Search

Intelligent systems in interventional healthcare depend on the reliable perception of the environment. In this context, photoacoustic tomography (PAT) has emerged as a non-invasive, functional imaging modality with great clinical potential. Current research focuses on converting the high-dimensional, not human-interpretable spectral data into the underlying functional information, specifically the blood oxygenation. One of the largely unexplored issues stalling clinical advances is the fact that the quantification problem is ambiguous, i.e. that radically different tissue parameter configurations could lead to almost identical photoacoustic spectra. In the present work, we tackle this problem with conditional Invertible Neural Networks (cINNs). Going beyond traditional point estimates, our network is used to compute an approximation of the conditional posterior density of tissue parameters given the photoacoustic spectrum. To this end, an automatic mode detection algorithm extracts the plausible solution from the sample-based posterior. According to a comprehensive validation study based on both synthetic and real images, our approach is well-suited for exploring ambiguity in quantitative PAT.

The tyrosyl radicals generated in reactions of ethyl hydrogen peroxide with both native and indomethacin-pretreated prostaglandin H synthase 1 (PGHS-1) were examined by low-temperature electron paramagnetic resonance (EPR) and electron nuclear double resonance (ENDOR) spectroscopies. In the reaction of peroxide with the native enzyme at 0 degrees C, the tyrosyl radical EPR signal underwent a continuous reduction in line width and lost intensity as the incubation time increased, changing from an initial, 35-G wide doublet to a wide singlet of slightly smaller line width and finally to a 25-G narrow singlet. The 25-G narrow singlet produced by self-inactivation was distinctly broader than the 22-G narrow singlet obtained by indomethacin treatment. Analysis of the narrow singlet EPR spectra of self-inactivated and indomethacin-pretreated enzymes suggests that they reflect conformationally distinct tyrosyl radicals. ENDOR spectroscopy allowed more detailed characterization by providing hyperfine couplings for ring and methylene protons. These results establish that the wide doublet and the 22-G narrow singlet EPR signals arise from tyrosyl radicals with different side-chain conformations. The wide-singlet ENDOR spectrum, however, is best accounted for as a mixture of native wide-doublet and self-inactivated 25-G narrow-singlet species, consistent with an earlier EPR study [DeGray et al. (1992) J. Biol. Chem. 267, 23583-23588]. We conclude that a tyrosyl residue other than the catalytically essential Y385 species is most likely responsible for the indomethacin-inhibited, narrow-singlet spectrum. Thus, this inhibitor may function by redirecting radical formation to a catalytically inactive side chain. Either radical migration or conformational relaxation at Y385 produces the 25-G narrow singlet during self-inactivation. Our ENDOR data also indicate that the catalytically active, wide-doublet species is not hydrogen bonded, which may enhance its reactivity toward the fatty-acid substrate bound nearby.

Electron spin resonance spectra of solutions used to make conductive bilayer lipid membranes reveal that free radical formation is a precursor to membrane conductivity. Tetracyanoquinodimethane (TCNQ) forms a stable radical that self-aggregates in nonaqueous media, yet will diffuse into the aqueous phase as a monomer, potentially allowing interfacial exchange. Tetracyanoethylene (TCNE) will undergo hydrolysis via a radical intermediate, and the decomposition product is ESR silent. The electrochemical response of an electrochemical cell partitioned by a membrane modified by these dopants is ionic for TCNE. Although the mechanism is less clear for TCNQ, the role of a stable radical is clearly implicated., (Copyright 1997 Academic Press.)

The optical and electron paramagnetic resonance (EPR) spectroscopic properties of a transferrin from the cockroach Blaberus discoidalis have been investigated to determine the relation of this protein to vertebrate transferrins. Difference spectrophotometry substantiates the involvement of tyrosyl residues in iron binding, and confirms the specific binding of two equivalents of iron per molecule. The far-UV CD spectrum also indicates a secondary structure with marked similarity to those of vertebrate transferrins. EPR studies show a dependence of iron binding on (bi)carbonate, consistent with the absolute requirement of transferrins for a synergistic anion in binding iron. Continuous wave (CW) and pulsed EPR studies of the cupric complex of the protein implicate a histidyl nitrogen ligand in metal coordination, as in human transferrin. Additional studies establish that the pH-dependent release of iron is similar to that of human serum transferrin. The present data confirm cockroach transferrin as an authentic member of the transferrin superfamily, thereby suggesting an ancestral relationship of insect to vertebrate transferrins.

