Fit-for-purpose characterization of air-liquid-interface (ALI) in vitro exposure systems for e-vapor aerosol.

Air-liquid-interface (ALI) exposure systems deliver aerosol to the apical surface of cells which mimics the in vivo inhalation exposure conditions. It is necessary, however, to quantify the delivered amount of aerosol for ALI-based in vitro toxicity assessment. In this study, we evaluated two commercially available ALI exposure systems, a Vitrocell® Ames 48 (Ames 48) and a Vitrocell® 24/48 (VC 24/48), and the Vitrocell® VC1/7 smoking machine using a cig-a-like cartridge-based e-vapor device with a prototype formulation (containing 4% nicotine by weight). We characterized aerosol particle-size distribution, aerosol mass, and major chemical components (nicotine, propylene glycol, and glycerol) at the generation source and verified the repeatability of the aerosol generation. We determined aerosol delivery at the ALI by gravimetric analysis of mass collected on Cambridge filter pads and analytical quantitation of the buffer medium which showed that both aerosol mass and nicotine to an exposure insert linearly increased up to 400 puffs. The delivered aerosol mass covered a wide range of 0.8-3.4 mg per insert in the Ames 48 with variability (relative standard deviation, RSD) up to 12% and 1.1-6.4 mg per insert in the VC 24/48 with variability up to 15%. The delivered nicotine ranged approximately up to 200 µg per insert in both exposure systems. These results provided operation and aerosol delivery information of these ALI exposure systems for subsequent in vitro testing of e-vapor aerosols.

White matter hyperintensity and stroke lesion segmentation and differentiation using convolutional neural networks.

White matter hyperintensities (WMH) are a feature of sporadic small vessel disease also frequently observed in magnetic resonance images (MRI) of healthy elderly subjects. The accurate assessment of WMH burden is of crucial importance for epidemiological studies to determine association between WMHs, cognitive and clinical data; their causes, and the effects of new treatments in randomized trials. The manual delineation of WMHs is a very tedious, costly and time consuming process, that needs to be carried out by an expert annotator (e.g. a trained image analyst or radiologist). The problem of WMH delineation is further complicated by the fact that other pathological features (i.e. stroke lesions) often also appear as hyperintense regions. Recently, several automated methods aiming to tackle the challenges of WMH segmentation have been proposed. Most of these methods have been specifically developed to segment WMH in MRI but cannot differentiate between WMHs and strokes. Other methods, capable of distinguishing between different pathologies in brain MRI, are not designed with simultaneous WMH and stroke segmentation in mind. Therefore, a task specific, reliable, fully automated method that can segment and differentiate between these two pathological manifestations on MRI has not yet been fully identified. In this work we propose to use a convolutional neural network (CNN) that is able to segment hyperintensities and differentiate between WMHs and stroke lesions. Specifically, we aim to distinguish between WMH pathologies from those caused by stroke lesions due to either cortical, large or small subcortical infarcts. The proposed fully convolutional CNN architecture, called uResNet, that comprised an analysis path, that gradually learns low and high level features, followed by a synthesis path, that gradually combines and up-samples the low and high level features into a class likelihood semantic segmentation. Quantitatively, the proposed CNN architecture is shown to outperform other well established and state-of-the-art algorithms in terms of overlap with manual expert annotations. Clinically, the extracted WMH volumes were found to correlate better with the Fazekas visual rating score than competing methods or the expert-annotated volumes. Additionally, a comparison of the associations found between clinical risk-factors and the WMH volumes generated by the proposed method, was found to be in line with the associations found with the expert-annotated volumes.

Instantiated mixed effects modeling of Alzheimer's disease markers.

The assessment and prediction of a subject's current and future risk of developing neurodegenerative diseases like Alzheimer's disease are of great interest in both the design of clinical trials as well as in clinical decision making. Exploring the longitudinal trajectory of markers related to neurodegeneration is an important task when selecting subjects for treatment in trials and the clinic, in the evaluation of early disease indicators and the monitoring of disease progression. Given that there is substantial intersubject variability, models that attempt to describe marker trajectories for a whole population will likely lack specificity for the representation of individual patients. Therefore, we argue here that individualized models provide a more accurate alternative that can be used for tasks such as population stratification and a subject-specific prognosis. In the work presented here, mixed effects modeling is used to derive global and individual marker trajectories for a training population. Test subject (new patient) specific models are then instantiated using a stratified "marker signature" that defines a subpopulation of similar cases within the training database. From this subpopulation, personalized models of the expected trajectory of several markers are subsequently estimated for unseen patients. These patient specific models of markers are shown to provide better predictions of time-to-conversion to Alzheimer's disease than population based models.

