DNA binding of in vitro activated Stat1 alpha, Stat1 beta and truncated Stat1: interaction between NH2-terminal domains stabilizes binding of two dimers to tandem DNA sites.

Stat1 alpha, Stat1 beta and a proteolytically defined truncated Stat1 (132-713, Stat1tc) have been prepared from recombinant sources. All three proteins were specifically phosphorylated on Tyr701 in vitro and the phosphoprotein purified to homogeneity. This was achieved by employing a new isolation scheme that does not include DNA affinity steps and readily allows for the isolation of tens of milligrams of activated Stat protein. The purified phosphoprotein was free of traces of unphosphorylated polypeptide as detected by mass spectrometry. The phosphorylated Stat1 preparations bound to various DNA recognition sites with the same Keq of approximately 1 x 10(-9) M; distinction between 'weak' and 'strong' binding sites is determined by the very rapid dissociation (< 30 s, t1/2) from 'weak' sites compared with 'strong' sites (approximately 3 min, t1/2). Reports of 'weak' tandem binding sites in a natural gene caused us to examine binding to tandem sites leading to the finding that the Stat1 alpha or beta (38 amino acids shorter on the C terminus) bound to two tandem sites (but not two head-to-head sites) with a higher stability than to a single recognition site. The N-terminally truncated protein Stat1tc did not show this cooperative binding, thus implicating the N-terminal domain in promoting Stat1-Stat1 dimer interaction.     

Ultrastructure and function of dimeric, soluble intercellular adhesion molecule-1 (ICAM-1)

Previous studies have demonstrated dimerization of intercellular adhesion molecule-1 (ICAM-1) on the cell surface and suggested a role for immunoglobulin superfamily domain 5 and/or the transmembrane domain in mediating such dimerization. Crystallization studies suggest that domain 1 may also mediate dimerization. ICAM-1 binds through domain 1 to the I domain of the integrin alpha(L)beta(2) (lymphocyte function-associated antigen 1). Soluble C-terminally dimerized ICAM-1 was made by replacing the transmembrane and cytoplasmic domains with an alpha-helical coiled coil. Electron microscopy revealed C-terminal dimers that were straight, slightly bent, and sometimes U-shaped. A small number of apparently closed ring-like dimers and W-shaped tetramers were found. To capture ICAM-1 dimerized at the crystallographically defined dimer interface in domain 1, cysteines were introduced into this interface. Several of these mutations resulted in the formation of soluble disulfide-bonded ICAM-1 dimers (domain 1 dimers). Combining a domain 1 cysteine mutation with the C-terminal dimers (domain 1/C-terminal dimers) resulted in significant amounts of both closed ring-like dimers and W-shaped tetramers. Surface plasmon resonance studies showed that all of the dimeric forms of ICAM-1 (domain 1, C-terminal, and domain 1/C-terminal dimers) bound similarly to the integrin alpha(L)beta(2) I domain, with affinities approximately 1.5--3-fold greater than that of monomeric ICAM-1. These studies demonstrate that ICAM-1 can form at least three different topologies and that dimerization at domain 1 does not interfere with binding in domain 1 to alpha(L)beta(2).     

Targeting Stat3 with G-quartet oligodeoxynucleotides in human cancer cells

Stat3 is an oncogene that is activated in many human cancer cells. Genetic approaches that disrupt Stat3 activity result in inhibition of cancer cell growth and enhanced cell apoptosis supporting the development of novel drugs targeting Stat3 for cancer therapy. G-quartet oligodeoxynucleotides (ODNs) were demonstrated to be potent inhibitors of Stat3 DNA binding activity in vitro with the G-quartet ODN, T40214, having an IC(50) of 7 microM. Computer-simulated docking studies indicated that G-quartet ODNs mainly interacted with the SH2 domain of Stat3 and were capable of inserting between the SH2 domains of Stat3 dimers bound to DNA. We demonstrated that the G-rich ODN T40214, which forms a G-quartet structure at intracellular but not extracellular K+ ion concentrations, is delivered efficiently into the cytoplasm and nucleus of cancer cells where it inhibited IL-6-stimulated Stat3 activation and suppressed Stat3-mediated upregulation of bcl-x and mcl-1 gene expression. Thus, G-quartet represents a new class of drug for targeting of Stat3 within cancer cells.     

