor subtype and number. Alterations in beta-adrenoceptor phenotype promoted by either butyrate or dexamethasone were functionally correlated with cAMP accumulation in these cells. Comparison of the effects of butyrate and dexamethasone on beta-adrenoceptor expression suggests that these two agents regulate beta-adrenoceptors by different mechanisms.

Glucocorticoid regulation of beta-adrenergic receptors in 3T3-L1 preadipocytes.

Treatment of 3T3-L1 preadipocytes (fibroblasts) with 250 nM dexamethasone for 48 hr caused a doubling of total beta-adrenergic receptors and an increase in beta 2-adrenergic receptor subtype proportion from approximately 50% in controls to 85% in treated cells. The responses to epinephrine and norepinephrine in a whole cell cAMP accumulation assay reflected these changes. The effects of dexamethasone on beta-adrenergic receptors were mediated through the glucocorticoid receptor and were time and dose dependent with an EC50 of 2.77 +/- 0.73 nM for an increase in the proportion of beta 2-adrenergic receptors. The rank order of potency of steroids to effect these changes (betamethasone = dexamethasone greater than fludrocortisone greater than hydrocortisone = triamcinolone greater than aldosterone) correlated with their glucocorticoid potency. [3H]Dexamethasone binding to intact cells yielded a KD value of 3.47 +/- 0.38 nM for binding to the glucocorticoid receptor which correlated well with the EC50 for dexamethasone to alter beta-adrenergic receptors. Inhibition of [3H]dexamethasone binding by other steroids confirmed that the ability of steroids to regulate beta-adrenergic receptors correlated with the affinity of each compound for the 3T3-L1 glucocorticoid receptor. Progesterone, which can bind to the glucocorticoid receptor but has only weak agonist activity, competitively inhibited the ability of dexamethasone to alter beta-adrenergic receptors. Protein synthesis, RNA synthesis, and N-linked glycosylation appeared to be necessary for the change in receptor subtype expression and the increase in beta-adrenergic receptor number induced by dexamethasone. The present study suggests that regulation of beta-adrenergic receptor expression in 3T3-L1 preadipocytes by dexamethasone is a glucocorticoid-specific effect which may require gene activation.

Pancreatic and intestinal response to dietary guar gum in rats.

Rats were fed a fiber-free semi-purified diet or one containing 10% guar gum, a hydrophilic galactomannan, or a laboratory stock diet. The presence of guar gum in the diet decreased food intake and body weight gain perhaps due to distension of the gastrointestinal tract. Relative to the group fed fiber-free diet, liver weight was smaller and pancreas weight larger in the groups consuming guar gum or the stock diet. The latter 2 diets were hypocholesterolemic relative to the fiber-free diet. In both the unfed and fed state the wet weight of the intestine was significantly greater in the rats consuming guar gum. The greatest difference was in the wet weights of the small intestine in the fed animals. In the pancreas, there were no notable differences in digestive enzyme activity between the groups fed guar and fiber-free diet. However, in the intestine, lipase, amylase and total proteolytic activity were significantly greater in the rats fed guar gum. This elevation of enzyme activity in the intestine could be due to a slower rate of enzyme degradation or to enhancement of the enzyme secretion. The ability of guar gum to increase the volume of intestinal contents may be important in its slowing of absorption.