Trade-offs between enzyme fitness and solubility illuminated by deep mutational scanning.

Proteins are marginally stable, and an understanding of the sequence determinants for improved protein solubility is highly desired. For enzymes, it is well known that many mutations that increase protein solubility decrease catalytic activity. These competing effects frustrate efforts to design and engineer stable, active enzymes without laborious high-throughput activity screens. To address the trade-off between enzyme solubility and activity, we performed deep mutational scanning using two different screens/selections that purport to gauge protein solubility for two full-length enzymes. We assayed a TEM-1 beta-lactamase variant and levoglucosan kinase (LGK) using yeast surface display (YSD) screening and a twin-arginine translocation pathway selection. We then compared these scans with published experimental fitness landscapes. Results from the YSD screen could explain 37% of the variance in the fitness landscapes for one enzyme. Five percent to 10% of all single missense mutations improve solubility, matching theoretical predictions of global protein stability. For a given solubility-enhancing mutation, the probability that it would retain wild-type fitness was correlated with evolutionary conservation and distance to active site, and anticorrelated with contact number. Hybrid classification models were developed that could predict solubility-enhancing mutations that maintain wild-type fitness with an accuracy of 90%. The downside of using such classification models is the removal of rare mutations that improve both fitness and solubility. To reveal the biophysical basis of enhanced protein solubility and function, we determined the crystallographic structure of one such LGK mutant. Beyond fundamental insights into trade-offs between stability and activity, these results have potential biotechnological applications.

Conjugation of Amphiphilic Proteins to Hydrophobic Ligands in Organic Solvent.

Protein-ligand conjugations are usually carried out in aqueous media in order to mimic the environment within which the conjugates will be used. In this work, we focus on the conjugation of amphiphilic variants of elastin-like polypeptide (ELP), short elastin (sEL), to poorly water-soluble compounds like OPPVs ( p-phenylenevinylene oligomers), triarylamines, and polypyridine-metal complexes. These conjugations are problematic when carried out in aqueous phase because hydrophobic ligands tend to avoid exposure to water, which in turn causes the ligand to self-aggregate and/or interact noncovalently with hydrophobic regions of the amphiphile. Ultimately, this behavior leads to low conjugation efficiency and contamination with strong noncovalent "conjugates". After exploring the solubility of sEL in various organic solvents, we have established an efficient conjugation methodology for obtaining covalent conjugates virtually free of contaminating noncovalent complexes. When conjugating carboxylated ligands to the amphiphile amines, we demonstrate that even when only one amine (the N-terminus) is present, its derivatization is 98% efficient. When conjugating amine moieties to the amphiphile carboxyls (a problematic configuration), protein multimerization is avoided, 98-100% of the protein is conjugated, and the unreacted ligand is recovered in pure form. Our syntheses occur in "one pot", and our purification procedure is a simple workup utilizing a combination of water and organic solvent extractions. This conjugation methodology might provide a solution to problems arising from solubility mismatch of protein and ligand, and it is likely to be widely applied for modification of recombinant amphiphiles used for drug delivery (PEG-antibodies, polymer-enzymes, food proteins), cell adhesion (collagen, hydrophobins), synthesis of nanostructures (peptides), and engineering of biocompatible optoelectronics (biological polymers), to cite a few.

Joint neutron crystallographic and NMR solution studies of Tyr residue ionization and hydrogen bonding: Implications for enzyme-mediated proton transfer.

Human carbonic anhydrase II (HCA II) uses a Zn-bound OH(-)/H2O mechanism to catalyze the reversible hydration of CO2. This catalysis also involves a separate proton transfer step, mediated by an ordered solvent network coordinated by hydrophilic residues. One of these residues, Tyr7, was previously shown to be deprotonated in the neutron crystal structure at pH 10. This observation indicated that Tyr7 has a perturbed pKa compared with free tyrosine. To further probe the pKa of this residue, NMR spectroscopic measurements of [(13)C]Tyr-labeled holo HCA II (with active-site Zn present) were preformed to titrate all Tyr residues between pH 5.4-11.0. In addition, neutron studies of apo HCA II (with Zn removed from the active site) at pH 7.5 and holo HCA II at pH 6 were conducted. This detailed interrogation of tyrosines in HCA II by NMR and neutron crystallography revealed a significantly lowered pKa of Tyr7 and how pH and Tyr proximity to Zn affect hydrogen-bonding interactions.

Producing glucose 6-phosphate from cellulosic biomass: structural insights into levoglucosan bioconversion.

