Large-scale experimental studies show unexpected amino acid effects on protein expression and solubility in vivo in E. coli.

The biochemical and physical factors controlling protein expression level and solubility in vivo remain incompletely characterized. To gain insight into the primary sequence features influencing these outcomes, we performed statistical analyses of results from the high-throughput protein-production pipeline of the Northeast Structural Genomics Consortium. Proteins expressed in E. coli and consistently purified were scored independently for expression and solubility levels. These parameters nonetheless show a very strong positive correlation. We used logistic regressions to determine whether they are systematically influenced by fractional amino acid composition or several bulk sequence parameters including hydrophobicity, sidechain entropy, electrostatic charge, and predicted backbone disorder. Decreasing hydrophobicity correlates with higher expression and solubility levels, but this correlation apparently derives solely from the beneficial effect of three charged amino acids, at least for bacterial proteins. In fact, the three most hydrophobic residues showed very different correlations with solubility level. Leu showed the strongest negative correlation among amino acids, while Ile showed a slightly positive correlation in most data segments. Several other amino acids also had unexpected effects. Notably, Arg correlated with decreased expression and, most surprisingly, solubility of bacterial proteins, an effect only partially attributable to rare codons. However, rare codons did significantly reduce expression despite use of a codon-enhanced strain. Additional analyses suggest that positively but not negatively charged amino acids may reduce translation efficiency in E. coli irrespective of codon usage. While some observed effects may reflect indirect evolutionary correlations, others may reflect basic physicochemical phenomena. We used these results to construct and validate predictors of expression and solubility levels and overall protein usability, and we propose new strategies to be explored for engineering improved protein expression and solubility.

In vitro evaluation of right ventricular outflow tract reconstruction with bicuspid valved polytetrafluoroethylene conduit.

Conduits available for right ventricular outflow tract (RVOT) reconstruction eventually become stenotic and/or insufficient due to calcification. In order to reduce the incidence of reoperations we have developed and used a bicuspid valved polytetrafluoroethylene (PTFE) conduit for the RVOT reconstruction. The purpose of this study is to investigate the hemodynamic performance of the new design using a pediatric in vitro right heart mock loop. PTFE conduit has been used for the complete biventricular repair of 20 patients (age 1.7±6 years) with cyanotic congenital defects. To account for the large variability of conduit sizes, 14, 16, 22, and 24-mm conduit sizes were evaluated using an in vitro flow loop comprised of a pulsatile pump with cardiac output (CO) of 1.2-3.2L/min, bicuspid valved RVOT conduit, pulmonary artery, venous compartments, and the flow visualization setup. We recorded the diastolic valve leakage and pre- and post-conduit pressures in static and pulsatile settings. In vitro valve function and overall hemodynamic performance was evaluated using high-speed cameras and ultrasonic flow probes. Three-dimensional flow fields for different in vivo conduit curvatures and inflow regimes were calculated by computational fluid dynamics (CFD) analysis to further aid the conduit design process. The average pressure drop over the valved conduits was 0.8±1.7mm Hg for the CO range tested. Typical values for regurgitant fraction, peak-to-peak pressure gradient, and effective office area were 23±2.1%, 13±2.4mm Hg, and 1.56±0.2 cm(2) , respectively. High-speed videos captured the intact valve motion with asymmetrical valve opening during the systole. CFD simulations demonstrated the flow skewness toward the major curvature of the conduit based on the pulmonic curvature. In vitro evaluation of the bicuspid valved PTFE conduit coincides well with acceptable early clinical performance (mild insufficiency), with relatively low pressure drop, and intact valve motion independent from the conduit curvature, orientation or valve location, but at the expense of increased diastolic flow regurgitation. These findings benchmark the baseline performance of the bicuspid valved conduit and will be used for future designs to improve valve competency.

Understanding the physical properties that control protein crystallization by analysis of large-scale experimental data.

Crystallization is the most serious bottleneck in high-throughput protein-structure determination by diffraction methods. We have used data mining of the large-scale experimental results of the Northeast Structural Genomics Consortium and experimental folding studies to characterize the biophysical properties that control protein crystallization. This analysis leads to the conclusion that crystallization propensity depends primarily on the prevalence of well-ordered surface epitopes capable of mediating interprotein interactions and is not strongly influenced by overall thermodynamic stability. We identify specific sequence features that correlate with crystallization propensity and that can be used to estimate the crystallization probability of a given construct. Analyses of entire predicted proteomes demonstrate substantial differences in the amino acid-sequence properties of human versus eubacterial proteins, which likely reflect differences in biophysical properties, including crystallization propensity. Our thermodynamic measurements do not generally support previous claims regarding correlations between sequence properties and protein stability.

Comparisons of NMR spectral quality and success in crystallization demonstrate that NMR and X-ray crystallography are complementary methods for small protein structure determination.

X-ray crystallography and NMR spectroscopy provide the only sources of experimental data from which protein structures can be analyzed at high or even atomic resolution. The degree to which these methods complement each other as sources of structural knowledge is a matter of debate; it is often proposed that small proteins yielding high quality, readily analyzed NMR spectra are a subset of those that readily yield strongly diffracting crystals. We have examined the correlation between NMR spectral quality and success in structure determination by X-ray crystallography for 159 prokaryotic and eukaryotic proteins, prescreened to avoid proteins providing polydisperse and/or aggregated samples. This study demonstrates that, across this protein sample set, the quality of a protein's [15N-1H]-heteronuclear correlation (HSQC) spectrum recorded under conditions generally suitable for 3D structure determination by NMR, a key predictor of the ability to determine a structure by NMR, is not correlated with successful crystallization and structure determination by X-ray crystallography. These results, together with similar results of an independent study presented in the accompanying paper (Yee, et al., J. Am. Chem. Soc., accompanying paper), demonstrate that X-ray crystallography and NMR often provide complementary sources of structural data and that both methods are required in order to optimize success for as many targets as possible in large-scale structural proteomics efforts.