Seeing gene expression in cells: the future of structural biology

Although much is known about the machinery that executes fundamental processes of gene expression in cells, much also remains to be learned about how that machinery works. A recent paper by O'Reilly et al. reports a major step forward in the direct visualization of central dogma processes at submolecular resolution inside bacterial cells frozen in a native state. The essential methodologies involved are cross-linking mass spectrometry (CLMS) and cryo-electron tomography (cryo-ET). In-cell CLMS provides in vivo protein interaction maps. Cryo-ET allows visualization of macromolecular complexes in their native environment. These methods have been integrated by O'Reilly et al. to describe a dynamic assembly in situ between a transcribing RNA polymerase (RNAP) and a translating ribosome - a complex known as the expressome - in the model bacterium Mycoplasma pneumoniae ¹. With the application of improved data processing and classification capabilities, this approach has allowed unprecedented insights into the architecture of this molecular assembly line, confirming the existence of a physical link between RNAP and the ribosome and identifying the transcription factor NusA as the linking molecule, as well as making it possible to see the structural effects of drugs that inhibit either transcription or translation. The work provides a glimpse into the future of integrative structural cell biology and can serve as a roadmap for the study of other molecular machineries in their native context.

Cholera toxin inhibits SNX27-retromer-mediated delivery of cargo proteins to the plasma membrane

Cholera toxin (CT) causes severe diarrhea by increasing intracellular cAMP leading to a PKA-dependent increase in Cl^- secretion through CFTR and decreased Na⁺ absorption through inhibition of Na⁺/H⁺ exchanger 3 (NHE3; also known as SLC9A3). The mechanism(s) by which CT inhibits NHE3 is partially understood, although no drug therapy has been successful at reversing this inhibition. We now describe that CT phosphorylates an amino acid in the PDZ domain of SNX27, which inhibits SNX27-mediated trafficking of NHE3 from the early endosomes to the plasma membrane (PM), and contributes to reduced basal NHE3 activity through a mechanism that involves reduced PM expression and reduced endocytic recycling. Importantly, mutagenesis studies (Ser to Asp) showed that the effect of this phosphorylation of SNX27 phenocopies the effects seen upon loss of SNX27 function, affecting PM trafficking of cargo proteins that bind SNX27-retromer. Additionally, CT destabilizes retromer function by decreasing the amount of core retromer proteins. These effects of CT can be partially rescued by enhancing retromer stability by using 'pharmacological chaperones'. Moreover, pharmacological chaperones can be used to increase basal and cholera toxin-inhibited NHE3 activity and fluid absorption by intestinal epithelial cells.This article has an associated First Person interview with the first author of the paper.

Latrepirdine (dimebon) enhances autophagy and reduces intracellular GFP-Aβ42 levels in yeast

Latrepirdine (Dimebon), an anti-histamine, has shown some benefits in trials of neurodegenerative diseases characterized by accumulation of aggregated or misfolded protein such as Alzheimer's disease (AD) and has been shown to promote the removal of α-synuclein protein aggregates in vivo. An important pathway for removal of aggregated or misfolded proteins is the autophagy-lysosomal pathway, which has been implicated in AD pathogenesis, and enhancing this pathway has been shown to have therapeutic potential in AD and other proteinopathies. Here we use a yeast model, Saccharomyces cerevisiae, to investigate whether latrepirdine can enhance autophagy and reduce levels of amyloid-β (Aβ)42 aggregates. Latrepirdine was shown to upregulate yeast vacuolar (lysosomal) activity and promote transport of the autophagic marker (Atg8) to the vacuole. Using an in vitro green fluorescent protein (GFP) tagged Aβ yeast expression system, we investigated whether latrepirdine-enhanced autophagy was associated with a reduction in levels of intracellular GFP-Aβ42. GFP-Aβ42 was localized into punctate patterns compared to the diffuse cytosolic pattern of GFP and the GFP-Aβ42 (19:34), which does not aggregate. In the autophagy deficient mutant (Atg8Δ), GFP-Aβ42 showed a more diffuse cytosolic localization, reflecting the inability of this mutant to sequester GFP-Aβ42. Similar to rapamycin, we observed that latrepirdine significantly reduced GFP-Aβ42 in wild-type compared to the Atg8Δ mutant. Further, latrepirdine treatment attenuated Aβ42-induced toxicity in wild-type cells but not in the Atg8Δ mutant. Together, our findings provide evidence for a novel mechanism of action for latrepirdine in inducing autophagy and reducing intracellular levels of GFP-Aβ42.

Anchoring a cationic ligand: the structure of the Fab fragment of the anti-morphine antibody 9B1 and its complex with morphine

The crystal structures of an anti-morphine antibody 9B1 (to 1.6A resolution) and its complex with morphine (to 2.0 A resolution) are reported. The morphine-binding site is described as a shallow depression on the protein surface, an unusual topology for a high-affinity ( Ka approximately 10(9) M(-1)) antibody against a small antigen. The polar part of the ligand is exposed to solvent, and the cationic nitrogen atom of the morphine molecule is anchored at the bottom of the binding site by a salt-bridge to a glutamate side-chain. Additional affinity is provided by a double cation-pi interaction with two tryptophan residues. Comparison of the morphine complex with the structure of the free Fab shows that a domain closure occurs upon binding of the ligand.