The most abundant carbohydrate product of cellulosic biomass pyrolysis is the anhydrosugar levoglucosan (1,6-anhydro-ß-d-glucopyranose), which can be converted to glucose 6-phosphate by levoglucosan kinase (LGK). In addition to the canonical kinase phosphotransfer reaction, the conversion requires cleavage of the 1,6-anhydro ring to allow ATP-dependent phosphorylation of the sugar O6 atom. Using x-ray crystallography, we show that LGK binds two magnesium ions in the active site that are additionally coordinated with the nucleotide and water molecules to result in ideal octahedral coordination. To further verify the metal binding sites, we co-crystallized LGK in the presence of manganese instead of magnesium and solved the structure de novo using the anomalous signal from four manganese atoms in the dimeric structure. The first metal is required for catalysis, whereas our work suggests that the second is either required or significantly promotes the catalytic rate. Although the enzyme binds its sugar substrate in a similar orientation to the structurally related 1,6-anhydro-N-acetylmuramic acid kinase (AnmK), it forms markedly fewer bonding interactions with the substrate. In this orientation, the sugar is in an optimal position to couple phosphorylation with ring cleavage. We also observed a second alternate binding orientation for levoglucosan, and in these structures, ADP was found to bind with lower affinity. These combined observations provide an explanation for the high Km of LGK for levoglucosan. Greater knowledge of the factors that contribute to the catalytic efficiency of LGK can be used to improve applications of this enzyme for levoglucosan-derived biofuel production.

New Twists and Turns for Actinide Chemistry: Organometallic Infinite Coordination Polymers of Thorium Diazide.

Two organometallic 1D infinite coordination polymers and two organometallic monometallic complexes of thorium diazide have been synthesized and characterized. Steric control of these self-assembled arrays, which are dense in thorium and nitrogen, has also been demonstrated: infinite chains can be circumvented by using steric bulk either at the metallocene or with a donor ligand in the wedge.

Gram-scale synthesis and efficient purification of 13C-labeled levoglucosan from 13C glucose.

(13)C-Labeled levoglucosan has been synthesized and purified in good yield, and on the gram scale in one step from commercially available (13)C glucose. This one-step protocol uses 2-chloro-1,3-dimethylimidazolinium chloride that serves to selectively activate the anomeric carbon toward substitution reactions. The labeled glucose is then smoothly converted to the anhydroglucose. Purification is efficiently achieved on large scale by chromatography on silica gel.

Backbone and side chain chemical shift assignment of diisopropyl fluorophosphatase (DFPase) from Loligo vulgaris, an organophosphorus-degrading enzyme.

NMR chemical shift assignments are reported for backbone (15N, 1H) and partial side chain (13Cα and ß, side chain 1H) atoms of diisopropyl fluorophosphatase (DFPase), a calcium-dependent phosphotriesterase capable of hydrolyzing phosphorus - fluorine bonds in a variety of toxic organophosphorus compounds. Analysis of residues lining the active site of DFPase highlight a number of residues whose chemical shifts can be used as a diagnostic of binding and detection of organophosphorus compounds.

Conformational control via sequence for a heteropeptoid in water: coupled NMR and Rosetta modelling.

We report a critical advance in the generation and characterization of peptoid hetero-oligomers. A library of sub-monomers with amine and carboxylate side-chains are combined in different sequences using microwave-assisted synthesis. Their sequence-structure propensity is confirmed by circular dichroism, and conformer subtypes are enumerated by NMR. Biasing the ψ-angle backbone to trans (180°) in Monte Carlo modelling favors i to i + 3 naphthyl-naphthyl stacking, and matches experimental ensemble distributions. Taken together, high-yield synthesis of heterooligomers and NMR with structure prediction enables rapid determination of sequences that induce secondary structural propensities for predictive design of hydrophilic peptidomimetic foldamers and their future libraries.

Rectifying system-specific errors in NMR relaxation measurements.

15N spin relaxation parameters provide a powerful tool for probing the internal dynamics and thermodynamics of proteins. The biological insight provided by these experiments often involves interpretation of small changes in relaxation parameters. This, in turn, requires careful data analysis, especially in the identification and treatment of systematic error. While progress continues on reduction of experiment-specific errors associated with pulse sequences, system-specific sources of error have received far less attention. The impact of these errors varies between facilities, spectrometers, and biological samples. We demonstrate that performing a series of control experiments along with relaxation measurements can help identify, quantify, and isolate sources of system-specific error, and, in some cases, correct for systematic changes. We further demonstrate that control experiments can be performed without significant loss of spectrometer time, and lead to more accurate relaxation parameter values.

Comparison of vibrational spectroscopy to biochemical and flow cytometry methods for analysis of the basic biochemical composition of mammalian cells.