The cDNA encoding the 182 amino acid long precursor stellacyanin from Cucumis sativus was isolated and characterized. The protein precursor consists of four sequence domains: I, a 23 amino acid hydrophobic N-terminal signal peptide with features characteristic of secretory proteins; II, a 109 amino acid copper-binding domain; III, a 26 amino acid hydroxyproline- and serine-rich peptide characteristic of motifs found in the extension family, extracellular structural glycoproteins found in plant cell walls; and IV, a 22 amino acid hydrophobic extension. Maturation of the protein involves posttranslational processing of domains I and IV. The copper-binding domain (domain II), which shares high sequence identity with other stellacyanins, has been expressed without its carbohydrate attachment sites, refolded from the Escherichia coli inclusion bodies, purified, and characterized by electronic absorption, EPR, ESEEM, and RR spectroscopy. Its spectroscopic properties are nearly identical to those of stellacyanin from the Japanese lacquer tree Rhus vernicifera, the most extensively studied and best characterized stellacyanin, indicating that this domain folds correctly, even in the absence of its carbohydrate moiety. The presence of a hydroxyproline- and serine-rich domain III suggests that stellacyanin may have a function other than that of a diffusible electron transfer protein, conceivably participating in redox reactions localized at the plant cell wall, which are known to occur in response to wounding or infection of the plant.

Preparation and characterization of two new site-directed mutant copper-zinc superoxide dismutase proteins from Saccharomyces cerevisiae, i.e., His46Cys (H46C) and His120Cys (H120C), in which individual histidyl ligands in the copper-binding site were replaced by cysteine, are reported here. These two mutant CuZnSOD proteins may be described as type 2 (or normal) rather than type 1 (or blue) copper-cysteinate proteins and are characterized by their yellow rather than blue color, resulting from intense copper-to-sulfur charge transfer bands around 400 nm, their type 2 EPR spectra, with large rather than small nuclear hyperfine interactions, and their characteristic type 2 d-d electronic absorption spectra. An interesting difference between these two copper site His-to-Cys mutations is that the imidazolate bridge between the two metal sites that is characteristic of the wild-type protein remains intact in the case of the H46C mutant but is not present in the case of the H120C mutant.

Lenticular opacities can be induced by numerous external agents that coincide with those that catalyze oxidative damage to lipids. One of the consequences of lipid peroxidation is that the affected membrane is rendered more permeable to protons. A proton leak in the tight epithelium of lens would uncouple the Na+/K(+)-ATPases that regulate the water and ionic content of the bounded tissue. Once regulatory control of the osmotic pressure is lost, the phase state of the cell's soluble proteins would change, leading to refractive changes or, in extreme cases, precipitation. The same does not occur in cornea because the stroma is an extracellular polymer blend rather than solution of soluble polymers. Polymeric phase transitions in the cornea require that divalent cations pass the epithelial membrane, which can occur only through the action of ionophores.

Phenylalanine hydroxylase from Chromobacterium violaceum (CVPAH) is known to bind an equivalent of divalent copper. The "metal-free" form of the protein is fully active, and Cu(II) is now shown to be an inhibitor of CVPAH rather than an activator of the enzyme [Carr, R. T., & Benkovic, S. J. (1994) Biochemistry 32, 14132-14138]. On the basis of amino acid sequence homology, the metal binding site may be related to those of rat liver PAH and other eukaryotic pterin-dependent hydroxylases, which require Fe(II) for activity. The conserved histidines at that site in CVPAH, histidines 138 and 143, were each mutated to serines. The mutant enzymes H138S and H143S were both catalytically inactive, but still able to bind Cu(II). Binding studies further demonstrated that both mutant enzymes still bind L-phenylalanine. Electron spin echo envelope modulation (ESEEM) studies on each of the mutants showed the presence of only a single copper-coordinating histidine, rather than the two histidine ligands suggested for the wild-type protein. This result supports a model in which Cu(II) is equatorially ligated to only two histidines in the Cu(II)-inhibited protein and allows us to unambiguously assign histidines 138 and 143 as these ligands. That the enzyme is inactive when these histidines are either bound with copper or when replaced with serines suggests that these histidines perform a catalytic function. Possible catalytic roles for these histidines in the hydroxylation mechanism of pterin-dependent monooxygenases are discussed along with potential future applications of the combination of ESEEM with site-directed mutagenesis.