Manifold population modeling as a neuro-imaging biomarker: application to ADNI and ADNI-GO.

We propose a framework for feature extraction from learned low-dimensional subspaces that represent inter-subject variability. The manifold subspace is built from data-driven regions of interest (ROI). The regions are learned via sparse regression using the mini-mental state examination (MMSE) score as an independent variable which correlates better with the actual disease stage than a discrete class label. The sparse regression is used to perform variable selection along with a re-sampling scheme to reduce sampling bias. We then use the learned manifold coordinates to perform visualization and classification of the subjects. Results of the proposed approach are shown using the ADNI and ADNI-GO datasets. Three types of classification techniques, including a new MRI Disease-State-Score (MRI-DSS) classifier, are tested in conjunction with two learning strategies. In the first case Alzheimer's Disease (AD) and progressive mild cognitive impairment (pMCI) subjects were grouped together, while cognitive normal (CN) and stable mild cognitive impaired (sMCI) subjects were also grouped together. In the second approach, the classifiers are learned using the original class labels (with no grouping). We show results that are comparable to other state-of-the-art methods. A classification rate of 71%, of arguably the most clinically relevant subjects, sMCI and pMCI, is shown. Additionally, we present classification accuracies between CN and early MCI (eMCI) subjects, from the ADNI-GO dataset, of 65%. To our knowledge this is the first time classification accuracies for eMCI patients have been reported.

A combined manifold learning analysis of shape and appearance to characterize neonatal brain development.

Large medical image datasets form a rich source of anatomical descriptions for research into pathology and clinical biomarkers. Many features may be extracted from data such as MR images to provide, through manifold learning methods, new representations of the population's anatomy. However, the ability of any individual feature to fully capture all aspects morphology is limited. We propose a framework for deriving a representation from multiple features or measures which can be chosen to suit the application and are processed using separate manifold-learning steps. The results are then combined to give a single set of embedding coordinates for the data. We illustrate the framework in a population study of neonatal brain MR images and show how consistent representations, correlating well with clinical data, are given by measures of shape and of appearance. These particular measures were chosen as the developing neonatal brain undergoes rapid changes in shape and MR appearance and were derived from extracted cortical surfaces, nonrigid deformations, and image similarities. Combined single embeddings show improved correlations demonstrating their benefit for further studies such as identifying patterns in the trajectories of brain development. The results also suggest a lasting effect of age at birth on brain morphology, coinciding with previous clinical studies.

Combining morphological information in a manifold learning framework: application to neonatal MRI.

MR image data can provide many features or measures although any single measure is unlikely to comprehensively characterize the underlying morphology. We present a framework in which multiple measures are used in manifold learning steps to generate coordinate embeddings which are then combined to give an improved single representation of the population. An application to neonatal brain MRI data shows that the use of shape and appearance measures in particular leads to biologically plausible and consistent representations correlating well with clinical data. Orthogonality among the correlations suggests the embedding components relate to comparatively independent morphological features. The rapid changes that occur in brain shape and in MR image appearance during neonatal brain development justify the use of shape measures (obtained from a deformation metric) and appearance measures (obtained from image similarity). The benefit of combining separate embeddings is demonstrated by improved correlations with clinical data and we illustrate the potential of the proposed framework in characterizing trajectories of brain development.

Role of macromolecular hydration in the binding of the Escherichia coli cyclic AMP receptor to DNA.