We have conducted an extensive comparison of cellular biochemical composition obtained from infrared and Raman spectra of intact cells with measurements using standard extraction and chemical analysis (including NMR), and flow cytometric assay on fixed cells. Measurements were conducted on a rat fibroblast carcinogenesis model consisting of normal and tumorigenic cells assayed as exponentially growing and plateau-phase cultures. Estimates of protein, DNA, RNA, lipids, and glycogen amounts were obtained from a previous publication in which vibrational spectra were fit to a set of basis spectra representing protein, DNA, RNA, lipids, and glycogen. The Raman spectral estimates of absolute cellular composition were quite similar to the independent biochemical and flow cytometric assays. The infrared spectra gave similar results for protein, lipid, and glycogen but underestimated the DNA content while overestimating the RNA level. When ratios of biochemical concentrations in exponential and plateau-phase cultures were examined, the Raman spectroscopic results were the same, within errors, as the independent methods, in all cases. Several changes in relative biochemical composition due to tumorigenic and proliferative status previously reported using vibrational spectroscopy were confirmed by the independent methods. These results demonstrate that vibrational spectroscopy can provide reliable estimates of the biochemical composition of mammalian cells.

Comprehensive Sequence-Flux Mapping of a Levoglucosan Utilization Pathway in E. coli.

Synthetic metabolic pathways often suffer from low specific productivity, and new methods that quickly assess pathway functionality for many thousands of variants are urgently needed. Here we present an approach that enables the rapid and parallel determination of sequence effects on flux for complete gene-encoding sequences. We show that this method can be used to determine the effects of over 8000 single point mutants of a pyrolysis oil catabolic pathway implanted in Escherichia coli. Experimental sequence-function data sets predicted whether fitness-enhancing mutations to the enzyme levoglucosan kinase resulted from enhanced catalytic efficiency or enzyme stability. A structure of one design incorporating 38 mutations elucidated the structural basis of high fitness mutations. One design incorporating 15 beneficial mutations supported a 15-fold improvement in growth rate and greater than 24-fold improvement in enzyme activity relative to the starting pathway. This technique can be extended to improve a wide variety of designed pathways.

Unusual alkyl group activation and cationic complex formation from a novel lutetium dialkyl complex supported by a tridentate monoanionic ligand.

We report herein the synthesis and characterization of a lutetium dialkyl complex supported by a multidentate, anilido-pyridine-imine ligand and its subsequent transformation into an unprecedented cationic monoalkyl derivative.

Synthesis of 2'beta-Deoxy-[8-(13)C;amino,9-(15)N(2)]adenosine: Unusual Annulation Conditions To Assemble the Purine Core.

Synthesis of 2'beta-Deoxy-[8-(13)C;amino,9-(15)N(2)]adenosine has been accomplished using a five-step process which employs a novel annulation of [(13)C]formamide [N-(4-[(15)N]amino-6-chloro-5-pyrimidinyl)] 2. To effect this dehydration, a complex of triethyl phosphite and TiCl(2)(i-OPr)(2) was used. 6-Chloro-[8-(13)C; 9-(15)N]purine was then converted in two steps to 2'beta-deoxy-[8-(13)C;amino,9-(15)N(2)]adenosine.

Lewis base assisted B-H bond redistribution in borazine and polyborazylene.

Lewis bases react with borazine and polyborazylene, yielding borane adducts. In the case of NH3 (l), ammonia-borane (AB) is formed and quantified using NMR spectroscopy against an internal standard. Calculations indicate that the formation of B(NH2)3 may provide the driving force of this redistribution. Given the complexity and expense of currently known spent AB regeneration pathways, it is suggested that this redistribution chemistry be used to recover AB and improve regeneration methods.

Purification and characterization of rhodobactin: a mixed ligand siderophore from Rhodococcus rhodochrous strain OFS.

The siderophore produced by Rhodococcus rhodochrous strain OFS, rhodobactin, was isolated from iron-deficient cultures and purified by a combination of XAD-7 absorptive/partition resin column and semi-preparative HPLC. The siderophore structure was characterized using 1D and 2D (1)H, (13)C and (15)N NMR techniques (DQFCOSY, TOCSY, NOESY, HSQC and LR-HSQC) and was confirmed using ESI-MS and MS/MS experiments. The structural characterization revealed that the siderophore, rhodobactin, is a mixed ligand hexadentate siderophore with two catecholate and one hydroxamate moieties for iron chelation. We further investigated the effects of Fe concentrations on siderophore production and found that Fe limiting conditions (Fe concentrations from 0.1 microM to 2.0 microM) facilitated siderophore excretion. Our interests lie in the role that siderophores may have in binding metals at mixed contamination sites (containing metals/radionuclides and organics). Given the broad metabolic capacity of this microbe and its Fe scavenging ability, R. rhodochrous OFS may have a competitive advantage over other organisms employed in bioremediation.

Structural energetics and base-pair opening dynamics in sarcin-ricin domain RNA.

