A Novel Polar Core and Weakly Fixed C-Tail in Squid Arrestin Provide New Insight into Interaction with Rhodopsin

Photoreceptors of the squid Loligo pealei contain a G-protein-coupled receptor (GPCR) signaling system that activates phospholipase C in response to light. Analogous to the mammalian visual system, signaling of the photoactivated GPCR rhodopsin is terminated by binding of squid arrestin (sArr). sArr forms a light-dependent, high-affinity complex with squid rhodopsin, which does not require prior receptor phosphorylation for interaction. This is at odds with classical mammalian GPCR desensitization where an agonist-bound phosphorylated receptor is needed to break stabilizing constraints within arrestins, the so-called “three-element interaction” and “polar core” network, before a stable receptor–arrestin complex can be established. Biophysical and mass spectrometric analysis of the squid rhodopsin–arrestin complex indicates that in contrast to mammalian arrestins, the sArr C-tail is not involved in a stable three-element interaction. We determined the crystal structure of C-terminally truncated sArr that adopts a basal conformation common to arrestins and is stabilized by a series of weak but novel polar core interactions. Unlike mammalian arrestin-1, deletion of the sArr C-tail does not influence kinetic properties of complex formation of sArr with the receptor. Hydrogen–deuterium exchange studies revealed the footprint of the light-activated rhodopsin on sArr. Furthermore, double electron–electron resonance spectroscopy experiments provide evidence that receptor-bound sArr adopts a conformation different from the one known for arrestin-1 and molecular dynamics simulations reveal the residues that account for the weak three-element interaction. Insights gleaned from studying this system add to our general understanding of GPCR–arrestin interaction.     

Epitopes of human interferon-alpha defined by the reaction of monoclonal antibodies with alpha interferons and interferon analogues

Five monoclonal antibodies reactive with human interferon (HuIFN)-alpha 2, but not with HuIFN-alpha 1, have been analyzed for their reaction with a series of IFN analogues and hybrid IFN molecules. Using analogues containing alpha 1 or gamma substitutions, it was shown that amino acids in the 107 to 113 region are implicated in the epitopes recognized by four of the five antibodies tested. Surprisingly, two of the antibodies that did not react with [alpha 1(113 to 114)112 to 113]alpha 2 also did not react with a truncated IFN-alpha 2(4 to 155). The presence of an epitope determined by amino acids at 112 and 113 and by the amino and carboxyl ends of the molecule supports a model for IFN where the carboxyl- and amino-terminals are adjacent to the proposed reverse turn around amino acid 110 to 115. The fifth alpha 2 specific antibody whose reaction with HuIFN-alpha 2 is not affected by the above substitutions and truncations recognizes IFN or IFN hybrids that, like alpha 2, have arginine at position 120, but does not react with IFN that, like alpha 1, have lysine at position 120. Amino acids 107 to 113 and 120 lie in regions of the molecule that have high hydrophilicity and are probably structurally involved in the epitopes recognized by the antibodies. Under conditions of antibody excess, the antibodies described here inhibit binding of HuIFN-alpha 2 to both human and bovine cells.     

Characterization of a second gene product related to rabbit cytochrome P-450 1

A cDNA, p1-88, was cloned from a library constructed using rabbit liver mRNA. Sequence analysis indicates that p1-88 is highly similar (congruent to 95%) to the cDNA, p1-8, that encodes rabbit liver cytochrome P-450 1 and that had been isolated from the same library. The predicted amino acid sequence of the protein encoded by p1-88, P-450 IIC4, differs at 25 of 487 amino acids from that encoded by p1-8. P-450 IIC4 was synthesized in vitro using rabbit reticulocyte lysate primed with RNA transcribed from the coding sequence of p1-88 using a bacteriophage T7 RNA polymerase/promoter system. P-450 IIC4 reacts with two monoclonal antibodies that recognize P-450 1 and exhibits the same relative electrophoretic mobility as P-450 1. In contrast, the reactivity of a third monoclonal antibody recognizing P-450 1, 1F11, toward P-450 IIC4 synthesized in vitro is greatly diminished. The latter antibody extensively inhibits hepatic progesterone 21-hydroxylase activity and recognizes phenotypic differences among rabbits in the microsomal concentration of P-450 1. This difference in the immunoreactivity of P-450 IIC4 and P-450 1 with the 1F11 antibody suggests that P-450 IIC4 does not contribute significantly to hepatic progesterone 21-hydroxylase activity. S1 nuclease mapping demonstrates that the expression of mRNAs corresponding to p1-88 are expressed to equivalent extents in rabbits exhibiting high and low expression of mRNAs corresponding to p1-8. Thus, P-450 1 differs from the protein encoded by p1-88, in its regulation, immunoreactivity, and by inference its catalytic properties although the amino acid sequences of P-450 1 and P-450 IIC4 are highly similar (congruent to 95%).     