Electron spin echo envelope modulation spectroscopy identified two ligand 14N interactions with the mixed-valence, Fe(II/III) diiron center of methane monooxygenase hydroxylase from Methylococcus capsulatus (Bath). Characteristic features of the spectra obtained at 9 and 10 GHz were analyzed and fit by simulation. One of the nitrogens possessed superhyperfine parameters (Aiso = 0.8 MHz, reff = 3.2 A, e2Qq = 1.8 MHz, eta = 0.35) consistent with a non-coordinating amino nitrogen of a histidine imidazole ligand to a Fe(III). The second, more strongly interacting nitrogen (Aiso = 5.0 MHz, reff = 2.2 A, e2Qq = 3.0 MHz, eta = 0.3) corresponds to the N delta directly bound to the effective Fe(II). These findings extend the previous electron nuclear double resonance results on the Methylosinus trichosporium hydroxylase (Hendrich, M.P., Fox, B.G., Andersson, K.K., Debrunner, P.G., and Lipscomb, J.D. (1992) J. Biol. Chem. 267, 261-269), which identified the N delta-Fe(II) interaction but failed to quantify its magnitude. Measurement of the linear electric field g shift of this mixed-valence species indicated that the site is charge-polarized on to one of the iron atoms, and its symmetry suggests that either charge is shifted away from the Fe-Fe axis (if gmax is defined by the Fe-Fe axis) or that gmid and gmax are perpendicular to the Fe-Fe axis (charge strongly localized at Fe(III) and axis taken as gmin).

Electron nuclear double resonance (ENDOR) and electron spin echo envelope modulation (ESEEM) spectroscopies were used to study whether protons in the immediate protein environment around CuA in cytochrome c oxidase are susceptible to solvent exchange. The enzyme was incubated in buffered D2O under resting or turnover conditions for 90 min and then frozen to quench the hydrogen/deuterium-exchange process. ENDOR spectra of the deuterated sample were essentially identical to those of control samples. The ESEEM spectra, however, provided a clear indication of the introduction of deuterium into the CuA environment following incubation in buffered D2O. The extent of deuterium incorporation was not affected by enzyme turnover. An analysis of the ESEEM data indicated that water is in reasonably close proximity to the CuA site, but not in the immediate coordination sphere of the metal(s). We estimate a minimum distance of 5.4 A between the CuA center and the protein/water interface. This relatively short surface separation distance is consistent with the role of CuA as the immediate oxidant of cytochrome c in the cytochrome oxidase (Hill, B. C. (1991) J. Biol. Chem. 266, 2219-2226).

The structural and spectrochemical effects of the replacement of Met44 in the hydrophobic surface patch of azurin from Pseudomonas aeruginosa by a lysine residue were studied as a function of the ionization state of the lysine. In the pH range 5-8, the optical absorption, resonance Raman, EPR and electron spin-echo envelope modulation spectroscopic properties of wild-type and Met44-->Lys (M44K) azurin are very similar, indicating that the Cu-site geometry has been maintained. At higher pH, the deprotonation of Lys44 in M44K azurin (pKa 9-10) is accompanied by changes in the optical-absorption maxima (614 nm and 450 nm instead of 625 nm and 470 nm) and in the EPR gII value (2.298 instead of 2.241), indicative of a change in the bonding interactions of Cu at high pH. The strong pH dependence of the electron self-exchange rate of M44K azurin supports the assignment of Lys44 as the ionizable group and demonstrates the importance of the hydrophobic patch for electron transfer. The pH dependence of the midpoint potentials of wild-type and M44K azurin can be accounted for by the ionizations of His35 and His83 and by the additional electrostatic effect of the mutation.

The nuclear modulation effect in pulsed EPR spectroscopy was used to study the type 2 copper binding site in the mercury derivative of laccase (MDL) in which the type 1 copper is substituted by Hg(II). By comparing the three-pulse electron spin-echo modulations and Fourier transform spectra of MDL and several model compounds, we conclude that the imidazole groups of two histidyl amino acid residues are equatorially coordinated to Cu(II) in the type 2 site. Computer simulations of these data suggest that the remote nonbonding nitrogens of the two imidazoles possess nuclear quadrupole parameters e2qQ = 1.47 MHz and eta = 0.83. A(iso) values of these two nitrogens are not identical, being 1.5 and 2.0 MHz. We have also used samples of the enzyme exchanged with D2O to examine the coordination of the water to the type 2 copper site. The deuterium modulation that is resolved by taking the ratio of the time domain ESEEM data from native and D2O-exchanged enzyme indicates that there is an equatorial water ligand, and further data show that this water is displaced by azide.