The osmotic stress technique was used to measure the changes in macromolecular hydration that accompany binding of the Escherichia coli CAP protein to its transcription-regulatory site (C1) in the lactose promoter and that accompany the transfer of CAP from site C1 to nonspecific genomic DNA. Formation of the C1 complex is accompanied by the net release of 79 +/- 11 water molecules. If all water molecules were released from macromolecular surfaces, this result would be consistent with a net reduction of solvent-accessible surface area of 711 +/- 189 A2. This area is only slightly smaller than the solvent-inaccessible macromolecular interface in crystalline CAP-DNA complexes. The transfer of CAP from site C1 to nonspecific sites is accompanied by the net uptake of 56 +/- 10 water molecules. Taken with the water stoichiometry of sequence-specific binding, this value implies that formation of a nonspecific complex is accompanied by the net release of 2-44 water molecules. The enhanced stabilities of CAP-DNA complexes with increased osmolality (decreased water activity) may contribute to the ability of E.coli cells to tolerate dehydration and/or high external salt concentrations.

Slow-binding inhibition of 6-phosphogluconate dehydrogenase by zinc ion.

6-Phosphogluconate dehydrogenase (EC 1.1.1.44) has been purified from Cryptococcus neoformans, an encapsulated yeast that is an opportunistic pathogen of AIDS patients. The dimeric enzyme had a subunit molecular weight of 50,000, a specific activity of 50 units mg-1, and Km values of 13 microM for 6-phosphogluconate and 0.89 microM for NADP. This enzyme, like many fungal dehydrogenases, was inhibited by Zn2+, with the inhibition pattern being competitive versus the nonnucleotide substrate. In the presence of micromolar Zn2+, the reaction was biphasic, with the reduction of NADP proceeding initially at the control rate, but, over the time course of 20-300 s, this initial nonlinear phase reached a final, linear steady state with a slower velocity. This pattern is indicative of a slow binding inhibition process, for which we have calculated the following kinetic constants: k6, the limiting rate constant for the transition from initial to final steady state was 0.0024 s-1, corresponding to a half-time of 300 s; Ki*, the overall equilibrium constant for the dissociation of E*Zn2+ to E + Zn2+ was 0.021 microM.

A kinetic comparison of the homologous proteases astacin and meprin A.

Astacin, a 23-kDa monomeric metalloprotease from the crayfish digestive tract, and meprin A, a 360-kDa tetrameric metalloprotease from the mouse kidney, are 30% identical in the amino acid sequence of their protease domains (Dumermuth et al., 1991, J. Biol. Chem. 266, 21381). The two were compared kinetically using a variety of substrates and inhibitors. Both enzymes degraded azocasein; meprin A had a twofold higher molar specific activity. Succinyl-Ala-Ala-Ala-p-nitroanilide was cleaved by both enzymes at the Ala-p-nitroanilide bond, indicating that in contrast to many metallopeptidases, peptidases of the astacin family are capable of arylamidolysis. Several peptides from a series of chromogenic and fluorogenic bradykinin analogs, originally designed for mapping the active site of meprin A, were found to be excellent substrates for astacin. Both enzymes cleaved most substrates at the site corresponding to the Phe5-Ser6 bond in native bradykinin (Arg-Pro-Pro-Gly-Phe-Ser-Pro-Phe-Arg). The peptide 2ABz-Arg-Pro-Ile-Phe decreases Ser-Pro-Phe(4-nitro)-Arg was found to have the highest kcat/Km ratio for both peptidases (4.5 x 10(5) M-1 s-1 for meprin A and 2.7 x 10(6) M-1 s-1 for astacin). The two enzymes were found to have several other substrates in common, indicating that some of these peptides could be substrates for the other putative proteases in the astacin family. Acetyl-Arg-Pro-Gly-Tyr-NHOH, an inhibitor which likely binds to the S subsites, was found to act on both enzymes via a predominantly non-competitive mechanism, whereas NHOH-succinyl-Pro-Phe-Arg, which likely binds to the S' subsites, was competitive. Significant specificity differences between astacin and meprin A were seen in substrates and inhibitors with bulky groups in the P1' position. Substrates with Arg, Lys, or Phe in the P1' position were cleaved 10(3)-10(4) times faster by meprin A than by astacin. Actinonin, a naturally occurring peptide hydroxamate with a pentyl group in P1', was a very potent competitive inhibitor of meprin A (Ki = 1.35 x 10(-7) M), but had a 1000-fold weaker affinity for inhibition of astacin. These kinetic differences indicate that the S1' subsite is smaller in astacin than in meprin A and correlate with structural differences seen in the three-dimensional models of the two enzymes.

