Isolation, expression, and characterization of fully functional nontoxic BiP/GRP78 mutants.

Mammalian BiP/GRP78 and Escherichia coli DnaK belong to the highly conserved hsp70 family and function as molecular chaperones in the endoplasmic reticulum or the cytosol, respectively. Induction of murine BiP/GRP78 expression in E. coli leads to growth arrest and cell death, independent of the bacterial strain and vector used. Analysis of various BiP constructs and mutants shows that the dominant-lethal phenotype is induced specifically by the expression of the 13.7-kDa C-terminal domain and abolished by a single substitution in that region. Deletion of that region also results in nontoxic gene products that can be overexpressed and purified to homogeneity. The nontoxic mutants are highly expressed in E. coli, representing up to 20% of the soluble fraction. They are catalytically active, depolymerize upon binding ATP or synthetic peptide, and interact with the J-domain of the DnaJ-like accessory protein, MTJ1, with near wild-type affinity. Our data indicate that the cytotoxic effect encountered during overexpression of recombinant proteins can be caused by a single domain and can be alleviated by a specific mutation or deletion in that region without altering the catalytic properties of the enzyme.

A metabolite of equine estrogens, 4-hydroxyequilenin, induces DNA damage and apoptosis in breast cancer cell lines.

Estrogen replacement therapy has been correlated with an increased risk of developing breast or endometrial cancer. 4-Hydroxyequilenin (4-OHEN) is a catechol metabolite of equilenin which is a minor component of the estrogen replacement formulation marketed under the name of Premarin (Wyeth-Ayerst). Previously, we showed that 4-OHEN autoxidizes to quinoids which can consume reducing equivalents and molecular oxygen, are potent cytotoxins, and cause a variety of damage to DNA, including formation of bulky stable adducts, apurinic sites, and oxidation of the phosphate-sugar backbone and purine/pyrimidine bases [Bolton, J. L., Pisha, E., Zhang, F., and Qiu, S. (1998) Chem. Res. Toxicol. 11, 1113-1127]. All of these deleterious effects could contribute to the cytotoxic and genotoxic effects of equilenin in vivo. In the study presented here, we examined the relative toxicity of 4-OHEN in estrogen receptor (ER) positive cells (MCF-7 and S30) compared to that in breast cancer cells without the estrogen receptor (MDA-MB-231). The data showed that 4-OHEN was 4-fold more toxic to MCF-7 cells (LC(50) = 6.0 +/- 0. 2 microM) and 6-fold more toxic to S30 cells (LC(50) = 4.0 +/- 0.1 microM) than to MDA-MB-231 cells (LC(50) = 24 +/- 0.3 microM). Using the single-cell gel electrophoresis assay (comet assay) to assess DNA damage, we found that 4-OHEN causes concentration-dependent DNA single-strand cleavage in all three cell lines, and this effect could be enhanced by agents which catalyze redox cycling (NADH) or deplete cellular GSH (diethyl maleate). In addition, the ER(+) cell lines (MCF-7 and S30) were considerably more sensitive to induction of DNA damage by 4-OHEN than the ER(-) cells (MDA-MB-231). 4-OHEN also caused a concentration-dependent increase in the amount of mutagenic lesion 8-oxo-dG in the S30 cells as determined by LC/MS-MS. Cell morphology assays showed that 4-OHEN induces apoptosis in these cell lines. As observed with the toxicity assay and the comet assay, the ER(+) cells were more sensitive to induction of apoptosis by 4-OHEN than MDA-MB-231 cells. Finally, the endogenous catechol estrogen metabolite 4-hydroxyestrone (4-OHE) was considerably less effective at inducing DNA damage and apoptosis in breast cancer cell lines than 4-OHEN. Our data suggest that the cytotoxic effects of 4-OHEN may be related to its ability to induce DNA damage and apoptosis in hormone sensitive cells in vivo, and these effects may be potentiated by the estrogen receptor.

Interaction of murine BiP/GRP78 with the DnaJ homologue MTJ1.

The activity of Hsp70 proteins is regulated by accessory proteins, among which the most studied are the members of the DnaJ-like protein family. BiP/GRP78 chaperones the translocation and maturation of secreted and membrane proteins in the endoplasmic reticulum. No DnaJ-like partner has been described so far to regulate the function of mammalian BiP/GRP78. We show here that murine BiP/GRP78 interacts with the lumenal J domain of the murine transmembrane protein MTJ1 (J-MTJ1). J-MTJ1 stimulates the ATPase activity of BiP/GRP78 at stoichiometric concentrations. The C-terminal tail of BiP/GRP78 is not required for the interaction with J-MTJ1, leaving the function of this portion of the molecule still unclear. Physical interactions between J-MTJ1 and BiP/GRP78 are stable and can be abolished by a single histidine --> glutamine substitution in the highly conserved HPD motif shared by all DnaJ-like proteins. The J-MTJ1 fragment, but not the mutant J-MTJ1:H89Q fragment, stimulates the ATPase activity of Escherichia coli DnaK, although at a higher concentration than its genuine partner DnaJ. Full-length DnaJ does not stimulate BiP over the range of concentrations investigated. These results indicate that the J domain of MTJ1 is sufficient for its interaction with BiP/GRP78 and cannot be substituted by E. coli DnaJ.

Substrate binding induces depolymerization of the C-terminal peptide binding domain of murine GRP78/BiP.

To investigate the role of each domain in BiP/GRP78 function, we have used a full-length recombinant BiP engineered to contain two enterokinase sites; one site is located after an N-terminal FLAG epitope, and a second site has been inserted at the junction between the N- and C-terminal domains (FLAG-BiP.ent). FLAG-BiP.ent oligomerizes into multiple species that interconvert with each other in a slow, concentration- and temperature-dependent equilibrium. Binding of ATP or AMP-PNP (adenosine 5'-(beta, gamma-imino)triphosphate), but not ADP, or of a peptidic substrate induces depolymerization of FLAG-BiP.ent and stabilization of monomeric species. Enterokinase cleavage of monomeric, nucleotide-free BiP.ent results in the physical dissociation of the 44-kDa N-terminal ATPase fragment (N44.ent) from the 30-kDa C-terminal substrate binding domain (C30.ent). Upon dissociation, the freed C-terminal substrate binding domain readily undergoes self-association while N44.ent remains monomeric. Enterokinase cleavage performed in the presence of a synthetic peptide prevents oligomerization of the freed C30.ent domain. Addition of ATP during enterokinase cleavage has no effect on C30.ent oligomerization. Our data clearly indicate that binding of a specific peptide onto the C-terminal domain, or ATP onto the N-terminal domain, induces internal conformational change(s) within the C30 domain that result(s) in BiP depolymerization.