Effect of transportation stress on ovarian function in superovulated Hereford heifers

A study was conducted to determine the effect of transportation stress on ovarian function in superovulated heifers. Thirty cyclic Hereford heifers of similar age and weight and in good body condition were randomly assigned to control and stress-treated groups. All animals received two daily injections of 5 mg follicle stimulating hormone (FSH) for 4 d beginning on Day 10 to 12 of the estrous cycle. A blood sample was collected at each FSH injection. On the fourth day of injections, heifers were given 25 mg prostaglandin F(2)alpha (PGF(2)alpha) in the morning and a second injection of 15 mg PGF(2)alpha in the afternoon. During superovulation, the stressed heifers were transported to a different location every 12 h whereas control animals remained at the pretrial site. Following 4 d of intermittent transporting and FSH treatment, stress-treated heifers were recombined with control animals. Ovaries were examined 8 d following the onset of standing estrus to determine length, width, thickness, and number of corpora lutea (CL). Peripheral plasma levels of cortisol were higher in the stressed group (P< 0.1). Least squares means for numbers of CL were 20.4 +/- 2.1 and 15.4 +/- 1.7 for control and treated heifers, respectively (P< 0.1). There were no treatment differences (P> 0.1) between length, width, or thickness of ovaries when the number of CL was held constant. These data suggest that stress of the type, intensity, and duration imposed in this study increased plasma levels of cortisol and reduced ovulation rate as determined by CL formation in superovulated heifers.     

Solid-state NMR study of trehalose/1,2-dipalmitoyl-sn-phosphatidylcholine interactions

31P and 2H solid-state NMR studies of dry trehalose (TRE) and 1,2-dipalmitoyl-sn-phosphatidylcholine (DPPC) mixtures are reported. 31P spectra are consistent with a rigid head group above and below the calorimetric phase transition for both dry DPPC and a dry 2:1 TRE/DPPC mixture. In addition, 2H spectra of DPPC labeled at the 7-position of the sn-2 chain (2[7,7-2H2]DPPC) show exchange-narrowed line shapes with a width of 120 kHz over the temperature range 25-75 degrees C. These line shapes can be simulated with a model involving two-site jumps of the deuteron. In contrast, the 2H NMR spectrum of a dry 2:1 TRE/2[7,7-2H2]DPPC mixture above the phase transition (Tc = 46 degrees C) is narrowed by a factor of approximately 4 to a width of 29 kHz. Simulation of this spectrum requires a model involving four-site jumps of the deuteron and is indicative of highly disordered lipid acyl chains similar to those found in the L alpha-phases of hydrated lipids. Thus, TRE/DPPC mixtures above their transition temperatures exist in a new type of liquid crystalline like phase, which we term a lambda-phase. The observation of the dynamic properties of this new phase indicates the mechanism by which anhydrobiotic organisms maintain the integrity of their membranes upon dehydration.     

Inhibition of guanidinobenzoatase by a substrate for trypsin-like enzymes

Guanidinobenzoatase is a proteolytic enzyme capable of degrading fibronectin and is a tumour associated enzyme. Guanidinobenzoatase has been shown to be an arginine selective protease and is distinct from trypsin, plasmin and thrombin, the latter enzymes can be assayed with bis(carbobenzyloxycarbonyl-L-argininamido)-Rhodamine or BZAR. Guanidinobenzoatase is inhibited by BZAR when the enzyme is assayed in free solution and when the enzyme is cell-bound in frozen sections of tumour containing tissues. It is proposed that BZAR and its analogues may be of value in inhibiting tumour cell invasion in vivo and also that the selectivity of BZAR may be used to direct cytotoxic drugs to tumour cells possessing active guanidinobenzoatase.     