The astacin family of metalloendopeptidases.

Molecular cloning of a human intestinal brush border metalloendopeptidase (N-benzoyl-L-tyrosyl-p-aminobenzoic acid hydrolase, PPH) and a mouse kidney brush border metalloendopeptidase (meprin A) has revealed 82% identity in the NH2-terminal amino acid sequences (198 residues) of the mature enzymes. Furthermore, searching of protein sequence data bases with the inferred peptide sequences as probes revealed strong similarities to astacin, a crayfish digestive protease, and an NH2-terminal domain of a human bone morphogenetic protein (BMP-1). Meprin A and PPH both have approximately 30% identity with astacin and BMP-1. Multiple alignment analysis indicated that 37 residues, including 3 cysteine residues, are strictly conserved for the four proteins in a sequence frame equivalent to the complete 200-amino acid astacin sequence. The four proteins contain a zinc-binding motif (HEXXH), found at the active site of most metalloendopeptidases, within an extended sequence of HEXXHXXGFXHE which is unique to this subgroup of metalloendopeptidases. In addition, the four proteins have 54% identity in a 24-amino acid sequence that includes the putative active site. A fifth protein, Xenopus laevis developmentally regulated protein UVS.2, also shares sequence identity with the metalloendopeptidases. These data provide strong evidence for an evolutionary relationship of these proteins. It is suggested that this new family of metalloendopeptidases be called the "astacin family."

Meprin-A and -B. Cell surface endopeptidases of the mouse kidney.

The proteinase meprin-A is a disulfide-linked tetramer of 90-kDa glycoprotein subunits. It is expressed at high levels in kidney brush border membranes of random bred and certain inbred strains of mice. Some mouse strains (e.g. C3H/He) do not express meprin-A subunits, but do produce a similar but less well characterized metalloendopeptidase, meprin-B. In the present study, meprin-B was purified from C3H/He mouse kidneys to electrophoretic homogeneity, and the relationship between it and meprin-A was investigated. The papain-solubilized form of meprin-B was similar to meprin-A in amino acid composition, molecular mass, secondary, and quaternary structure. However, immunoblots indicated that the enzymes have some common and some distinct epitopes. Lectin blots indicated both enzymes have high mannose and/or complex biantennary oligosaccharides, but there are differences in the complex-type glycosylation. Peptide maps and sequencing of cyanogen-bromide fragments of the enzymes revealed some different amino acid sequences. Thermal inactivation studies indicated that meprin-B was much less stable than meprin-A; the half-life for inactivation at 58 degrees C for meprin-A was 50 min, whereas for meprin-B it was less than 3 min. Both enzymes hydrolyzed azocasein and insulin B chain, but limited proteolysis of the enzymes with trypsin activated meprin-B 5-20-fold, whereas meprin-A was activated 2-fold at most. Analysis of hydrolysis products of the oxidized insulin B chain revealed some common and some distinct sites of cleavage. Bradykinin was a good substrate for meprin-A, while it was not hydrolyzed by meprin-B. A synthetic peptide, YLVC(SO3-)GERG, derived from insulin B chain was hydrolyzed faster by meprin-B than meprin-A, and neither enzyme was activated by trypsin treatment against this substrate. Taken together, the data indicate that the two metalloendopeptidases have many similarities but are distinct enzymes.

Mapping the active site of meprin-A with peptide substrates and inhibitors.