Solution structure, stability, and nucleic acid binding of the hyperthermophile protein Sso10b2

The Sso10b (or Alba) family of proteins is a conserved group of archaeal and eukaryotic proteins which are thought to play a role in both chromatin organization and RNA metabolism. We describe here the solution structure and properties of Sso10b2 from Sulfolobus solfataricus. NMR data including residual dipolar couplings and (15)N relaxation data demonstrated that the protein adopts a beta(1)alpha(1)beta(2)alpha(2)beta(3)beta(4) topology with an IF-3-like fold. The protein dimerizes in solution at 30 degrees C via a hydrophobic surface defined by the C-terminal alpha(2)beta(3)beta(4) elements with a structure similar to one of the putative dimers indicated by previous crystal structures. DSC and circular dichroism data demonstrated an unusual two-state structural transition near the growth temperature which led to an increase in beta-sheet content without dissociation of the dimer. The cooperativity of the transition exceeded that of a dimer at pH 7, demonstrating the presence of higher order oligomers near the growth temperature at pH 7. Reverse titrations of Sso10b2 with nucleic acid showed that the protein binds single-stranded DNA (K(d) of 3 x 10(-)(7) M) with higher affinity than RNA (1.3 x 10(-)(6) M) or double-stranded DNA (1.5 x 10(-)(5) M) in 10 mM KH(2)PO(4) (pH 7.0, 20 degrees C). NMR chemical shift perturbation data indicated that single-stranded DNA and RNA binding occurred across the same dimer interface and encompassed a surface defined by the C-terminal ends of the beta(1), beta(2), and beta(3) strands of each monomer.     

Effects of antioestrogens on the DNA binding activity of oestrogen receptors in vitro.

We have investigated the effect of a series of steroidal oestrogen antagonists, related to ICI 164,384, on the DNA binding activity of mouse oestrogen receptors expressed in insect cells. The analogues possess different side chains at the 7-position of the B ring in the steroid. Inhibition was observed when the length of the side-chain was 15-16 carbon atoms but not 10 or 20 carbon atoms and only when the 7 alpha isomer was used. The DNA binding activity of receptors expressed in COS-1 cells was also inhibited after extended periods of incubation with antioestrogens but not that of the human receptor in breast cancer cell-extracts. We have proposed that ICI 164,384 might disrupt receptor dimerisation, and therefore the variation in its ability to inhibit DNA binding activity may reflect differences in dimer stability. Since the DNA binding activity of in vitro translated receptors was inhibited when they were translated in the presence of the antioestrogen we suggest that ICI 164,384 might prevent the formation of receptor dimers without necessarily being able to disrupt preformed dimers.     

Beta gamma dimers of G proteins inhibit atrial muscarinic K+ channels

It has been proposed that beta gamma dimers of signal-transducing G proteins mediate muscarinic activation of atrial K+ channels. We examined this hypothesis by testing the effects of beta gamma dimers from four sources (human erythrocytes, human placenta, bovine brain, and bovine retina) on single channel muscarinic K+ (K+[acetylcholine (ACh)]) currents in inside-out membrane patches of adult guinea pig atria. None of the four beta gamma dimer preparations stimulated K+[ACh] currents; on the contrary, each inhibited the currents whether the currents were activated with GTP alone (agonist-independent activity) or with GTP plus a muscarinic agonist (agonist-dependent activity). Detergents at concentrations used to suspend erythrocyte, brain, and placental beta gamma dimers had no effect by themselves, and detergents were not used with the retinal beta gamma dimers. We conclude that beta gamma dimers do not mediate stimulatory effects of the endogenous G protein that regulates the K+ channels. In fact beta gamma dimers appear to inhibit activation by the endogenous G alpha subunits. Further insight into the role of beta gamma dimers came from the observation that agonist-independent GTP-activated K+[ACh] currents were inhibited by beta gamma dimers at about one-tenth the concentration required to inhibit agonist-dependent activation. One possibility is that dimeric beta gamma may have a higher affinity for free alpha subunits than for alpha subunits associated with agonist-occupied receptors. Thus, in addition to the known requirement of beta gamma dimers for the interaction of alpha subunits with receptors, beta gamma dimers may also improve the signal-to-noise ratio for agonists by reducing agonist-independent background activities.     

Structural requirements for the transmembrane activation of the insulin receptor kinase

Tetrameric insulin holoreceptor (alpha 2 beta 2) was reduced with dithiothreitol into alpha beta dimers such that they maintain up to 50% of insulin binding at tracer ligand concentrations. Scatchard analysis of insulin binding to dimers revealed that they had a reduced affinity for ligand by a factor of 3-6 compared to holoreceptor, whereas the maximum number of high affinity binding sites was not affected. The alpha beta dimers can be separated from holoreceptor by sucrose density gradient centrifugation, and hence, they are not associated by noncovalent interactions. Insulin-dependent autophosphorylation of alpha beta dimers isolated from low ionic strength sucrose density gradients was minimal and was always accompanied by reoxidation of dimers to the tetrameric holoreceptor. The reformed tetramer exhibited a strong insulin-dependent autophosphorylation reaction. Reoxidation was prevented by isolating alpha beta dimers in sucrose density gradients containing 0.15 M NaCl. Under these conditions, no insulin-dependent autophosphorylation was observed. When insulin receptor was first autophosphorylated and then reduced, receptor kinase activity, as assayed by histone phosphorylation, was not affected. Also, the insulin-independent, basal autophosphorylation was maintained after reduction into alpha beta dimers. We conclude that alpha beta-alpha beta interaction is not necessary for the maintenance of basal kinase activity or for insulin-activated kinase activity once autophosphorylation occurs. However, dimer-dimer interaction appears critical for the insulin-dependent activation of the receptor's intrinsic kinase activity.     