Inhibition of the aminopeptidase from Aeromonas proteolytica by L-leucinephosphonic acid. Spectroscopic and crystallographic characterization of the transition state of peptide hydrolysis

The nature of the interaction of the transition-state analogue inhibitor L-leucinephosphonic acid (LPA) with the leucine aminopeptidase from Aeromonas proteolytica (AAP) was investigated. LPA was shown to be a competitive inhibitor at pH 8.0 with a K(i) of 6.6 microM. Electronic absorption spectra, recorded at pH 7.5 of [CoCo(AAP)], [CoZn(AAP)], and [ZnCo(AAP)] upon addition of LPA suggest that LPA interacts with both metal ions in the dinuclear active site. EPR studies on the Co(II)-substituted forms of AAP revealed that the environments of the Co(II) ions in both [CoZn(AAP)] and [ZnCo(AAP)] become highly asymmetric and constrained upon the addition of LPA and clearly indicate that LPA interacts with both metal ions. The X-ray crystal structure of AAP complexed with LPA was determined at 2.1 A resolution. The X-ray crystallographic data indicate that LPA interacts with both metal centers in the dinuclear active site of AAP and a single oxygen atom bridge is absent. Thus, LPA binds to the dinuclear active site of AAP as an eta-1,2-mu-phosphonate with one ligand to the second metal ion provided by the N-terminal amine. A structural comparison of the binding of phosphonate-containing transition-state analogues to the mono- and bimetallic peptidases provides insight into the requirement for the second metal ion in bridged bimetallic peptidases. On the basis of the results obtained from the spectroscopic and X-ray crystallographic data presented herein along with previously reported mechanistic data for AAP, a new catalytic mechanism for the hydrolysis reaction catalyzed by AAP is proposed.

Preliminary crystallographic studies of a protease resistant botulinum neurotoxin associated protein Hn-33

Botulinum neurotoxin (BoNT) is one of the most potent toxins known. BoNT is also a food poison, which means that the toxin must survive the protease action and acidity of the gut. A group of neurotoxin-associated proteins which are only beginning to be identified and characterized are believed to be responsible for this protection. Hn-33 is a 33 kDa polypeptide which is a major component of the type A botulinum neurotoxin complex. Crystals of Hn-33 have been grown by vapour-diffusion techniques. They belong to a primitive orthorhombic space group and diffract to a resolution of 2. 6 A, with unit-cell parameters a = 130.3, b = 122.2, c = 37.2 A.

Substrate inhibition of D-amino acid transaminase and protection by salts and by reduced nicotinamide adenine dinucleotide: isolation and initial characterization of a pyridoxo intermediate related to inactivation

D-Amino acid transaminase, a pyridoxal phosphate (PLP) enzyme, is inactivated by its natural substrate, D-alanine, concomitant with its alpha-decarboxylation [Martinez del Pozo, A., Yoshimura, T., Bhatia, M. B., Futaki, S., Manning, J. M., Ringe, D., and Soda, K. (1992) Biochemistry 31, 6018-6023; Bhatia, M. B., Martinez del Pozo, A., Ringe, D., Yoshimura, T., Soda, K., and Manning, J. M. (1993) J. Biol. Chem. 268, 17687-17694]. beta-Decarboxylation of d-aspartate to d-alanine leads also to this inactivation [Jones, W. M., van Ophem, P. W., Pospischil, M. A., Ringe, D., Petsko, G., Soda, K., and Manning, J. M. (1996) Protein Sci. 5, 2545-2551]. Using a high-performance liquid chromatography-based method for the determination of pyridoxo cofactors, we detected a new intermediate closely related to the inactivation by d-alanine; its formation occurred at the same rate as the inactivation and upon reactivation it reverted to PLP. Conditions were found under which it was characterized by ultraviolet-visible spectral analysis and mass spectroscopy; it is a pyridoxamine phosphate-like compound with a C2 fragment derived from the substrate attached to the C'-4 of the pyridinium ring and it has a molecular mass of 306 consistent with this structure. In the presence of d-serine, slow accumulation of a quinonoid intermediate is also related to inactivation. The inactivation can be prevented by salts, which possibly stabilize the protonated aldimine coenzyme complex. The reduced cofactor, nicotinamide adenine dinucleotide, prevents D-aspartate-induced inactivation. Both of these events also are related to formation of the novel intermediate.