The extended substrate-binding site of meprin-A, a tetrameric metalloendopeptidase from brush border membranes of mouse kidney proximal tubules, was mapped with a series of peptide substrates. Previous studies led to the development of the chromogenic substrate Phe5(4-nitro)bradykinin for meprin-A. With this substrate, several biologically active peptides were screened as alternate substrate inhibitors, and, of these, bradykinin (RPPGFSPFR) was found to be the best substrate with a single cleavage site (Phe5-Ser6). Three types of bradykinin analogues were used for a systematic investigation of substrate specificity: (1) nonchromogenic bradykinin analogues with substitutions in the P3 to P3' subsites were used as alternative substrate inhibitors of nitrobradykinin hydrolysis, (2) analogues of nitrobradykinin with variations in the P1' position were tested as substrates, and (3) intramolecularly quenched fluorogenic bradykinin analogues with substitutions in the P1 to P3 sites were tested as substrates. A wide variety of substitutions in P1' had little effect on KM (174-339 microM) but markedly affected kcat (51.5 s-1 = A greater than S greater than R greater than F greater than K greater than T greater than E = 0). Substitutions in P1 had a greater effect on KM (366 microM-2.46 mM) and also strongly affected kcat (98.5 s-1 = A greater than F much greater than L greater than E greater than K = 2.4 s-1). The variety of allowed cleavages indicates that meprin-A does not have strict requirements for residues adjacent to the cleavage site. Substitutions farther from the scissle bond also affected binding and hydrolysis, demonstrating that multiple subsite interactions are involved in meprin-A action.(ABSTRACT TRUNCATED AT 250 WORDS)

Determination of enzyme specificity in a complex mixture of peptide substrates by N-terminal sequence analysis.

A method has been developed to determine preferred residue substitutions in the P' position of peptide substrates for proteolytic enzymes. The method has been validated with four different enzymes; the angiotensin I-converting enzyme, atrial dipeptidyl carboxyhydrolase, bacterial dipeptidyl carboxyhydrolase, and meprin A. A mixture of N-acylated potential peptide-substrates for each of the enzymes was prepared in a single synthesis procedure on the same solid-phase synthesis resin. The peptides were identical in all residue positions except the P' position to be studied, into which numerous amino acid residues were incorporated on a theoretical equimolar basis. After cleavage and extraction of the peptides from the resin, no attempt was made to purify them individually; the exact concentration of each peptide in the mixture was determined by quantitative amino acid analysis. Incubation of an enzyme with its peptide-substrate mixture at [S] much less than Km yielded peptide hydrolytic products with newly exposed N-termini. The identity and amount of each hydrolysis product was determined by automated N-terminal sequence analysis. One cycle of sequencing revealed preferred amino acid substitutions in the P'1 position, two cycles the P'2 position, and so forth. Comparison of the rates of production of the various products indicates the preferred substitution in that particular P' position. New information on the substrate specificities of each of the enzymes tested was obtained and it is clear that this approach can be applied to any protease with a defined (or suspected) point of cleavage in a peptide substrate.

Soluble metalloendopeptidase (MMP-7ase) activity in mouse kidney cytosol.

Previously, MMP-7ases were isolated from rat skeletal muscle by gel filtration and anion exchange chromatography. The enzyme that hydrolyzed succinyl-Ala-Ala-Pro-Phe-AMC (AMC: 7-amino-4-methyl-coumarin) was inhibited by EDTA. In this study we attempted to isolate MMP-7ase from mouse kidney. The isolation procedure was the same as that previously used for skeletal muscle. Kidneys of ICR mice were homogenized and, after centrifugation, the supernatant fraction was subjected to gel filtration chromatography. The fraction with the highest activity (Mr 67-72 kDa) was subjected to anion exchange chromatography, which showed three peaks of activity. The second peak hydrolyzed succ-Ala-Ala-Pro-Phe-AMC, but had low activity against Arg- or Ala-AMC. This peak was a single protein (Mr 68-72 kDa) and its activity could be inhibited with EDTA. Several tri- and tetrapeptide derivatives were tested as substrates for this enzyme and the best was found to be succ-Ala-Ala-Pro-Phe-AMC. We can conclude that mouse kidney cytosol contains a metalloendopeptidase similar to muscle MMP-7ase.

Phe5(4-nitro)-bradykinin: a chromogenic substrate for assay and kinetics of the metalloendopeptidase meprin.

