The auxin-inducible degradation system in C. elegans allows for spatial and temporal control of protein degradation via heterologous expression of a single Arabidopsis thaliana F-box protein, transport inhibitor response 1 (AtTIR1). In this system, exogenous auxin (Indole-3-acetic acid; IAA) enhances the ability of AtTIR1 to function as a substrate recognition component that adapts engineered degron-tagged proteins to the endogenous C. elegans E3 ubiquitin ligases complex (SKR-1/2-CUL-1-F-box (SCF)), targeting them for degradation by the proteosome. While this system has been employed to dissect the developmental functions of many C. elegans proteins, we have found that several auxin-inducible degron (AID)-tagged proteins are constitutively degraded by AtTIR1 in the absence of auxin, leading to undesired loss-of-function phenotypes. In this manuscript, we adapt an orthogonal auxin-derivative/mutant AtTIR1 pair (C. elegans AID version 2 (C.e.AIDv2)) that transforms the specificity of allosteric regulation of TIR1 from IAA to one that is dependent on an auxin derivative harboring a bulky aryl group (5-Ph-IAA). We find that a mutant AtTIR1(F79G) allele that alters the ligand binding interface of TIR1 dramatically reduces ligand-independent degradation of multiple AID*-tagged proteins. In addition to solving the ectopic degradation problem for some AID targets, addition of 5-Ph-IAA to culture media of animals expressing AtTIR1(F79G) leads to more penetrant loss-of-function phenotypes for AID*-tagged proteins than those elicited by the AtTIR1-IAA pairing at similar auxin analog concentrations. The improved specificity and efficacy afforded by the mutant AtTIR1(F79G) allele expands the utility of the AID system and broadens the number of proteins that can be effectively targeted with it.

Sickle hemoglobin nucleation occurs in solution as a homogeneous process or on existing polymers in a heterogeneous process. We have developed an analytic formulation to describe the solution crowding and large nonideality that affects the heterogeneous nucleation of sickle hemoglobin by using convex particle theory. The formulation successfully fits the concentration and temperature dependence of the heterogeneous nucleation process over 14 orders of magnitude. Unlike previous approaches, however, the new formulation can also accurately describe the effects of adding nonpolymerizing agents to the solution. Without additional adjustable parameters, the model now describes the data of M. Ivanova, R. Jasuja, S. Kwong, R. W. Briehl, and F. A. Ferrone, (Biophys. J. 2000, 79:1016-1022), in which up to 50% of the sickle hemoglobin is substituted by cross-linked hemoglobin A, which does not polymerize, and which substitution causes the rates to decrease by 10(5). The success of this approach provides insight into the polymerization process: from the size-dependence of the contact energy deduced here, it also appears that various contacts of unknown origin are energetically significant in the heterogeneous nucleation process.

The homogeneous and heterogeneous nucleation kinetics of sickle hemoglobin (HbS) have been studied for various degrees of solution crowding by substitution of cross-linked hemoglobin A, amounting to 50% of the total hemoglobin. By cross-linking hemoglobin A, hybrid formation between hemoglobin A and hemoglobin S was prevented, thus simplifying the analysis of the results. Polymerization was induced by laser photolysis, and homogeneous nucleation kinetics were determined by observation of the stochastic behavior of the onset of light scattering. Heterogeneous nucleation was determined by observing the exponential growth of the progress curves, monitored by light scattering. At concentrations between 4 and 5 mM tetramer (i.e., approximately 30 g/dl), the substitution of 50% HbA for HbS slows the reaction by a factor of 10(3) to 10(4). Using scaled particle theory to account for the crowding of HbA, the observed decrease in the homogeneous nucleation rate was accurately predicted, with no variation of parameters required. Heterogeneous nucleation, on the other hand, is not well described in the present formulation, and the theory for this process appears to require modification of the way in which nonideality is introduced. Nonetheless, the accuracy of the homogeneous nucleation description suggests that such an approach may be useful for other assembly processes that occur in a crowded intracellular milieu.

No abstract

Nicotinamide nucleotide transhydrogenase from Escherichia coli is composed of two subunits, the a and the b subunits, each of which contains a hydrophilic domain, domain I and III, respectively, as well as several transmembrane helices, collectively denoted domain II. The interactions between domain I from Rhodospirillum rubrum (rrI) and the intact or the protease-treated enzyme from E. coli was investigated using the separately expressed and purified domain I from R. rubrum, and His-tagged intact and trypsin-treated E. coli transhydrogenase.Despite harsh treatments with, e.g. detergents and denaturing agents, the a and b subunits remained tightly associated. A monoclonal antibody directed towards the a subunit was strongly inhibitory, an effect that was relieved by added rrI. In addition, rrI also reactivated the trypsin-digested E. coli enzyme in which domain I had been partly removed. This suggests that the hydrophilic domains I and III are not in permanent contact but are mobile during catalysis while being anchored to domain II.Replacement of domain I of intact, as well as trypsin-digested, E. coli transhydrogenase with rrI resulted in a markedly different pH dependence of the cyclic reduction of 3-acetyl-pyridine-NAD 1 by NADH in the presence of NADP(H), suggesting that the protonation of one or more protonable groups in domain I is controlling this reaction. The reverse reaction and proton pumping showed a less pronounced change in pH dependence, demonstrating the regulatory role of domain II in these reactions.Keywords: transhydrogenase; NADP, NAD; membrane protein; proton pump. (Fig. 1). The active forms of the enzymes are dimers (220 kDa), either as a homodimer in the case of the mitochondrial H 1 -TH, as (a,b) 2 in the case of the Escherichia coli H 1 -TH or as (a1,a2,b) 2 in the case of Rhodospirillum rubrum H 1 -TH (1). As shown in Fig. 1, the a subunit (54 kDa) of E. coli H 1 -TH comprises domain I (40 kDa) and four transmembraneous helices, which form one component of dII. The b subunit (49 kDa) is composed of domain III (20 kDa) and nine helices spanning the membrane which constitute the second part of dII. The a1 subunit of R. rubrum corresponds only to domain I and is therefore purified as a soluble homodimer of 81 kDa [3]. The structural properties of domain I (dI) and domain III (dIII), from the three sources mentioned previously, have been characterized by expressing them separately. The determination of their structures has been realized by structural predictions [4±6], NMR [6±9] and X-ray crystallography [10,11]. From these studies, it was found that dI was made of two similar domains (dIa and dIb) each of which comprise a parallel b sheet surrounded by mainly b helices. The structure of domain III also exhibited a classical dinucleotide-binding fold [10,11].