Specificity of androgen resistance inMus caroli kidney

Androgen controls the expression of beta-glucuronidase and several other proteins in the kidney of the standard laboratory mouse, Mus musculus. Other species within the genus Mus exhibit a variety of response patterns for kidney beta-glucuronidase and other markers of androgen action. We have investigated the mechanism of androgen action in M. caroli, a Mus species that does not produce beta-glucuronidase in response to testosterone. The failure of testosterone to induce beta-glucuronidase in M. caroli females cannot be overcome by treatment with dihydrotestosterone, with pharmacological doses of testosterone propionate or dihydrotestosterone propionate, or with a variety of potent androgen analogues. All of these compounds induce kidney beta-glucuronidase in M. musculus females and kidney ornithine decarboxylase, submandibular gland renin, and submandibular gland epidermal growth factor in both M. caroli and M. musculus females. Furthermore, kidney androgen receptor proteins from M. caroli and M. musculus animals have the same sedimentation characteristics on sucrose density gradients. These data indicate that androgen resistance in M. caroli is not due to deficient 5 alpha-reductase or aberrant hormone metabolism producing suboptimal levels of functional androgen and is not caused by a defective androgen receptor. They suggest that the resistance of beta-glucuronidase in M. caroli kidney to induction by androgen occurs at the level of the beta-glucuronidase gene.     

Expression of the cytosolic and particulate forms of transglutaminase during chemically induced rat liver carcinogenesis

Transglutaminase (EC 2.3.2.13) activity in chemically induced rat hepatocellular carcinomas was reduced by some 65% when compared to normal rat livers. The majority of the remaining activity (approx. 85%) was found in the particulate fraction. The use of non-ionic detergent to extract the transglutaminase activity present in both normal and tumour tissue followed by its separation on a Mono-Q column revealed two distinct peaks of activity. These peaks of activity were equivalent to those previously identified as a membrane-bound transglutaminase and the more characteristic cytosolic or tissue transglutaminase. The ratio of the activity of the cytosolic enzyme to that of the membrane-bound enzyme in normal liver was calculated as 5:1. In hepatocellular carcinomas, this ratio was reduced to 0.4:1. No significant change in the activity of the membrane-bound enzyme was detectable in tumour tissue. Comparison of the cytosolic enzyme found in hepatocellular carcinomas with that found in normal liver indicated no change in its molecular weight, Km,app for putrescine incorporation into N,N'-dimethylcasein and sensitivity to activation by Ca2+. These observations suggest that the reduction in transglutaminase activity observed in the hepatocellular carcinoma is due to a selective reduction in the expression of the cytosolic transglutaminase.     

Transglutaminase-catalysed cross-linking of proteins phosphorylated in the intact glucose-stimulated pancreatic beta-cell

Incubation of intact islets in the presence of [32P]Pi and stimulatory levels of glucose followed by separation of phosphorylated islet proteins by SDS-polyacrylamide gel electrophoresis revealed the presence of a high molecular weight phosphopolymer which did not transverse a 3% (w/v) acrylamide gel. The majority of this phosphopolymer (approx. 70%) was present in the 600 x g sedimented fraction of islet homogenates. Islet homogenates obtained from intact islets previously incubated with [32P]Pi and stimulatory levels of glucose when incubated under conditions that activated the islet transglutaminase resulted in an increase in the amount of phosphopolymer present in the 600 x g sedimented fraction. Inhibitors of transglutaminase activity which are known to inhibit glucose-stimulated insulin release led to a significant reduction in the fraction of phosphopolymer present in the glucose-stimulated intact islet. These findings suggest that protein cross-linking and phosphorylation reactions may be closely linked in the pancreatic beta-cell.