The ubiquitous cofactor NADH protects against substrate-induced inhibition of a pyridoxal enzyme

In the usual reaction catalyzed by D-amino acid transaminase, cleavage of the alpha-H bond is followed by the reversible transfer of the alpha-NH2 to a keto acid cosubstrate in a two-step reaction mediated by the two vitamin B6 forms pyridoxal 5'-phosphate (PLP) and pyridoxamine 5'-phosphate (PMP). We report here a reaction not on the main pathway, i.e., beta-decarboxylation of D-aspartate to D-alanine, which occurs at 0.01% the rate of the major transaminase reaction. In this reaction, beta-C-C bond cleavage of the single substrate D-aspartate occurs rather than the usual alpha-bond cleavage in the transaminase reaction. The D-alanine produced from D-aspartate slowly inhibits both transaminase and decarboxylase activities, but NADH or NADPH instantaneously prevent D-aspartate turnover and D-alanine formation, thereby protecting the enzyme against inhibition. NADH has no effect on the enzyme spectrum itself in the absence of substrates, but it acts on the enzyme.D-aspartate complex with an apparent dissociation constant of 16 microM. Equivalent concentrations of NAD or thiols have no such effect. The suppression of beta-decarboxylase activity by NADH occurs concomitant with a reduction in the 415-nm absorbance due to the PLP form of the enzyme and an increase at 330 nm due to the PMP form of the enzyme. alpha-Ketoglutarate reverses the spectral changes caused by NADH and regenerates the active PLP form of the enzyme from the PMP form with an equilibrium constant of 10 microM. In addition to its known role in shuttling electrons in oxidation-reduction reactions, the niacin derivative NADH may also function by preventing aberrant damaging reactions for some enzyme-substrate intermediates. The D-aspartate-induced effect of NADH may indicate a slow transition between protein conformational studies if the reaction catalyzed is also slow.

Interaction of pyridoxal 5'-phosphate with tryptophan-139 at the subunit interface of dimeric D-amino acid transaminase

The crystal structure of dimeric bacterial D-amino acid transaminase shows that the indole rings of the two Trp-139 side chains face each other in the subunit interface about 10 angstroms from the coenzyme, pyridoxal 5'-phosphate. To determine whether it has a role in the catalytic efficiency of the enzyme or interacts with the coenzyme, Trp-139 has been substituted by several different types of amino acids, and the properties of these recombinant mutant enzymes have been compared to the wild-type enzyme. In the native wild-type holoenzyme, the fluorescence of one of the three Trp residues per monomer is almost completely quenched, probably due to its interaction with PLP since in the native wild-type apoenzyme devoid of PLP, tryptophan fluorescence is not quenched. Upon reconstitution of this apoenzyme with PLP, the tryptophan fluorescence is quenched to about the same extent as it is in the native wild-type enzyme. The site of fluorescence quenching is Trp-139 since the W139F mutant in which Trp-139 is replaced by Phe has about the same amount of fluorescence as the wild-type enzyme. The circular dichroism spectra of the holo and the apo forms of both the wild-type and the W139F enzymes in the far-ultraviolet show about the same degree of ellipticity, consistent with the absence of extensive global changes in protein structure. Furthermore, comparison of the circular dichroism spectrum of the W139F enzyme at 280 nm with the corresponding spectral region of the wild-type enzyme suggests a restricted microenvironment for Trp-139 in the latter enzyme. The functional importance of Trp-139 is also demonstrated by the finding that its replacement by Phe, His, Pro, or Ala gives mutant enzymes that are optimally active at temperatures below that of the wild-type enzyme and undergo the E-PLP --> E-PMP transition as a function of D-Ala concentration with reduced efficiency. The results suggest that a fully functional dimeric interface with the two juxtaposed indole rings of Trp-139 is important for optimal catalytic function and maximum thermostability of the enzyme and, furthermore, that there might be energy transfer between Trp-139 and coenzyme PLP.

Catalytic ability and stability of two recombinant mutants of D-amino acid transaminase involved in coenzyme binding

Of the major amino acid side chains that anchor pyridoxal 5'-phosphate at the coenzyme binding site of bacterial D-amino acid transaminase, two have been substituted using site-directed mutagenesis. Thus, Ser-180 was changed to an Ala (S180A) with little effect on enzyme activity, but replacement of Tyr-31 by Gln (Y31Q) led to 99% loss of activity. Titration of SH groups of the native Y31Q enzyme with DTNB proceeded much faster and to a greater extent than the corresponding titration for the native wild-type and S180A mutant enzymes. The stability of each mutant to denaturing agents such as urea or guanidine was similar, i.e., in their PLP forms, S180A and Y31Q lost 50% of their activities at a 5-15% lower concentration of urea or guanidine than did the wild-type enzyme. Upon removal of denaturing agent, significant activity was restored in the absence of added pyridoxal 5'-phosphate, but addition of thiols was required. In spite of its low activity, Y31Q was able to form the PMP form of the enzyme just as readily as the wild-type and the S180A enzymes in the presence of normal D-amino acid substrates. However, beta-chloro-D-alanine was a much better substrate and inactivator of the Y31Q enzyme than it was for the wild-type or S180A enzymes, most likely because the Y31Q mutant formed the pyridoxamine 5-phosphate form more rapidly than the other two enzymes. The stereochemical fidelity of the Y31Q recombinant mutant enzyme was much less than that of the S180A and wild-type enzymes because racemase activity, i.e., conversion of L-alanine to D-alanine, was higher than for the wild-type or S180A mutant enzymes, perhaps because the coenzyme has more flexibility in this mutant enzyme.