Phe5(4-nitro)-bradykinin has been identified as a good synthetic substrate to study the kinetics and mechanism of action of the metalloendopeptidase meprin. No convenient substrate for kinetic analysis of the enzyme had been previously described. HPLC analyses indicated that meprin cleaved bradykinin and nitrobradykinin between Phe5 (or Phe5(NO2)) and Ser6. Reaction rates for bradykinin were determined by quantitative HPLC analyses, whereas rates for nitrobradykinin were measured by continuous monitoring of the spectral change that occurs at 310 nm when the Phe(NO2)-Ser bond is hydrolyzed. For nitrobradykinin and unmodified bradykinin, respectively, Km values were 281 and 425 microM, kcat values were 28 and 22 s-1, and kcat/Km values were 9.7 x 10(4) and 5.1 x 10(4)M-1. The two products of bradykinin hydrolysis were not substrates for the enzyme, but they were inhibitors. The initial rates of hydrolysis of nitrobradykinin increased linearly with enzyme concentration (0.09-2.2 micrograms/ml), and increased linearly with temperature in the range from 15 to 55 degrees C. Hydrolysis of the substrate was optimal at alkaline pH values. The cysteine endopeptidases papain and cathepsin L and the metalloproteases thermolysin, angiotensin-converting enzyme, and neutral endopeptidase (EC 3.4.24.11) also cleaved nitrobradykinin, but at different peptide bonds than meprin. The single cleavage of nitrobradykinin at the Phe(NO2)-Ser bond and the concomitant spectral shift that occurs at alkaline pH makes this a particularly suitable substrate for meprin.

Thiol containing compounds and amino acid hydroxamates as reversible synthetic inhibitors of Astacus protease.

Reversible synthetic inhibitors are characterized for Astacus protease, a 22,614-Da zinc containing neutral endopeptidase from the digestive tract of crayfish. Effective inhibition was demonstrated for several simple thiol containing compounds and a series of amino acid hydroxamates. Both classes of inhibitors had ID50 values ranging from 10(-2) to 10(-4) M for inhibition of hydrolysis of succinyl-Ala-Ala-Ala-p-nitroanilide. Tyrosine hydroxamate was found to be the most effective inhibitor with an ID50 of 175 microM and the mode of inhibition by this compound was determined to be of the simple noncompetitive type. In contrast to the other inhibitors tested, cysteine was seen to partially inactivate the enzyme in a time-dependent manner. The kinetics of this process was studied in detail using progress curve analysis. It was determined that cysteine was acting as a weak chelator and slowly establishing an equilibrium between metallo- and apoenzyme. In the presence of the strong zinc scavenger EDTA, cysteine can, in effect, function as a catalyst in transferring the metal from the protein to the secondary chelator at a rate 10,000 times faster than the rate of unassisted zinc dissociation. The series of amino acid hydroxamates served as probes into the microenvironment of the active site. Possible binding modes of the inhibitors are discussed on the basis of the relationship between the chemical nature of the inhibitor side chains and the strength of inhibition.

Kinetic evidence for cooperative binding of two ortho-phenanthroline molecules to astacus protease during metal removal.

Kinetic evidence is presented that introduces a new possibility for a mechanism of metal removal from a protein by a chelator. Astacus protease is a 22,614 dalton zinc-metalloendopeptidase from the digestive tract of the freshwater crayfish. Recent studies have shown that it contains a single zinc atom and that removal of this metal yields inactive apo-enzyme, which can be reactivated upon readdition of zinc, cobalt, or copper. The enzyme is inactivated by metal chelators in a time and concentration dependent manner. The inactivation of Zn-Astacus protease by 1,10-phenanthroline (OP) can be monitored continuously in the presence of substrate. The concentration of substrate was found to have no effect on the inactivation rate, indicating that the chelator binding during inactivation is of the noncompetitive type. First-order rate constants for the inactivation process are seen to depend on the concentration of chelator in a sigmoidal manner. Based on mathematics analogous to that for cooperativity in enzyme-substrate kinetics, the deduction is made that there are two OP binding sites on the protein and that the rate of inactivation is related to the saturation of both sites with ligand. If one uses this model, the limiting rate constant of inactivation upon saturation of both sites with ligand is 6.76 x 10(-3) sec-1, and the half maximal rate occurs at an OP concentration of 6.52 mM. A mechanism is proposed wherein both protein bound chelators can cooperate during metal removal either by direct chelation of the metal or by allosteric means. The proposed model and the noncompetitive binding of chelator and substrate are discussed in relation to a recently proposed metal binding site.