Mechanism of DnaB helicase of Escherichia coli: structural domains involved in ATP hydrolysis, DNA binding, and oligomerization.

We describe the delineation of three distinct structural domains of the DnaB helicase of Escherichia coli: domain alpha, amino acid residues (aa) 1-156; domain beta, aa 157-302; and domain gamma, aa 303-471. Using mutants with deletion in these domains, we have examined their role(s) in hexamer formation, DNA-dependent ATPase, and DNA helicase activities. The mutant DnaBbetagamma protein, in which domain alpha was deleted, formed a hexamer; whereas the mutant DnaBalphabeta, in which domain gamma was deleted, could form only dimers. The dimerization of DnaBalphabeta was Mg(2+) dependent. These data suggest that the oligomerization of DnaB helicase involves at least two distinct protein-protein interaction sites; one of these sites is located primarily within domain beta (site 1), while the other interaction site is located within domain gamma (site 2). The mutant DnaBbeta, a polypeptide of 147 aa, where both domains alpha and gamma were deleted, displayed a completely functional ATPase activity. This domain, thus, constitutes the "central catalytic domain" for ATPase activity. The ATPase activity of DnaBalphabeta was kinetically comparable to that of DnaBbeta, indicating that domain alpha had little or no influence on the ATPase activity. In both cases, the ATPase activities were DNA independent. DnaBbetagamma had a DNA-dependent ATPase activity that was kinetically comparable to the ATPase activity of wild-type DnaB protein (wtDnaB), indicating a specific role for C-terminal domain gamma in enhancement of the ATPase activity of domain beta as well as in DNA binding. Mutant DnaBbetagamma, which lacked domain alpha, was devoid of any helicase activity pointing to a significant role for domain alpha. The major findings of this study are (i) domain beta contained a functional ATPase active site; (ii) domain gamma appeared to be the DNA binding domain and a positive regulator of the ATPase activity of domain beta; (iii) although domain alpha did not have any significant effect on the ATPase, DNA binding activities, or hexamer formation, it definitely plays a pivotal role in transducing the energy of ATP hydrolysis to DNA unwinding by the hexamer; and (iv) all three domains are required for helicase activity.     

DNA interaction and dimerization of eukaryotic SMC hinge domains

The eukaryotic SMC1/SMC3 heterodimer is essential for sister chromatid cohesion and acts in DNA repair and recombination. Dimerization depends on the central hinge domain present in all SMC proteins, which is flanked at each side by extended coiled-coil regions that terminate in specific globular domains. Here we report on DNA interactions of the eukaryotic, heterodimeric SMC1/SMC3 hinge regions, using the two known isoforms, SMC1alpha/SMC3 and the meiotic SMC1beta/SMC3. Both dimers bind DNA with a preference for double-stranded DNA and DNA rich in potential secondary structures. Both dimers form large protein-DNA networks and promote reannealing of complementary DNA strands. DNA binding but not dimerization depends on approximately 20 amino acids of transitional sequence into the coiled-coil region. Replacement of three highly conserved glycine residues, thought to be required for dimerization, in one of the two hinge domains still allows formation of a stable dimer, but if two hinge domains are mutated dimerization fails. Single-mutant dimers bind DNA, but hinge monomers do not. Together, we show that eukaryotic hinge dimerization does not require conserved glycines in both hinge domains, that only the transition into the coiled-coil region rather than the entire coiled-coil region is necessary for DNA binding, and that dimerization is required but not sufficient for DNA binding of the eukaryotic hinge heterodimer.     

Regulation of Stat3 nuclear import by importin alpha5 and importin alpha7 via two different functional sequence elements

Regulated import of STAT proteins into the nucleus through the nuclear pores is a vital event. We previously identified Arg214/215 in the coiled-coil domain and Arg414/417 in the DNA binding domain involved in the ligand-induced nuclear translocation of Stat3. In this study, we investigated the mechanism for Stat3 nuclear transport. We report here that among five ubiquitously expressed human importin alphas, importin alpha5 and alpha7, but not importin alpha1, alpha3, and alpha4, bind to Stat3 upon cytokine stimulation. Similar results were observed for Stat1, but not for Stat5a and 5b, which were unable to interact with any of the importin alphas. The C-terminus of importin alpha5 is necessary but not sufficient for Stat3 binding. Truncation mutant of Stat3 (aa1-320) that contains Arg214/215 exhibits specific binding to importin alpha5, and an exclusive nuclear localization. Point mutations of Arg214/215 in this mutant destroy importin alpha5 binding and its nuclear localization. In contrast, the truncation mutant (aa320-770) including Arg414/417 fails to interact with importin alpha5 and is localized in the cytoplasm. However, both sequence elements are necessary for the full-length Stat3's interaction with importin alpha5. These results suggest that Arg214/215 is likely the binding site for importin alpha5, whereas Arg414/417 may not be involved in the direct binding, but necessary for maintaining the proper conformation of Stat3 dimer for importin binding. A model for Stat3 nuclear translocation is proposed based on these data.