The sarcin-ricin domain is a universal element of the RNA from the large ribosomal subunit. The domain is part of the binding site for elongation factors and is specifically cleaved by the toxins alpha-sarcin and ricin. In this work, we have mapped the energetics and dynamics of individual structural motifs in a 29-mer RNA oligomer containing the sarcin-ricin domain. The stability of individual base pairs in the structure was characterized from measurements of the exchange rates of imino protons using nuclear magnetic resonance spectroscopy at 10 degrees C. The measurements also provided the rates of opening and closing for selected base pairs. The results reveal that the structural stabilization free energies in the sarcin-ricin domain are broadly distributed between 2.9 and 10.6 kcal/mol. One of the least stable sites in the structure is the noncanonical G-A base pair located next to the phosphodiester bond that is cleaved by alpha-sarcin. The low stability of this base pair supports the proposal that cleavage by alpha-sarcin occurs by a base flipping mechanism. The opening dynamics of other base pairs is affected by elements of the structure such as the bulged-G motif and its cross-strand stacking. Participation in these motifs increases the lifetimes of the bases in an open, solvent-accessible conformation.

Iron(III) coordination properties of a pyoverdin siderophore produced by Pseudomonas putida ATCC 33015.

The iron complexation of a fluorescent green pyoverdin siderophore produced by the environmental bacterium Pseudomonas putida was characterized by solution thermodynamic methods. Pyoverdin binds iron through three bidentate chelate groups, a catecholate, a hydroxamate, and an alpha-hydroxycarboxylic acid. The deprotonation constants of the free pyoverdin and Fe(III)-pyoverdin complex were determined through a series of potentiometric and spectrophotometric experiments. The ferric complex of pyoverdin forms at very low pH (pH < 2), but full iron coordination does not occur until neutral pH. The calculated pM value of 25.13 is slightly lower than that for pyoverdin PaA (pM = 27), which coordinates iron by a catecholate and two hydroxamate groups. The redox potential of Fe-pyoverdin was found to be very pH sensitive. At high pH (approximately pH 9-11) where pyoverdin coordinates Fe in a hexadentate mode the redox potential is -0.480 V (NHE); however, at neutral pH where full Fe coordination is incomplete, the redox potential is more positive (E(1/2) = -0.395 V). The positive shift in the redox potential and the partial dissociation of the Fe-pyoverdin complex with pH decrease provides a path toward in vivo iron release.

APART: automated preprocessing for NMR assignments with reduced tedium.

MOTIVATION: High-throughput NMR structure determination is a goal that will require progress on many fronts, one of which is rapid resonance assignment. An important rate-limiting step in the resonance assignment process is accurate identification of resonance peaks in the NMR spectra. Peak-picking schemes range from incomplete (which lose essential assignment connectivities) to noisy (which obscure true connectivities with many false ones). We introduce an automated preassignment process that removes false peaks from noisy peak lists by requiring consensus between multiple NMR experiments and exploiting a priori information about NMR spectra. This process is designed to accept multiple input formats and generate multiple output formats, in an effort to be compatible with a variety of user preferences. RESULTS: Automated preprocessing with APART rapidly identifies and removes false peaks from initial peak lists, reduces the burden of manual data entry, and documents and standardizes the peak filtering process. Successful preprocessing is demonstrated by the increased number of correct assignments obtained when data are submitted to an automated assignment program. AVAILABILITY: APART is available from http://sir.lanl.gov/NMR/APART.htm CONTACT: npawley@lanl.gov; rmichalczyk@lanl.gov SUPPLEMENTARY INFORMATION: Manual pages with installation instructions, procedures and screen shots can also be found at http://sir.lanl.gov/NMR/APART_Manual1.pdf.

Structural studies on the hairpins at the 3' untranslated region of an anthrax toxin gene.

Three proteins, namely, protective antigen (PA), edema factor (EF), and lethal factor (LF), encoded by the pX01 plasmid of Bacillus anthracis play a major role in the pathogenesis of target host cells. PA combines with EF and LF to form bipartite PA-EF and PA-LF toxins and facilitates intracellular delivery of EF and LF both of which cause cytotoxicity to the host. Since the level of PA is crucial to pathogenesis by anthrax toxins, it is important to understand how the host environment regulates the expression of the PA (or pagA) gene by utilizing the 5' and 3' untranslated regions (UTR). The 5' UTR sequence determines the initiation of transcription, whereas the 3' UTR sequence determines the efficient termination and stability of the transcript. Although, the role of the 5'UTR sequence of pagA has been investigated, little is known about the role of the 3' UTR. Since hairpin formation at the 3'UTR of a gene is an established mechanism for efficient termination and stability of the transcript, we carried out structural studies, including gel electrophoresis, circular dichroism, and two-dimensional nuclear magnetic resonance spectroscopy, to determine whether the 3' UTR sequences of pagA also form hairpin structures. Our results unequivocally demonstrate that both the coding and the noncoding 3' UTR sequences form stable hairpin structures. It is quite likely that the hairpins at the 3'UTR may contribute to efficient termination and stability of the pagA transcript.